Hazards Posters

The possibility of use the productions of Earth Resource Satellite (ERS-1/2) and Advanced Environment Satellite ENVISAT SAR (Synthetic Aperture Radar) C-band have given the potential to detect and estimate the time series of dynamic ground deformation within high spatial and temporal resolution. Additionally assess the details scale ground degradation effects on environment. The Larissa National Airport in Central Greece is suffering from continued ground deformation as evidenced by the presence of ground fissures and sinkholes as well as observed land subsidence. Synthetic Aperture Radar interferometric techniques (InSAR) were used to detect short- and long-term ground deformation dynamics in the airport using the GAMMA Software (S/W). The results indicate complex subsidence and uplift processes at ranges between – 9 and + 18 mm a^-1 to co-occur in different parts of the study region. The observed ground deformation processes are likely to result from volume changes in expansive soils and human induced groundwater lowering with possible micro-tectonic movements.

Taiwan is hit by approximately three to four typhoons every year. In such a concentrated visit, surface with vegetation could easly change their land cover type to bare land. Remote sensing data can quiclky detect such land transformation due to Normalized Difference Vegetation Index (NDVI). Net Primary Production (NPP) plays an important role in reducing atmospheric carbon dioxide concentration in the carbon cycle. NPP can be estimate by models base on light-use effeciency and NDVI, which represents the absorbed photosynthetically active radiation, using remote sensing data. This research use SPOT and MODIS data to compare the lose of NPP cause by typhoon landslides and forest fire. Results show that the study area NDVI spent four months to return to the original level in wild fire incidents. While in typhoon caused landslide, it may take near one year to recover. The resilience of the surface NDVI difference would lead to the change on NPP estimation.

A new horizontal and vertical crustal velocity fields of Alpine Mediterranean area was determined by continuous long time series (6.5 years) of 113 Global Navigation Satellite System (GNSS) permanent stations and long time Permanent Scatters Syntetic Aperture Radar (PS SAR) series from ERS and ENVISAT.

The GNSS processing was performed using state-of-the-art absolute antenna phase center correction model and recomputed precise IGS orbits available since April 2014. Moreover, a new more accurate tropospheric mapping function for geodetic applications was adopted.

The vertical velocity field were compared and integrated with long time series PS SAR data. The dataset is composed by ERS data that cover the a period from 1995 to 2000, and ENVISAT from 2002 to 2010.

Results show differences in directions and amplitudes of movements respect previous GPS static campaigns. Alps chain, Corsica, Sardinia and the Tyrrhenian confirm their stability to Eurasian plate. The whole Apennines axis discriminates two different velocity patterns, the Adriatic and the Tyrrhenian area. The area around Messina Strait, which separates peninsular Italy and Sicily, represents a poorly understood region. Results identify an important boundary zone between two different domains, Calabria and Sicily, which are characterized by different crustal motions. The northeastern part of Sicily and Calabria move like Adriatic area, whilst the rest of Sicily, Malta and Lampedusa are dominated by African motion.

The launching of satellites platform with synthetic aperture radar (SAR) sensors onboard, purposely built for differential interferometry (D-InSAR) applications opened new opportunities for the mapping and monitoring of slow ground deformations and even high speed of landslides, for the updating of regional landslide inventories as well as retrieving time series of landslide movements. The present work has been done by using DInSAR method, with objection of performance assessment of two radar sensors in order to detection, monitoring and landslide inventory mapping in Dena Mountain of Zagros located in south of Isfahan province.

Landslides play an important role in the evolution of landforms and represent a serious hazard in many mountainous regions of Iran. Remote sensing techniques represent a valuable tool for landslide identification and mapping, which are necessary landslide risk assessment and hazard zonation. RADAR remote sensing imagery is a powerful tool to detection, monitoring, measuring and preparing landslide inventory map because of its synoptic view, which can be repeated at different time intervals. The launching of satellites platform with synthetic aperture radar (SAR) sensors onboard, purposely built for differential interferometry (D-InSAR) applications, opened new opportunities for the mapping and monitoring of slow ground deformations and even High speed of landslides, for the updating of regional landslide inventories and for retrieving time series of landslide movements. The present work is deal with landslide detection, monitoring and risk assessment based on RADAR remote sensing technique over the Dena area of Zagros range. The study area lied    between 51° 12´-51° 54´E and 30° 38´-31° 12´N with total area of 1700.97 km². The work has been done using a pragmatic method based on set imagery and field data, which is focused on describing the factors and conditions for landslide hazards incidence in terms of zonation and assessment. The study’s main objective is to identify and prepare an inventory of landslide type base on D-InSAR technique as well as field investigation. The set of aerial photographs, optical satellite images (ASTER2001, ETM+2002) alone with RADAR data form two C band and L band (ASAR & PALSAR) in period of year 1996 to 2011 and almost high accurate digital elevation model (DEM) have been used for landslide inventory maps  and landslide conditioning factors extraction (15 layers in 87 subclasses) .

Based in two mention techniques, landslide inventory map was generated using D-InSAR technique where the weighted of evidence bivariate statistical approach was applied for landslide susceptibility zonation and landslide hazard map to extract the risk over the area. The pixel values of outcome maps (Landslide susceptibility zonation) was analysis for classified into low, moderate, high and very high susceptibility groups based on natural break classification scheme to determine the class intervals in the landslide susceptibility map in observed and predictive models.

In other to landslide detection and measuring the set of two difference frequency in including 17 frame of L band (PALSAR data) and 21 frame of C band (ASAR data) have processed. The preliminary processing result was shown out of all data sets were applied, 16 interferometer images were suitable for the next processing stages. The outcome of interfrogram analysis along with field checking shown 914 segments marked as landslides anomaly over the area of 13577.98 hectares. The maximum land sliding activities was identified as 1.58 meter vertical displacement during October 2010 to September 2011.

Farthermore, The results showed differential interferometry method applied for above mention frequencies have sufficient accuracy and conform efficiency for landslide detection, monitoring and distribution mapping compared with other methods (field survey in extend area, aerial photo and optical satellite images). Out of this two frequency low frequency of PALSAR images have better ability in the landslide distribution due its capability such as longer wavelength and reduce incoherency. Where the high frequency of ASAR images is can offer more details of the landslides due its shorter wavelength.

In order to validation of two susceptibility produced by two scenarios (radar interferometry and field survey approaches) training data (70 percent of landslide area) and tested data (30 percent of landslide area) was compared by using area under curve (AUC) receiver operating characteristic (ROC) and success rate curves (SR), frequency ratio (FR) and seed core area indices (SCAI). In basis of mentioned indices values, landslide susceptibility zonation using radar interferometry scenario was more favorable than field survey scenario. Landslide hazard zonation map was prepared by multiplying event probability (Pe), temporal probability (Pt), spatial probability (Ps) and overlaying temporal period landslide inventory maps. Then spatial vulnerability map was created by using final weighting and element of risk maps including population, Building and road, natural and agricultural land use. Finally landslide risk evaluation was comprised four landslide risk classes, such as low, moderate, high, and very high. The areal extents of these sub-classes were found to be 12.6%, 71.7%, 15.4%, and 0.2%, respectively. From 125084 persons population in inhabitant zones in basis census population and habitation was settled in landslide risk classes from low, moderate, collectively high and very high, 26.2%, 73.3%, 0.5%, respectively. The mentioned outline can conclude that landslide inventory map based on RADAR images appeared as high capability technique in form accuracy, time consuming and cost effectiveness comparing with other approaches (spread area field survey, aerial photographs and optic satellite images). It also absolved that this technique can apply for many wide vulnerable areas such as Zagros. Providing the scientific technique and a set of validated tools for the preparation and the optimal use of landslide maps, landside prediction models, and landslide forecasts to reduce this shortcoming of this subject in Iran can mark as a main goal and outcome of this research.

The CSES space mission will study the temporal stability of the inner Van Allen radiation belts, investigating precipitation of trapped particles induced by magnetospheric, ionosferic and tropospheric EM emissions, as well as by seimo-electromagnetic disturbances.
CSES satellite will be launched in September 2016 and inserted into a circular Sun-syncronous orbit with 98 degrees inclination and 500 km altitude. Expected lifetime is 5 years. CSES hosts several instruments onboard: 2 magnetometers, an electrical field detector, a plasma analyzer, a Langmiur probe and a High Energy Particle Detector (HEPD). The HEPD detector consists of two layer plastic scintillators, one segmented, for the trigger and a calorimeter constituted by a tower of plastic scintillator counters and a LYSO plane. The direction of the incident particle is provided by two planes of double-side silicon microstrip detectors placed in front of the trigger.
HEPD detector will measure electrons (3 - 100 MeV) and protons (30 - 300 MeV) along CSES orbit. The angular and energy resolution and the detector acceptance are optimized to accurately detect the expected low short-term time variations of the particle flux from the radiation belts.
Topic of this talk is the technical description of the HEPD and its main characteristics.

This paper presents a fast and transferable procedure for rapid mapping of landslides, with the advantage of requiring only post-event SAR imagery. The methodology was developed using dual-polarimetric TerraSAR-X imagery acquired over a landslide at the Yeager Airport near Charleston, West Virginia, USA. Then, the transferability of the methodology was successfully demonstrated by applying it to a further landslide event, a huge mining waste landslide which occurred on 1 April 2015 near Bolshaya Talda, Kemerovo Oblast, Russia.

On 12 March 2015, a large-scale landslide occurred at an artificial slope at the Yeager Airport. The airport, completed in 1947, was constructed atop seven semi-connected hilltops. In 2006, due to new Federal Aviation Administration (FAA) safety regulations, the construction of an Engineered Material Arrestor System (EMAS) was necessary. Therefore, the Yeager Airport had to be extended, leading to the construction of a large artificial slope, being the tallest geosynthetic reinforced slope in North America (of ratio horizontal/vertical 1:1).

Very High Resolution (VHR) archive SAR imagery recorded shortly before a landslide event is commonly not available. Modern VHR SAR missions such as TerraSAR-X, COSMO-SkyMed or Radarsat-2 do not systematically cover the entire world. Each acquisition has to be programmed manually. Furthermore, due to limited disk space on board the satellites and especially due to limited downlink transmission rates, these sensors are not able to provide worldwide coverage within a short time period – i.e., no archive image recorded shortly before the event. Therefore, we present a methodology for rapid landslide mapping based solely on post-event SAR imagery. Shortly after the first failure of the slope at the Yeager Airport a dual-polarimetric TerraSAR-X HighResolution SpotLight dataset was programmed. About one month later a second failure of the slope occurred, which was then recorded by a second TerraSAR-X scene acquired using the same imaging geometry as the first one.

Assuming land cover changes due to the landslide event, i.e. destruction and removal of the vegetation cover – as it was the case at the Yeager Airport landslide – the first step of our object-oriented procedure is the pre-selection of formerly vegetated areas based on the Normalized Difference Vegetation Index (NDVI) of freely available pre-event optical imagery (e.g. Landsat-8 or the Sentinel-2). Next, after polarimetric speckle filtering, the entropy/anisotropy/alpha (H/A/alpha) decomposition is applied to the post-event polarimetric SAR image to detect (within the pre-selected areas) regions characterized by low entropy values, i.e. an evidence of bare soil or rock (landslide material). Then, assuming a certain minimum slope value as a necessary requirement for a landslide event, the landslide detection map is refined accordingly. The slope information was derived from the 30 m resolution version of the Shuttle Radar Topography Mission (SRTM) Digital Elevation Model (DEM).

The developed methodology was applied independently to both aforementioned TerraSAR-X images showing the landslide at different stages of development. As a second failure of the landslide occurred between the two acquisition dates, change detection procedures show high accuracies in mapping the active landslide.

ACKNOWLEDGEMENTS

This work has been funded by ESA in the framework of the ASAPTERRA project (Contract No.: 4000112375/14/I-NB).

Indonesia leads the world in many volcano statistics. More than 75% of its residents live within 100 km of a Holocene volcano. The combination of densely packed population and large amount of active volcanoes has made Indonesia suffering the highest number of eruption produced fatalities and damage to human infrastructures. With the improved revisit time, swath coverage and higher resolution over ERS-1/2 and Envisat ASAR, Sentinel-1 makes it more achievable to implement large-scale InSAR survey and resolve more detailed and transient deformation process. Here we present the latest time series InSAR result on volcanic area in Indonesia with Sentinel-1 Interferometric Wide (using TOPSAR technique) Swath mode data. Our purpose is to better define the potential of Sentinel-1 data for dynamic volcanic processes.

Land subsidence is generally caused by human and natural activities such as alterations to the earth’s surface and underground geologic processes. The Nigerian coastal geosyncline is subsiding not only because it was formed in a tectonic setting but because of the continuing dewatering and compaction of its sediments which were deposited rapidly. Subsidence in this area of Nigeria is presently being accelerated by the indiscriminate withdrawal of fluids, including oil and gas particularly in Warri, Yenagoa and Port Harcourt cities from underground aquifers and reservoir strate.

This uncontrolled exploitation of the groundwater, oil and gas has led to progressive decline of the aquifer level and a continuous need for opening deeper drillings to exploit deeper aquifers.

In this study Multi -Temporal Interferometry (MT -InSAR) Technique has been applied to monitor subsidence in the study area in the period 2003 to 2010 with Envisat data. From the analysis of the results, Lagos state appears to be subsiding conically, The velocity rates of subsidence in the surrounding cities like Lekki, Badagry, Ikorodu, Epe and others are much higher than that of Lagos city, these are the cities from the preliminary investigation results reveals heavy  structures, particular buildings were seen constructed mostly on the sand filled areas where the sediments compaction rates is very high.

In recent years, the Canadian Space Agency (CSA) has formed partnerships with industry, universities, and governments to support the integration of operational Earth Observation (EO) technologies and services, including data processing, methods, applications, demonstrations, in the field of civil security and disaster risk management in Canada. These partnerships focus on the needs of organizations concerned with civil security, and the utilization of Earth Observation technologies has contributed to advancements in that field. These innovations and solutions help us better understand disaster risks and mitigate the consequences of disasters, such as ice jams and floods. The investments that CSA has made with the industry, academia and governments since the launch of RADARSAT-1 in 1995 in ice jam & floods related thematics (soil moisture, river ice, snow, water extent, water levels, etc.) led to the successful integration of EO products in the Ministère de la Sécurité Publique du Québec (MSP) (Province of Quebec, Canada) operations. The MSP has the mandate to coordinate the actions of the Government of Quebec in mitigation, preparedness and response to disasters in order to assure the safety of the population, infrastructures and services. Since 2010, the MSP is monitoring rivers at risk of ice jams and floods and generates its own river ice maps automatically, just a few minutes after the acquisition and reception of SAR imagery (i.e. RADARSAT-2). The methods and algorithms used by the MSP were developed by the Institut National de Recherche Scientifique - Centre Eau Terre Environnement (INRS-ETE) and supported by the CSA in close collaboration with Public Safety Canada and Natural Resources Canada. The public health community is also interest in floods information to mitigate risks of water contamination. CSA EO Applications & Utilizations (EOAU) programs such as the Government Related Initiatives Programs (GRIP), Earth Observation Applications Development (EOADP) and the Rapid Information Products and Services (RIPS) initiatives played a major role in promoting the utilization of SAR data for ice jams and flood monitoring by the MSP. The MSP is planning to use the SENTINEL-1 data as a complement of information with RADARSAT-2 and the up-coming RADARSAT Constellation Mission. Other CSA supported EO missions could be used by the MSP such as SWOT (Surface Water and Ocean Topography).

The presentation will (1) illustrate CSA & MSP past and on-going activities linked to the development of products and services in support of ice jams and floods monitoring; (2) demonstrate how the SENTINEL-1 data are relevant and can be used by MSP in its monitoring operations in complement with RADARSAT-2 and in preparation to the RADARSAT Constellation Mission; (3) demonstrate how the SWOT data could be integrated in MSP monitoring operations.

The world's 1500 active subaerial volcanoes pose a diverse set of hazards to both people living in their vicinity and the global community.  Advanced warnings of the possibility of an eruption can mitigate these risks, but many volcanoes are not routinely monitored due to the cost and logistical difficulties of installing instruments.  However,  satellite based geodesy provides a unique opportunity for routine global volcanic monitoring. 

Currently, technical limitations of operative satellites greatly reduce their ability to perform routine global monitoring.  However, the short revisit times, routine acquisitions and fast data availability of the Sentinel-1 satellites provide a unique opportunity to overcome these technical difficulties, and expand routine monitoring to most of the world's 1500 active volcanoes.  Due to the large number of interferograms that this will produce, an algorithm to automatically detect signs of unrest in a time series of interferograms will be required. 

Here, we present initial progress on the construction of this algorithm through a study of the use of blind signal separation methods to decompose interferograms into components (or parts).  We have synthesised several time series of interferograms that are a linear combination of several components, such as deformation due to a magmatic source, and delay due to an atmospheric phase screen.  We have then implemented non-negative matrix factorisation, principal component analysis, and independent component analysis to recover the original magmatic signals, and the time courses for their use throughout the time series.   

We discuss the relative merits of each method, and if the data used satisfies the various assumptions required for these methods to be applied.  We also present results from the application of these methods to a real time series of data acquired by Sentinel 1-a. 

This paper examines the possibility of fire detection in Zabaikalsky Kraj in april 2015 by Sentinel 1 SAR images. Fires are detected due to the ecological changes in the areas of fire burning after fire. A satellite imagery of an operational monitoring system “Kosmosnimki - Fires” served as additional information.

The most informative parameters for burned area detection are the values of sigma 0 and Haralick’s textural feature ‘contrast’.

 On the basis of the ratio of the sigma 0 after and before the fire it was shown the increase in the values of sigma 0 and ‘contrast’ in the burning areas compared with the areas outside the fire. Sigma 0 increasing along the fire area profiles is from 4 to 5 dB for VV and VH polarizations, as for areas outside the fire is from 0.6 to 3.4 dB. The mean ‘contrast’ value along the fire area profiles increased from 2 to 2.7 times (VV polarization after speckle filtering), while outside fire area there is contrast reduction or increase up to 1.4 times.

Comparison of the images for two dates before and after fire based on 1) unsupervised classification, k-means, 2) dual polarizations and 3) texture segmentation does not allow to interpret the classification results.

The use of optical information is required for a reliable determination of the fire scars.

Flood extent maps derived from Synthetic Aperture Radar (SAR) data can be a key information source for an effective disaster management, helping humanitarian relief organizations and decision makers to obtain spatially-explicit information about inundated areas in a time- and cost-efficient manner [1]. In comparison to manual or semi-automatic flood mapping approaches often utilized in the framework of rapid mapping activities, automatic SAR-based processing chains can substantially reduce the critical time delay between the delivery of satellite data after a crisis event and the subsequent provision of satellite derived crisis information (e.g. the extent of a flood situation) to emergency management authorities. In this work, an automated Sentinel-1 based processing chain designed for flood detection and monitoring in near real-time (NRT) is presented. The work is based upon a TerraSAR-X based flood service [2] which has been adapted to Sentinel-1, including a number of enhancements for improving both robustness and thematic accuracy. Since no user intervention is required at any stage of the flood mapping procedure, the processing chain allows deriving time-critical disaster information in less than 45 minutes after a new dataset is available on the ESA Sentinel Data Hub. Due to the systematic acquisition strategy and high repeat rate of Sentinel-1, the processing chain can be set up as a web-based service which regularly informs users about the current flood conditions in a given area of interest. The processing chain is composed of the following main elements: a) automatic data ingestion through a Python-based script which routinely queries the ESA Sentinel Data Hub for new acquisitions matching user-defined criteria and subsequently downloads them, b) geometric correction and radiometric calibration using the graph processing tool (GPT) of the ESA Sentinel-1 toolbox (S1TBX) which is embedded in the Sentinel Application Platform (SNAP), c) initial classification using an automatic thresholding methodology, d) fuzzy-logic based classification refinement, e) final classification including auxiliary data, and f) dissemination and visualization of the results using a dedicated web client. The thematic accuracy of the processor has been assessed for two test-sites of a flood situation at the border between Greece and Turkey with encouraging accuracies. The accuracy assessment, which was performed separately for both standard polarizations (VV/VH) of the interferometric wide swath (IW) mode of Sentinel-1, further indicates that the thematic accuracy of VV polarization is slightly higher than that of VH polarization under calm wind conditions.

Sunsari district in Nepal has been inundated many times in recent decades, with the most extensive inundation taking place after the breach of the Koshi embankment at Kusaha on 18 August 2008. Historically, land has been simply washed away by the unregulated stream flow and strong currents, and after floods people are often unable to identify their land. Thus it is not possible to determine the vegetation status of land that has been washed away. For damage assessment, disaster preparedness, and land use planning, it is important to know what type of vegetation was established prior to flooding, how much was destroyed especially by the 2008 flood, and how the plant succession has developed in the flood damaged areas. In the present study, land cover dynamics and plant succession information for 2007, 2008, 2014, and 2015 were investigated using maps derived from Advanced Land Observing Satellite Phased Array type L-band Synthetic Aperture Radar (ALOS PALSAR) images supported by Advanced Visible and Near Infrared Radiometer type (AVNIR) images. Open access NEST (Next ESA SAR Toolbox) tools were used for reading, post-processing (import/export, calibration, filtering, resampling, co-registration, and orthorectification), and visualizing the ALOS PALSAR images. Object based image analysis in eCognition was used to determine land cover classes such as, forest, shrubland, grassland, crop land, water bodies, and settlements. The ALOS-PALSAR data training areas were chosen using the AVNIR imagery. Ground reference data were collected from sites throughout the study area for validation. 

The contribution focuses on the results of the ESA SEOM study entitled “INSARAP: Sentinel-1 InSAR Performance study with TOPS data”. The study investigates the performance of the interferometric wide swath (IW) mode of Sentinel-1, implemented using the Terrain Observation by Progressive Scans (TOPS) mode. In this regard, first analyses on Sentinel-1 time series will be presented, focusing on the comparison with in-situ geodetic measurements on a volcanic test site (Campi Flegrei/Vesuvius area, Italy) identified in the framework of the study. In particular, Campi Flegrei, west of Naples, are undergoing an uplift phase started at the beginning of 2005, with an overall inflation of about 7 cm in the time span October 2014 - October 2015 and clearly detected with Sentinel-1 data. The assessment of the results will be performed by exploiting the geodetic data from the INGV-OV (Istituto Nazionale di Geofisica e Vulcanologia - Osservatorio Vesuviano) monitoring networks located in the area, namely, from continuous GPS (cGPS) stations and discrete leveling measurements. The chance to project the 3D GPS data into the Sentinel-1 LOS, besides the availability of InSAR data from both ascending and descending tracks to be splitted for comparison with vertical measurements from leveling, will allow such a comparison, providing also a dataset for building up the 3D deformation field acting in the area of interest and afterwards the modeling of the deformation source.

In this paper we present a new method for fully automatic detection and
derivation of areas endangered by heavy rainfall based on digital elevation
models and multispectral satellite data. Tracking news show that the majority of
occuring natural hazards are flood events. So already many flood prediction
systems were developed. But most of these existing systems for deriving areas
endangered by flooding events are based only on horizontal and vertical
distances to existing rivers and lakes. Typically such systems take not in
account dangers arising directly from heavy rain events. In a study conducted by
us together with a german insurance company a new approach for detection of
areas endangered by heavy rain was proven to give a high correlation of
the derived endangered areas and the losses claimed at the insurance company.

Following the experience from this study we present here a system to detect
areas endangered by heavy rain and perform further vulnerability analysis based
on digital elevation models (DEM) and medium to high resolution satellite
imagery.  The work shown here is based on medium resolution DEMs like the
SRTM-C-band, the downsampled EUROMAPS DEM from the indian Cartosat satellite or
any other DEM representing a digital terrain model resembling ground elevations
with a ground sampling distance in the range of about 25 to 30~m. In a first
step the so called terrain positioning index (TPI) is derived using values for
the inner and outer radii calibrated in the insurance project mentioned
above. The terrain positioning index used in this approach classifies the
digital elevation model to six classes: summit, top hillside, steep slope,
bottom hillside, depression and plain. The study shows that most of the heavily
affected properties are located in the depression areas, fewest -- as can easily
be figured out -- on summit areas. In the next step the vulnerabiltiy analysis
is performed. For this the catchment areas for each depression area found are
derived. For each of these catchment areas a land type classification is derived
from the corresponding multispectral satellite imagery. Also the depth of each
depression area is calculated. Combining these information together to a model
allows the prediction of vulnerability depending on the amount of rainfall and
the size and classification of the catchment area.

In this paper we describe the above mentioned method used for automatic
detection and vulnerability analysis for areas endangered by heavy
rainfall. Afterwards some examples using different DEMs and different
multispectral satellite imagery are presented and validated using
insurance data where possible.

Achieving food security for all is at the heart of FAO's (Food and Agricultural Agency of the UN) efforts – to make sure people have regular access to enough high-quality food to lead active, healthy lives. FAO’s three main goals are: the eradication of hunger, food insecurity and malnutrition; the elimination of poverty and the driving forward of economic and social progress for all; and, the sustainable management and utilization of natural resources, including land, water, air, climate and genetic resources for the benefit of present and future generations.

In the framework of supporting developing countries in their fight against emergencies (diseases, disasters, etc.), we help to develop adoptable, adaptable and feasible tools based on satellite data to help in forecasting of tropical human, and livestock disease as well as nature emergences.

Moreover in order to help prepare countries to be ready for REDD+, the UN-REDD Programme assists developing countries to prepare and implement national REDD+ strategies. For the monitoring, reporting and verification, FAO supports the countries to develop national satellite forest monitoring systems that allow for credible measurement, reporting and verification (MRV) of REDD+ activities. These are among the most critical elements for the successful implementation of any REDD+ mechanism. The UN-REDD Programme through a joint effort of FAO and Brazil's National Space Agency, INPE, is supporting countries to develop cost- effective, robust and compatible national monitoring and MRV systems, providing tools, methodologies, training and knowledge sharing that help countries to strengthen their technical and institutional capacity for effective MRV systems.

To develop strong nationally-owned forest monitoring systems, technical and institutional capacity building is key. The UN-REDD Programme, through FAO, has taken on intensive training together with INPE, and has provided technical help and assistance for in-country training and implementation for national satellite forest monitoring.

In this talk, some concrete examples of early warning systems, its implementation and its impact on food security will be presented.

Volcanic unrest which produces a variety of geological hazard is sometimes difficult to predict and capable of triggering natural disasters on wide areas. Therefore it is important to monitor volcano continuously. The monitoring of active volcanoes requires the reliable measurement of surface deformation before, during and after volcanic activities and it helps for the better understanding and modeling of the involved geophysical processes. Space-borne synthetic aperture radar (SAR) interferometry (InSAR), persistent scatterer interferometry (PSI) and small baseline subset algorithm (SBAS) provide a powerful tool for observing the eruptive activities and measuring the surface changes of millimeter accuracy [1-13]. All the mentioned techniques address the challenges by exploiting medium to large SAR image stacks. The process of selecting, ordering, downloading, storing, logging, extracting and preparing the data for processing is very time consuming task and has to be done manually for every single data-stack. In many cases it is even an iterative process which has to be done regularly and continuously. The activities of data retrieval and data preparation require a significant part of the processing time and shortening those leads to much quicker results.

In the framework of remote sensing task of MED-SUV project, DLR-IMF is responsible for developing an automated SAR satellite-based system which will automate the entire time consuming tasks and allows an operational monitoring for ground deformations over active volcanoes. The automated volcano monitoring system keeps all logging updated and prepares reports showing the data processing results. Furthermore, the system will deliver specified reports and maps to a database for review and used by specialists. The interaction of the person doing the processing will be minimized and iterative processes will be totally avoided. Users should get in contact with DLR personnel and agree on a test-area, time period and acquisition geometry and once the system starts producing results, there will be an automated upload to a dedicated ftp-server to which the user will get access, including E-Mail alert.

In this presentation, a prototype (system concept, interfaces and examples) of automated SAR Satellite based High Resolution Data Acquisition System, developed and operated by DLR, will be described in detail. The workflow of the developed system is described which allow a meaningful contribution of SAR for monitoring volcanic eruptive activities. A more robust and efficient InSAR data processing in DLR@IWAP processor will be introduced. The system reduces the load of manual work necessary to perform interferometric stacking and quickly gain first information on evolving geophysical processes at the, but not limited to the Italian Supersites. An application of the developed prototype system to a historic eruption [14-15] in Mount Etna and Piton de la Fournaise will be depicted in the last part of the presentation.

Fires and floods are among the natural hazards with the higher social impacts in the 21st century.Both cause economic costs of the order of billions of euros. When these occur in urban and periurban areas, the loss of human lives, the destruction of private and public properties, the degradation of health and quality of life, as well as the disruption of economic activities are among the impacts that cause. Floods that occur after the manifestation of fires, are extremely catastrophic, especially in periurban areas. The study of both hazards is based on the same background data and Earth Observation (EO) is a crucial information source, as from the same satellite imagery up-to-date fuel map can be derived in case of fire modeling in urban and periurban areas, while the parameterization of flood modeling in different scales (hydrological modeling in catchment basin level and hydraulic model in the urban area) need dedicated land cover/use information, updatable when needed and suitable for the specifications of the models.

The investigation of both fire and flood hazards traditionally has been conducted separately even if the same data are needed. This approach overlook the “collect once – use for many purposes” model which when is adopted, result in the increase of the accuracy and economies, as these phenomena are tightly interrelated; fires exacerbate the flood risk and the preceding flood dramatically reduce the fire risk.

In the framework of the LIFE+ project FLIRE (LIFE11ENV/GR/975), an integrated Decision Support System (DSS) was developed for both floods and fires risk assessment and management by adopting the model of “once collect – use for many”, by using existence infrastructure and by incorporating extensively EO sources in different phases of the project. 

The FLIRE DSS is consists of three modules and seven applications unified under the FLIRE Server. The modules for fire management, flood management and weather forecasting have been implemented as webservices. The system has been designed as web-based solution which integrates the abovementioned tools. FLIRE adopt the distributed architecture of the components of the system while the DSS is accessible from the web (www.fliredss.eu) under a password protected scheme due to the sensitive information that serve. A user group with people from departments that are related to the planning, management and confrontation of the natural hazards in local and regional level has been established and only these have access. For demonstration reasons, a free version of the FLIRE DSS has been designed. It has scenarios for each tool and is available to each interested from the same website.

The FLIRE server uses FTP and HTTP communication protocols and web service technologies. Visual basic, JavaScript, Google Maps API and Ajax have been used for the design and implementation of the FLIRE DSS. The user’s interface has been designed and developed based on the user's requirements, goals and needs. ESA’s Sentinels missions (1 & 2) is expected to have a crucial role for these steps due to the high temporal and spatial resolution, the high data quality, as well as the free data access policy.

This study focuses on evaluating the potential of ALOS/PALSAR time-series

data to analyze the activation of deep-seated landslides in the foothill zone of the high

mountain Alai range in the southern Tien Shan (Kyrgyzstan). Most previous field-based

landslide investigations have revealed that many landslides have indicators for ongoing

slow movements in the form of migrating and newly developing cracks. L-band

ALOS/PALSAR data for the period between 2007 and 2010 are available for the 484 km2

area in this study. We analyzed these data using the Small Baseline Subset (SBAS)

time-series technique to assess the surface deformation related to the activation of

landslides. We observed up to ±17 mm/year of LOS velocity deformation rates, which

were projected along the local steepest slope and resulted in velocity rates of up to

−63 mm/year. The obtained rates indicate very slow movement of the deep-seated

landslides during the observation time. We also compared these movements with

precipitation and earthquake records. The results suggest that the deformation peaks

correlate with rainfall in the 3 preceding months and with an earthquake event. Overall, the

results of this study indicated the great potential of L-band InSAR time series analysis for

efficient spatiotemporal identification and monitoring of slope activations in this region of

high landslide activity in Southern Kyrgyzstan.

Many areas in the world are known as being affected by slow gravitational soil movements, called soil creeping, and Slovenia is no exception in this regard. A fact that creeping is one of the pre-failure processes in landslide-prone slopes a monitoring and analysis of creeping is crucial for planning any remediation measures or even detecting not so obvious slope movements. Although soil creeping gives less attention to the media and decision makers as landsliding does and although it is also harder to detect, it causes steadily increasing damage to buildings and infrastructure in the long term. Among the remote sensing monitoring techniques radar interferometry is one of the most promising tools that provide valuable data for mapping, monitoring and updating regional landslides cadastre. Various upgrades of radar interferometry technique, including persistent scattering radar interferometry (PSI), enable monitoring and especially identifying slope instability worldwide. The persistent scatterers interferometry allows us to detect the stability/instability for very large areas with a millimetre precision and the possibility to monitor historical events dating back to 1992 when interferometry data acquisition began. This method provides high value data in areas where other techniques (i.e. optical remote sensing techniques) for monitoring the instability fails. PSI is often used complementary with other techniques (e.g. ground control point). In the field of engineering geology using the radar interferometry method has also a great potential, especially as its ability to accurately measure the movements in range of mm opens new domains of effective expertise implementation.

In Slovenia, PSI campaigns have been used to detect slope mass movements in the Škofjeloško Cerkljansko area, Maribor and Ljubljana surrounding and in Potoška planina. Displacement rates at the first three territories have been obtained by natural point targets, referred to as persistent scatterer (PSs) like buildings, antennas, bridges, urban structures or stable natural outcrops, meanwhile the deformation rates in Potoška planina have been monitored by artificial scatterers in a form of compact active transponders (CATs). Historical SAR images were acquired from ESA, ERS-1, ERS-2 and Envisat that operated in C-band with a wavelength of 5.6 cm.

The results of the PSI survey in the Škofjeloško-Cerkljansko area show that the PSI method is suitable for assessing the temporal evolution of slow and extremely slow landslides with constant deformation velocities, hereafter called the soil creep. Due to the high PS density in some smaller areas within the broader study area it was possible to identify three areas with a constant deformation (area above Cerkno, Planina and Čeplez). In those areas, the analysis of average annual creeping rates revealed that lithology and slope inclination are among the key precondition factors for the occurrence of the slope creeping process. Similarly the PSI revealed unstable slopes in the vicinity of Maribor and Ljubljana but to smaller extents. Due to the ability of PSI methods to monitor the deformations in time, the assessment was made about possible climatic triggering factors, especially rainfall events. The results show, that soil creeping is induced not only by heavy precipitation events (>100 mm/day), but also by less intense precipitation (>20 mm/day), or cummulative of 50 mm of rain in three days. First event produced 0.4-0.76 mm of elevation change, and events with higher precipitation the average elevation decrease of -1.2 mm.

In the case of Potoška planina a field trial was set up to test a novel device that was developed in the frame of European Union funded project “Integrated Interferometry and GNSS for Precision Survey (I2GPS)” found by Seventh Framework programme (FP7-GALILEO-2008-GSA-1). The devices combine a Compact Active Transponder (CAT) and a Global Navigation Satellite System (GNSS) antenna and integrated two technologies InSAR and global navigation satellite system (GNSS). This approach provided 3D displacement assessments of the monitored locations on the landslide and its vicinity. InSAR and GNSS results for the monitoring period (02/2011 – 08/2011) showed relatively large (up to 32 mm horizontal and up to 15 mm) vertical displacements, indicating a displacements of the central-upper and south-eastern parts of the landslide body. 

Thanks to the all-weather, day-night capability to detect and measure small ground surface deformations, multi-temporal Synthetic Aperture Radar (SAR) Interferometry (InSAR) techniques are attractive for landslide investigations. This Multi-temporal InSAR (MTI) application poses challenges related to the limited spatial extent of the phenomenon, the complexity of the ground deformations, and the occurrence in mountainous and vegetated areas that cause visibility problems.

Nowadays, several satellite missions are available providing interferometric SAR data at different wavelengths, spatial resolutions, and revisit time. High-resolution X-Band SAR sensors, such as the COSMO-SkyMed constellation, have recently provided data with spatial resolution reaching metric values, and revisit time up to few days leading to increase the density of the measurable targets as well as to improve the detection of non linear movement. Medium resolution C-band SAR data have been thoroughly exploited in the last two decades thanks to the ERS-1/2,  ENVISAT-ASAR and Radarsat missions. A new interesting opportunity is recently provided by Sentinel-1 mission, which has just started to deliver data to the user community.  Although the spatial resolution is comparable to previous ESA C-band missions the revisit time is reduced to 12 and 6 days, by considering, respectively, one or two satellites. It is envisioned that by offering regular globe-scale coverage, improved temporal resolution and freely available imagery, Sentinel-1 will guarantee an increasing use of MTI in landslide investigations. These background missions are necessary for long-term, systematic mapping of unstable unstable slopes and regional scale assessment of landslide processes.

According to these different SAR space-borne missions, the present work discusses current and future opportunities of MTI applications to slope instability monitoring by addressing,  through a simple theoretical model,  issues related to slope visibility, sensitivity to ground displacement, coherent target detection, mean velocity precision. In particular a comparative analysis is carried out aimed at investigating specific advantages of different satellite missions with respect to wavelength, resolution, revisit time, and clutter backscatter. According to this analysis, we show that a reliable estimates of the displacement rates are possible in a narrower time span by using X-band than C-band, but Sentinel-1 constellation will improve the performance of C-band.
 Moreover, for a certain time span the expected velocity STD is better for X-band than for C-band but, again, the Sentinel-1 constellation will improve the performance of C-band. This impacts on the monitoring capabilities in high risk situations by providing results in a shorter time, and allows monitoring mountainous areas during short period avoiding snow coverage, thus maximizing the chance to detect coherent targets.

Finally, the work provides examples of multi-sensor and multi-scale slope instability investigations obtained by processing real data from both C-band medium resolution and X-band high resolution SAR data. In particular, the differences between displacement maps in terms number of images, observation time span, number of coherent targets, and estimated displacement precision, are used to support the outcomes of previous model.   

ACKNOWLEDGMENTS

Work supported by the project “APULIA SPACE” (PON03PE_00067_6), PON Ricerca e competitività 2007-2013. 

Fires in general and forest fires specific are a major concern in terms of economical and biological looses. Fires can burn annually more than 4 million km² worldwide causing close to 2 Gt of yearly carbon emissions into the atmosphere. The causes of fire activity are mainly anthropogenic, because of land management, pest control or negligence. Remote sensing technologies have been focussed on developing several algorithms, adapted to different sensors, platforms and regions in order to locate hotspots as fast as possible. Hotspots detection, in the remote sensing field, has been mainly carried out by two different kinds of platforms: geostationary and polar orbiting, both having advantages and disadvantages. The aim of this study is to establish an automatic methodology to develop hotspots detection algorithms with Spinning Enhanced Visible and Infrared Imager (SEVIRI) sensor on board Meteosat Second Generation platform (MSG) based on machine learning techniques that can be exportable to others geostationary platforms and sensors and to any area of the Earth. In order to perform the model with a minimum of complexity only two SEVIRI channels have been used: the 3.9 μm and 10.8 μm channels. The ground truth fire data have been acquired and maintain by the Ministerio de Agricultura, Alimentación y Medio Ambiente of Spain (MAGRAMA) and distributed by www.españaenllamas.es and www.civio.es. The analysed data comprises all the fires reported during the year 2012 in peninsular Spain. There are 15115 analysed fires in total which have been used to train and test the performance of the machine learning algorithm. Several machine learning classification algorithms have been applied in order to solve the problem under different data management that include choosing the predictive variables and the optimum number of the fire and non fire cases in order to train avoiding overfitting. The confusion matrix has been calculated for each case: this is the number of fires and non fires correctly identified, true positives (TP) and true negatives (TN) and the number of fires incorrectly identifies and non fires incorrectly identifies, false positives (FP) and false negatives (FN) respectively. The sensitivity (SE), specificity (SP) and accuracy (AC) indicators are also calculated. These parameters have been analyzed in order to develop the final machine learning taking into account the preferences and final use of the predicted data. A concrete model has been performed in order to detect the forest fires in the peninsular Spain and the guidelines to choose the performance of the final model have been stabilised in order to make the procedure applicable to other platforms and sensors on any world area. 

Separation of small magnitude linear and non-linear land subsidence in the presence of seasonal displacement using InSAR is challenging, especially using short time series. We attempt to identify these different displacement components using a three-year InSAR time series, processed using combined small-baseline (SBAS) and persistent scatterer (PS) methods, supported by results from independent geodetic and hydrogeological data.

Subsidence due to groundwater withdrawal from aquifers beneath the Perth Basin (Western Australia) for residential and horticultural use has been occurring at varying rates since the mid-1970s. A lack of a dedicated monitoring system over this period has resulted in limited datasets from which to analyse and quantify this subsidence, but the recent establishment of a monitoring system within this project has started to yield useable observations from InSAR, repeat levelling, continuous GPS and groundwater level data from monitoring bores.

Initial processing and analysis of a two-year TerraSAR-X (TSX) time series of 56 stripmap scenes acquired under a German Space Centre (DLR) science project has identified subsidence in some areas of the Perth Basin in the period from October 2012 to November 2014. However, the small-magnitude of the subsidence (<6 mm/yr from continuous GPS) makes it challenging to detect significant surface displacement, primarily because the uncertainty of atmospheric phase delay in the processed PS is at a similar (or larger) magnitude than two-years’ linear displacement. This is exacerbated by seasonal displacement indicated by repeat levelling (up to 30 mm/yr) and non-linear displacement indicated by continuous GPS stations and groundwater level analysis. The TSX time-series has detected this seasonal variation and/or non-linearity at specific spatial points, but at different orders of magnitude to the levelling, which is a discrepancy that remains unresolved and is the focus of this work.

The three-year TSX time series of 79 TSX stripmap scenes has been processed using the SBAS and PS methods in StaMPS. The atmospheric phase delay contribution to the uncertainty in the PS velocities is reduced because (1) the linear displacement component is expected to now exceed the variance due to the atmospheric delay component because of the increased observation period and the large number of acquisitions, and (2) the three-year time series allows for modelling of seasonal displacement.

Analysis is conducted at continuous GPS stations, repeat levelling benchmarks and groundwater monitoring boreholes to identify any correlation between InSAR time series and those from the continuous GPS, levelling and groundwater. From the combination of these independent time series, separate linear, non-linear and seasonal displacement components are mapped, in the process resolving the discrepancies identified in the two-year TSX time series.

Since its creation in October 2000, the International Charter on Space and Major Disasters has been activated about 450 times, half of them being due to floods; this unified system of space data already allowed more than 110 countries affected by natural or man-made disasters to take prompt and efficient actions to protect populations and their environment. Among the Charter resources is Sentinel-1A satellite: based on C-band imaging radar, it provides an all-weather day-and-night supply of imagery for Copernicus user services. The mission’s high frequent revisit and availability makes it particularly suitable for emergency support. This was illustrated in the beginning of August 2015 with its response to the activations of a Charter and of a Copernicus EMS, both triggered for floods in Myanmar.

Sentinel-1’s C-band synthetic aperture radar instrument (C-SAR) operates in four nominal operational modes: Stripmap (SM), Interferometric Wide swath (IW), Extra Wide swath (EW) and Wave mode (WV). Besides the difference in the way they operate and the type of surface they are intended to cover, their specific swath imply different levels of coverage and thus different revisit time for a given area. The purpose of this study, conducted at ESA, is to assess - for SM, IW and EW modes - the revisit time for any point of the Earth. Observations in the WV mode are too sparse and therefore not considered in this study.

In order to compute the revisit time, orbit propagation and satellite pointing calculations were performed with the Earth Observation CFI software, a collection of libraries made available by ESA-EOP System Support Division. The number of acquisitions per cycle varies significantly in latitude and repeats every 2° in longitude due the repeat-track nature of S-1's orbit. The computation of the revisit time was therefore made sweeping latitudes between [−90°; 90°] and for longitudes ranging between 0° and 2°. The study was conducted for two cases: first one being only Sentinel-1A operating, and second case corresponding to the whole constellation. For each operational mode studied, the maximum, average, and minimum revisit time were calculated.

Coverage maps were first produced to illustrate graphically the impact of different swaths on the potential number of acquisitions [Figure1]. As expected, the number of acquisitions of Sentinel-1A over one cycle increases when getting closer to the poles, reaching respectively 24 (SM), 53 (IW) and 64 (EW). The second step consisted in generating plots that show the evolution of the minimum, average and maximum revisit time along with latitude, depending on the C-SAR mode selected (longitude being fixed) [Figure2]. The graphs obtained are not smooth but serrated, which is due to the diamond-pattern observed on the coverage maps that results from successive swaths crossing one another [Figure3]. A study performed over a longitude band (2° large) instead of a discrete longitude value causes those fluctuations to lessen. While the maximum revisit time reaches 12 days with the SM mode and 11.5 days for the IW mode, it only reaches 7 days for the EW mode. This shows the impact of the operational mode selected on the revisit time, a critical parameter in emergency situations. Results obtained with both Sentinel-1A and 1B show that half the revisit time obtained with one satellite is a good approximation.

Eventually, revisit time is indeed a very important parameter in the resource selection process, but other parameters such as the type of data and their resolution are also essential; for earthquake or landslides, high-resolution optical sensors are the most effective and interferograms obtained with SAR measurement allow quantifying ground displacement. SAR data are also very useful for volcanic eruptions, flood monitoring and oil spills.

Landslides cause considerable human and economic losses around the world. They can be triggered by factors including intense or persistent rainfall, thawing of snow or ice, earthquakes and human interventions in the landscape. Landslide inventory maps are crucial for disaster relief and risk management because they record the current state of the landscape and provide evidence where future landslides may occur. The importance of landslide data, and its acquisition using Earth Observation techniques, is demonstrated here through two contrasting situations. The first comprises landslide inventory mapping undertaken to determine the current and future landslide hazard in the Caribbean, while the second was initiated as a rapid response in the midst of the emergency relief effort in Nepal.

The Caribbean is heavily affected by natural hazards with over 5 billion US$ in losses in the last 20 years (source: CRED database). In October 2010 Hurricane Tomas hit St Lucia resulting in seven deaths with 5,952 people severely affected. The cost of the damage was estimated at US$336.2 million (43.4% of GDP; ECLAC, 2011). The ESA-funded British Geological Survey (BGS) eoworld2 project utilised high-resolution, multi-temporal satellite imagery to deliver ‘risk information services’, including landslide inventories of St. Lucia for 2010-2014 based on image interpretation using a combination of high-resolution optical Pleiades and RapidEye satellite imagery (accessed through the ESA Third Party Mission scheme) and supplemented by reconnaissance field mapping. These inventories have been combined with information such as land cover/land use (also derived from eoworld2 satellite imagery) and geology in a World Bank CHARIM project led by ITC to generate a landslide susceptibility map detailing the future hazard posed by landslides. Frequent interaction with stakeholders such as planners and engineers has helped to ensure that the results are fit-for-purpose and will have sustainable application in the island states.

The 2015 earthquake sequence in Nepal, including the M_w 7.8 Gorkha earthquake of 25 April and the M_w 7.3 Dolakha aftershock of 12 May, triggered several thousand landslides. The BGS and Durham University responded to an urgent UK Government request to provide advice on the impact of co-seismic landslides. The results were disseminated freely and openly (e.g. via the ESA Geohazards Exploitation Platform and the Humanitarian Data Exchange) to ensure maximum utilisation of the results. Satellite imagery interpretation comprises the optimum method to produce timely inventories and maps of landslides e.g. https://www.disasterscharter.org/image/journal/article.jpg?img_id=157332&t=1431084341836. Imagery from the International Charter, Space and Major Disaster (and from other suppliers) was used by the UK team working with other agencies such as ICIMOD, USGS, NASA, CSA and MDA. The resulting inventories (>3500 landslides) have been used by relief organisations (e.g. World Food Programme, MapAction, UNOSAT) to help plan and deliver aid. This mapping effort helped identify the main areas affected by landsliding, but it was also hampered by a number of issues, including availability of suitable cloud-free imagery, difficulty in identifying landslides, and coordination between the different mapping teams, data providers and potential end-users. It was apparent that the hazard posed by landslides would persist due to monsoon rainfall triggering. Therefore the mapping effort, led by Earth Observation, continued into 2016 with funding from the UK Department for International Development (UK Aid) to produce monthly landslide inventories using imagery provided by ESA, CEOS Seismic Pilot, UNOSAT and CSA/MDA. Pleiades and WorldView II data were tasked monthly throughout the monsoon and manually interpreted. Radarsat-2 data were acquired every 24 days in the knowledge that cloud cover would be a particular challenge, with change detection algorithms used to identify landslides on the radar data.

The two case studies highlight the contribution of satellite imagery to real world applications of disaster risk reduction, both as part of a wider study (e.g. in the Caribbean) and as a rapid response to a natural disaster (e.g. Nepal). Lessons learnt include the importance of international coordination and the provision of the right type of derived information in a timely manner to potential end-users.

Interferometric Synthetic Aperture Radar (InSAR) is a proven geodetic imaging technique that makes use of remotely sensed radar imagery to map spatial patterns of ground surface movement and their temporal evolution. One application of the InSAR technique is to monitor human interactions with the landscape, such as the extraction of resources from the crust.

The increasing demand for gas in Australia has led to increased extraction of unconventional coal seam gas (CSG) reserves, particularly in the Surat Basin in south-east Queensland. Proved and Probable reserves of CSG now exceed 32,000 Petajoules, making the Surat Basin the largest onshore gas reserve in Australia. The geological target of CSG extraction in the Surat Basin is the Walloon subgroup of the Jurassic period, which is typically between 300 to 600 metres depth. Production of CSG from the Walloon subgroup began in 2006 and reserves are currently being extracted by several operators, with combined extraction exceeding 160 Petajoules in 2013-2014. Predictions of the magnitude of subsidence in the Surat Basin based on analytical poroelastic models and quoted CSG production rates indicate that total subsidence on the order of a decimetre may occur. In this contribution we will present new InSAR analysis of the Surat Basin using multi-sensor SAR imagery spanning the 2006-2015 time period.

Should patterns of subsidence be detected over the producing gas fields, we will use a geophysical inversion scheme to characterise the objective function between the spatial InSAR observations and predictions of a simple analytical model. Our methodology will make use of a Monte-Carlo sampling algorithm run on High Performance Computing architecture to efficiently sample the multi-dimensional parameter space. The homogenous poroelastic model we employ has dependence on the depth and thickness of the target geological unit as well as on the unit’s rock properties (porosity, Young’s Modulus, Poisson’s Ratio and Shear Modulus). Given that limited information about these properties is generally publically available for the Surat Basin, the geophysical inversion scheme will enable a sensitivity analysis to be conducted that will allow us to understand uncertainties and what parameters have the most significant impact on the system. This in turn will enable more accurate predictions of future subsidence using the poroelastic model.

In 2014, Geoscience Australia installed a regional geodetic network over a sub-region of the north-eastern Surat Basin in the vicinity of the towns of Dalby, Miles and Chinchilla in Queensland. The network covers a region of approximately 20,000 km² and consists of 40 co-located corner reflectors and survey marks. Ongoing SAR imaging of the corner reflectors and periodic campaign GNSS surveys on the survey marks will enable InSAR analysis to be combined with ground-based geodetic measurements and as a result, refine the geodetic reference datum in this region. Preliminary analysis of the persistent scatterer response of the corner reflector network will form a part of this contribution.

A dense archive of Interferometric-Wide-Swath (IWS) and Extra-Wide-Swath (EWS) Sentinel-1A images is currently being acquired over the region since the permanently deployed corner reflectors are being used as targets for ongoing geometric and radiometric calibration of the Sentinel-1A SAR sensor. InSAR analysis of this Sentinel-1A data will also form a part of this contribution.

Landslides including rock falls and debris flows expose great dangers to people’s lives and property. Many regions of the Three Gorges area are prone to landslides due to a weak stratigraphic layer. Furthermore, the large increase of water level since 2003 significantly affects the stability of slopes along Yangtze River. Moreover, the periodical fluctuations of water level and seasonal rainfall contribute to additional unstability. Therefore, slope stability monitoring in the Three Gorges area becomes a vital task for the safety of local residents and infrastructure.

Traditional displacement monitoring methods such as Global Positioning Systems (GPS) and geodetic leveling cannot be widely applied in inaccessible mountainous regions. For this reason, satellite SAR observation is sometimes the only available tool for slope stability monitoring. As such, it is important to make full use of the information within SAR images. Interferometric Synthetic Aperture Radar (InSAR) methods relying on the phase information from SAR images are widely applied in slow-moving landslide monitoring with millimeter-level precision. Advanced InSAR methods such as Persistent Scatters InSAR (PS-InSAR) and Small baselines Subsets (SBAS) have been proposed to overcome the effect of decorrelation and atmospheric turbulence. However, phase unwrapping also becomes a challenging task if large deformation occurs. In comparison with InSAR methods, the amplitude-based pixel offset tracking and point-like targets offset tracking (PTOT) are more suitable for large displacement measurement, having precision of more than 1/10 pixels if high correlations are guaranteed. Usually, high correlations can be obtained on point-like targets in vegetated areas, thus, PTOT was adopted for our study of fast moving landslide displacement extraction.

In comparison with X band and C band SAR datasets, L band SAR data with a longer wavelength is usually more suitable for displacement measurement in rural areas due to much less decorrelation effects. Thus, three paths of ALOS PALSAR data were used to map the slow-moving landslides along Yangtze River from Fengjie to Zigui. The SBAS method which is more suitable for rural scenarios was used. Considering the mountainous topography and seasonal rainfall, a linear topography-related model combined with spatial and temporal filters was used to estimate the atmospheric turbulence. It was determined that areas of more than 48 km² were identified as active. Most of the landslides identified were found to show linear movement.

34 TerraSAR-X StripMap (SM) images of approximately 3 m resolution and 36 TerraSAR-X High resolution Spotlight (HS) images of approximately 1 m resolution were collected for fast moving landslide monitoring. With these two datasets, line of sight (LOS) and azimuth displacements were obtained on Shuping landslide with PTOT method. Although both datasets were acquired from descending orbits, three dimensional displacements were obtained by making use of the different look angles applied in both datasets.

After we retrieved the time series deformation results for landslides, triggering factors of landslide and deformation mechanisms will be analyzed. Based on our initial results, water level decline and abundant rainfall were the most dangerous impact factors for the stability of landslides. Correlations between landslide movement and potential contributing factors are investigated in detail.

Natural hazards pose significant threats to cultural heritage worldwide. A number of monuments and sites that are inscribed on the World Heritage List (WHL) of UNESCO are affected and continuously impacted and weathered by hazards, including both shallow and deep geological and landscape processes (e.g. landslides, subsidence, ground collapses, coastal retreat, floods, earthquakes, volcanic activity), some of which are influenced or triggered by extreme meteorological events, or exacerbated by climate change and anthropogenic interaction. Only part of these endangered sites is inscribed on the list of World Heritage in Danger, which was defined in accordance with Article 11 of the Convention concerning the Protection of the World Cultural and Natural Heritage [UNESCO, 1972], and mainly includes sites threaten by armed conflicts, accelerated deterioration, calamities and cataclysms, and rapid urban or tourist development.

An innovative contribution towards the analysis of geo-hazards in areas of cultural heritage in Europe is being made by the project PROTHEGO: PROTection of European cultural HEritage from GeO-hazards. Funded in the framework of the Joint Programming Initiative on Cultural Heritage (JPICH) – Heritage Plus, the project is coordinated by the Italian Institute for Environmental Protection and Research (ISPRA), and carried out with the British Geological Survey of the Natural Environment Research Council (NERC), Cyprus University of Technology (CUT), University of Milano-Bicocca (UNIMIB) and the Geological and Mining Institute of Spain (IGME).

PROTHEGO is now identifying and mapping geohazards affecting the 400+ WHL sites in Europe. In the UK, only 1 out of 29 WHL sites has been recognised as in danger, whilst it is renown that many others are affected and could be threatened by geohazards. Examples are Maritime Greenwich, the Tower of London, the Palace of Westminster and Westminster Abbey in Greater London, where natural and anthropogenic land instability processes have been observed in 1992-2010 [Cigna et al. 2014]. By exploiting ground motion information from satellite Interferometric Synthetic Aperture Radar (InSAR) techniques, PROTHEGO will fill this knowledge gap and provide an enhanced understanding of geological threats to the European WHL sites. For those identified as critical, detailed geological interpretation, local-scale monitoring, advanced modelling and field surveying will be carried out to confirm the cause for the observed motions and identify to what extent the instability is affecting the endangered sites and to support risk management and site conservation.

References

-   UNESCO, 1972. Recommendation concerning the Protection at National Level, of the Cultural and Natural Heritage: http://whc.unesco.org/en/conventiontext/, accessed on: 14/10/2015.

-   Cigna, F., Jordan, H., Bateson, L., McCormack, H. & Roberts, C. 2014. Natural and anthropogenic geohazards in Greater London observed from geological and ERS-1/2 and ENVISAT Persistent Scatterers ground motion data: results from the EC FP7-SPACE PanGeo project. Pure and Applied Geophysics, 1-31. doi: 10.1007/s00024-014-0927-3

The British Geological Survey has a long track record of using satellite Interferometric Synthetic Aperture Radar (InSAR) ground motion data to better understand surface and subsurface geological processes.

In a number of study areas in the UK, geological hazards due to natural and anthropogenic causes have been identified by exploiting long-term archives of historical radar data acquired by ESA’s ERS-1/2 and ENVISAT satellites during the 1990s and 2000s. Ground motion patterns and trends characterising Greater London, Stoke-On-Trent, Newcastle and Durham, Welsh coalfields and peatlands, Lancashire and Greater Manchester have been analysed and interpreted [Banton et al. 2013, Jordan et al. 2013, Cigna et al. 2014b, 2014c, Bateson et al. 2015, BGS 2015]. These allowed the recognition and delineation of land processes including natural compaction of compressible deposits, groundwater level variations in shallow and deep aquifers, active and abandoned mining-related land instability and engineering works.

Two decade-long (1992-2010) snapshots of ground motion for these areas are crucial to improve our understanding of these dynamic processes and to establish a baseline understanding of the natural environment. This historical dataset, compared with more recent and future monitoring data, will provide evidence on whether and to what extent current exploitation of energy resources, land management practices and climate change are affecting land stability in the UK.

A comprehensive nationwide assessment of the feasibility of satellite InSAR methods to be used across Great Britain was also carried out [Cigna et al. 2014a]. This allowed BGS not only to quantify topographic and land cover constraints to undertake the baseline ground motion analysis, but also to predict InSAR performances and identify novel image processing solutions to overcome constraints in rural areas [Bateson et al. 2015]. This study provided a framework for generating ground motion products anywhere across the landmass, and has also paved the way for establishing a nationwide baseline based on processing of the ERS-1/2 and ENVISAT archives.

Parallel and cloud computing systems are also been tested to take on the required InSAR processing demand to create such a baseline by reducing image processing times. First tests with the Parallel-Small Baseline Subset (P-SBAS) method implemented into ESA’s G-POD platform focussed on the areas of Greater London and Newcastle using ERS-1/2 1992-2000 and ENVISAT 2002-2008 image archives [Cigna 2015], and revealed a decrease of the time demand from several days to only ~8 hours per image frame.

References

Banton, C., Bateson, L., McCormack, H., Holley, R., Watson, I., Burren, R., Lawrence, D. & Cigna, F. 2013. Monitoring post-closure large scale surface deformation in mining areas. Proc. Mine Closure 2013, 18-20 Sept 2013, Cornwall, UK. pp. 97-108.

Bateson, L., Cigna, F., Boon, D. & Sowter, A. 2015. The Application of the Intermittent SBAS (ISBAS) InSAR Method to the South Wales Coalfield, UK. Int. J. Applied EO and Geoinformation, 34, 249-257. doi: 10.1016/j.jag.2014.08.018

BGS 2015. Environmental baseline monitoring in Lancashire: ground surface change analysis: http://www.bgs.ac.uk/research/groundwater/shaleGas/monitoring/groundSurfChangeLancashire.html. Accessed on: 13/10/2015

Cigna, F. 2015. Getting ready for the generation of a nationwide ground motion product for Great Britain using SAR data stacks: feasibility, data volumes and perspectives. Proc. IEEE Int. Geoscience & Remote Sensing Symp. (IGARSS), 26-31 Jul 2015, Milan, Italy. pp. 1464-1467.

Cigna, F., Bateson, L.B., Jordan, C.J. & Dashwood, C. 2014a. Simulating SAR geometric distortions and predicting Persistent Scatterer densities for ERS-1/2 and ENVISAT C-band SAR and InSAR applications: Nationwide feasibility assessment to monitor the landmass of Great Britain with SAR imagery. Remote Sensing of Environment, 152, 441-466. doi: 10.1016/j.rse.2014.06.025

Cigna, F., Jordan, H., Bateson, L., McCormack, H. & Roberts, C. 2014b. Natural and anthropogenic geohazards in Greater London observed from geological and ERS-1/2 and ENVISAT Persistent Scatterers ground motion data: results from the EC FP7-SPACE PanGeo project. Pure and Applied Geophysics, 1-31. doi: 10.1007/s00024-014-0927-3

Cigna, F., Sowter, A., Jordan, C.J. & Rawlins, B.G. 2014c. Intermittent Small Baseline Subset (ISBAS) monitoring of land covers unfavourable for conventional C-band InSAR: proof-of-concept for peatland environments in North Wales, UK. Proc. SPIE 9243, SAR Image Analysis, Modeling, and Techniques XIV, 22 Oct 2014, Amsterdam, the Netherlands. ID 924305, 6 pp. doi: 10.1117/12.2067604

Jordan, H., Cigna, F. & Bateson, L. 2013. GeoHazard Description for Stoke-On-Trent, Version 1.0, 122 pp.: http://www.pangeoproject.eu/pdfs/english/stoke/Geohazard-Description-stoke.pdf. Accessed on: 13/10/2015

Interferometry is one of the most modern techniques of acquisition earth surface height information and it has a wide range of applications such as surface monitoring, volcanic hazards, seismic events etc. This work focus on exploitation of Sentinel-1 data for the monitoring of an active landslide in a village of Ilia Prefecture, Greece. Sentinel-1 mission provides timely, with global coverage, operational and easily accessible data with satisfactory spatial resolution. These advantages make Sentinel-1 data the best solution for the observation of landslides. In particular, at an initial level of landslide’s observation, interferometry contributed to Digital Surface Model (DSM) generation, utilizing the phase difference between the representations of the interferometric pairs. Thereby eighteen images before the manifestation of the landslide in March of 2015 were acquired during October of 2014 till March of 2015 for DSM extraction. Additionally, eighteen images following the landslide on 11 March 2015, during March of 2015 to May of 2015 were subjected to interferometric process yielding a second DSM. It is worth mentioning that during the interferometric process the affect of the baseline, the influence of different tracks (ascending or descending) and the effect of the stronger coherence between the interferometric pairs on the final products of the procedure were examined, in both cases. The accuracy of the interferometric DSMs was assessed using a respective high precision DSM from the Greek Cadastral as well as ground control points collected with a Differential GPS. Other points of certified elevation have been used and statistical values such as 2D RMSE, correlation and the percentile values were computed. Finally, the two DSMs were compared to each other in order to identify height differences and ground subsidence and map the landslide zone. The results are presented in the current paper.  

In order to measure and understand the relationship between water and landslides in tropical environment, we studied the correlation between annual rainfall and ‘Mare à Poule d’Eau‘landslide located in the Salazie erosion basin in the Island of La Reunion. This island is characterized by annual cumulative rainfall locally exceeding 7000cm. Among the three major techniques used to measure landslide displacement (GNSS measurements, InSAR techniques and image correlation), we chose correlation of SPOT 5 images acquired between 2002 and 2014 which gives a synoptic view of landslide displacements on a period of several months to years combined with GNSS measurements. The combination of these two techniques avoids the smoothing of the temporal signal produced by image correlation only and preserves a synoptic view of the landslide displacement. We used 10 panchromatic SPOT 5 images acquired with similar incidence angles and ortho-rectified by CNES (French spatial agency) at a spatial resolution of 2.5m. The dates of these images are 02/2002, 07/2003, 05/2004, 06/2005, 06/2006, 05/2008, 02/2010, 08/2012, 01/2014, 05/2014. The correlation, realized with a precision of 0.1pixel, was realized using MicMac open source software (IGN – French geographical institute). Every pixel was correlated with a correlation window of 9 pixels. The GNSS data were acquired by the BRGM since 2003 at a daily period. The rainfall data have been acquired by Meteo France on 7 stations located among the Salazie erosion basin. The first results of this study shows that the maximum displacement of the landslide measured by image correlation is 1.70 m.y^-1. There is a large spatial variability among the landslide: the minimal displacement is 0.4m.y^-1. These values matche well with GNSS data. There is not a good correlation between averaged yearly rainfall values and landslide displacement; however there is a significant correlation between intense meteorological events and displacement. In 2007, the tropical cyclone GAMEDE (02/2007) produced intense rainfalls. The velocity of the landslide was the largest of this study (1.70m.y^-1). When the rainfall was less intense (between 10/2002 and 07/2003, when no intense event happened) the maximum velocity decreased to 0.50m.y^-1. This dataset shows a good correlation between punctual intense rainfalls and landslide displacement.

This work will be focused on the potentiality of Sentinel-1 Persistent Scatterer Interferometry (PSI) and Differential SAR Interferometry (DInSAR) to perform landslide inventories and landslide monitoring. It will be based on the Interferometric Wide Swath SAR data of the Sentinel-1 mission. These data, in most of the areas covered by the Sentinel-1 mission, started to be acquired, in a regular way, in the first half of 2015. They offer interesting characteristics and open new opportunities for the inventory and monitoring of landslides. The first characteristic is the relatively short repeat cycle (12 days). This represents a remarkable improvement with respect to previous C-band missions (e.g. the 35 days of ERS and Envisat), which results in an increase of the coherence of the interferometric Sentinel-1 data. This aspect is very important and can be fully exploited in the applications related to landslide deformation measurement and monitoring. The second important characteristic is the wide area coverage of Sentinel-1 imagery, which in the Interferometric Wide Swath SAR mode is of 250 by 250 km with a single image frame. This aspect opens the possibility to perform deformation monitoring over wide areas, covering different types of deformation phenomena, including landslides. This paper will illustrate the first experiences of the authors in landslide inventory and monitoring by using Sentinel-1 SAR imagery. In particular, it will describe the tools developed to perform landslide inventories and landslide monitoring: they include both simple Differential SAR Interferometry (DInSAR) tools and Persistent Scatterer Interferometry (PSI) tools, which allow us deriving deformation maps and time series over the selected area of interest. Taking advantage of both methods, the effectiveness of these tools will be illustrated by analysing different study areas, mainly located in Italy and in Spain.

This work will be focused on deformation measurement and monitoring using Sentinel-1 SAR imagery. It will firstly address the data processing and analysis procedure used by the authors. This include both Persistent Scatterer Interferometry (PSI) and Differential SAR Interferometry (DInSAR) tools to analyse both short and large stacks of SAR images. The work will be then focused on the characteristics of the main products derived by using Sentinel-1 DInSAR and PSI. This will include the analysis over different types of land use area, e.g. urban, peri-urban and non-urban areas. Sentinel-1 deformation monitoring will be compared with the monitoring based on data from pre-existing missions, e.g. ERS, Envisat, TerraSAR-X, CosmoSkyMed, etc. The comparison will concern different study areas, mainly located in Italy and in Spain.

The Trondheim city is situated in central Norway on the Trondheim Fjord, 70 km away from the open sea. The Gulf Stream causes a moderate climate during the winter period. Climate scenarios for Trondheim predict an increase in temperature of 2.5 °C and annual precipitation of 10-20% in 2100. Because of its special location, climate change and rapid urbanization, Trondheim is an interesting area for ground deformation analysis. Recently, a massive landslide took place just south of Trondheim leading to evacuation order of people from a rural area. Unusual warm winter weather as well as violent storm across Scandinavia, which dumped massive amounts of rain on Trondheim, may have contributed to the landslide.

Ground deformation is the most evident expression of landslide collapse. Monitoring of ground movements can be effectively used to map and characterize active sliding masses before collapse. In the regions like Trondheim, risk managers and local administrations need this information to plan effective prevention measures and implement warning systems. Even if ground deformation cannot be prevented or stopped, it must be accounted for in new construction planning.

Identification and monitoring of ground deformation can be accomplished using a number of surveying techniques. These techniques have been evolved using Global Navigation Satellite Systems (GNSS), but for large areas, the use of GNSS and leveling are time consuming, expensive and laborious.

In contrast to the surveying techniques that rely on point measurements at the Earth surface, Interferometric Synthetic Aperture Radar (InSAR) overcomes many practical limitations and readily provides high-resolution measurements of surface displacement at sub-cm accuracy over relatively large areas.

This study presents deformation analysis of Trondheim city using observations from InSAR. 31 SAR images acquired in a descending track from TerraSAR-X in stripmap mode are used to assess deformations during 2012–2014. The interferometric processing was achieved using the repeat-pass technique implemented in DORIS software. The Digital Terrain Model was provided by the Norwegian Mapping Authority with a resolution of 10 m for topography phase correction and geocoding. The interferometric dataset includes 86 differential interferograms with minimal spatial, temporal and Doppler baselines, which are processed using InSAR time series technique of small baseline subset approach, implemented in StaMPS software. The processing in StaMPS is done in three main steps. They are (1) generation of small baseline interferograms, (2) selection of coherent pixels, and (3) phase unwrapping and least squares inversion. The results show a clear indication of deformations, especially on the harbour area of city in 2012–2014. A correlation analysis of InSAR derived deformation with metrological information are also investigated.

The Black Sea coast of the Great Caucasus has always been a region of high landslide risk due to widely spread clays and marls seasonally saturated by abundant rainfalls. In recent years landslide risk assessment has become vital because of strongly increased human-induced impact dealt with construction of the Sochi-2014 Olympic Games facilities which triggered landsliding. Thus, identification and monitoring of slowly moving landslides is of particular interest. Our investigations demonstrated efficiency of INSAR application to locate and monitor activity of slowly moving landslides there. We studied the landslide activity in the area of the Bolshoy Sochi (Big Sochi) using the StaMPS software. We incorporated radar images from the satellites with different wavelengths from ascending and descending orbits: 13 Envisat images (track 35D, descending orbit, 14.08.2003-09.02.2006), 12 Envisat images (track 35D, descending orbit, 2007.11.01-08.07.2010), 13 Envisat images (track 35D, descending orbit, 29.11.2010-23.03.2012), 11 Envisat images  (track 85A, ascending orbit, 01.08.2004-01.02.2009), 17 ALOS images (track 588A, ascending orbit, 22.01.2007-17.09.2010), 17 TerraSAR-X images (track 54A, ascending orbit, 24.12.2011-13.09.2012).

We present results of processing of all these data sets for the Moldovka landslide which is a typical representative of slowly moving landslides on the Black Sea coast of the Northern Caucasus being of particular interest as it is situated  nearby theSochi airport.  The Moldovka landslide is an ancient landslide in the Chvizhepse folded area. Displacements of this landslide became visible  and were fixed by ground methods after heavy rainfalls in winter-spring 2012. The boundaries of the Moldovka landslide were clearly determined applying the Stamps technique to the above mentioned  data sets. Analyzing the time series it was found out that displacements of the landslide took place starting at least in 2004. Analyzing  all the deformation maps and time series we determined the LOS displacement rates in 2004-2013 being about 20-30 mm/Y with several accelerations induced by rainfalls. Maximum estimates of displacement rates in the down the slope direction were revealed in the first months of 2012 being about 150 mm/Y .

For atmospheric corrections we used the TRAIN software which appeared to be the most effective for the TSX data. Using the TRAIN software improved the results of the processing of the TSX data - boundaries of the landslide on the deformation map became more clearly seen and several new slowly moving zones became visible. 

Sinkholes commonly form by subsurface dissolution cavities that collapse after the overlying layers become mechanically unsupported. Sinkholes along the Dead Sea shorelines are preceded by, associated with, and followed by gradual surface subsidence that accompanies the cavities' growth. We exploit satellite radar interferometry (InSAR) to resolve temporal and spatial relationships between gradual subsidence and sinkhole collapse. The geometry of the deflating cavity roof is determined by elastic inverse modeling of the surface displacements. A Coulomb failure stress criterion is applied to calculate the stress field induced by the deflating cavity at the ground surface. We find that the induced stress field favors generation of sinkholes at the perimeters of the subsiding areas rather than at their centers, in agreement with field observations, providing important information for sinkhole hazard assessment. Further, our analysis suggests that short-term deformation in consolidated gravel layers at shallow depths could be approximated by simple elastic modeling.

We present precise deformation velocity maps for the two years period from September 2011 to July 2013 of the northern Taiwan area, Taipei, by using persistent scatterer interferometry (PSI) technique for processing 18 high resolution X-band synthetic aperture radar (SAR) images archived from COSMO-SkyMed (CSK) constellation. According to the result, the highest subsidence rates are found in Luzou and Wuku area in which the rate is about 15 mm/yr and 10 mm/yr respectively in the whole dataset. However, dramatic change from serve subsidence to uplift in surface deformation was revealed in the Taipei Basin in two different time spans: 2011/09–2012/09 and 2012/09–2013/07. This result shows good agreement with robust continuous GPS measurement and precise leveling survey data across the central Taipei Basin. Moreover, it also represents high correlation with groundwater table. This high correlation indicated that one meter groundwater level change could induce about 8 and 16 mm surface deformation change in Luzou and Wuku area respectively, which is about eight times faster the long-term tectonic deformation rate in this area. Thus, to access the activity of the Shanchiao Fault, it is important to discriminate tectonic movement from anthropogenic or seasonal effect in Taipei Basin to better understand the geohazards and mitigation in the Taipei metropolitan area.

Pipeline networks for gas and water are the lifelines of our society. Most pipelines are buried, which impedes direct monitoring of the structural reliability of the networks. The behavior of the soil surrounding the pipes may impose hazardous loads or deformations threatening the integrity of the network or even cause failure of pipes. This is especially true for subsidence due to shallow compaction.

The paper describes the outline and first results from a Dutch joint research project with partners from applied research, consultancy, gas and drinking water suppliers.  In this project the deformation of the terrain surface is measured by InSAR and translated to a risk map to assess the structural reliability of all pipe segments of the network. For this risk mapping, heterogenous data sources are integrated using a probabilistic approach. Examples of these data sources are soil layering, information on civil works, and physical and structural properties of the pipelines. The mapping involves geological, geotechnical and structural models. One of the main aims of the project is to construct a nation-wide risk map that is expected to play a key role in future risk-based asset management strategies.
InSAR is an important and relatively new data source for reducing the uncertainties of soil displacements around pipelines. The paper will focus on the wide-area InSAR processing of very high resolution imagery and its application for asset management of pipeline networks.

The western margin of South America is characterized by an eastward subduction of the Nazca plate underneath the South American plate with moderate to high convergence rates. Large megathrust earthquakes with short recurrence times are typical in this tectonic setting. On September 16, 2015, the central part of the convergent Chilean continental margin was struck by the M_w =8.3 Illapel earthquake. After the Pisagua event in 2014 (M_w =8.2) and the Maule event in 2010 (M_w =8.8)this earthquake constitutes the third largest earthquake above magnitude 8.0 within the past five years.

In this study, we present co- and postseismic ground displacements for the Illapelearthquake observed by InSAR and GPS.Theco-seimicSentinelinterferogram in descending orbit covers the time period24 days before and one day after the event.In order to obtain thepost-seismic ground displacements we aim to process Sentinel and also ALOS-2data.In addition, we presentGPSdataofseven continuous stationsin a radius of 200km around the epicenter.

The rupture nucleated near the coast (31.5°S, 71.7°W) in a depth of about 21 km, close to the city of Illapel and propagated updip to the north. The ground deformation is approximately circular shaped with a diameter of about 100 km centered beyond the middle slope of the forearc. Peak slip reaches about 7m. Both, theInSAR and GPSdata are used toderive a first co-seismic rupture model.

Almost all segments along this highly active plate boundary broke within the last decades. The last earthquake with comparable along-strike extent in the Illapel region itself took place in 1943. The accumulated slip deficit matches the approximate maximum slip, which could indicate that the Illapel area is a candidate for characteristic earthquakes with short recurrence times, especially because the 1943 event was preceded by another event with similar rupture mechanisms in 1880. The degree of coupling in this area along the coast is generally very high and the hypocenter of the Illapel earthquake is colocated with a local peak in locking. The area south of the main rupture is thought to be partially interseismic creeping, whereas to the north, locking gradually decreases. Most likely, the rupture terminated against this zone of reduced locking. Altough the Illapel earthquake could relieve much of the accumulated stress in the local segment, the level of observational attention remains high - especially because the segment to the north of the Illapel area last ruptured in 1922. Our analysis contributes to the ongoing research in this region of increased seismic hazard potential.

In this paper a new approach to creation the short-term forecasting systems of river floods is introduced. The proposed approach extends the traditional separate methods based on satellite monitoring or modeling of river physical processes, by integration of different models and technologies such as satellite image and hydrological data joint processing, digital maps of a relief and river terrain, crowdsourcing, hydrological models and geo-simulation, inundation visualization, and duly warning of stakeholders.

The real-time flood monitoring and forecasting system which was created within the described approach is shown on the fig.1.

Fig. 1. Floods short-term forecasting system

Forecasting of flood areas and depths is performed on a time interval of 12 to 48 hours to be able to perform the necessary steps to alert and evacuate the population. Forecast results are available as Web-services.

A choice of software architecture type for the practical implementation of the multidisciplinary approach is one of the most fundamental issues. To solve this problem an Open Source Code and Service Oriented Architecture were used and that allowed to integrate different models and technologies into a single software complex. A key element of the proposed architecture is an intelligent interface for adaptation and selection of the model to forecast in each specific situation.

Using of remote sensing data in the flood forecasting allows to compensate the lack of ground-based data and to perform additional adaptation of the models. There are the following areas of remote sensing data application in short-term forecasting system constructing:

1). Actualization of cartographic information on modern infrastructure in river valleys.

2). Obtaining information on the river valleys relief.

3). Calibration and verification of hydrodynamic models on the base of the information on the flooding areas outlines obtained by satellite images.

It should be noted that the most reliable modeling results can be achieved only by a joint, integrated processing of various data received both in-situ and space measurements.

The developed system has been successfully tested in forecasting of floods on the riverNorthern Dvina (Russia) and on the river Daugava (Latvia) during the implementation of project ESTLATRUS/2.1/ELRI-184/2011/14 “Integrated Intelligent Platform for Monitoring the Cross-border Natural-Technological Systems”

In particular, a LISFLOOD hydrological model was adapted to simulate water flows in the Daugava river bed and within the channel network by integrating the digital map of the relief of the specified area and obtained hydrological characteristics of the river. The calibration of the model has been performed in two steps: by using the images received from the satellite RadarSat to precise the current state of the river channel and by using crowdsourcing technology.

Flood forecasting results were automatically published at the Geo-Portal for prior notification about emergency situation. The web service provides possibilities of viewing layers with flood contours from the beginning of the modelling process up to 12-hour forecast.

It should be noted the designed system after installation does not require any interaction with the operator. System automatically requests data from hydrological stations, processes them and sends the results to the database server. This operation is done once an hour, and the user receives updated flood forecast each hour by web service.

It is necessary to mention that the inundation areas were evaluated by the model with the real accounting to the features of water movement, not by simply raising the water level as it is done usually in well-known geoapplications. So the good accuracy was achieved. The performed real-time experiments allowed achieving about 90-95% confidence in floods forecasts regarding significant objects which were actually inundated later on.

THEME: HAZARDS

PRESENTATION: ORAL

Sentinel 1A Time series SAR Interferometry monitoring of Santorini volcano during post unrest period (2014-2015)

Christos Bountzouklis ¹, George Benekos², Issaak Parcharidis ², Pierre Briole³

¹ Lund University, Department of Physical Geography and Ecosystem Science, chris-boun@hotmail.com

² Harokopio University of Athens, Dep. of Geography, benekosgis@gmail.com; parchar@hua.gr

³ Ecole Normale Supérieure, Laboratoire de Géologie, Paris, briole@ens.fr

Abstract

Mitigation of volcanic risk is feasible and thus reducing damages can be achieved by knowing in detail about structure and history of the volcanoes, eruption mechanisms, unrest behavior etc. The identification, analysis and evaluation of risk comprise the basis for timely, well oriented and essential disaster management. It is clear that reducing risk for volcanoes requires many steps (hazard and vulnerability assessment, exposure, coping capacity) to be addressed. Long-term hazard assessment presents the basic tool for the behavior of a volcano especially in the case of dormant volcanoes due to the lack of plethora historical data.

Ground deformation monitoring is one of the main geoindicators that should be considered to assess volcanic hazard. Satellite Earth Observation data are used for different facets of risk management concerning volcanic hazards. Space-borne SAR interferometry has been used continuously since 1992 to measure or study the temporal evolution of surface deformation in volcanic areas in conjunction with ground-based geodetic measurements.

Santorini Volcanic Complex the period 1992–2010 is characterize by the gradual deflation signal over Nea Kameni volcano however, at the beginning of 2011 the volcano showed signs of unrest with increased microseismic activity and significant ground uplift. A gradual decrease of inflation rates within the first quarter of 2012 was confirmed from subsequent observations.

The goal of this study was to identify eventual surface deformation in the post unrest period (10-2014/3-2015) over Nea Kameni volcanic center and validate interferometric results with ground base geodetic observations.. To address this issue a hybrid method of multitemporal SAR interferometry has been used as method using Copernicus satellite Sentinel 1A SAR scenes. A set of 16 SLC Sentinel 1A ascending scenes covering the period March to October of 2015, frame 29 has been used. Additional repeated pass interferometry was applied (Figure 1) using suitable interferometric pairs by choosing scenes from the above dataset. The fact that the volcano exhibits high coherence independently of the time distance of the two images (master and slave) gave us this ability. Both Gamma and Sentinel toolbox have been used for multitemporal and conventional interferometry.

Interferometric results show that Nea Kammeni Volcano shows a relative uniform surface deformation with low uplifting rates. Interferometric results are compared with existing GPS measurements for the same period

The Pacaya volcano complex, located 25 km south of Guatemala City, is intermittently active and deforming. Knowing the edifice has collapsed in the past, most recently between 0.6-1.6 ka, it is crucial to better understand the long-term flank deformation and potential associated magmatic processes at Pacaya for hazard mitigation purposes. New Interferometric Synthetic Aperture Radar (InSAR) measurements from the RADARSAT-2 C-band satellite show deformation associated with the last 2010 and 2014 eruptions. Persistent scatterer and small baseline methods (e.g., StaMPS) are also applied to produce InSAR time series using RADARSAT-2 ascending and descending orbits datasets covering Pacaya from 2010 to 2015. This approach is effective for detecting spatially and temporally complicated volcanic deformation and can mitigate topographic and atmospheric effects, commonly encountered when applying InSAR in tropical areas such as Guatemala. InSAR measurements, combined with observations from high-resolution optical satellite imagery, are used to identify and interpret surface deformation at Pacaya. InSAR-derived products can indeed be used to model sub-surface processes causing the observed deformation. The most significant deformation events occurred along the southern flank of the MacKenney cone in 2010 and 2014, as well as on the northeastern flank in 2014.

Fire is an ubiquitous source of vegetation disturbance and a source of greenhouse gases and particulate matter emissions. Through the removal of vegetation and deposition of char, fire also changes surface albedo and affects the energy balance of land surfaces. Under a context of climate change, the meteorological conditions that enhance fire activity, such as higher temperatures and long dry spells, are predicted to be more prevalent in many areas, making fire monitoring at global scales a necessity. This can only be realistically done with monitoring systems based on space-borne sensors. A number of global burned area products derived from optical data are available, but these provide only an estimation of burned area and time of burning, with no description of the impact of fire on the vegetation. Moreover, such binary burned area maps, derived from moderate resolution data, either record “burn” or “no burn”, with no concept of partially burned pixels.

In this contribution, we introduce and demonstrate a simple spectral model of short-term changes of reflectance due to fire, based on (Roy and Landmann 2005). It is assumed that a fire removes part of the vegetation, and so the post-fire scene is a combination of a part of the scene that has not been affected by the fire, and one that has, and that is basically characterised by a spectral behaviour which is a combination char, ash and exposed soil.  Using spectral databases, we develop a simple two parameter model of burned material. The inversion of the model results in the estimation of 3 parameters: two parameters that describe the burn signal, a0 and a1 (that broadly correspond with a dark char and a dry soil spectral behaviour respectively), and fcc, the combined of effect of the fraction of the pixel affected by the fire (f) and the (radiometric) combustion completeness, cc. To invert the model, pre- and post-fire reflectance measurements with identical illumination/acquisition geometries are required. Inversion of the model is then straightforward through a linear least squares approach. Note that the three parameters are spectral invariants, that is, they can be retrieved from sensors with different spectral characteristics. In the case of wide swath sensors like MODIS, angular normalisation of the directional reflectance needs to be carried out before the model can be applied, but this requirement is met for sensors like Landsat 8, SPOT4/5 and Sentinel2, which tend to have a constant acquisition geometry.

We demonstrate the use of the proposed method to Landsat 8 and Spot5 data acquired during the Take5 campaign. We show that the retrieved parameters are consistent with expectations of burned signals measured in situ, and compare the retrievals from Landsat with SPOT. We also apply the proposed method to the global MODIS archive, and show parameter distributions derived from different burned area products, as well as show comparisons with the medium resolution estimates of fcc, a0and a1. The temporal evolution of fccshows a clear seasonal variation, whereas the spectral components appear to be locally quite stable over time. We note that this approach can be used to evaluate the performance of burned area products, as fcc can be taken to be a measure of the change signal, and thus, an indicator of the “smallest” detectable change by a particular product/algorithm.

Acknowledgements

Funding for this work was provided under NERC NCEO.

References

Roy, D. P., and T. Landmann. 2005. “Characterizing the Surface Heterogeneity of Fire Effects Using Multi-Temporal Reflective Wavelength Data.” International Journal of Remote Sensing 26 (19). Taylor & Francis: 4197–4218.

The Himalayan range is a high-risk area; natural hazards can cause damages, destructions, injuries and deaths at any time (SDC, 2009). Landslides, rockfalls, floods and earthquakes cause numerous deaths each year and destroy villages, access roads and other important infrastructures. As the Himalayas is located in a tectonically active zone, landslides and rockfalls triggered the earthquakes represent one of the most dangerous hazards. More than 3000 landslide, for example, were triggered by the earthquake of April and May 2015 (BGS, 2015).

The hazard assessment represents one of the most important action in the disaster risk management strategy. Because a lack of diffuse field data (i.e. diffuse landslide mapping), landslide hazard map are not available for vast regions of the Himalayas.

Based on satellite EO data valuable information to perform landslide hazard map for two areas in Nepal (Lukla region) and Bhutan (Chomolhari/Paro region) were collected. Landslide hazard maps were performed based on geomorphological and quantitative (InSAR analysis) approach through a compilation of a landslide inventory map at regional scale by means of satellite photo-interpretation and the assessment of the state of activity of mapped phenomena based on surface displacement rates quantified from differential SAR interferometry. Regarding Lukla region, the 7.5% of the area (total of 724 km²) is affected by landslides. The 40% of mapped landslides were classified as active. The 6% of the Chomolhari area (total of 620 km²) is affected by landslides; 45% of them were classified as active. Rockslide represents for both region the most of the mapped landslide

The landside hazard assessment over large regions based on EO products represents an important aspect for the disaster risk reduction not only in the whole Himalayan region but also in other mountain areas in absence of detailed landslide inventory maps.

Monitoring urban infrastructure using InSAR requires the use of SAR data at the highest possible spatial and temporal resolution. In this work, we demonstrate how using the highest resolution imagery for monitoring civil engineering installations of the size of a few meters yields a dramatic increase in information. In specific, we focus on the Hessigheim lock structure situated on the river Neckar, north of Stuttgart, Germany (Figure 1). As a consequence of subsurface leaching processes, these lock structures are  being extensively monitored since 1970, combining groundwater observations, in situ geodetic and extensometer measurements, all performed at least at yearly intervals.  

To increase spatial and temporal resolution, we monitor these locks using SAR data available since 1992. We show the results obtained using the full spectrum of available SAR resolutions -- ERS, Envisat, TerraSAR-X StripMap and SpotLight modes (Figure 2) -- comparing them with the available in situ measurements. We further discuss the methodology, challenges and benefits of using high resolution SpotLight data, both in cases of point and distributed scatterers for this application.

This brings the end-users a more complete coverage of the objects of interest, increasing the likelihood of discovering instabilities early on, improving the ability to isolate actual signal from surrounding scatterers and finally improving the localization of the signal. The additional measurement density is shown to increase the usability of InSAR on small-sized objects.

In future, we plan to perform similar analyses using Sentinel-1 imagery. We will compare the added benefit of the coherence obtained from a 6- to 12-day repeat cycle in offsetting the effect of larger pixel sizes, and demonstrate its usability for infrastructure monitoring.

We acknowledge ESA for the ERS, Envisat and Sentinel-1 imagery, and Airbus for the partial donation of SpotLight imagery.

On 24 October 2002, a M_w 6.2 earthquake occurred in the central part of the Lake Kivu basin, western branch of the East African Rift. This is the largest event recorded in the Lake Kivu area since 1900. An integrated analysis of radar interferometry (InSAR), seismic and geological data, demonstrates that the earthquake occurred due to normal-slip motion on a major pre-existing east-dipping rift border fault. A Coulomb stress analysis suggests that diking events, such as the January 2002 dike intrusion, could promote faulting on the western border faults of the rift in the central part of Lake Kivu. We thus interpret that dike-induced stress changes can cause moderate to large-magnitude earthquakes on major border faults during continental rifting. Continental extension processes appear complex in the Lake Kivu basin, requiring the use of a hybrid model of strain accommodation and partitioning in the East African Rift.

In the Czech republic, as well as in other areas, the bridges are of various age and various condition. It seems that more problems can be found on recently-built bridges, but this is due to the fact that the old ones are usually not monitored in such a detail or not at all.

The aim of monitoring bridges is to find possible deformations before the deformations get large/significant to endanger the people and/or traffic.

However, there are more deformation causes of the bridges, and most of them are reversible, such as thermal dilation, bending with load etc. The problem is to recognize these components from the irreversible deformations, and one needs a large dataset, covering at least 18 month data (to separate the thermal dilation component from the linear deformation component).

The direction of the irreversible movements can be arbitrary, and due to the easily accessible Sentinel-1 data, we try to monitor the bridge from various directions.

For improving the accuracy we plan to make experiment with corner reflectors to enhance the intensity, and therefore to lower the standard deviation of the movement, however this experiment cannot be evaluated yet in May 2016.

Our project monitors few bridges in the Czech republic, of various length and age.

Earth observations by means of space-born geoscientific instruments provide new methods and comprehensive data systems being available to reduce the risks for natural disasters such as earthquakes and powerful tropical cyclones - hurricanes (typhoons). Although the long-range forerunners of strong earthquakes were preliminary described decades ago (Nersesov and Latynina, 1992) the first reliable observational results have been obtained owing to our ground-based and satellite data joint complexions (Dubrov M.N. et al, 2010). The phenomenon explanation, as an involvement of hurricanes in the earthquake triggering, is becoming to be understood together with an evidence that hurricane transient features are influenced by the solid Earth activity during earthquake preparing processes. Otherwise, powerful events in atmosphere having large horizontal scale are necessary to interact not only with the Earth surface, but also with higher layers of the thermosphere and near-Earth environment. In this paper, we present the results of comparison of geophysical field variations and seismic activity of the Earth, which show the correlation between lithosphere-atmosphere interactive disturbances, tropical cyclonic activity in the World Ocean, and seismic processes in the solid Earth. The found correlation can be interpreted as appearing or increase in amplitude the wide-band oscillations disturbed by typhoons and hurricanes which together with quasi-static pressure loading on the ocean bottom provoke powerful earthquakes through the triggering effect. The spatial and temporal tracks of tropical cyclones are coupled with place and time of occurring earthquake.

The examples of geophysical disturbances having near-Earth environment origin are presented and discussed too. The comet C/2011 L4 (PanSTARRS) transit on March 2013 has been recorded by spatially distributed and simultaneously operating ground-based instruments. We used the system of wide-band geophysical laser strainmeters, pendulum gravimeters, and tiltmeters in our observations. Data from four measuring sites in the East Europe are presented. The distance between separate instruments varies from a few hundred kilometers within local site installations, and up to thousands kilometers for a different remote sites and observatories. The datasets of geostationary satellites GOES13 and GOES15 of Space Weather Prediction Center (NOAA USA, 1998-2015) have been used in complementation to the ground-based instrument recordings. Earthquake and hurricane coupling effect during the recent strong earthquakes (Nepal, 25.04.2015, and Chile, 16.09.2015) are presented and analyzed.

Investigation of the observed phenomena and deployment the detailed interaction mechanisms of the atmosphere, lithosphere, and other adjacent geospheres would give a chance to find the regularity and origins of such natural disasters as earthquakes and hurricanes. The ground-based laser interferometer and gravity-inertial techniques being supplemented by satellite observational systems can be considered as promising methods for common earthquake and hurricane monitoring and prediction.

References:
Nersesov I. L. and Latynina L. A., 1992, Tectonophysics v.202, N2-4, pp. 221-225
Dubrov M.N. et al, Living Planet Symposium 2010, ESA Special Publications, SP-686, CDROM, 2010
NOAA USA, 2015, Space Weather Prediction Center (available at www.swpc.noaa.gov/index.html)

The aim of this work is to perform analysis of vertical displacements in the Upper Silesian Coal Basin (USCB) using different Synthetic Aperture Radar (SAR) systems mounted on satellites ENVISAT, ERS1, ERS2, ALOS, and TerraSAR‑X in relation with land use map and vegetation index from Landsat satellites images. Vertical displacements are detected using two satellite radar interferometry methods: Differential Interferometry Synthetic Aperture Radar (DInSAR) and Persistent Scatterer Interferometric Synthetic Aperture Radar (PSInSAR). Upper Silesian Coal Basin is a highly urbanized industrial area where about 40 active underground mines are located. Vegetated areas also occur in the USCB which causes frequent decorrelation of the SAR systems signal, rendering it unusable for obtaining vertical displacements map. Longer wave (L-band) used in the PALSAR system mounted on the ALOS satellite allows to detect deformations in vegetated areas, while C‑band wave used in SAR systems from ENVISAT, ERS1 and ERS2 satellites or X-band wave used by system on TerraSAR-X satellite usually do not detect deformations in such areas. Areas are selected for analysis on the basis of the vegetation index calculated from Landsat images and land use map from the OpenStreetMap Foundation. Vegetation index is calculated to show differences according to seasons of the year. Present study shows capabilities of different SAR systems for detection of vertical displacements in the USCB in different areas (mainly vegetated, but also urbanized, rocky etc.). Influence of temperature, humidity, and precipitation (obtained from the stationary measuring station and from Landsat images) on vegetation index is also presented.

Data were provided by: 1) European Space Agency (ENVISAT, ERS1, ERS2), 2) Japan Aerospace Exploration Agency (ALOS), 3) German Aerospace Center [DLR] (TerraSAR-X), 4) OpenStreetMap Foundation (land use map), 5) National Aeronautics and Space Administration (Landsat), 6) United States Geological Survey (Landsat), 7) National Oceanic and Atmospheric Administration (temperature, humidity and precipitation), 8) Institute of Meteorology and Water Management  National Research Institute (IMGW-PIB) (temperature, humidity and precipitation).

The aim of this work is to show the relation between water factors (precipitation, humidity, and soil moisture) and vertical deformations detection in the Upper Silesian Coal Basin (USCB) when using different Synthetic Aperture Radar (SAR) systems. SAR systems mounted on satellites ENVISAT, ERS1, ERS2, ALOS, and TerraSAR-X are used for the purpose of this analysis. Vertical displacements are detected using two satellite radar interferometry methods: Differential Interferometry Synthetic Aperture Radar (DInSAR) and Persistent Scatterer Interferometric Synthetic Aperture Radar (PSInSAR). Precipitation, humidity, and temperature are obtained from Landsat images and stationary measurements stations. Separate analysis is performed according to seasons of the year (on the basis of optical Landsat images and time of acquisition of radar images). The observations are compared with images taken in the same season for different SAR systems and years. Analysis is performed separately for each season as, for example, in winter the area is covered with snow which strongly influences the detection of vertical displacements, compared to other seasons, when no snow coverage is observed. The soil moisture (dependent on humidity, precipitation, temperature, and the type of soil or terrain analysed) influences backscattered signal of radar systems and vertical displacements processes. Terrain type map is provided by the OpenStreetMap Foundation. Relations between water factors and vertical deformations in USCB are showed in the present study. Analysis was performed on the basis of data  from different sensors (SAR systems and Landsat), complementing each other.

Data were provided by: 1) European Space Agency (ENVISAT, ERS1, ERS2), 2) Japan Aerospace Exploration Agency (ALOS), 3) German Aerospace Center [DLR] (TerraSAR-X), 4) OpenStreetMap Foundation (land use and terrain map), 5) National Aeronautics and Space Administration (Landsat), 6) United States Geological Survey (Landsat), 7) National Oceanic and Atmospheric Administration (temperature, humidity, and precipitation), 8) Institute of Meteorology and Water Management National Research Institute (IMGW-PIB) (temperature, humidity and precipitation).

This study aims to evaluate ALOS-2 and SENTINEL-1 wide swath observation capability to investigate subduction processes along the entire Mexican subduction zone. This highly seismogenic subduction is also the place of large Slow Slip Events (SSE). The role and significance of those SSE in the subduction earthquake cycle is still a matter of debate. ENVISAT InSAR combined with GPS have shown some potential for better characterizing SSE and interseismic coupling along subduction zone, but ALOS-2 and SENTINEL-1 data open new perspectives for time-series analysis at regional scale. ALOS-2 and Sentinel-1 interferograms covering large parts of the subductions zone will be presented and discussed, with respect to previous results, in terms of capabilities to map the interseismic coupling along the subduction interface and to catch one of the large SSE, which periodically occurred in the area.

The resources of Differential Radar Interferometry (DInSAR) are used for investigating the crustal deformations in the area of Bulgaria’s capital Sofia. The town is located in southern part of Sofia valley – an active graben between Vitosha Mt. and Stara planina (Balkan) Mt. A number of faults cut the graben and the surrounding horsts including the system of active Vitosha and Zheleznitsa faults at the outskirts of Vitosha Mt and set of smaller faults passing through the city centre. As a result, the valley is characterised by block fragmentation covered by alluvial sediments produced by dense river system and mainly by Iskar River to the east of the city. Also a number of mineral water sources along the ground cracks give an influence over the ground dynamics in the frames of the city. To investigate the surface changes in the studied area affected by urban subsidence and micro seismicity a set of X-band (TerraSAR-X) and C-band (Sentinel-1 and ENVISAT) SAR images are used. The purpose is to determine areas with different ground movement behaviour in time triggered by the complex geological system in the investigated region. On the base of line-of-sight deformations velocity map is generated. The results are compared with levelling and GPS data conducted in the area. Several areas with different ground movement cheracteristics are deliminated. 

The Global Inundation Extent from Multi-Satellites (GIEMS) provides multi-year monthly variations of the global surface water extent at about 25kmx25km resolution, from 1993 to 2007. It is derived from multiple satellite observations. Its spatial resolution is usually compatible with climate model outputs and with global land surface model grids but is clearly not adequate for local applications that require the characterization of small individual water bodies. There is today a strong demand for high-resolution inundation extent datasets, for a large variety of applications such as water management, regional hydrological modeling, or for the analysis of mosquitos-related diseases.

This paper presents three approaches to downscale GIEMS. The first one is based on an image-processing technique using neighborhood constraints. The second approach uses a PCA representation to perform an algebraic inversion. The PCA-representation is also very convenient for temporal and spatial interpolation of complex inundation fields. The third downscaling method is based on topography information from the HydroSHED database. A new floodability index is introduced and an innovative procedure is developed to ensure the smooth transition, in the high-resolution maps, between the low-resolution boxes from GIEMS. Topography information is pertinent for natural hydrology environments controlled by elevation, but is problematic for more complex basins such as anthropized ones. However, this approach can merge satellite information in these difficult regions such as the Sentinel 1 and 2 (visible/infrared and SAR) recently available data, or SWOT (Surface and Water Ocean Topography) to be launched in 2019.

The resulting GIEMS-D3 database is the only long-term (1993-2007), dynamic (monthly), and high spatial resolution inundation database available at the global scale. GIEMS-D3 is assessed by analysing its spatial and temporal variability, and by comparison with other independent satellite observations, over the Niger, Amazon, and Mekong basins.

REF HYDROSHED

Aires, F., F. Papa, C. Prigent, J.-F. Cretaux and M. Berge-Nguyen, Characterization and downscaling of the inundation extent over the Inner Niger delta using a multi-wavelength retrievals and Modis data, J. of Hydromete- orology, in press, 2014.

Aires, F., F. Papa and C. Prigent, A long-term, high-resolution wetland dataset over the Amazon basin, downscaled from a multi-wavelength retrieval using SAR, JoH, 14, 594-6007, 2013.

Fluet-Chouinard, E., B. Lehner, L.-M. Rebelo, F. Papa, and S. K. Hamilton, 2015: Development of a global inundation map at high spatial resolution from topographic downscaling of coarse-scale remote sensing data. RSE, 158, 348–361.

Miolane, L, F. Aires, C. Prigent, E. Fluet-Chounard, B. Lehner, Global, high spatial-resolution, dynamic and long-term inundation extent dataset, downscaled from GIEMS using topography information, To be submitted.

Prigent, C., F. Papa, F. Aires, C. Jimenez, W.B. Rossow, and E. Matthews. Changes in land surface water dynamics since the 1990s and relation to population pressure. GRL, 39(L08403), 2012. 

Abstract. Since the launch of RADARSAT-2 in 2007 important capabilities, methods and assets have been developed for improved land-based change detection and monitoring. They include:

New unique image modes that combine high resolution and wide swath widths.

Methods for systematic change detection that benefit from the exactly repeated radar illumination geometries of spaceborne SAR.

A systematically collected archive of acquisitions using the new image modes for the purpose of broad area change detection in particular XF (Extra-Fine) with a nominal resolution of 5 m and a swath width of 125 km.

Perhaps one of the greatest strengths of exactly repeated SAR data is that change detection methods can be applied reliably, efficiently and robustly compared to optical imagery as there is virtually no influence of varying solar illumination, clouds or geometry. A global or regional regularly updated archive of SAR imagery of good resolution can be used for land surveillance applications such as:

1)       Forest canopy change detection to detect forest activity.

2)       Land slide detection (inventory).

3)       Non-coherent detection of creeping landslides

4)       Earthquake rupture zone location and 3-D co-seismic displacement

5)       Infrastructure damage assessment.

6)       Permafrost melting.

7)       Construction and demolition activity such as urban sprawl.

8)       Subtle deformation and coherent change detection using InSAR.

This paper gives an overview of methods, data sources, application examples and validation activity for various international locations.

Seville, with a metropolitan population of about 1.5 million, is the capital and largest city of Andalusia (S Spain). It is the 30^th most populous municipality in the European Union and contains three UNESCO World Heritage Sites. The Seville harbour, located about 80 kilometres from the Atlantic Ocean, is the only river port in Spain. The city is located on the plain of the Guadalquivir river.

Using Multi-Temporal InSAR with ERS-1/2 and Envisat data a subsidence behavior is detected in the period 1992-2010. The geometry of the subsiding areas suggests that it should be conditioned by the fluvial dinamics of the Guadalquivir river and its tributaries. Facies distribution along the fluvial system (paleochannels, flood plains...), with different grain size and matrix proportion, may explain the relative subsidence between the different sectors.

Glacial Isostatic Adjustment (GIA) is the reaction of the solid Earth to the large-scale glaciation and deglaciation of its surface, leading to vertical land motion. In Northern Britain, today’s GIA signal is predominantly influenced by the disappearance of the Pleistocene British-Irish ice-sheet and also to a lesser extent by far-field deglaciation effects of the Laurentide and Fennoscandian ice-sheets. The melting of the ice load resulted in uplift of the formerly depressed lithosphere in the glaciated areas, mainly Scotland with about 1-2 mm/yr, accompanied by subsidence in southern Britain [1].

The investigation of GIA in Scotland is critical for understanding the dynamics of relative sea-level (RSL) change at the coast. By determining the GIA-induced vertical land motion, RSL trends measured by tide-gauges can be corrected to determine change in eustatic sea-level. The complex interaction between near- and far-field GIA effects contribute to a significant temporal and spatial heterogeneity and variability of the RSL change in Scotland [2]. Against the backdrop of climate change and the global rise of sea-levels caused by the ocean’s thermal expansion and global melt-water influx, determining the modern rate and spatial distribution of GIA is crucial. The process of GIA has generally slowed throughout the Holocene and recent measurements of rising RSL at tide gauges around northern Britain suggest that GIA-related land uplift in Scotland might well be being outpaced by eustatic sea-level change. This is cause for new concerns about the impact on Scottish coasts and it contradicts the belief that sea-level rise in Scotland is “automatically” mitigated by GIA.

This study examines the applicability of differential SAR Interferometry and deformation time-series inversion techniques to measure GIA-related vertical land movement in Northern Britain. The aim is to establish a spatially more detailed picture of GIA and help identify the present location of the centre of maximum uplift within Scotland. The InSAR observations thereby inform recent GIA modelling efforts of crustal uplift [3] and complement conventional geodetic techniques, such as Continuous GPS [4], which rely on interpolation from point measurements.

SAR data sets from ESA’s ERS-1/2, ESA’s Envisat, JAXA’s ALOS-1 PALSAR and ESA’s Sentinel-1 are being used to establish time-series of surface movements across Scotland over the past two decades with the Small Baseline InSAR technique [5]. A range of possible error sources within the InSAR processing chain lead to challenges that require to be addressed before the extraction of any GIA signal is made possible. These include the sufficient removal of orbital and atmospheric artefacts from the total phase signal in the differential interferograms before inverting the deformation signal. This is especially problematic when it comes to measuring very small, long-wavelength and wide-scale ground deformation (GIA) due to its similar spatial patterns to orbital ramps and long-wavelength atmospheric signals. Various techniques are explored to separate the error signals from desired surface deformation signals. Apart from integrating the InSAR data with Continuous GPS station time-series as additional constraints, these methods comprise more sophisticated statistical approaches, such as Principal Component Analysis and Independent Component Analysis in order to exploit the signals’ different temporal characteristics.

Results of this study will eventually inform stakeholders and environmental agencies, namely Scottish Natural Heritage, for future coastal planning purposes.

Bangladesh is a low-lying country with over 160 million people, and located at the confluence of three large rivers – Ganges, Brahmaputra and Meghna. It is prone to monsoonal flooding, sea-level rise, potentially aggravated by more frequent and intensified cyclones resulting from anthropogenic climate change. In Bangladesh, about 46% of the Bangladesh population lives within 10 meters of the average sea level and 80% of the country is flood plain. Because of the flat topography of the floodplain, on average ~20% or more of the country is inundated annually by floods. In the 1960s, Bangladesh government constructed 123 embankments or polders around much of the low-lying coastal regions to mitigate relative sea-level rise and flood risks. Sediment load and groundwater extraction induced land uplift/subsidence, and riverbank erosion, have exacerbated Bangladesh’s coastal vulnerability. The subsidence rate of the polders due primarily to erosions tidally-driven sediment compaction and loading are arguably not well known at the spatial and temporal scale needed to mitigate or adapt to the Bay of Bengal coastal vulnerability. In addition, the polders in coastal Bangladesh can easily breach and be damaged by riverbank erosion that caused by high monsoon wind, waves, strong tidal actions and storm surges, and most of embankments in coastal region have experienced breaching more than once since their completion. It is estimated that about 5% of the total floodplain of Bangladesh is directly affected by erosion and millions of people are affected by erosion that destroys standing crops, farmland and homestead land every year.  Here we use the Persistent Scatterer Interferometric SAR (PSInSAR) techniques employing long-term ALOS-1/-2 PALSAR L-band data to estimate the land subsidence over example polders. PSInSAR is a new InSAR processing method, which relies on isolating pixels with consistent and stable radar reflections and is designed to accurately measure small spatial-scale land deformation signals in non-urban areas, allowing one to more accurately mitigate errors towards more accurately constructing land subsidence time series.  In addition, we use optical imageries from the Planet Labs constellation of over 100 cube-sats at daily sampling and 3–5 m spatial resolution, to monitor Bangladesh coastal riverbank erosions.

Surface deformation is an expected consequence of the geothermal fluid production. The effects of reservoir deformation (compaction) propagate to the ground surface, causing both vertical and horizontal ground displacements. The dynamic of these displacements is a complicated process both in time and space. Surface deformation monitoring can provide valuable information for assessing the environmental impact of the geothermal fluid production, understanding the fluid flow in the subsurface, evaluating volume/pressure changes in a reservoir, as well as estimating other geophysical parameters.

The Differential Synthetic Aperture Radar Interferometry (DInSAR), providing high-quality, remotely acquired measurements of surface deformation affecting large areas over long periods of time, is considered to be one of the most valuable and cost-effective technique for surface deformation monitoring.

In the Cerro Prieto geothermal field, located in the Mexicali Valley, northwest Mexico, the extraction of fluids produces deformation of large magnitude (Glowacka et al., 1996, 1999, 2005; Carnec and Fabriol, 1999; Sarychikhina et al., 2011). This surface deformation, mainly subsidence, and related ground fissures cause considerable damages to local infrastructure like roads, railroad tracks, irrigation channels, and agricultural fields (e.g. Glowacka et al., 2010).

This study presents an application of conventional DInSAR together with the interferogram stacking method to investigate the surface deformation in the Mexicali Valley near Cerro Prieto Geothermal Field for the period 1993 – 2014. The Synthetic Aperture Radar (SAR) data used in this study consists of five independent data sets from: ERS-1/2 descending track, spanning April 1993 – July 2000, ENVISAT ALOS descending and ascending tracks, IS2 mode, together spanning December 2003 –December 2009, and RADARSAT-2 ascending and descending tracks, mode MF1 and MF4N, respectively, together spanning November 2012–March 2014. Gamma ISP and DIFF/GEO software packages were used to calculate differential interferograms from SLC data and for differential interferograms stacking (Wegmüller and Werner, 1997). Since here we attempt to present results of only aseismic character, the interferograms which span the occurrence of large and moderate sized seismic events were excluded from the analytical procedure. Average LOS (line of sight) displacement velocity maps were generated for different periods: 1993-1997, 1998-2000, 2004, 2005, 2007, 2009, and 2012-2014. The changes in the surface deformation pattern and rate were identified. Thes main changes occur between 2000 and 2005. We suggest that these changes are mainly caused by production development in the Cerro Prieto Geothermal Field.

For the periods where the datasets from ascending and descending track were available, 2007 and 2012-2014, the average LOS displacement velocity maps from ascending and descending tracks were compared. This comparison give the important evidence of horizontal component presence. Average LOS displacement velocities estimated from tracks with different viewing geometries were used for a decomposition into horizontal and vertical displacement velocities. This study points out that the existence of horizontal component have to be taken into account for more precise evaluation of the surface deformation in the study area. 

The formidable advantage of space-borne Interferometric Synthetic Aperture Radar (InSAR) is its ability to monitor small deformations over wide areas without the need for human or special equipment presence, while providing accuracy similar to the conventional terrestrial techniques. InSAR has currently been applied in a number of applications encompassing the use of space-borne sensors operating at a range of wavelengths and resolutions. Results from multiple systems that are operational since 1992 often reveal different features in the exact same scene. The repeat-pass nature and different system characteristics give rise to the low coherence due to geometrical and temporal decorrelation in some areas. The wide variety of currently available spaceborne SAR sensors allows for the combined use of different frequencies and spatial resolutions as well as in-situ measurements for completeness and comparison reasons. Emphasizing the operational safety of anthropogenic structures as common radar scattering objects and usual targets of conventional monitoring techniques, the InSAR technology appears to be a credible candidate for providing continuous deformation monitoring throughout the historical decades until now. Following the ongoing research, satellite radar data acquired by ERS, Envisat, TerraSAR-X and Radarsat-2 satellites are utilised in order to improve the knowledge about the spatial and temporal evolution of the buildings and infrastructures within urbanised area of Bratislava, capital city of Slovakia. Thanks to the relentless boom in construction, that city is going through, floods occurring every years on average, geology containing number of tectonic faults and considerable amount of heritage structures in the centre and industrial complexes on the outskirts, structures safety concerns gain in importance. Moreover, with the evidence of buildings affected by static problems and components of Gabcikovo-Nagymaros waterworks (e.g. Cunovo Dam, Appendix 1), the efficient asset management to meet stringent demands on safety and reliable performance of such objects becomes a discipline of capital interest.

The objective of this study is to investigate the behaviour of new and ageing infrastructures under varying or hazardous environment, focusing on the determination of technical feasibility and economic viability of the integrated use of multiple space assets for deformation monitoring with the exploitation of advanced multi-temporal InSAR techniques implemented in SARPROZ software. Quality control analysis is performed by incorporating existing ground truth data from conventional geodetic techniques (GNSS, levelling). With already operational Sentinel-1A mission and free accessible near-real time data, the experiments and comparisons on the performance of the system for urban deformation monitoring are carried out in order to develop effective monitoring strategies that would be able to continuously collect the physical and dynamic parameters of the areas of interest.

Forecasting and monitoring of floods are currently the key objectives in ensuring life safety in residential areas located in river valleys. Timely notification of emergency response teams and population on imminent danger is crucial for saving people’s health and life and for minimizing economic losses in potentially dangerous regions. Prompt estimation of flood progress allows making weighed managerial decisions in emergency response and rescue operations and mitigation of consequences.

Forecasting of floods implies application of complex analytical formulae and processing of large quantity of statistical data: meteorological, hydrological, hydromechanical and thermophysical.

Methods of operational (short-term) forecasting of floods for obtaining highly accurate estimations of flood data in several hours’ to several days’ timeframe are based on numerical modeling of seasonal and flash flood wave propagation. As of today a great number of such models have been developed and used in practice in various countries. Models primarily differ in employed computation schemes and methods. With that said there is no such thing as a universal model suitable for describing flood development process across different in terms of length and configuration river sections. Calibration of almost all models requires complex parameter setting incorporating considerable number of integral characteristics of a particular river section. Extra complexity is introduced by the fact that a river is a dynamic structure and its characteristics may change significantly due to various reasons. Methods of forecasting based on mathematical models imply analysis and processing of a vast volume of hydrometeorological information both in method development and in forecast issuing.

In Russia with its vast territories and lack of detailed and continuous hydrological research, just a few sections of river systems are covered by models adapted to their conditions. The number of flood forecasting systems that de-facto work in real time is catastrophically low.

Estimation of possible boundaries of water flood zones at potential or actual water rise levels utilizes geoinformational method of modeling since this method is the least sensitive to the number of input data. It is based on geometrical approach: boundaries of water surface are defined by comparison of inclined water level and terrain elevation. Then certain rules are applied to use those boundaries for shaping flood zone polygons and defining flood depths.

The proposed procedure is based on the geoinformational approach. Highly detailed terrain data (DEM) are used as input data for modeling water flood. Geometrical water level is modeled with consideration of river fall and parameters calculated on the basis of DEM. It’s advisable to use this procedure for estimating potential danger and consequences of floods. If water levels are promptly available from an upstream gauging station the procedure could be adapted for a short-term forecast.

The procedure was validated by examples of floods occurred in Russia in 2013-2015. Accuracy of modeling was verified using series of space images of flood regions that demonstrated actual levels of area flooding.

Coal is a major source of energy in developing nations including India. Unmonitored extraction
of coal through methods like open cast mining in Jharia has adverse effect on the environment.
Having exposed to the air, coal can easily catch fire creating a catastrophe that can burn for
centuries and have a global impact. Coal fires can be identified using different conventional
methods like borehole temperature measurement or thermo-graphic measurement. The remote
sensing being an easy and more convenient technology can also be applied for seam fire
detection. In this research multi-date remote sensing data (1994, 2004 and 2014) of Landsat5 and
Lansat8 were analyzed. Using thermal remote sensing techniques Landsat5(TM) and Landsat
8(TIRS&OLI) change in surface temperature was measured using mono/multi-split window
algorithm. Fire pixels were identified from the non-fire background depending on the separation
in data clusters in maximum kinetic temperature v/s mean kinetic temperature scatter-plot,
obtained using simple random sampling of pixel-blocks (25×25 pixels) in the coal fields. Using
cut off temperature based on the scatter plot anomaly, fire pixels can thus be delineated and then
verified on the field. The study shows comparatively significant rise in the seam fires over a
period of two decades that pose an environmental threat if left uncontrolled.

The use of SAR data for flood mapping is presently well-established in operational services for flood management. Very high resolution SAR data, as those supplied by TerraSAR-X and the COSMO–SkyMed (CSK) constellation, allow for the generation of very detailed flood maps useful for emergency managers. In addition, taking advantage of the present (CSK) and forthcoming (Sentinel-1) constellations of satellites, the problem of the long revisit time of SAR images is considerably mitigated, thus enabling to monitor also the temporal dynamics of floods. However, some events might be missed because of the limited area that can be observed through a SAR image and the need of programming SAR acquisitions in advance. To tackle these problems, it is possible to set up a system able to trigger the SAR acquisitions based on flood forecasts and to take advantage of the various satellite SAR sensors that are presently operating. To provide the Italian Civil Protection Department with a rapid flood mapping service, a prototype of this kind of system has been set up and preliminary tested, using CSK and Sentinel-1 (S-1) data, to monitor the Po River (Northern Italy) flood occurred in November 2014.

The system supports an early programming of SAR acquisitions based on the forecast of FEWS-Po, the operational flood forecast model of the Po River Basin Authority (AIPo) available through the Flood Early Warning System platform; it supports also a combined exploitation of different SAR data to produce maps of flood extent. This study describes the prototype processing chain, which can be in principle exported worldwide, provided flood forecast models are available, taking advantage of the large number of spaceborne SARs available at present and in the near future. In the description, the attention is focused on an automatic algorithm, originally developed for a quick classification of CSK stripmap images [1], which is modified  in order to profitably implement it in the prototype system by increasing its reliability, as well as its robustness and transferability, to be used for S-1 interferometric wide swath data too. The algorithm is based on an automatic thresholding method and on a region growing technique used to increase the spatial homogeneity of the flood maps.

The reliability of the flood predictions used for the early programming of CSK acquisitions (S-1 follows a systematic and preprogrammed acquisition scenario) is assessed in this study, by comparing the predicted discharges with actual ones measured by some stations placed in the Po River Basin. In addition, SAR derived flood maps are compared with reference maps provided by AIPo, which refer to the critical reference flood, i.e., the depth produced by the discharge with return period of 200 years. Finally some achievements, obtained through a joint analysis of CSK and S-1 data for the purpose of a correct interpretation of some ambiguous radar signatures, are highlighted.

 [1] L. Pulvirenti, F. S. Marzano, N. Pierdicca, S. Mori, and M. Chini, “Discrimination of water surfaces, heavy rainfall and wet snow using COSMO-SkyMed observations of severe weather events,” IEEE Trans. Geosci. Remote Sens., vol. 52, no. 2, pp. 858–869, Feb. 2014.

Glacial lake outburst floods (GLOFs) are the most far reaching processes within glacier hazards. In the course of glacier retreat and vanishing in most glacierized regions of the world, growth of existing and formation of new lakes is currently observed and further expected for the future. Hazards related to glacier lakes therefore are of increasing concern, in particular in densely populated mountain regions, such as the European Alps, Central Asia, the Himalayas, and the Andes.

Considering the remoteness and often difficult access to such lakes, remote sensing provides important approaches for their detection and monitoring. Here we present the Slope Stability and Glacier Lake Monitoring (S:GLA:MO) service for hazard assessments of glacier lakes based on Earth Observation (EO) products. High and very-high resolution radar and optical EO data is used to derive products that provide the basis for an integrative, initial hazard assessment of glacial lakes. These data products comprise digital elevation models (DEMs), glacier and glacial lake maps, glacier velocity fields, and landslide inventories based on different types of InSAR analysis and interpretation in combination with high-resolution optical data. A particular focus of the service is the use of information on surface displacements derived by SAR interferometry (InSAR). Although slope stability is a crucial factor for the evaluation of the current hazard, the use of spaceborne surface displacement measurements is rather novel to glacial lake investigations. If available, auxiliary data such as geological maps, glacier bed topography, lake bathymetry, and permafrost distribution are considered as well.

As GLOFs often are involved in chains of mass-movement processes, integrative approaches are required for the assessment of the hazard emerging form glacial lakes. This not only involves a profound analysis of the current situation of the lake dam, and the up and downstream situation, but also requires an estimation of potential future developments of the lake and its surroundings, as well as the use of different modelling approaches to simulate the impacts and effects of potential GLOF scenarios. In order to provide a reference for the state-of-the-art, a checklist for glacial lake hazard assessments has been developed within the S:GLA:MO project. The structure of this checklist has been employed in the two pilot studies of the service in the European Alps, and hazard assessments of an ice-dammed lake in Greenland; three glacial lakes in the Cordillera Blanca, Peru; and a major lake in Tajikistan, including smaller upstream lakes in its catchment. In addition to the assessment of the current and potential future GLOF hazards, a case-specific monitoring strategy is recommended for each case, involving both EO products that can be provided by the service, and ground-based observations and investigations to be performed by the user.

The seven case studies highlighted the potential of EO data and related products for the use in first-order hazard assessments. Also, they revealed some limitations of InSAR products, for instance related to the investigation of very steep glacierized mountain faces, or the determination of the lake level elevation. However, in combination with ground-based investigations, SAR and InSAR products have a large potential to provide highly valuable information for the evaluation of the hazard situation, timely anticipation of adverse developments, and the planning of adequate hazard and risk management strategies.

Subsidence hazard and risk within the USCB are usually connected with the deep coal mining. In such cases, the surface becomes pitted with numerous collapse cavities or basins which depth may even reach tens of meters. They may remain dry or be filled with water (land surface inundation) depending on the local hydro geological conditions. The subsidence is particularly dangerous because of causing severe damage to gas and water pipelines, electric cables, and to sewage disposal systems. In Upper Silesia it is common to find houses strengthened with iron bars anchored in the walls in order to prevent further damage or collapse, but even such reinforced buildings could show cracks and joints on their walls. Nowadays, the surface of not active mining areas is around 2000 km² of the Upper Silesia. Collection of systematic information on the ground instability in these parts of the mining basin is very important because of the ongoing changes in spatial planning politics.

The first InSAR dataset analysed by PGI for particular investigations in the northern-central part of USCB, was delivered by Terrafirma project in 2004 and included PSI processing of 54 ERS-1 and ERS-2 scenes. Science then PGI has performed various analysis on InSAR data from further Terrafirma stages and different research grants, including all three SAR bands processed in DInSAR, PSInSAR and SqueeSAR techniques. These analysis of both conventional and advanced DInSAR approaches has proven to be effective to detect the extent and the magnitude of mining subsidence impact on urban areas. Conventional DInSAR was useful to detect the most active mining subsidence areas, whereas advanced DInSAR permitted to delimit residual mining subsidence surrounding the detected subsidence troughs.

In this study an analysis of two series of subsequent differential interferograms obtained in the DInSAR technique are presented. SAR scenes are covering two periods and were acquired by two different satellites: ALOS-PALSAR data from 22/02/2007-27/05/2008 and TerraSAR-X data from 05/07/2011-27/01/2013 (previous results monitoring period extended from 21/06/2011 till 27/01/2013.) The data were used to identify terrain motion caused by underground coal mining in the area of 15 active mines located in a northern and central part of USCB between Piekary Śląskie (N), Zabrze (W), Tychy (S) and Mysłowice (E). For both periods the subsidence basins were distinguished and manually digitized. The analysis included determination of the direction and development of subsidence movement in relation to the mining front and statistic comparison between range and value of maximum subsidence detected for each mining area. Furthermore comparison between the two period was made, revealing a decrease in the terrain changes in the newest period.

The observed motion on more than 80 subsiding areas reached  1 meter in one year and could cover over 20% of mining area. The decreasing of the subsidence basins number between the two period is related most probably to the reduction of the hard coal exploitation. On the other hand interferometric data which were compared, are characterized by different  wavelength, L-band 25 cm (lower resolution) and X-band 3.1 cm. It could have some impact on the interpretation and identification of the “fringes” indicating the subsidence basins. The obtained results successfully proof the usefulness of SAR scenes in long term and wide area monitoring of terrain changes in active underground mining areas. The data can help in estimation not only the range of the subsidence events, but also its value, direction of changes and character of the motion.

Stromboli volcano is highly accessible and well-monitored by ground instruments. Nevertheless, observational gaps remain and satellite imagery has proven useful for providing synoptic views of activity while the eruption is ongoing. During the 2014 effusive eruption at Stromboli, we exploited the benefits of integrating high spatial resolution detailed products extracted from EO optical data acquired by Landsat-7 ETM+, Landsat-8 OLI and EO-1 ALI with EO products characterized by a high frequency of refresh, global view but poor spatial resolution (MODIS and SEVIRI), in order to provide, with high precision, geographic information on the location of the active volcanic feature, the fraction and temperature of both ‘hot’ and ‘crust’ components present in each pixel of the anomaly, the radiant flux distribution and - finally - the effusion rate distribution of lava flows. In another side, the combined exploitation of various multispectral data acquired in day and night time modes offers great promise to provide a high rate of information refresh and to permit a complete comprehension of volcanic feature evolution and chronology. The 2014 Stromboli eruption started on 6 August with several overflows from the N rim of the crater terrace, located at ~750 m a.s.l. Explosive activity at the summit craters continued for two more days, while lava flows formed several branches spreading N and reaching the sea up to half August. After this the activity gradually decreased and eventually ended by mid-November. Here, we use the HOTSAT volcano hot spot detection system that works with satellite thermal infrared data to monitor the evolution of the eruptive activity. Data acquired by SEVIRI and MODIS were promptly analyzed to output hot spot location, lava thermal flux, and effusion rate estimation. The satellite-derived erupted volume is between 5.5 and 10.7 millions of cubic meters for the whole eruption. It is worth noting that these values are affected by uncertainties being computed from the radiance emitted at the surface, neglecting flows superimposition. At the same time, higher spatial resolution data acquired by Landsat-7 ETM+ and Landsat-8 OLI and TIRS sensors were used to validate and complement the lower resolution satellite products. Indeed, due to the favorable weather conditions, we exceptionally dispose of a huge quantity of high resolution data (about 20 images in the three months period) to map and track the emplacement of the lava flow field along the different phases of the eruption. By analyzing these high spatial resolution images we could follow the lava flow field evolution in the Sciara del Fuoco area and we found that this is in agreement with the radiant heat flux measured from lower resolution sensors. Moreover, after the climaxing initial phase (maximum active lava flow area of 0.72 km²), the effusive activity remained low and stationary for the whole eruptive period, with active lava flow area between 0.01 and 0.2 km².

The spatial and temporal evolution of postseismic deformation following large earthquakes is important to understand for two key reasons. Firstly, postseismic deformation plays a significant role in determining future regional seismic hazard. Patches of the coseismic fault plane that failed to rupture in the initial earthquake may fail in future damaging earthquakes or release stress by slipping aseismically. Postseismic motion may also increase the stress on other parts of the fault and influence the locations of future earthquakes.

Secondly, postseismic deformation can provide important constraints for models of the earthquake deformation cycle. For example, afterslip models can provide insight in to the mechanical properties of fault zones, whilst viscoelastic relaxation models can provide an estimate of the viscosity of the lower crust/upper mantle. The appropriate kind of model for any earthquake is informed by the postseismic deformation data.

The April 2015 M7.8 Gorkha (Nepal) earthquake ruptured a major thrust fault beneath central Nepal. Rupture began at ~15 km depth and appears to have propagated laterally along a shallow decollement. Geodetic and seismological observations suggest that most coseismic slip occurred on the deepest half of the fault plane, between the ramp under the high Himalaya and Kathmandu. 50 km of the fault south from Kathmandu to the Main Frontal Thrust did not fail coseismically, consistent with a lack of clear surface ruptures.

Analysis of GPS data suggests the entire Main Himalayan Thrust (MHT) is locked from the surface to the depth of the Gorkha earthquake. Large parts of this locked region did not rupture in the earthquake. The behaviour of the MHT to the west and up-dip of the Gorkha earthquake in the postseismic period is essential for our understanding of the ongoing seismic hazard. These sections of the fault may be relieving stresses aseismically or still be accumulating stresses which could be released in future, large earthquakes. Attempting to answer why these sections of the fault are behaving as they are will also place constraints on fault zone rheology.

The launch of ESA's Sentinel-1A allows for monitoring of postseismic deformation every 12 days over a large area. This short repeat time and large coverage allows us to investigate the spatial and temporal evolution of postseismic deformation in great detail and complements GPS results from the region.

We will present a preliminary analysis of postseismic deformation in the months following the earthquake using data from Sentinel-1A.These data will be used to understand the effects of this postseismic deformation on the regional seismic hazard and place constraints on postseismic deformation models for the earthquake.

Mashhad plain is a large intra-mountain basin surrounded by Kopeh Dagh and Binaloud mountains, in northeastern part of Iran. Mashhad city, with a population about 3 million people is located in this plain. Based on the results of six GPS surveys station, since 2005 up to 2013, a land subsidence with an average rate of about -10 cm/year and a maximum of about -26 cm/year in Tous station in the north west of the city war reported.  The land subsidence in the plain has been caused by over exploitation of groundwater during last decades. It is found that there is a strong correlation between the rates of groundwater level lowering and the GPS-derived rates of land subsidence in Mashhad Plain.

The DInSAR technique is applied for C-band ENVISAR ASAR data acquired between 1998 and 2010 in order to determine the extent and amount of land subsidence information in Mashhad plain. For retrieve time-dependent deformation in Mashhad Valley we utilize 31 images in a descending orbit during Jun. 2003-Nov. 2010, from 39 differential interferograms spanning different long- and short-term intervals, and carried out a time series analysis to extract deformation signals out of differential interferograms. Time series analysis suggests that the subsidence occurs within a northwest-southeast elongated elliptical shaped bowl along the axis of Mashhad valley. The DInSAR results are compared with historical precise leveling data (1995-2005) and detailed geological information to define the extension, limits and rate of land subsidence in the study area. The results in this study detected significant subsidence in the agricultural areas, where very large volumes of groundwater are usually extracted.

In the framework of different FP7 projects (SENSUM [1], RASOR [2] and MARSITE [3]), a set of open-source tools [4] [5] has been developed and refined with the aim to monitor the exposure component in risk assessment. This is done through fully automated extraction of selected vulnerability and exposure proxies, defined based on an extensive literature search both in the remote sensing and the risk assessment domains. Proxy extraction algorithms were designed to process medium resolution Landsat imagery, looking for the extension and the expansion in time of urban areas [6] [7], while very high-resolution optical imagery was used to zoom into detected urban areas and extend the analysis to single-building level. The code is currently distributed open source on GitHub [8] and as official QGIS plugins [9].

The recent launch of Sentinels 1 and 2 with their open data license attracted a lot of attention in the remote sensing community stimulating new research and applications. To our tools, this meant Open Data feed into Open Source code, which composed a nice example of “all-open” information production chain.

The goal of this paper is to present some tests made on “fresh” Sentinel-2 data using the Earth Observation tools –in particular the built-up area extraction algorithms- and assess their usability on data fed from the Copernicus Space Segment. The similarity in both spectral and geometric resolutions between Landsat and Sentinel-2 satellites raises good hopes for reusability of the developed tools.

Modifications to the original algorithms will also be suggested according to the experiments outcome, in view of a new generation of Open Tools in line with the new Copernicus-aware Earth Observation scenario.

[1] SENSUM project, 2015, available online at: http://www.sensum-project.eu/

[2] RASOR project, 2015, available online at: http://rasor-project.eu/

[3] MARSITE project, 2015, available online at: http://marsite.eu/

[4] D. De Vecchi, M. Harb, D. A. Galeazzo, F. Dell’Acqua, “Exposure Monitoring from Optical Earth Observation data: an Open-Source and integrated set of Tools”, Earth Observation Open Science 2.0, 12-14 October 2015, Frascati, Italy.

[5] D. De Vecchi, M. Harb, D. A. Galeazzo, F. Dell’Acqua, “An integrated, open-source set of tools for Urban Vulnerability Monitoring from Earth Observation data”, European Geoscience Union General Assembly 2015, 12-17 April 2015, Vienna, Austria. 

[6] M. Harb, D. De Vecchi, F. Dell'Acqua. “Automatic Hybrid-Based Built-Up Area Extraction from Landsat 5, 7 and 8 Data Sets”. Joint Urban Remote Sensing Event (JURSE) 2015, Lausanne, Switzerland.

[7] D. De Vecchi, M. Harb, F. Dell'Acqua. “A PCA-based hybrid approach for Built-Up Area Extraction from Landsat 5, 7 and 8 datasets”. International Geoscience and Remote Sensing Symposium IGARSS, 26-31 July, Milan, Italy, 2015.

[8] SENSUM GitHub repository, available online at: https://github.com/SENSUM-project/sensum_rs_qgis

[9] SENSUM QGIS plugin, available online at: https://plugins.qgis.org/plugins/sensum_eo_tools/

Spaceborne Synthetic Aperture Radar (SAR) systems represent a powerful tool to monitor floods because of their all-weather capability, the very high spatial resolution of the new generation of instruments and the short revisit time of the present and future satellite constellations. However, mapping flooded vegetation still a represents a challenging problem because the radar signatures of this target are often ambiguous, depending on several parameters (plant water content, structure and geometry of the vegetation) that are generally unknown. The possibility to exploit mission data provided by radars working at different frequencies and/or polarization represents a unique opportunity to perform a thorough investigation on how to improve the accuracy of flooded vegetation mapping. This investigation is currently carried out within the framework of two projects involving different space agencies, namely the EO-based CHange detection for Operational Flood Management – ECHO-FM project, funded by the Italian Space Agency (ASI) in the framework of its agreement with the Japan Aerospace Exploration Agency (JAXA), and the joint COSMO-SkyMed - Radarsat2 Announcement of Opportunity, issued by ASI and the Canadian Space Agency.

To perform the study, we have focused on an area placed in Northern Italy where rice fields are often artificially flooded for a significant period of time during the growing season. The area has been monitored in the period April-July 2015 through five X-band COSMO-SkyMed images in the standard Stripmap mode and four C-band Radarsat-2 fully polarimetric images. Through the ECHO-FM project, the availability of also three L-band ALOS-2 fully polarimetric images is also expected. By analysing this dataset, the capability of the different instruments to detect the double-bounce effect involving standing water and rice stems will be studied. The intensity of the double bounce backscattering is generally enhanced by floodwater, but this increase is not always detectable by a standard single polarization SAR image, as discussed in detail in [1]. Conversely, it is expected that through polarimetric decompositions, such as that based on the eigenvectors-eigenvalues of the coherency matrix, a better discrimination of the double bounce can be achieved.

The findings derived from the analysis of the aforementioned multifrequency and multipolarization dataset will be used to update a near real time flood mapping algorithm that, as most of the algorithms available in the literature, searches for regions of low backscatter only, so that it is generally unable to detect flooded vegetation.

To enable the algorithm to detect also this target, it will consider quantities like the mean alpha angle, which is the parameter used to characterize the results of the polarimetric decomposition based on the eigenvectors/values of the coherency matrix that represents the mean dominant scattering mechanism.

Located at the junction of the Philippine and Eurasian tectonic plates converging at a rate of about 8 cm per year, Taiwan is one of the most deforming areas in the world.  With a subtropical environment, it is faced to different hazards, including earthquakes, debris flow, landslides, and flooding. The precise measurement of the present-day ground displacements at the scale of the whole Taiwan Island is thus essential for researchers in several domains of Earth Sciences, in particular for earthquake cycle study and earthquake hazard assessment, but also subsidence, landslide and mountain uplift monitoring. Taiwan benefits from a very dense GPS network, and is one of the best instrumented areas in the world. However, this density is not high enough for many applications, in particular to detect precisely active tectonic structures.

SAR interferometry (InSAR) has proven its capacity to spatially densify the measurement of ground deformation in Taiwan, with a dramatically increase of the spatial information in some areas. However, previous studies conducted on Taiwan only focused on local part of the island, using different processing approaches.

Using data from ALOS-1 satellite and NSBAS method, a method based on the small baseline approach, we propose a first complete coverage of Taiwan by InSAR. ALOS L-band data are the most effective in the vegetated and hilly Taiwan environment. Our dataset consists of a total of about 300 images, the whole ALOS-1 images available on the island. They are distributed on 4 ascending tracks covering the island, with about 20 dates for each track on the 2007-2011 period. This homogeneous dataset is processed consistently over the whole island, contrasting with the previous INSAR studies, and allows generating consistent time series and dense map of ground displacements over the period 2007-2011.

SAR images are processed through a small baseline approach with NSBAS chain. This chain, developed at ENS Paris in collaboration with ISTerre is based on ROI PAC modules, but includes new specific ones. In particular, digital elevation model errors are estimated and removed before unwrapping. It has also a specific module to mitigate the effects of the atmospheric phase delay and remove residual orbit errors, and a module to construct the phase change time series from small baseline interferograms.

On all tracks, our results are consistent with previous studies using PSI approaches: e.g. large subsidence of the western coastal plain studied by Tung and Hu (2012), and the Longitudinal Valley Fault creep already studied by Champenois et al. (2012). However, compared to these previous studies, our work provides a better spatial continuity and coverage, leading to new geological interpretation. In particular, time series analysis of SW of Taiwan provides a far better coverage in the foothills area, revealing very large active deformation of tectonic features that were overlooked in previous tectonics studies. There are specific evidences of a series of en-echelon Folds that seems to be controlled by a N-S detachment ramp with a gradient of aseismic deformation from south to North, suggesting creep on the detachment.

Fire Radiative Power (FRP) retrieved by infrared sensors, such as flown on several polar orbiting and geostationary satellites, has been shown to be proportional to fuel consumption rates in vegetation fires, and hence the total radiative energy released by a fire (Fire Radiative Energy, FRE) is proportional to the total amount of biomass burned. However, due to the sparse temporal coverage of polar orbiting and the coarse spatial resolution of geostationary sensors, it is difficult to estimate fuel consumption for single fire events. Here we explore an approach for estimating FRE through temporal integration of MODIS FRP retrievals over MODIS-derived burned areas. Temporal integration is aided by statistical modelling to estimate missing observations using a generalized additive model (GAM) and taking advantage of additional information such as land cover and a global dataset of the Canadian Fire Weather Index (FWI), as well as diurnal and annual FRP fluctuation patterns. Based on results from study areas located in savannah regions of Southern and Eastern Africa and Brazil, we compare this method to estimates based on simple temporal integration of FRP retrievals over the fire lifetime, and estimate the potential variability of FRP integration results across a range of fire sizes. We compare FRE-based fuel consumption against a database of field experiments in similar landscapes. Results show that for larger fires, this method yields realistic estimates and is more robust when only a small number of observations is available than the simple temporal integration. Finally, we offer an outlook on the integration of data from other satellites, specifically FireBird, S-NPP VIIRS and Sentinel-3.

This work proposes methodologies aiming at evaluating the sensitivity of optical and SAR change features obtained from satellite images with respect to the damage grade. The test case is the Mw 7.0 earthquake that hit Haiti on January 12, 2010. The disastrous shock caused the collapse of a huge number of buildings and widespread damage with its epicentre located about 25 kilometres West-South-West of the city of Port-au-Prince.

The proposed methods are derived from the literature and here we investigate on possible parameters that can affect the robustness and sensitivity of such methods. The main novelty of the proposed analysis concerns the estimation of these derived features at object scale.

The analysis of change detection indicators is based on ground truth information collected during a post-earthquake survey and available from Joint Research Centre database. In order to exploit the object-oriented approaches, it is therefore necessary to perform a segmentation of the study area into homogeneous regions. Aiming at testing an operational approach, we have generated the damage map of Port-au-Prince by considering a set of polygons extracted from the open source Open Street Map (OSM) geo-database. The resulting damage map was calculated in terms of collapse ratio: this quantity is defined considering a city block, and by calculating the number of collapsed buildings with respect the total number of buildings within the block.

The available satellite dataset is composed of several optical and SAR images, collected before and after the seismic event.

A preliminary study allowed us to identify some features having a good sensitivity with damage at object scale. Regarding the optical data, we selected the Normalised Difference Index (NDI), and two quantities coming from the Information Theory, namely the Kullback-Libler Divergence (KLD) and the Mutual Information (MI). In addition, for the SAR data we picked out the Intensity Correlation Difference (ICD) and the KLD parameter.

In order to analyse the ability of these parameters to correctly detect damaged areas, a Naive Bayes (NB) classifier was used. The classification results demonstrate that the simultaneous use of several change features from EO observations can improve the damage estimation at object scale.

Studies of interseismic strain accumulation are crucial for our understanding of continental deformation, the earthquake cycle and seismic hazard. The interseismic period is the longest portion of the earthquake cycle, often spanning decades to centuries. It has therefore, been difficult to measure the variation in strain accumulation at different stages of the interseismic period. In this study we address this issue by investigating the surface deformation along the North Anatolian Fault (NAF) in Turkey.

The NAF is a 1300 km long, mature right-lateral continental transform fault that accommodates a significant fraction of the motion between central Turkey and Eurasia. Together with the East Anatolian Fault it facilitates the motion of Anatolia away from the Arabia-Eurasia collision zone westwards towards the Hellenic subduction zone.

In the last century the fault has accommodated 8 large earthquakes (Mw 6.7 and above), with a dominant westward progression in seismicity culminating in the M7.4 Izmit and M7.2 Duzce earthquakes in 1999. This earthquake sequence provides a unique opportunity for us to investigate the surface deformation and strain accumulation as each previous rupture segment is at a different stage of the earthquake cycle.

In this study, we perform a time series InSAR analysis of 22 Envisat tracks in both ascending and descending geometries that cover the entire continental extent of the North Anatolian Fault. Our InSAR dataset spans an 8 year time window from 2003 to 2010. We use a small baseline processing strategy using the StaMPS software, where we also implement an iterative unwrapping procedure that tracks well-unwrapped pixels for each subsequent iteration. The latter relies on checking for unwrapping errors by summing the residuals around closed interferometric loops.

By combining our InSAR results with existing GPS measurements, and assuming negligible vertical motion, we create a high resolution map of the horizontal velocity field in northern Turkey. Our results confirm the presence of two aseismically creeping segments at the Izmit and Ismetpasa sections of the NAF. We find that the remainder of the fault appears to be fully locked and accumulating strain throughout the crust.

We calculate the strain rate from our velocity field and use elastic dislocation models to determine the fault slip rate and locking depth at various locations along the fault. We investigate how these parameters vary at each of the previous rupture segments.

The discovery and monitoring of natural and anthropogenic induced mass movements such as land subsidence, landslides and sinkholes have become powerful assistance by Radar Remote Sensing. Detection of area wide land subsidence and landslide monitoring is an important information for in time hazard mitigation. The intention of this presentation is to show the information content of SAR Interferometry results and the information values with respect to an optimized hazard mitigation approach. Based on Persistent Scatterer Interferometry (PSI) and Small Baseline Subset (SBAS) processing of satellite data from ERS-1/-2 and Envisat-ASAR, TerraSAR-X and Sentinel 1 and ALOS-Palsar ground motion data are investigated at different locations in Germany and worldwide. The structure of a processing, validation and value adding scheme will be presented.

Natural gas extraction leads to a decrease in reservoir pressure which can cause land subsidence and earthquakes. This process can cause damage to buildings, infrastructure and may affect ecosystems and agriculture by influencing the hydro(geo)logical setting. A case study focusing on the application of the PSI technique for the detection of land subsidence at the northern German natural gas extraction province is performed. Results show that land subsidence is concentrated at several gas extraction sites. For a geohazard approach in the German federal state of Rhineland-Palatinate PSI and SBAS analysis of ERS/ENVISAT and TerraSAR-X data is performed. The extent and the timelines of the mass movements show that the majority of landslides perform a clear correlation to the local geology and geomorphology. Another area for detailed investigation is located in an abandoned underground basalt mining area. The near-surface cavities are still supported by remaining basalt pillars. Due to tension changes, the supporting pillars are damaged and locally sinkholes occur already. The project investigates if further indicators for ground motion and geohazards can be derived from remote sensing data, as ground based monitoring did not provide sufficient information. The impact of climate change and sea level rise for coastal areas is indisputable. To mitigate impacts for the living environment and coastal protection anthropogenic and natural influences need to be understood. Thus a national ground motion database is considered.

Following natural calamities, the risk assessment of the roads blockage is a crucial aspect to plan search and rescue operations by the agencies in charge of civil protection activities. With particular reference to seismic crisis, the geometry and the physical vulnerability of the buildings, together with the position and size of the roads, represent the key parameters to evaluate the road network performance especially after hazardous earthquakes in urban areas. In this framework, the risk assessment mapping of the transit routes occlusion allows a better planning of the pathways for emergency activities (search/rescue, safe evacuation, medical aid, etc.). This can be possible effectively by classifying the roads according to the likelihood to be blocked by the presence of debris due to the damaging or collapse of surrounding buildings.

Based on this rationale, in this work, a procedure to map the risk of road blockage in urban environments is presented. It has been effectively applied to the city of Cosenza, located in a highly seismic area of Calabria Region (Southern Italy), analyzing the roads of both the historical town and the new-settlement built-up area. The methodology to assess the building loss and consequently the blockage risk attributed to the road network can be briefly described in three phases.

In the first one, the geo-database of the city has been developed through the integration of the results obtained by analysis of space-borne multispectral images and airborne Light Detection and Ranging (LiDAR) data. In particular, the land-use based on the Geoeye-1 satellite imagery, as well as the merging and the classification of the point clouds carried out by RIEGL LMS-Q680i sensor, allowed to outline the digital terrain model for topographic layout, the 3-dimensional (3D) geometry of buildings and the characteristics of road network. All elements have been then merged as raster or vector files in a 2D/3D Geographic Information System (GIS) platform. The same geo-database also contains the information on both the seismic hazard and the construction period of build-up areas, provided by consecutive urban plans.

The second phase of the proposed methodology have concerned the modeling of the building vulnerability, based on the corresponding 3D geometry and construction time information. The latter parameter allows to take into account techniques and materials construction.

Finally, the road network blockage risk of the selected test city is mapped through an ad-hoc index defined as ratio between a characteristic height, linked to the building placed on the road, and the width of the same road. To make the analysis more fitting to the real damaging scenario, the proximity analysis has been further performed taking into account that the deposition threshold distance of building debris can be partially covered by elements between building and road (such as sidewalks). Furthermore, the 3D mapping of the buildings with the expected damaging levels has been provided together with the related spatial distribution of the road block index.

The presented work has been carried out in the framework of PON MASSIMO[i]  (Monitoraggio in Area Sismica di SIstemi MOnumentali) project for the cultural heritage monitoring in a seismic area. Some meaningful experimental results of the proposed activity are presented in this work.

[i] The present work is supported and funded by Ministry of Education, University and Research (MIUR) under the project PON01-02710 "MASSIMO" - "Monitoraggio in Area Sismica di SIstemi MOnumentali".

In view of the recent and serious flood events occurred in latest years worldwide, the interest towards accurate methodology for the evaluation of flood prone areas and flood monitoring is dramatically increased. To address this issue the mapping of reliable flooded areas is of utmost importance. On this context the launch of new ESA Earth observation Sentinels missions is expected to provide a contribute improving spatial and temporal resolutions of the current operating missions. The Sentinel-1 space-borne microwave sensor with its all-weather, day and night capabilities, might increase the capacity for worldwide repeated flood extent mapping. Due to the  typical backscattering characteristics of active radar signal on plain water surfaces, the use of SAR data for high-resolution flood mapping is relatively straightforward. Moreover, the Sentinel-2 optical sensor can characterize land surfaces changes thanks to different spectral responses in the visible-infrared domain. Therefore, the combined use of the two information promises to bring to an even more exhaustive picture of inundation characteristics due to the complementary information provided by the two satellite sensors.

We propose a method based on: 1) the recognition and modelling of spectral changes, due to floods occurred in the past, and 2) exploitation of the obtained models to recognize and classify new floods. In the first phase, a library of signatures (i.e. the functions modelling the probability density distributions) representative of changes due to floods is implemented, based on the change detection techniques. Successively, such library is used for classifying  flood areas when new images will be available. Respect to conventional classification approaches, commonly requiring a supervised training for each flood event, the proposed procedure requires a supervision only during the library implementation. This allows to skip the manual step of training areas selection when mapping floods from new images.

In this study, we present the results of first step concerning the implementation of the library, through the exploiting of the signatures in both spectral ranges.

The activity is carried out under the project “Land cover change detection and monitoring methodologies based on the combined use of S1 and S2 for natural resources and hazard management” in the ESA framework SEOM SY 4Sci Synergy.

Warsaw agglomeration is characterized by complex geological structures and by various ground deformations. Some of them are geological-based large-area deformations. Additionally there are many places of local man-made subsidence caused by the construction of new apartments and office buildings, metro lines and other underground works. Many of local deformations are located close to the natural or artificial escarpments.

The presented work was based on the two sets of radar data. History of movements of earth surface in Warsaw agglomeration in the period 2002 – 2010 was re-created on the basis of analysis of archival radar images collected by ENVISAT satellites derived from 9 different ascending and descending orbits. The prepared method of joint analysis of ENVISAT images collected with various angles will be used for analysis of Sentinel-1 images, which will be performed in the next step of the project. High-resolution COSMO-SkyMed stripmap images registered in the period 2011 – 2014 were used both for analysis of large area and local deformation in the study area. Comparison of results from two data sets was performed.

The conducted research indicates a limited possibility to use ENVISAT data for subsidence studies in Warsaw agglomeration.  The radial shape of the agglomeration outside of the main part of the city and a large amount of vegetation within the city was the main obstacle. As a consequence of this situation there were problems with correct phase unwrapping especially in the external districts of the agglomeration, low density of persistent scatterers and problems of applicability of some methods of selection of candidates for persistent scatterers. Small set of data from the certain orbits was the other obstacle.  As a consequence many local deformations which existed within the study area were omitted and there were difficulties to obtain the correct results for the whole study area. Despite of this some places of local subsidence were correctly identified and measured. The advantage of use of multi-angle ENVISAT data was a possibility to identify some movements located close to the escarpments which were not visible on the set of COSMO-SkyMed data due to the direction of existing movements.

The results obtained using COSMO-SkyMed data show many local man-made deformations existing within the study area, related to construction of first and second metro lines, new buildings and with location of some constructions very close to the natural and artificial escarpments. Some of deformations were not known before of this study. Identified subsidence was up to 1 cm/year. There was also a possibility to identify some large-area deformations within the study area. The reasons of their existence are under investigation.

Presented work shows great usefulness of high-resolution COSMO-SkyMed data for deformation studies in Warsaw agglomeration, however both ascending and descending orbits have to be used to identify and measure all existing deformations.

Presented work was performed within the project “The integrated system of surface deformation monitoring based on Persistent Scatterer Interferometry, measurements from permanent GNSS stations and precise leveling” financed by The National Centre for Research and Development from Poland within the Applied Research Programme. The Envisat data were obtained within Cat-1 ESA project ID 19409.

According to the statistics provided by the European Join Research Centre in Ispra/Italy Poland is a third European country after Portugal and Spain in number of fires reported annually. The National ForestFire Information System functioning currently in Poland is based explicitly on the ground data collection. TheEO data, which is recognized as valuable source of information about fire event is not included into the existing system. It is known that ground data collection is usually time consuming and cost ineffective especially at large scale as national and sub-national levels. Therefore, the main aim of this study was to examine and assess the suitability of EO data for providing additional information on the fire regime in Poland. The main objectives are: 1) detection of fires using active fire products, 2) validation of hotspots against ground-based fires reported in the National Forest Fire Information System, 3) dynamics of post-fire vegetation regrowth, 4) assessment of fire intensity based on MODIS Fire Radiative Power (FRP), and 5) calculation of emissions from burning biomass.

The national satellite-based Active Fire Database contains ‘hotspots’ (thermal anomaly) from MODIS sensor on board Terra/Aqua (viewing the Earth’s surface twice a day in complimentary orbit), from ATSR sensor on board of ENVISAT satellite (viewing the Earth’s surface at night) and from AVHRR/NOAA. The satellite-based Active Fire Database covers the period 2001-2014 and contains only ‘vegetation fires’. The comparison of hotspot and ground-based fires revealed that on average 32,3% of the remotely detected fires were recorded in national database. The highest agreement was obtained for the year 2012 (50%), 2008 (43%), 2003 (41%), and 2009 (39%). On average, about 68% of detected hotspots was not reported in national database, which indicate high commission error. Out of all hotspots coupled with fires recorded in national database, 13% of hotspots correspond to fires equal to 1 hectare, 42% to fires covering an area between 1-5 hectares 24% to fires between 5 and 10 hectares, and 21% to fires greater than 10 hectares. The results indicates that satellites has missed the vast majority of vegetation fires. The limitation of both databases was also discussed.

Information on fire extent for selected fires was derived using bi-temporal Normalised Burn Ratio index. The dynamics of post-fire vegetation regrowth was assessed using the Percentile Rank Order Differentiation (PROD) method based on Normalized Difference Vegetation Index (NDVI).In order to better characterise the post-fire vegetation regrowth over the burned wetlands more detailed analysis of LIDAR data and high resolution Pleiadas image were performed. The LIDAR cloud points confirmed the difference in the height of post-fire vegetation.The MODIS FRP was used to derive information on fire intensity. FRP of a fire represents the amount of radiant heat energy liberated per time unit from the burning (Wooster et al., 2003, Kaufman et al., 1998), which is directly proportional to the combustion rate (Wooster et al., 2005). The analysis of fire intensity, for the fires detected both remotely and on the ground, demonstrated that forest fires were the greatest and the most intensive fires compare to the agricultural burning or to fires on the other non-wooded natural land. The FRP values were used to calculate biomass losses (top-down approach) and then to obtain the quantitative estimation of emissions from burning biomass for the wetland fire. Investigation of fire intensity and quantification of biomass/carbon losses and emissions from burning is essential to mitigate climate changes.

Hunga Tonga Hunga Ha’apai (HTHH) is a newly-formed island within the Kingdom of Tonga that formed rapidly between December 19, 2014 and January 26, 2015. The dominant formation process was a classical surtseyan eruption dominated by tephra. The eruption ended 5 weeks after it started, leaving behind a classical oceanic island volcano (tephra cone), approximately 2 km^2 in area with an un-measured subaerial volume.   The island is dominated by a quasi-circular tephra cone with an off-center summit crater complex, and heavily gullied flank slopes. The gullied slopes are most probably a consequence of late-stage pyroclasic flows, incising narrow gullies into the unconsolidated tephra that forms the bulk of the subaerial landscapes. The underwater expression of the island is unknown, but may be in the form of a platform as at Surtsey, the canonical hydro-volcanic island in the sub-arctic off the coast of Iceland that formed in the 1960s.  Immediately after its formation HTHH island experienced fast-paced erosion and marine abrasion of the weakly consolidated tephra. After a few months (May 2015) the interior crater lake rim wall was breached to the Pacific ocean and only a “bar” of unconsolidated fragmental debris of mainly tephra remains. The Pacific Ocean is now infiltrating the crater and triggering enhanced marine abrasion.  Monitoring the erosion via satellite optical and SAR imaging at meter resolution is ongoing, thanks to partners from CSA and NGA. 

An important dimension in quantifying island evolution via erosional processes is topography in the form of geodetic-quality digital elevation models (DEM) that capture the geometry of the island and which enables measurement of volumes of key facies. In this study, we use bistatic TanDEM-X/TerraSAR-X acquisitions of different modes to generate high resolution DEMs and quantify rates of change (dV/dt) by means of differential SAR interferometry (DInSAR).  In addition, we have used optical stereogrammetric DEM’s from the early evolutionary stages of the island to document initial conditions, in partnership with NGA (via DigitalGlobe Worldview data).

The results from the bistatic TSX/TDX measurements will allow quantitative assessment of:

1)      Stages of evolution in the post-formational geologic development of an oceanic volcanic island formed by means of classical surtseyan eruption styles

2)      Whether or not compositional control of island stabilization and hence prolonged lifetime (in the face of oceanic erosion) is evident

3)      Geometric progression of island evolutionary stages as it transitions from construction to destructive erosion (and eventually total submergence)

4)      Depositional patterns and rates relevant to other oceanic volcanic island settings on Earth and extrapolated to Mars (where rates are unknown)

From current experience the initial estimates of island lifetime range from as short at ~1.1 years to at least 3.65 years. However, these estimates are poorly constrained and require many assumptions. The inclusion of TDX HDEM data and DInSAR will readily lead to improved predictions of lifetime and geomorphic evolution, and permit development of a new class of landscape model for such surtseyan oceanic islands.

Starobin deposit is one of the largest potassium-bearing basins in the world, and in this regards it is of great importance for the economic development of the Republic of Belarus. Its mining is carried out by underground methods since the early 1960s and it has a massive impact on the ground surface over an area of 50 square kilometers.

Before 2011, the monitoring of subsidence was performed based on traditional surveying methods: leveling and tacheometry with partial use of GPS-measurements. Above surveying give us possibility to create the mechanical model of rock massif based on viscoelastic and transversely isotropic laws. Finite element method was used for numerical modelling. Coulomb-Mohr and maximal tension deformation criteria were used for determination of discontinuous zones. It was established that for the subsidence estimation problem it is enough to choose several (namely three) basic sets of rock’s beds with different physical-mechanical parameters. The conventionally selected three basic sets of rock’s beds can be called sedimentary bed, argillo-marlaceous strata and salt bed.

Even traditional surveying demonstrated that subsidence process extremely depend from speed of excavation. The reason is that during mining several types of excavation techniques are used, that leads to wide interval variations of subsidence speed. Simultaneously it is possible to observe on Starobin deposit the rate of subsidence from about 10^-4 m/day to 0.03 m/day. Above features forced us to consider all parameters of our mechanical model (elastic properties, cohesion, friction, etc.) for each rock’s beds as reological and dependent from speed of excavation. Unfortunately, the traditional surveying of subsidence don't provide the possibility to reconstruct reological dependence with accepted accuracy.

In 2011 year, the ground surface of Starobin deposit was observed with the RADARSAT-1 interferometric synthetic aperture radar (InSAR). First experience has shown that the practical use of this method for the condition of the Starobin deposit, along with access to a fundamentally new and require further analysis of the results, inevitable confronted with a number of adverse factors, among them in the first place should include intensive agricultural activities. Unfortunately the duration of time period between RADARSAT-1 expositions (14 days) was too long to observe big rate subsidence.

In 2015 year we are starting the project concern to the monitoring of subsidence of Starobin deposit based on TerraSAR-X platform with 11 days period of SAR expositions. The new platform sufficient increased the accuracy of monitoring and give possibility for reconstruction the reological dependences of properties of mechanical model for each of rock’s beds. Our experience shows practical example to use radar interferometry for validation of geomechanical models.

Greece is located directly north of the boundary between the Eurasian and African plates. The Greek territory is characterized by tectonic deformation, showing a north south displacement. The Aegean and the Ionian region are the most seismically active parts of Europe characterized by a high number of seismic events, most of them located offshore. The seismicity of the Greek mainland and the related induced surface deformation has been in depth studied. Unlike in the case of offshore seismic events, onshore deformation induced has never been mapped except in cases of very strong events with serious consequences.

The aim of this study concerns the mapping of onshore co-seismic deformation provoked by offshore events. On the possibility of the occurrence of surface deformation, provoked by offshore events, two are the main parameters influencing (a) the characteristics of earthquake-induced deformation (b) the distance from the coast line. For implementing this aim we selected three seismic offshore events in three different areas, namely the following seismic event occurred (source NOA/GEIN)

(i)                on 16-4-2015 40 km to south-east of Crete Island with Mw=6.0, depth=36.8 km

(ii)             on 17-11-2014 in Eubea Gulf  70 km NNW from Athens double event with Mw 5.2 and 5.1 depth= 23 km

(iii)           on 8-11-2014 3 km SSW of Kefalonia island with Mw=5.0, depth=18.4 km

The approach method is based on repeat-pass interferometry using a rich multi-temporal dataset SENTINEL 1A SAR IW mode scenes covering the above three sites. Two factors had to face in the frame of the interferometric processing (a) to form suitable interferometric pairs in terms of temporal and spatial decorrelation and (b) to avoid the possible presence of atmospheric phase in the final product. The first issue is easily solve as a large archive of Sentinel 1 SAR images already stored in Sentinel hub tool. The second one precisely because of the rich archive multiple suitable pairs have been formed and therefore a significant number of interferograms have been created allowing by interpreting different pairs to recognize and exclude scenes that induced atmospheric phase in the final interferogram.

Investigation of onshore co-seismic deformation in terms of pattern and rate, as well expressed faults in onshore regime continue in sea depth, can be used to improve and infer the offshore earthquake generation process and can be also be utilized in seismic hazard assessment.

The SAR Interferometry (InSAR) application has shown great potential in monitoring of land terrain changes and in detection of land deformations such as subsidence. It has a great welcome by geodesists - it is a high resolution measuring technique capable for achieving spatial information instead of typically pointwise information (as most of geodetic methods are). Within spatial information in high sensitivity for subtle movements, experts can interpret extents of subsidence, temporal development of subsidence troughs, evaluate accuracy (or specify more accurate parameters) of subsidence models or to compute the deformation slope angle and rate within the subsiding area - that can help in a warning against significant deformations of buildings and other infrastructure.

On the other hand, notion of “Disaster Risk Management” is accepted all around the world especially within earth science society in recent years. Advanced and almost developing countries are trying to develop their plans and policies with particular emphasis on this aspect. Disaster Risk Management refers to determining, mitigating and sharing the studies about risks and principles of planning over country to region, region to city and city to local areas. Preparation of disaster scenarios, identifying priorities and the general policies and practices in order to reduce the risk of strategic planning, preparation and implementation of the application plans is part of this process.

The innovative character of the paper is thus also the focus into evaluation of applicability of InSAR to provide an early warning. To arrange a continuous monitoring system providing early warning against sudden changes or accelerations of land mass movements in both urban and non-urban areas, an appropriate multitemporal method should be chosen and tested in combination with different input dataset. In our paper we consider both low-resolution C-band SAR data of Sentinel-1 (and based also on previous experience from ERS and Envisat) and high-resolution X-band data (Cosmo-SkyMed) to understand which data are the most appropriate for automatic continuous early warning system. For continuous early warning system useful for practical application at General Directorate of Disaster Affairs centers (AFAD) of Turkey, data of guaranteed continuous acquisitions with an appropriately short temporal baseline should be selected. That is why we don’t consider any other SAR system for these purposes.

Usage of high frequency of acquisitions by Cosmo SkyMed and Sentinel-1 helps to identify faster terrain movements than it was ever possible before. Within this contribution we report our continuation of the work on monitoring Konya closed basin and show the differences and potential of new sensors w.r.t. past ERS, Envisat and Alos-1 missions.

The current state of work demonstrated in this paper is maps of the spatial and temporal patterns of deformation at ground subsidence (or uplift) by using InSAR and geodetic methods at specific areas surrounding Konya closed basin and its near vicinities. This assists significantly in time planning of preparation of susceptibility maps over Konya Basin subsidence using satellite based interferometry with these new satellite systems.

Cosmo SkyMed and Sentinel-1 data evaluated using PS InSAR processing and other multitemporal technique such as Quasi-PS InSAR, SBAS methods, based on SARPROZ software package abilities. Differences of application of other InSAR processing tools or techniques should be also tested. The processing results are detected movements of subsidence in Konya. The main objective is to evaluate if the method can be applied to provide an early warning in advance of local dangerous deformations, i.e. if the accelerated deformation changes can be detected confidently using high-resolution high-number X-band data.

Land subsidence is a well-recognized environmental problem in mining areas . Negative aspect of mining exploitation is visible on the surface in the form of surface deformation and subsidence. Such deformation pose a danger in highly urbanized areas and could result in significant damages to houses and urban infrastructure. For urban development and protection it is necessary to monitor and measure the subsidence with very high vertical accuracy and spatial continuity at regular time intervals.

Traditional monitoring techniques that use levels, total stations and GPS can only measure on a point-by-point basis and hence are comparatively costly and time-consuming. Differential radar interferometry is well suited to detect small relative displacements of the surface. It is also capable of obtaining dense information related to the deformation across a large area in an efficient and economic manner. However in central Europe interferometric analysis can be aggravated by extensively vegetated land surface. Longer temporal separation between two radar acquisitions typically reduce the ability to interpret the measured phase differences.

Sentinel-1A has a 12-day revisit time, which should resolve problems with time decorrelation. This paper shows possibilities for monitoring subsidence on mining areas with new Sentinel-1 data.

The purpose of this paper is to present work being carried out within an ESA General Support Technology Programme project called “Advancing SAR and optical methods for rapid mapping” (ASAPTERRA) that is led by the DLR with SERTIT as a partner. This project’s main objective is the development of robust and transferable, but at the same time fast and accurate methods to significantly enhance rapid mapping techniques for three types of natural hazards – floods, landslides and fires – using active and passive information sources.

This paper concentrates on fire, burnt area mapping and the identification and exploring new methods based on recent satellite sensors with particular focus on national missions in Europe and EO missions of ESA, such as the Pléiades, Sentinel-2 optical sensor missions  with the aim to better characterize the potential and limitations of these sensors and their synergy.

Work on improved satellite based rapid mapping geo-information extraction techniques is presented using relatively new algorithms leveraging existing partner R&D projects such as the Copernicus PREFER (Space-based Information Support for Prevention and REcovery of Forest Fires Emergency in the MediteRranean Area) project in the fire and fire impact mapping domain, capitalising on existing knowledge while trying to fill in holes.

In this paper potential methods for the improved (semi-) automatic and operational detection of burnt areas using VHR and HR optical data (Pléiades-1A/1B and Sentinel-2 type), plus ancillary data sets, are presented (e.g. digital elevation models, topographic information, reference data sets, land-cover information).  

The work explores change detection in burnt area mapping and moreover in an operational rapid mapping context. Methods are chosen to be easily repeated, require little adjustment, take into account pre and post fire imagery (multi-temporal) and use normalised indices to reduce terrain effects and eliminate clouds. The implementation of algorithms is through Orfeo Toolbox (OTB). OTB development is supported by the French space agency (CNES) and constitutes a library of open source remote sensing image processing tools.

The selected methodologies are tested over Reunion Island and Mediterranean sites from a thematic accuracy and speed of execution perspective, given the objective of incorporating them into operational rapid mapping work-flows. Furthermore, the HR optical imagery work is important in the light of the arrival of Sentinel-2 data whose distribution will complement existing data sources.

Finally, in the context of fire impact assessment this paper explores how the integration of 3D information can lead rapidly to the delivery of fire related soil erosion vulnerability mapping, widening the rapid mapping portfolio with the objective of meeting user needs. This kind of product would also be interesting within a risk and recovery context.

Mahabad dam, one of the ten largest dams in Iran, is an enbankment dam which has built on the Mahabad river close to the city of Mahabad. According to studies conducted in recent years based on campaign GPS measurements, an event of landslide has been detected in right side about one kilometer away from the dam crest. Since any movement of the crust around the lake, such as landslides can endanger the structural stability of the dam and the safety of residents near the dam, it is necessary to monitor these natural hazards continuously to predict the risk of hazards and prevent damages. Therefore, in this study we tried to evaluate the rate of landslide around Mahabad dam based on InSAR technology. To end this, single look complex (SLC) images of ascending and descending track of Envisat ASAR between 2004 and 2007 were analyzed. We used Small Baseline Subset(SBAS) algorithm and producing persistent scatterer (PS) in the area to estimate the rate of surface deformation. Based on the Envisat ASAR images we found the rate of deformation along line of sight of the satellite about 5 mm/yr.  Furthermore, few number of PALSAR, ALOS images over this area were processed analyzed.  We want to validate different InSAR against each other also against deformation derived from GPS measurements. The final result and comprehensive analysis about surface deformation in the study area will be provided.

Based on the case studies and statistical analysis of earthquake-related ionospheric disturbances mainly from DEMETER satellite, ground-based GPS and ionosounding data, the statistical characteristics of earthquake-related ionospheric disturbances are summarized, and the LAI coupling mechanisms are checked and optimized. According to the existing knowledge of seiesmo-ionospheric interaction as well as the requirements of earthquake monitoring and prediction research, China Seismo-Electromagnetic Satellite (CSES), the first space-based platform of Chinese earthquake stereoscopic observation system, was  proposed with the main objectives to check the LAI model and to distinguish earthquake-related ionospheric disturbances. And coincide with the mission objectives, the satellite decides to use the Circular Sun Synchronous Orbit with altitude 507km and descending node time at 14:00LT. The payload assemble includes 8 instruments, Search-Coil Magnetometer, Electric Field Detector, High precision Magnetometer, GNSS occupation Receiver, Plasma Analyzer, Langmuir Probe, Energetic Particle Detector, and Three-frequency Transmitter. According to the planned schedule, the satellite is due to be launched in the end of 2016 and the life time is 5 years.

Earth observation provides an increasingly stronger contribution to disaster prevention and crisis management. This helps both decision makers and emergency services in place. By the setup of the sentinel fleet, as well as the extension ofnationalGermanmissions, the number of available satellite sensors and the overflight frequency over risk- and disaster areas is increasing. Thus a higher spatial and temporal coverage is given for the derivation of crisis information. It is expected that inthefuture, user requirements can be better served and that an improved operational use of satellite systems with optimized products or services will be achieved. For this purpose it is essential to explore the possibilities of new sensors as well as the synergistic benefits of different systems in order to fall back in the case of crisis situations on a profound basis for decision making.

By „Dikes under pressure“ a profound basis for decision making will be created for the analysis of the data fromERS, ASAR andSentinel-1/2as well asfromGermannational missions (TerraSAR-X, RapidEye), allowing an optimized and synergistic use of these systems to monitor the technical and ecological vulnerability and resilience of dike landscapes. The focus is on the investigation of the stability of dikes and the dike surrounding area in presence of natural processes such as floods and tides, as well as anthropogenic changes using Synthetic Aperture Radar (SAR) interferometry. In the projectpossible vertical and horizontal movements/changes in the dike body and dike surrounding areaswill beobserved and therefore an efficient method for the monitoring of dikes will be developed. In the project methods based on remote sensing for the detection of possible land use changes as a result of fairway deepening in important river basins, along with a potential subsidence of the dike surrounding areas and a possible salt intrusion in the groundwater body will be developed. During prolonged flood events, the horizontal pressure of the water causes an increased risk potential on the dike body. In this project by the synergistic use of SAR and optical sensors, existing algorithms will be further developed to detect openwater bodiesas well as underflooded vegetation/water bodiesand dike percolations. Their size and volume will be used as an indicator for the condition of a dike and also for the analysis and validation of deformations identified by means of SAR interferometry

In this poster presentation, first results of the project dike under pressure will be shown along with an discussion of the potentials of this approach for a dike monitoring system from space.

Over the course of the last decade, large populations living in vulnerable areas have led to record damages and substantial loss of life in mega-disasters ranging from the deadly Indian Ocean tsunami of 2004 or the Haiti earthquake of 2010 to the Tohoku tsunami of 2011, and the astonishing extent of the environmental impact of the Deepwater Horizon explosion in 2009. These major catastrophes have widespread and long-lasting impacts with subsequent recovery and reconstruction costing billions of euros and lasting years. While satellite imagery is now commonly used immediately after many disasters to support rapid damage assessment, satellite data is rarely used to support recovery planning and monitoring. The Committee on Earth Observation Satellites (CEOS) led the creation of a Recovery Observatory Oversight Team, with a view to creating a Recovery Observatory. This team brings together major recovery stakeholders such as the United Nations Development Programme (UNDP), the World Bank/Global Facility for Disaster Reduction and Recovery, and the Joint Research Centre of the European Union under the Copernicus EMS Risk and Recovery project, as well as value-adding providers and leading space agencies. The principal aims of the Observatory are to:

Demonstrate the utility of a wide range of earth observation data to facilitate the recovery and reconstruction phase following a major catastrophic event;

Provide a concrete case to focus efforts in identifying and resolving technical and organizational obstacles to facilitating the visibility and access to a relevant set of EO data; and

Develop dialogue and establish institutional relationships with the Recovery phase user community to best target data and information requirements;

Foster innovation, R&D products around high-technology applications to support Recovery.

The CEOS Recovery Observatory Pilot will cover a multi-year period, beginning with a preparatory phase, in which satellite agencies establish the information technology infrastructure to receive products and data, determine data needs and define potential information products to support the recovery planning and monitoring process. This is undertaken in collaboration with international recovery stakeholders and value adders, and may be tested operationally through support to Post Disaster Needs Assessments (PDNAs) in order to prepare for the creation of a single large-scale Observatory. With preparation essentially complete, the ROOT is now monitoring global events with a view to triggering the creation of an Observatory in 2015/2016. This paper presents the work conducted in preparing for the Recovery Observatory both in terms of product definition and collaborative infrastructure. In addition to the preparatory phase, the Observatory includes:

Triggering;

Establishment; and

Operations.

Once triggered and established, the Observatory aims to offer free and open access to data and information useful in planning and monitoring recovery, but also to serve as a forum of exchange and collaboration on recovery related issues to foster resilience at the community level. A new opportunity will be given to scientists to develop/validate some promising methodologies specially from X-band and L-bands sensors before being demonstrated in this operational context. The Observatory is an open partnership of volunteer organizations and new members are welcome to propose contributions that enhance the functionality of the Observatory. After a six month period of operations, the Observatory will be evaluated and lessons learned will be documented to support broader use of EO during the recovery phase for future events.

Multi-temporal InSAR methods are effective tools for monitoring and investigating surface displacement on Earth based on conventional radar interferometry. These techniques allow us to measure deformation with uncertainties of one millimeter per year, interpreting time series of interferometric phases at coherent point scatterers (PS). Over the last decades, coastal areas in many parts of Spain have undergone a continuous urban expansion because of the growth of cities and development of new residential areas. The transgression of the sea, as a consequence of sea level rise and the subsidence of populated areas, may result in serious problems to many constructions situated in the coastline. This has an important impact on the economy, environment and society, representing a considerable natural hazard. We use ERS-1/2 and Envisat data in the period 1992-2010 to detect subsidence areas over the southern Spanish coast using time series analysis of SAR data. 

As reported by the European Environmental Agency (EEA) in several publications referring to the landscape of Austria and Switzerland, land-slides and flash-floods already cause more and more damages to infrastructure and settlements. Growing demand on land increases the competition for the limited resource. Areas at higher risk of flooding and land-slides are now used for housing and infrastructure.

Floods are presents. The last example are the floods from October 2015  caused by heavy rain and killing at least 19 people near France's Mediterranean coast. More than 350 mm (14 inches) of rainfall in the Var department in southern France in a few hours, triggering flooding that surged in some places to two metres over normal water level.

In the frame of the ESA – ARTES 20 feasibility study an international group of experts worked on the feasibility of an increasing flood forecasting system quality for the protection of critical infrastructures in Medium sized catchments.

According to the state of the art analysis for flood forecasting systems as well as the gap analysis, and the general structure usually used for such systems it is possible to define the specific EFFORS system and service as illustrated in the figure below. The colors used to distinguish the different modules are enumerated bellow.

Grey module: pre and post-processing

Green module: external model calibration

Yellow / red module: shell for real time flood forecasting

The team of experts mentioned in the sketch (abbreviations in red) are listed below:

JR: JOANNEUM RESEARCH Forschungsgesellschaft mbH. JR is responsible for new sensor, data communication and hydrological modelling.

SLU: Company for Remote Sensing and Environmental Research. SLU is responsible for Land Use EO data

TUG: Graz University of Technology, Institute of Hydraulic Engineering and Water Resources Management. TUG is responsible for the 2D hydraulic model calibration.

UWM: Ingenieurbüro für Umweltmanagement und Wasserwesen. UWM is responsible for programming the flood forecasting shell and the module intra and inter modelling.

ZAMG: Austrian Central Institute for Meteorology and Geodynamics. ZAMG is responsible for weather EO data, sensor, snow simulation and the warning dissemination in the frame of the flood forecasting centre management

External: partners not defined at the moment

As flood forecasting results should be understood and used from a very large user community (from private persons to firefighter) it was necessary to identify their major requirements. In total, five of them each representing a group of requirements must be fulfilled by the EFFORS system so that an added value for the users can be provided. This includes exemplarily:

Update of informationfor the users generaly several times a day.

Forecast of threshold values of water levels or of qualitative meteorological information.

Access via systems hosted by the service provider

Availability 24/7

Cost efficient for the user to receive enough benefit out of the measures to be taken 

The western branch of the East African Rift is threatened by a rare combination of several types of geohazards, while it is also one of the most densely populated regions of Africa. These geohazards can globally be classified as seismic, volcanic and landslide hazards. Landslides are possibly the most important geohazard in terms of recurring impact on the populations, causing fatalities every year and resulting in structural and functional damage to infrastructure and private properties, as well as serious disruptions of the organization of societies. This study makes part of the RESIST project, which contributes to the understanding of the source mechanisms driving volcanic eruptions and landslides in the area extending from North Tanganyika to North Virunga, based on remote sensing analyses with ground-based (climatic, seismic) networks for validation. The source mechanisms underlying landslide triggering and dynamics in the region of interest are still poorly understood, even though in recent years, some progress has been made towards appropriate data collection. Demand for mitigation strategies is high in this populated region, especially in the urban areas where demography is soaring. Here we focus on the city of Bukavu in the South Kivu region where more than 15% of the area is affected by landslides of various types and where population has more than doubled over the last two decades. To overcome difficulties of field accessibility in this area, we opted for a multiple sensor approach to monitor ground displacements, i.e., InSAR data from different satellites (Envisat, COSMO-SkyMed), and very high optical resolution images (Pléiades). First displacement results have been validated in the field using DGPS measurements and terrain observations. Patterns of the movements are irregular and range between few centimeters to several decimeters in a few months. To fully understand the mechanisms behind the landslides, acquisition of satellite images enlarging the multi-temporal timeframe for displacement analyses is needed.

Available and accessible natural and raw material resources are essential in today’s economy. The impact of mining activities at local, regional and national level is subject to scientific investigations and analyses, mainly related to environmental and societal impacts. Despite strict regulations and safety procedures in place by law at every stage of mining operations, there is an urgent need describing the potential and existing influences and changes using up-to-date, reliable and standardizes Earth Observation (EO) data during the different phases of the mining life cycle. This include not only topographic information for the mining site itself, moreover information on industrial infrastructure and urban development are of interest.

Since December 2010 the two German satellites TerraSAR-X and TanDEM-X are operating jointly in a helix tandem formation. The primary mission goal of the TanDEM-X mission is the generation of a seamless global multi-coverage DEM using single-pass interferometry in bi-static mode. The outstanding quality and resolution of the DEM will allow topographical applications at an unprecedented level of detail; therefore the possibility of monitoring mining activities with up-to-date and high resolution DEM is given. Thus, in this paper mining areas are investigated using different multi-date acquisitions of the TanDEM-X dataset.

For our investigations we concentrate on several mining sites with strong changes in topography, e.g. the Appalachian Mountains, whereas in the Northern and Central Appalachian Basin the mountaintop removal mining takes place and the open pit Mpumulanga Coalfields Region in South Africa. Both regions are highly dynamic coal mining area with continued growth and expansion of active mining and associated industry. For both sites a complete coverage for the years 2011 and 2012 and partly for the years 2013 and 2014 do exist.

The methodology is involving TanDEM-X mission data that has been acquired operationally since December 2010. The base idea of the presented approach is the usage of separately generated DEM. For this purpose, for every time stamp a separate DEM will be generated. For a mass estimation an accurate registration of the multi-date DEMs is crucial. We use the absolute calibration correction of the DEM Calibration and Mosaicking Processor calculated by a block adjustment of a larger region. That means the relative accuracy of the input DEM scenes is within some decimetre. This allows us to perform for comprehensive high-resolution analysis like DEM change comparisons. Values like mass balance are derived. To validate and compare these results current reference data are used, namely Laser DEMs and ICESat.

References:

(1)     Krieger, Gerhard, Alberto Moreira, Hauke Fiedler, Irena Hajnsek, Marian Werner, Marwan Younis,  Manfred Zink: TanDEM-X: A Satellite Formation for High-Resolution SAR Interferometry, In: IEEE Transactions on Geoscience and Remote Sensing, Vol. 45, No. 11, 2007.

(2)     Rossi, Cristian, Fernando Rodriguez Gonzalez, Thomas Fritz, Nestor Yague-Martinez, Michael Eineder: TanDEM-X calibrated Raw DEM generation, In: ISPRS Journal of Photogrammetry and Remote Sensing 73, 12-20, 2012. 

(3)     Gruber, Astrid, Birgit Wessel, Martin Huber, Achim Roth: Operational TanDEM-X DEM calibration and first validation results, In: ISPRS Journal of Photogrammetry and Remote Sensing 73, 39-49, 2012.

(4)     Ruppert, L.F., et al: The US geological Survey’s national coal resource assessment: the results, In: International Journal of Coal Geology 50, 247-274, 2002.

The project LIFE EBRO-ADMICLIM puts forwards pilot actions for adaptation to and mitigation of climate change in the Ebro Delta (Catalonia, Spain), an area vulnerable to sea level rise and subsidence.  The project proposes an integrated approach for managing water, sediment and habitats (rice fields and wetlands), with the multiple aim of optimizing ground elevation (through inputs of inorganic sediment and organic matter), reducing coastal erosion, increasing the accumulation (sequestration) of carbon in the soil, reducing emissions of greenhouse gases (GHG), and improving water quality.

The main pilot adaptation actions focus on sediment inputs from the Ebro River into its Delta. The aim is to demonstrate the feasibility of permanently restoring the sediment flow, both from a water treatment plant on the Ebro and from the reservoirs along the lowest stretches of the Ebro. For this purpose, it is of key importance to achieve an accurate estimation of the subsidence phenomenon affecting the area, up to now monitored at only a few points using levelling techniques.

This work proposes a study of the subsidence phenomenon over the Delta area based on the application of coherence-based differential interferometric techniques (DInSAR). To this end, the whole SAR archive available at ESA over the area of interest will be analyzed. It covers an observation time window of approximately 20 years, from 1992 to 2010, and is made up of 8 differential stacks: 3 descending and 1 ascending stacks for ERS1/2, and 2 descending and 2 ascending stacks for Envisat. For their processing, the Grid Processing on Demand (G-POD) service of the European Space Agency (ESA), based on the Small-Baseline (SBAS) technique, will be employed. Once retrieved a deformation profile of coherent pixels from each stack, a data fusion methodology will be put forward to merge the whole information and generate a final deformation rate map and an estimation of the deformation profile of the coherent points. Finally, the Differential Interferometric Software developed at the Institute Cartogràfic i Geològic de Catalunya (DIS-ICGC) will be used to process a stack of 12 ALOS-PALSAR images covering the time span between 2007 and 2011. The results of the two studies will be compared over the overlapping period of the C-band and L-band sensors, and the agreement between the two estimations will be analyzed.

Finally, preliminary results in terms of deformation process retrieval using Sentinel1 data will be shown. 

In 2015 the United Nations adopted the Sendai Framework for Disaster Risk Reduction and the Sustainable Development Goals. Both are accompanied by targets and will be further elaborated through indicators that are action oriented, global in nature and universally applicable. For several of these indicators adequate data is available only from remote sensing, in particular for human settlements. However, although there is a plethora of satellites producing images of the Earth at different spatial resolutions there is today no system that is monitoring globally and consistently human settlements.

This presentation describes an innovative concept for the mapping of human settlements at regional and global scales. The Global Human Settlement Layer (GHSL) concept was implemented and tested at the JRC using new, in-house developed methods for information extraction. The methods are based on image data sequencing and symbolic machine learning by association analysis. These techniques are well established tools in, for example, genome characterization studies, but so far not used in the remote sensing domain.

As a result of the open and free data access policies of the United States (for Landsat) and the European Union (for Sentinel) it is for the first time possible to produce with these methods cost-effective fine-scale global and regional settlement information. At the global scale the settlement maps were produced with historical Landsat data that go back to the beginning of civilian Earth Observation in the 70ies; collections of the year 1975, 1990, 2000, and 2014 were processed. The monitoring of future developments relies on the European Sentinel-1 and Sentinel-2 satellites of the Copernicus Programme with the first global update produced by the end of 2016.

However, providing global information is not sufficient for policy support. Therefore indicators were developed in collaboration with decision makers that are ready for uptake in international frameworks such as Sendai or the SDG’s. The presentation provides examples of such indicators and how they are integrated in the international frameworks.

Slope deformations are a significant geodynamic phenomenom for the Czech Republic territory. They are responsible for considerable damage to technical objects as well as on the property of the population. Neovolcanic range of the Ceske Stredohori Mts. in  northwestern Czech Republic represents the region with susceptibility to various types of landslides. The basic conditions for slope deformations of a landslide type in Ceske Stredohori Mts. is that at least two different rock complexes of uneven strength characteristics are found together.

Slopes in less resistant rocks keep inclinations higher than critical one which leads to permanent instability. Monitoring of their activity and causes has been carried out by IRSM ASCR for more than 30 years. Detailed geodetic and geotechnical in-situ monitoring in two localities of landslides (Trebenice, Dolni Zalezly) has started in 2013. We want to fit  upcoming measurements (mainly laser scanning, trigonometric measurements and close-range photogrammetry) to dates of SAR-image acquisitions.

Evaluation of landslide activity using InSAR can reveal valuable information both in spatial scale (tens to thousands square kilometers) and in temporal scale. On the other hand, we cannot obtain full spatial information in hilly terrain of Ceske Stredohori Mts., another limit is presence of dense vegetation. Hence, we would like to apply two approaches in our research, (1)  to delimit an area of few hundreds of square kilometers and to process as much as possible of SAR images using multitemporal InSAR techniques (PSI, SBAS, dInSAR and other) to acquire of spatiotemporal distribution of active landslides. We assume that we can identify a suitable amount of signal backscatters represented by buildings at foothills or rocky outcrops on the left bank Labe river valley. If some landslide locality will show clear pattern of activity we would like to perform additional processing for its active time period using differential interferometry; (2) For selected localities we want to choose SAR images with suitable dates of their origin (following the date of event, dates of monitoring etc.) always for every single processing for each of locality. After evaluation of preliminary InSAR results, presented here, we will aim to upgrade and optimize in-situ monitoring to be better verifiable and comparable with results of InSAR processing.

X-band radar signal determines a behaviour of objects (and man-made structures) which we can use for landslide monitoring as are the parts of railroads, roads, bridges and tunnels (for highway), buildings and rocky outcrops both in active and inactive area. L-band data penetrates through vegetation more in depth and can delimit whole landslide extent for each of selected locality. Sentinel-1 C-band guarantee us a continuation of an input of SAR data that are used for comparison with recent in-situ monitoring.

Conventional differential Interferometric Synthetic Aperture Radar (dInSAR) techniques are routinely used in accurate deformation mapping including landslide activities. However, several difficulties arise when attempting to use dInSAR in areas with steep topography, high humidity and dense vegetation cover such as the Three Gorges Region. In the case of very fast slope movements, due to the limitations of dInSAR with regard to the maximum detectable displacement (MMD), dInSAR techniques are not able to correctly measure these high landslide rates.

As an alternative method, sub-Pixel Offset Tracking (sPOT) method has recently been employed to derive centimetre-level land movement from artificial Corner Reflectors (CRs). Continuous landslide activity in the Shuping area was measured using the sPOT method applied using TerraSAR-X (TSX) 1m Spotlight data acquired from February 2009 – April 2010 and January 2012 – February 2013. The landslide rates show a consistent seasonal pattern with a dramatic increase of landslip from May to August in both 2009-2010 and 2012-2013, coinciding with substantial changes of the reservoir water level during the flooding season. These results suggest a strong connection between the formation of landslides and the operation of the Three Gorges Dam as well as seasonal rainfall. A further statistical analysis was carried out on the deformation measurements derived from artificial Corner Reflectors versus natural scatterrers in densely vegetated terrain. The results indicate that even sPOT measurements in areas suffering from lower dInSAR coherence, are still able to correctly measure ground deformation range in densely vegetated terrain.

With the availability of super high resolution TerraSAR Staring Spotlight (TSX-ST), the ground deformation monitoring of Shuping as well as Tanjihe landslide site in the Three Gorges Region is continued using the offset tracking techniques to achieve a higher accuracy. In addition, the potential of TSX-ST data of measuring surface deformation using dInSAR techniques is re-assessed in this region, in particular whether the improvement of the resolution of Staring Spotlight mode helps to address some of the issues that were encountered previously. The use of dInSAR techniques on Staring Spotlight mode in the densely vegetated terrain is evaluated in terms of the accuracy, temporal de-correlation effect, integer phase-cycle ambiguities present in phase unwrapping, maximum detectable displacement of dInSAR, etc.

This work is partially supported by the CSC and UCL through a PhD studentship at UCL-MSSL.

A tunnel in the surroundings of Dobrovskeho street serves to enhance transportation in Brno city. Two tunnels of length around 1200 m were minted in Neogene clays. Before the excavation, exploration drifts have added information about the geological structure of the rock mass and give first information about settlement of the overlying Neogene low reinforced rocks. Geotechnical survey was conducted with a broad complex methods in order to achieve the most complete and reliable information on the geological structure and the physical and mechanical properties of the mass along the tunnel and its surroundings.

Since the beginning of work with the wider vicinity of the tunnel is monitored by an extensive set of methods. Before the construction of the tunnel and during its construction, over 900 leveling points, 300 inclinometric points and 300 deformetric points were measured within 2008-2012. The task of monitoring is to monitor deformation of both the tunnel and also of all buildings on the surface, to ensure their safety. Currently, the warranty monitoring is performed that spans over 100 leveling points. Because of number of checkpoints was destroyed during the construction of the tunnel and other works in the surrounding area, the satellite radar interferometry (InSAR) techniques should be performed to extend the complex of monitoring methods.

For the work, we have achieved 20 Radarsat-2 acquisitions in fine beam mode within ESA project C1P.21629 - Evaluation of Potential Threats to Stability of Linear Structures using InSAR Technology. These acquisitions show deformations in Brno city between August 2014 and October 2015 with a regular step of 24 days temporal difference. Also, we have additionally achieved a series of 75 Cosmo SkyMed images with temporal step  every 16 days in average, for dates between June 2011 and July 2014. The Cosmo SkyMed dataset partially overlaps with the reference measurements of tilt and height changes. After the end of the intensive measurements, the PS InSAR time series can deliver knowledge about continuation of movement and depict the date of final stabilization of the area. The accuracy can be validated using the limited number of the continuing warranty levelling mission.

We have realized that the available dataset can be used also for monitoring of other events. In March 2015, a cavity was detected under a house at 28. dubna square in Brno-Bystrc. The cavity of radius around 5 m appeared due to wrongly constructed waste canals. A subsidence was detected due to the cavity and the house had to be evacuated due to real danger of destruction. It appears that also neighboring road was affected by subsidence due to presence of the cavity. Data from Radarsat-2 should detect such movement and indicate whether an early warning could have been delivered by detecting the cavity before it was visible by the family living in the house.

For this work, we use PS InSAR technique implemented in SARPROZ software. The processing is performed at supercomputer center IT4Innovations in Ostrava.

Systematic detection and monitoring of upland peat moorland wildfires in the UK is critical to conservation groups to understand the extent of damage and to monitor natural post-fire recovery and peatland restoration measures. The collection of burn scar perimeter data and the rate of vegetation recovery in the field is labour-intensive and time-consuming, limiting the spatial coverage of regular monitoring. Since upland peat moorland areas in the UK are regularly covered by cloud, burned area characterization based on existing optical satellite data is limited. With the launch of Sentinel-1A, Sentinel-2A and upcoming hyperspectral missions (e.g. PRISMA, EnMaP), regular monitoring of moorland areas for burn scar detection is possible using SAR and optical data in synergy.

The aim of this study is to use aerial high resolution hyperspectral data to validate multi-temporal and multi-sensor techniques to compare and contrast the ability of optical (Landsat 7 ETM+) and radar (ASAR and ERS-2) data to detect the small upland peat moorland burn scar located in the Peak District National Park (PDNP), in northwest England. A wildfire on 26 May 2008 resulted in a 0.10 km^{2 burn scar on the Kinder plateau near Edale, PDNP. The fire moved quickly across dwarf shrubs and moorland grasses, driven by easterly winds and burnt into the peat soil in places (McMorrow et al., 2010). Aerial hyperspectral images from the twin Eagle-Hawk sensors was collected five weeks after the wildfire consisted of imagery by the Natural Environment Research Council (NERC) Airborne Research and Survey Facility (ARSF). These imaging spectrometers operate in the VNIR-SWIR spectral range with a spatial resolution of 1.5m/pixel for Eagle. Unfortunately, atmospheric correction and mosaicking of the airborne imagery was not sufficient for a differenced Normalized Burn Ratio (dNBR) calculation. We carried out a supervised classification of the hyperspectral data by using a Support Vector Machines (SVM) algorithm, as implemented in the EXELIS-HARRIS ENVI 5.1 software to validate Edale burn scar products derived from ASAR and ERS-2 (25m/pixel) and Landsat 7 (30m/pixel). Four classes –burned area, vegetation, bare peat and cloud-covered vegetation –were selected to train the algorithm. A modal filter was applied to the classification result (Figure 1) and prior comparison with both satellite data types (Figure 2). Preliminary results show the feasibility of using SVM classification of hyperspectral image to validate burn scars in upland peat moorland environments delineated by ASAR, ERS-2 and Landsat 7 ETM+ satellite data. Data fusion of optical and radar data is explored for burn scar enhancement are now being explored.}

The new generation of space-borne Synthetic Aperture Radars (SAR) and optical sensors provides metric or sub-metric resolution imagery, thus allowing, in principle, the detection of single building damage after an earthquake. However, the complexity of the image forming mechanisms within urban settlements, especially of radar images, makes the automatic detection of damaged buildings  still a challenging task. The integration of different pieces of information, not only from remote sensing but also from geological and structural data sources, may help providing reliable damage assessment. Such an integration is foreseen in the APhoRISM (Advanced PRocedures for volcanic Seismic  Monitoring) FP7 project which is the framework of the present study.

In this work we will present a semi-automatic procedure exploiting Very High Resolution images acquired before and after an earthquake from both SAR and optical sensors for providing  damage assessment products at single building scale. In order to test the proposed methodologies we use optical images from QuickBird satellite and Spotlight COSMO-SkyMed SAR imagery of the seismic event that hit L’Aquila city (Italy) on April 6, 2009. For validation purposes two ground based damage maps are used. The first one refers to the survey performed by the Istituto Nazionale di Geofisica e Vulcanologia(INGV), while the second one is related to ground survey carried out by the Italian Civil Protection Department (DPC).

Dealing with metric and sub-metric resolutions, object-based change detection approaches are proposed. For segmenting optical images a GIS layer reporting building footprints is used. This allow the change analysis to be focused on the objects of interest, avoiding false alarms due for example to vegetation changes and temporary objects. As for SAR data, because of  the complexity and peculiarity of building appearance in radar images, an ad-hoc segmentation technique of the pre-event image has been developed. It is based on the use of morphological profiles to extract bright stripes and ridges caused by double bounce and/or layover mechanisms, the most distinctive features of the SAR building response. Looking at changes in these regions heavy damaged buildings can be identified. When a building collapses changes are also expected within the building footprint and in the shadow area. Typically an increase of the backscattering is observed in these regions due to the scattering contribution from debris and to the return coming from the ground previously occluded by the shadow. In order to single out such kind of changes a segmentation approach exploiting the Kullback-Leibler Divergence (KLD) is proposed.

A deep analysis of many change detection features, evaluated at object scale, is done by assessing their correlation with damage information provided by ground surveys. As for SAR data, the intensity ratio, the interferometric coherence, the intensity correlation and the KLD are analyzed. Regarding optical data, features describing texture and color changes are considered in addition to statistical similarity and correlation descriptors, such as the KLD and the Mutual Information. Exploiting these features, a non parametric classification approach based on the Bayesian Maximum A Posterior criterion is implemented for both SAR and optical data.

The classification performances are not excellent when tested using the available ground truth, but a similar uncertainty has been observed comparing the INGV ground truth with that provided by the Italian DPC demonstrating the challenge of an accurate damage assessment even on ground (considering the difficulties encountered after an earthquake). SAR and optical data allow comparable performances in terms of damage sensitivity. Better performances in terms of false alarms rate are found using optical data. An improvement of the results is expected from the data integration approach foreseen in the APhoRISM project, which is presently under development and will be also depicted.

RASOR (Rapid Analysis and Spatialization Of Risk) platform performs multi-hazard risk analysis for the full cycle of disaster management. A scenario-driven query system simulates future scenarios based on existing or assumed conditions and compares them with historical scenarios.

RASOR integrates diverse data and products across hazards, update exposure data quickly and make scenario-based predictions to support both short and long-term risk-related decisions.

Global applicability of such system is subjected to data availability at relative high resolution.   

The RASOR has access to high resolution datasetsat global scale,(such as 12m resolution tanDEM-Xby DLR),that allowsa fairly detailed implementation of hazard models embedded into the platform, aimed to predictsthe effects of earthquakes, tsunamis, floods, landslides, subsidence, etc.). However, globalscale exposuredatasetson exposurecontain information only at national/provincial scale and cannot be directly used on damage computation.

To sort out this inconsistency, RASOR incorporates downscaling methodologies for exposure data to be consistent to high-resolution hazard layers.

Considering buildings, lumped census data or EO-based imageries provide information on different construction parametersataggregated onlargescaleareas(percentage of buildings with). Through those parameters it is possible to define typical building classes and relate them to a vulnerability model. Data on those classes need to be disaggregated usingahigh resolution EO-based built-up mask layers(e.g. GlobalHumanSettlement Layer by Joint Research CenterRC ….. citare….)) to obtain a spatial distribution at a reasonable fine resolution. Mapping schemes techniques may also be used to spatially distribute disaggregated datato increase results’accuracy. 

The same downscaling procedure can be applied to other assets such as population, crops, GDPproduction.  Global population data(….. citare….)can be further disaggregated at settlement resolution (30m), taking into account country-based societal habits (working hours, etc.). Coarse resolution data on agricultural sites and crop production may be disaggregated cross-comparing with finer EO-based land cover layers. 

This methodology has been successfully implemented in data poor context, such as Malawi, allowing for a complete flood risk assessment. Exposure andThe study presents …floodhazard layersat a relative high resolutionform thebasis for arisk assessmentthatproducesfiguresaimedto increase scientific-supported awareness of risk at the national level and sub-country level. This can help to protect people and properties of Malawi against future floods. 

Robust estimates of kinematic earthquake source parameters (location, seismic moment, fault geometry) and knowledge about the estimation uncertainties are essential for earthquake physics in general, and especially for reliable seismic hazard assessment and the investigation of large-scale and complex tectonics. Today, source parameters are routinely estimated and also published in global catalogs (e.g. GCMT and ISC), using teleseismic data. The associated uncertainties, however, are usually not presented, although parameter estimates can vary greatly for the solutions of the source location and mechanism, especially for very shallow earthquakes. Modeling of earthquake source parameters by using Interferometric Synthetic Aperture Radar (InSAR) data is also a widely used technique, which has a comparably high spatial resolution and good resolution on shallow-depth fault slip. Comparisons and/or joint inversions between these two datasets are frequently made, although typically no full uncertainty analysis is performed. Knowledge about the case dependent uncertainty of modeling parameters is as important as the modeling process itself and enables the development of inversion techniques of both InSAR and seismic data for the calculation of more robust source models.

This study explores the sensitivity on earthquake source parameters estimated from InSAR and teleseismic data separately for the commonly used rectangular dislocation models and seismic moment tensors, respectively. We consider correlated data errors and estimate in a Bayesian framework model parameter trade-offs and uncertainties of the earthquake sources in comparison. Special regard is given to exploring the data-dependent resolvability for geometrically complex segmented faults.

We present two real-data case studies of the 06.04.2009 L'Aquila earthquake (Italy, ~Mw 6.3) and the more complex 25.08.2008 Zhongba earthquake (Tibet, ~Mw 6.7), as well as synthetic tests. For the InSAR data modeling anrectangular dislocation sourcein an homogeneous elastic half-spaceis assumed. For the seismological data we use a double-couple moment tensor and a standard 1D seismic velocity model,employingpre-calculatedGreen's Function databases.

We optimize the source model parameter with Monte-Carlo direct search algorithms. Gaussian initial probability densities are assumed for the linear inversion and the observational error and the mismodeling errors are estimated for both datasets.

Afterwards a coupled sensitivity analysis on a parameter basis between the two datasets is carried out.

A comparison of the source model parameter resolution from both datasets and their uncertainties is then carried out.

The project aims to answer the feasibility of InSAR and seismological datasets towards the analysis of structural more complex fault system on a case basis and deliver information for maybe transferable and reasonable weighting of future coupled or joint optimizations.

The SAR data for this study have been kindly provided by ESA through the SOAR-EU2 program, project #16736.

Nearly 30 years have passed since the first operational active fire detection systems were established using 1 km resolution NOAA Advanced Very High Resolution Radiometer (AVHRR) Local Area Coverage data. It was only recently that a global day/night active fire detection product was achieved with a spatial resolution better than 1 km using the Suomi-NPP Visible Infrared Imaging Radiometer Suite (VIIRS) 375 m data. During that time, application of complementary finer resolution Landsat-class data in support of active fire mapping was limited by data policy and availability constraints. Nonetheless, Landsat-class short- and long-wave infrared data have been successfully used to map high temperature thermal anomalies (e.g., volcanic activity, gas flares and biomass burning) across different geographic regions. Among the proposed applications, wildfire mapping is perhaps one of the areas with greatest potential for growth in a data-rich environment. That environment is gradually becoming a reality thanks to free-data policies and the availability of new Landsat-class data sets leading to improved sampling capabilities. Here we describe the development of fire products building on those new sensor data, namely Landsat-8 Operational Land Imager and Sentinel-2 Multispectral Imager, and their application in support of wildfire mapping and management. Our results demonstrate the greatly improved fire line resolving capability of those new data sets, and their use in decision support systems. We conclude with the discussion of future scenarios where Landsat-class data would play a major role in bridging the gap between strategic and tactical fire hazard mapping applications, highlighting specific mission requirements to support that.

The subsidence of Venice and its lagoon has been systematically studied since the 1970s’ in connection with the frequency increase of flooding events during exceptional high tides. It is a well-known case study not so much for the subsidence magnitude but because each millimeter lost of elevation seriously threats the conservation of the historical center and the surrounding lagoon considering their small ground elevation with respect to the mean sea level and the eustatic sea level rise.

During the last two decades, the use of satellites instrumented with SAR (Synthetic Aperture Radar) sensors and the interferometric processing have provided excellent information on the land displacements. The first SAR analysis on Venice was carried out by DInSAR (Differential Interferometric SAR) and provided several thousands of measurements, i.e. much more than those obtained by the few hundred benchmarks monitored by levelling since 2000 (Tosi et al., 2002). More recently the application of Persistent Scatterer Interferometry (PSI) has remarkably increased the number of measurable targets, achieving accuracies comparable to that of leveling and permanent GPS (e.g., Teatini et al., 2005; Teatini et al., 2007; Teatini et al., 2012). With the launch of the new generation X-band sensors, it becomes possible to obtain hundred thousands of data with high spatial resolution and a short revisiting time, and detect differential movements of single churches, palaces, bridges, etc. (Tosi et al., 2012; Tosi et al., 2013). Therefore, PSI opens new possibilities for a more accurate interpretation of the land subsidence and progressively takes over the in-situ traditional measurements (i.e. levelling and GPS), reducing their use just for the calibration of the SAR-based methodologies.

The aim of this work is to provide an overview of the ground movements of Venice obtained by SAR-based interferometry on more than 20 years of acquisition data from multi-generation satellites including ERS1/2, ENVISAT ASAR, Cosmo-SkyMed, TerraSAR-X, ALOS-PALSAR, RADARSAT-2, Sentinel-1. In particular, we discuss on the capability of the outcomes from the different satellites to quantify the ground movements of the historical center of Venice. An image of the subsidence evolution of the City since 1992, including details of single structures of particular interest is also given. Furthermore, the natural and anthropogenic causes of the ground displacements are highlighted.

References

Teatini P, Tosi L, Strozzi T, Carbognin L, Wegmuller U, Rizzetto F (2005). Mapping regional land displacements in the Venice coastland by an integrated monitoring system. Remote Sensing of Environment, vol. 98, p. 403-413..

Teatini P, Strozzi T, Tosi L, Wegmuller U, Werner C, Carbognin L (2007). Assessing short- and long-time displacements in the Venice coastland by synthetic aperture radar interferometric point target analysis. Journal of Geophysical Research. Earth Surface, vol. 112, F01012, doi: 10.1029/2006JF000656.

Teatini P, Tosi L, Strozzi T (2012). Comment on "Recent subsidence of the Venice Lagoon from continuous GPS and interferometric synthetic aperture radar" by Y. Bock, S. Wdowinski, A. Ferretti, F. Novali, and A. Fumagalli. Geochemistry, Geophysics, Geosystems, 13, Q07008, doi: 10.1029/2012GC004191.

Tosi L, Carbognin L, Teatini P, Strozzi T, Wegmuller U (2002). Evidence of the present relative land stability of Venice, Italy, from land, sea, and space observations. Geophysical Research Letters, vol. 29, 1562 , ISSN: 0094-8276, doi: 10.1029/2001GL013211.

Tosi L, Strozzi T, Teatini P (2012). Cosmo-skymed versus TerraSAR-X -based interferometry for monitoring the mose settlements at the Venice lagoon inlets. In: Proceeding of the 32nd IEEE International Geoscience and Remote Sensing Symposium. IEEE IGARS Proceedings, p. 2837-2840, Munich: IEEE Geoscience and Remote Sensing Society and IGARSS 2012 Organizing Committee, Munich, Germany, 22-27 July 2012, doi: 10.1109/IGARSS.2012.6350841.

Tosi L, Teatini P, Strozzi T (2013). Natural versus anthropogenic subsidence of Venice. Scientific Reports, 3, 2710, doi: 10.1038/srep02710.

The subsidence of Venice, one of the most beautiful and famous cities in the world, is well known not by reason of the magnitude of the ground movement but because it has seriously compromised the heritage and the safety of the city in relation of its small elevation above the sea. In particular, the challenge of the new researches is the quantification of the natural land subsidence and movements induced by anthropogenic activities. Subsidence and eustacy have both contributed to the loss of land elevation with respect to the mean sea level (or relative sea level rise, RSLR). The natural subsidence rate is strictly related to the reference period (i.e. geological- and modern- term) (Teatini et al., 2012). Its quantification in Venice was obtained through various methodologies and not uniquely quantified. Recent researches are aimed at distinguishing between regional land subsidence and local consolidation processes, as well as between long- and short- term displacements. Due to the present low elevation of Venice with respect to the sea level, it is much more interesting to evaluate the natural displacement over the last few decades, i.e. the present natural land subsidence, than that averaged values over geological periods. PSI provides the cumulative land displacements (natural plus anthropogenic) of the investigated area. Using decade-long observations from the C-band sensors and short-time (e.g., annual) displacement records by high resolution X-band satellites, the land movement component primarily ascribed to natural processes and those induced by human activities can be distinguished, as recently demonstrated by the combined use of ERS/ENVISAT and TerraSAR-X (Tosi et al., 2013).

This research is carrying out in the framework of the COSMO-SkyMed/RADARSAT-2 Initiative (Canadian and Italian Space Agencies) and the Flagship Project “RITMARE” (National Research Programme funded by the Italian Ministry of University and Research) and is aimed at distinguishing between natural land subsidence and movements induced by anthropogenic activities in Venice and the surrounding lagoon using the long- and short-term displacements obtained by PSI on C- and X-band images, respectively. RADARSAT-2 long-term analysis (C-band dataset from 2008 to 2015) is used for quantifying the natural component of the land subsidence and the COSMO-SkyMed short-term analysis (X-band dataset of 2013-2015) for highlighting the anthropogenic displacements. The higher pixel resolution and shorter revisiting time of the X-band data enhance the detection of small-scale short-time movements. The task is of paramount importance in the framework of the geo-hazard risk management of the Venice coastland. The interferometric analysis is carried out at regional (i.e. the lagoon) and at local scales (e.g., the historical center of Venice, lagoon inlets, salt marshes) with a special focus on the lagoon inlets where the MoSE project is under-construction.

References

Teatini P, Tosi L, Strozzi T (2012). Comment on "Recent subsidence of the Venice Lagoon from continuous GPS and interferometric synthetic aperture radar" by Y. Bock, S. Wdowinski, A. Ferretti, F. Novali, and A. Fumagalli. Geochemistry, Geophysics, Geosystems, 13, Q07008, doi: 10.1029/2012GC004191.

Tosi L, Teatini P, Strozzi T (2013). Natural versus anthropogenic subsidence of Venice. Scientific Reports, 3, 2710, doi: 10.1038/srep02710.

The continuously increasing fleet of optical satellites yields enhance the availability of satellite image time-series with high spatial (e.g. Pléaides, Spot 6/7, WorldView-3) and temporal resolution (e.g. Sentinel-2, RapidEye, Venμs); offering new opportunities for the monitoring of geomorphological and tectonic processes. In particular sub-pixel correlation of optical satellite images has been demonstrated to be a valuable tool for measuring surface deformation resulting from co-seismic slip (Leprince et al. 2007), glacier flow (Heid and Kääb 2012) and landslides (Delacourt et al. 2004; Stumpf et al. 2014) with accuracies that can approach 1/5th of a pixel under ideal conditions. There are, however, also numerous variables such as imperfect co-registration, changing view angels, atmospheric effects, cast shadows as well as changes of the illumination and the surface aspect which easily lead to spurious deformation signals and mismatches. As a consequence, it remains challenging to exploit image correlation techniques for the detection of surface motion where the affected zones are not known a priori. This is especially the case if the surface displacement is small with respect to the pixel size and noise level.

Existing image archives and dense multi- temporal acquisitions should in principal allow redundant measurements over the same area that could be exploited to reduce the measurement noise but still very few studies have investigated this potential (Dehecq et al. 2015, Jeong et al. 2015).

The presented work proposes a multi-pair image correlation approach which matches several combinations of image pairs with variable geometric and temporal baselines to obtain redundant and more robust observations. The resulting measurements are analyzed regarding the spatio-temporal consistency of the observed velocity vectors. Four different indicator metrics for the consistency of the displacement are compared regarding there capability to distinguish real ground motion from spurious detections. The influence of the main algorithm parameters are assessed experimentally.

The technique is evaluated at a landslide prone study site in the Southern French Alps with a time-series of Pléiades stereo images and ground truth on the landslide extent and velocity. The analysis shows that the techniques is effective for the detection of surface motion over wide areas with complex terrain. The derived velocity vectors are also consistent with permanent ground-based measurements. The obtained multi-temporal landslide motion maps at the basin scale provide important insights in the seasonal dynamics of the landslide in relation to meteorological factors and implications for hazard assessment and regional sediment budgets.

Especially the image correlation step of the developed processing chain is computational expensive and was executed on high-performance computing cluster to reduce the processing time by a factor of 20. Higher-level analytics of processing chain are implemented using the R open-source platforms and libraries for parallel computing. Our current work is dedicated to the optimization of the processing chain and its integration in ESA's Geohazards Thematic Exploitation Platform with a particular focus on the analysis of Sentinel-2 time series. The processing chain will also include modules for satellite stereo-photogrammetry which will enable end users the end-to-end processing from multi-temporal satellite images to the final ground motion maps.

Dehecq, A., Gourmelen, N., Trouve, E. (2015). Deriving large-scale glacier velocities from a complete satellite archive: Application to the Pamir–Karakoram–Himalaya. Remote Sensing of Environment, 162, pp. 55-66

Delacourt, C., Allemand, P., Casson, B., Vadon, H. (2004). Velocity field of the “La Clapière” landslide measured by the correlation of aerial and QuickBird satellite images. Geophysical Research Letters, 31, (15), pp. L15619

Heid, T., Kääb, A. (2012). Evaluation of existing image matching methods for deriving glacier surface displacements globally from optical satellite imagery. Remote Sensing of Environment, 118, pp. 339-355

Jeong, S., Howat, I. M. (2015). Performance of Landsat 8 Operational Land Imager for mapping ice sheet velocity, Remote Sensing of Environment, 170, pp. 90-101

Leprince, S., Barbot, S., Ayoub, F., Avouac, J.P. (2007). Automatic and Precise Orthorectification, Coregistration, and Subpixel Correlation of Satellite Images, Application to Ground Deformation Measurements. Geoscience and Remote Sensing, IEEE Transactions on, 45, (6), pp. 1529-1558

Stumpf, A., Malet, J.P., Allemand, P., Ulrich, P. (2014). Surface reconstruction and landslide displacement measurements with Pléiades satellite images. ISPRS Journal of Photogrammetry and Remote Sensing, 95, pp. 1-12

The contribution is within the activities of a regional project funded by the Regione Campania (Southern Italy) and focusing on ground deformation studies in Campi Flegrei, west of the city of Naples.

The project has a double target: research and monitoring activities. This latter is addressed to the strengthening of the already existing geodetic networks through the installation of new continuous GPS (cGPS) stations and Corner Reflectors (CRs) for InSAR data exploitation, to be installed on areas characterized by low or no coherence and/or a clear deformation signal.

A relevant role in the project is the availability of long time-series of COSMO-SkyMed data on the area, from 2009 to date, besides the GPS time-series of some stations already available starting from 2000.

The Campi Flegrei are now undergoing an uplift phase started at the beginning of 2005, with an overall inflation of more than30 cmup to date, although with different deformation rates in time, suggesting the Italian Civil Protection Department to move form green (base) to yellow (alert) level in the framework of the Campi Flegrei emergency plan.

The availability of such long GPS and InSAR time-series, besides the processing of new data, will allow not only a comparison between in-situ and spaceborne data, but also to better define and constrain the 3D deformation filed acting in the area of interest, in order to model the deformation source.

In order to improve the infrastructures for deformation measurements, and to develop a multidisciplinary approach, new cGPS stations have been setup with, in addition, CRs for ascending and descending CSK data. Such multisensor sites are the reference points where to estimate 3D movements and cross-validate them. 

FireBIRD is an infrared remote sensing mission of the German Aerospace Center (DLR). The mission consists of two satellites, TET-1 and BIROS, both based on the DLR Bispectral and Infrared Remote Detection (BIRD) satellite and respective infrared sensor (2001-2004). The aim of the mission is to detect hot temperature anomalies for fire detection and fire characterization as well as thermal patterns for the analysis of ecological processes.

The spectral bands of FireBIRD comprise the green (460 – 560 nm), red (565 – 725 nm), near infrared (790 – 930 nm), midwave infarred (3400 - 4200 nm)  and longwave infrared (8500 – 9300 nm) part of the electromagnetic spectra. The thermal bands have a ground resolution of 356 m while the other bands have a resolution of 40 m.

The mission FireBIRD is a scientific mission. By using FireBIRD data scientific research questions, new algorithm development, and applications shall be fostered. All data is processed and archived by DLR and will be public available to registered users for scientific use. With its spatial resolution the thermal sensors has the potential of covering unique observation possibilities. Use cases comprise besides monitoring of vegetation fires for example the detection of gas flares, monitoring of volcanoes or monitoring of irrigated lands. In this presentation data products,  use case examples as well as the way of data access will be shown.

The seismic source models are estimated using Gravity Recovery And Climate Experiment (GRACE) data for several large undersea earthquakes, including the 2004 Sumatra-Andaman earthquake, the 2010 Maule, Chile earthquake, the 2011 Tohoku earthquake, the 2012 Indian Ocean earthquakes, the 2007 Bengkulu earthquake, the 2013 Mw 8.3 Okhotsk deep-focus earthquake, and the 2015 Mw 8.3 Illapel, Chile earthquake. The innovative processing of GRACE data using only the north component of gravity change and its corresponding gravity gradient changes allows the enhancement of the spatial resolution for coseismic deformation signals. Single double couple and multiple double couples inversion are conducted, which demonstrates the unique constraint on source models from GRACE data as compared to models inverted using other data sources. For the 2004 Sumatra-Andaman earthquake, GRACE inverted model produce a shallower centroid depth (9 km), as compared to the depth (28 km) from GPS inverted model. For the 2011 Tohoku earthquake, the GRACE-estimated centroid location is southwest of the GPS/seismic solutions, and the slip azimuth is about 10° larger. The GRACE-estimated dip angles are larger than those from GPS/seismic data for three earthquakes, i.e., 18° larger for the 2004 Sumatra-Andaman earthquake, 8° larger for the 2010 Maule, Chile earthquake, and 3° to 13° larger for the 2007 Bengkulu earthquake. We concluded that GRACE data contribute to the improved coseismic source modeling for these recent undersea earthquakes.

Landslides represent a serious hazard to communities around the world, with effects ranging from infrastructure damage and transport network disruption through to destruction of properties and loss of life. Slopes across the UK are susceptible to landsliding due to the country’s temperate climate and high levels of rainfall. Consequently, the cost of repairing roads and railways affected by these processes is high and the disruption caused to communities, especially those isolated within rural areas with limited transport options, is a cause of concern to residents and businesses. Whilst a number of methods for monitoring landslides have been employed, these typically rely on the installation of in situ sensors, groundwater measurements and other time and resource intensive activities.

This project proposes a methodology based on multi-platform remotely sensed synthetic aperture radar interferometry (InSAR). Data from ESA’s Sentinel-1 satellite mission is used to detect movements on slopes that may be indicative of future slope failure. The strength of Sentinel-1 is its ability to collect a large amount of data over a wide area. However it is limited by the return period, look angle and resolution. For example, because of Sentinel-1’s 12 day repeat cycle it is likely that a slope failure event could occur between the first detection of movement and the subsequent acquisition. In this case, the only data available is a “snapshot” of the hillslope before failure, providing only limited information about the event. Additionally, the geometry of the ground and particularly the orientation of the hillslope has a large influence on the satellite’s ability to observe the event. On the other hand, FastGBSAR is a ground based radar system that can gather displacement measurements at sub-millimetre resolutions with each measurement taking only a few seconds. It can be deployed in the field to collect data in greater detail and build a complex picture of a slope failure event before, during and after it occurs. Furthermore, FastGBSAR can be moved between multiple locations and view directions, and thus produce 3D displacement data of a site.  However, prior knowledge of where and when a failure is likely to occur is required in order to target the deployment of the system, as the time and cost of installing permanent ground-based radar systems across the country is prohibitive. Such a priori information can be provided by the Sentinel-1 satellite.

In this study, Sentinel-1 observations are used to detect active landslides in Northern Britain, and FastGBSAR is deployed to collect further data over the relevant field sites. An artificial embankment in Northumberland, undergoing induced failure, is monitored firstly with Sentinel-1 data to detect initial movements and followed up with a ground based approach with FastGBSAR to detect finer detail. This approach is then transferred to a larger site with active landsliding at the British Geological Survey’s landslide observatory at Hollin Hill. For the first time, it is demonstrated that a multi-platform InSAR approach using complementary systems has the potential to improve understanding of land sliding mechanisms and help mitigate the risk of these disruptive processes.

SAR interferometery and GPS, used at Mount Etna since more than twenty years, show that the volcano is characterized by a relatively stable western domain and a mobile eastern flank. The boundary between both is well defined to the north with a unique discontinuity, the Pernicana left lateral fault. The south boundary is more complex with several right lateral faults, e.g. the Mascalucia, Trecastagni, S.Leonardello, and Timpe faults. Some of them produce a well visible topography, e.g. the Pernicana, while others do not, e.g. the Mascalucia fault. Their motion, as shown by the data and discussed in several articles, combines creep activity and episodic co-seismic slips. The seismicity on those faults is known not to be steady with time, and the geodetic data also suggest the existence of temporal changes. The depth of those faults is not well established. They might cut only the volcanic products and not the sedimentary basement, and therefore allow the eastward motion of the recent (<300kyr) volcanic products pushed away from the volcano axis by the episodic injection of dykes feeding the eruptions.

In this work, we study the spatio-temporal variability of the motion along those faults in the period 1992-2010. We produced a series of ascending and descending interferograms using the AMI/ERS and ASAR/ENVISAT archive, with maximum baseline of 250 m and maximum time span of 4 years. Using published fault maps and our interferograms, we refine the location of the fault and we discuss the status of the transitions between faults. We calculate accurately the average slip velocities and temporal changes and compare with those published, and with the available GPS constaints.

In a second step we analyze the time series derived from PS and SBAS inteferometry. For the Pernicana and both Mascalucia/Trecastagni fault, a specific PS processing method for non-urban areas is used separately, starting from a multi-reference stack of inteferograms, a technique that is suitable for areas affected by non-uniform motion, fast line of sight (LOS) deformation rate and high decorrelation resulting in coherence loss over long time intervals.

SENTINEL 1 TOPSAR data acquired on Mt. Etna between October 2014 and October 2015 were processed by the GAMMA software, using a spectral diversity method and a procedure able to co-register the TOPSAR SLC pairs with extremely high precision (< 0.01 pixel). The DInSAR results are analysed and successively used as input for the time series analysis using the StaMPS package. In order to optimize the time processing, a new software architecture based on the hypervisor virtualization technology for the x64 versions of Windows has been implemented.

The SENTINEL data were used to support and detail the ground deformation recorded by GPS on Mt. Etna, during the last year’s volcanic episodes. In particular the SENTINEL images have clearly depicted the ground deformation associated with the dike intrusion occurred on December 28, 2014, when the eruptive activity resumed at Mt. Etna with a fire fountain activity feeding two lava flows spreading on the eastern and south-western upper flanks of the volcano.

GPS surveys carried out a few days after the eruption, and DInSAR Sentinel-1A ascending interferogram detailed the ground deformation field affecting the area surrounding the summit, showing decimetric displacement around the craters and on the westernmost part of the Valle del Bove area.

Just after an adequate number of data were available, the SENTINEl-1 TOPSAR descending acquisitions have been analysed by using thePersistent Scatterers Interferometryapproach implemented into StaMPS package, in order to produce the time series and to filter the atmospheric artefacts. In a second step, we will perform the analysis of ascending TOPSAR acquisition.

The mean velocity map obtained by using the Persistent Scatterers Interferometry technique, were validated and integrated with the dense GPS (more than 100 benchmarks) geodetic in-situ data collected on Mt. Etna, applying the SISTEM approach.

The SISTEM approach simultaneously integrates all the available datasets (i.e. GPS displacement vectors on sparse benchmarks and SAR displacement maps), providing the high-resolution 3D displacement map by taking advantage of the positive features of each datasets, i.e. the availability of multiple view geometries and the high temporal and spatial resolution of the SENTINEL C-band interferometric data, and the 3D displacements component provided by GPS with sub-cm accuracy level.

Weapplied theSISTEMmethod also to compute 3D high-resolution surface displacement maps of Mt. Etna related to ground deformations referring to May 2015 eruption; the results are consistent with the geophysical and field volcanological observation, showing a slight deflation of the summit crater area.

The first Ground Based Radar (GB-RAR) survey was carried out at Vulcano Island (Aeolian archipelago, South Tyrrhenian Sea) in 2014, for measuring the northern slope of the La Fossa Crater. The GB-RAR measurement is able to detect small changes in the shape of the volcanic cone, allowing monitoring the flank dynamics. This slope, in fact, is made by a succession of oblique tuff strata that produced a very steep morphology, which in turn generates an evident gravity instability. In the past, indeed, this sector of the La fossa Crater has been affected by significant slope failure producing also little tsunamis. Furthermore, the fracture systems associated to the gravity instability intersect the shallow hydrothermal system causing a wide active high-temperature fumaroles field in the upper part of the northern side of the crater. The fumaroles alter the rocks, promote the continuous rock falls and in general increase the slope instability. For this reason, during the GBRAR surveys, thermal measurements by an infrared camera were also performed. The data from the two remote sensing techniques have been compared and integrated in order to investigate about possible links between thermal releases and flank movements

The city of Lisbon is the largest city in Portugal and has been hit by catastrophic events in the past, such as earthquakes and tsunamis. Also, the geological properties of its location and the proximity to the Tagus River turn it into a city prone to ground instability.

Multi-temporal interferometric techniques enable a regional analysis. In previous studies using ERS-1, ERS-2 and ENVISAT imagery datasets applied to Lisbon, ground subsidence linked to groundwater extraction has been identified. One of the analysed locations is a densely populated area in the city centre.

This study presents a time-based analysis of the ground subsidence phenomenon obtained by the application of multi-temporal interferometric techniques with TerraSAR-X, Cosmosky and Sentinel-1A imagery datasets. INSAR processing was improved by using atmospheric phase screen maps retrieved from GNSS data provided by permanent stations.

The application of InSAR techniques for detecting and measuring surficial deformation processes of ground and structures is becoming more and more popular as the number and variety of SAR satellites increase. Availability of SAR images with various resolutions, and the possibility to get both archive datasets and future acquisitions (for historical analyses and monitoring respectively) makes InSAR extremely interesting for geological and engineering applications. In this regard, the development of A-DInSAR methods effectively extended such applications in different fields, such as civil engineering, engineering geology as well as geomorphology, among others.

In this work we present several case studies related to successful and significant applications of interferometric (InSAR) and Advanced Differential InSAR (A-DInSAR) analyses by using the specifically developed Sarproz software tool.

Large-scale geophysical processes (such as the coseismic deformation induced by the 2014 South Napa earthquake (California) detected by using Sentinel-1 images), as well as extremely localized processes affecting portions of structures or small landslides, have been successfully investigated.

As well known, the subsidence is one of the processes better observable by InSAR. In this case study, an alluvial plain located about 20 km east of Rome (Italy) characterised by groundwater exploitation for various purposes has been analysed by means of extensive processing at different scales of ERS and ENVISAT satellite images by means of A-DInSAR technique was performed. A multidisciplinary approach based on geological, geotechnical and hydrogeological modelling, combined with the analyses of time series of ground displacements allowed us to describe the space and time distributions of the subsidence process, thus inferring the local subsidence mechanism. Specifically, the non linear deformation of measurement points, observed thanks to a specific InSAR algorithm, allowed us to derive that the groundwater level variations drive the timing of subsidence triggering over the area, whereas the local geological conditions control the magnitude of the deformation process.

Moreover, several case studies related to landslides events have been investigated following their evolution over more than 20 years. In the first case study related to landslide C band ERS and Envisat stacks of data have been extensively used by A-DInSAR methods to measure deformation in an Alpine rockslide affecting a small village, thus defining the most active portions of the slope. A-DInSAR has been also used in another case study to investigate a landslide-prone area in central Italy where more than 90 phenomena affect sandy and clay slopes of a large valley hosting many villages and rural areas. More than 200 ERS and Envisat images in ascending and descending orbital geometries have been processed to derive the state of activity of each landslide between 1992 and 2010. In addition, the characterization of slope kinematics has been performed for larger phenomena, thus deriving the real deformation process, based on line-of-sight information.

In addition to methods based on multi-image algorithms (such as Persistent Scatterers Interferometry - PSI) adopted for the cases above, a specific approach based on the use of differential interferograms with pairs of images have been put into practice. In this regard, a particularly interesting case is related to the investigation of a landslide located on a coastal region in Oman. This analysis allowed to get valid information on past deformations even with less than ten SAR images, also thanks to the use of L-band data acquired by ALOS PALSAR (JAXA) between 2006 and 2010 in single and double polarization.

Also more recent high-resolution SAR data (namely COSMO-SkyMed and TerraSAR-X) have been processed with A-DInSAR methods (namely, PSI) in order to take advantage of larger number of measurement points achievable from X band data. This aspect has been useful, for example, for some landslides located in Alpine region and hilly areas in the northern Italy analysed by using large datasets of images acquired by the COSMO-SkyMed constellation between 2010 and 2014. The high-density of measurements points allowed to define the state of activity of more than 25 landslides and their evolution in the analysed period.

In addition to deformation processes affecting the ground surface, A-DInSAR has been positively adopted as a reliable tool also for structural health monitoring. In particular, three case studies related to different cases of dams worldwide have shown the potential of such approach for long-term investigations and also for historical analyses, where other monitoring data are lacking.

The aforementioned high resolution (spatial and temporal) typical of current X-band sensors, proved to be a key-point to characterize with high detail the structural deformations of two extremely sensitive dams, namely, the Three Gorges Dam (Hubei region, China), analysed by using COSMO-SkyMed data and the Plover Cove dam (Hong Kong) investigated with a long time series of images provided by TerraSAR-X and TanDEM-X sensors. On the other hand, one of the main advantages of the technique (i.e. the unique opportunity to measure displacements occurred in the past) allowed to define the historical conditions of stability of an earth dam in Italy in the years 2002-2010 thanks to the data archive acquired by Envisat in double orbital geometry.

To conclude, again regarding civil structures, the high-resolution data showed the same advantages for the analysis of high-rise buildings. More specifically, we used TerraSAR-X data to detect deformation processes at different heights of a high building. In this latter case, the used software allowed to separate the different deformation contributions for all measuring points such as the thermal expansion of materials and long-term deformational trends. This approach allowed to properly infer the overall deformation process in order to demonstrate the potential of the technique as a valuable tool for highly detailed monitoring and control of structures and infrastructures. 

Transport pipelines of natural resources such as natural gas, oil and their products are considered one of the safest onshore transportation systems of fluids between multiple locations (Papadakis 1999). However, the potential and actual occurrence of pipeline leakage is a reality that requires (through mandated government regulations) the pipeline operator to routinely monitor the Right Of Way (ROW) for issues that may impact the integrity of the pipeline. As examples, un-authorized man-made structures, intense vegetation overgrowth, large animal presence, seepage due to microfractures and major failures of the pipeline are all issues within the ROW that require identification, monitoring and, potentially, immediate action.

Considering the multitude of potential issues that need to be addressed in ROW monitoring, this study has focused on the development of reflected hyperspectral imaging and broadband MWIR (3-5micron) techniques to aid in the monitoring process. The primary data for this study has been obtained using repeat airborne campaigns over known ROW’s during a period of one year. The study area of this multi-temporal research project is the Enbridge 9 pipeline located along the central and north eastern coast of Lake Ontario, Canada. Airborne hyperspectral imagery using a CASI (Compact Airborne Spectrographic Imager), covering the visible near infrared range (376 to 1048 nm) and SASI (Shortwave Infrared Airborne Spectrographic Imager) sensor, covering the near to shortwave infrared wavelength range (870 to 2500 nm), and a FLIR (Forward Looking InfraRed) SC-8303 broadband MWIR (3-5 micron) thermal data was collected over the research area. Data was collected over six flight lines from October 2014 to October 2015.

Preliminary results (at the time of abstract submission) from the airborne campaign highlight the relative ease of detecting man-made structures with a wide variety of hyperspectral bands as well as the MWIR. A variety of vegetation indices (i.e. NDVI, SGI, etc.) have been calculated and will show the progressive encroachment of nearly all vegetation species into the ROW. At this point in the analysis, there is no uniquely identifiable vegetation stress that would indicate any issue related to the pipeline presence. Furthermore, no incidents along the ROW under study have been reported or identified in the data thus far. We also note that from the MWIR imagery, a number of potential false positive leak locations have been analytically removed as the high resolution of the FLIR system has been adequate to provide an indicator of mammal presence within the ROW. 

As there are millions of km’s of pipeline routes throughout the World, the ultimate goal of this work is to provide methodologies and real-world tools to the pipeline monitoring institutions to perform assessments of ROW’s from a spaceborne platform.  Therefore, upscaling the results of the airborne trials to the level of satellite platforms will be shown and results discussed. Using Landsat 8 for October 2014 and June 2015 and potentially the Sentinel-2 mission data for August 2015 (if data is available) will provides an excellent analysis on the state-of-the-art in the ability of spaceborne platforms to carryout critical infrastructure monitoring.

References:

Papadakis, G. A. (1999). Major hazard pipelines: a comparative study of onshore transmission accidents. Journal of Loss Prevention in the Process Industry, 12(1), pp. 91–107

Zahra Sadeghi^1,2, Mohammad Javad Valadan Zoej¹, Jan-Peter Muller²

^{1  K.N.TOOSI University of Technology, Tehran, Iran}

^{2 Imaging Group,  Mullard Space Science Laboratory, University College London, Holmbury St Mary, UK}

Monitoring the deformation rate of ground subsidence is critical for hazard management.

Differential Interferometric synthetic aperture radar (dInSAR) provides precise measurements of surface deformation caused by land subsidence. In order to avoid decorrelation effects and other issues in the deformation estimation, all Advanced differential InSAR (AdInSAR) techniques start by selecting only those pixels for processing with persistent scattering behavior over time.

In general, AdInSAR methods need a large number of pixel candidates to work properly. Average coherence for a full set of interferograms is a signal parameter employed as a quality criterion for pixel-candidate selection [1], where coherence is computed over multi-look spatial windows in the interferograms [2]. Optimizing coherence by combining the available polarimetry channels provides the possibility to increase the number of pixel candidates. With the introduction of Polarimetric SAR Interferometry (POLInSAR), a coherence optimization technique was presented by Cloude and Papathanassiou [3]. This method is considered as the most general one since it allows different polarization states at the ends of the baseline. Navarro-Sanchez presented a general coherence optimization method (Exhaustive Search Polarimetric Optimization, ESPO) for increasing the number of pixel candidates [2].

This work is aimed at using the ESPO method to improve AdInSAR in high-rate Land Subsidence phenomena over Tehran, Iran. The study area is covered by vegetation and suffers from high-rate deformation. Therefore, a low density of coherent pixels is the biggest difficulties in this case study.

A set of 8 stripmap dual-polarisation SLC images was acquired by TerraSAR-X from July 2013 to October 2013 with an 11 days interval. The processing has been applied over a section of the image of 2600  2000 pixels, which mostly cover rural areas.

 A dual-pol formulation for polarimetric SAR interferomety has reflectivity information from two combinations of transmit/receive polarizations. For each resolution element, a scattering vector  is obtained as a vectorization of its 2  2 scattering matrix.  The ESPO approach is formulated for maximization of the average magnitude of the interferometric coherence, estimated using a multi factor scheme [2012].

A 9  9 multilook  was used for coherence computation. In this approach, 7 interferograms have been generated assuming an unique master image. In the Coherent pixel selection step, multilook pixels with coherence higher than 0.7 were selected. All remaining interferometric processes containing phase unwrapping and deformation rate estimation were applied on selected Coherent pixels.

 Density of selected Coherent pixels has been improved more than 2.5 times in optimized channel comparing standard AdInSAR in VV channel.

 Therefore, it can be concluded that adding polarimetry information to AdInSAR in this rural case study was able to increase the identified number of Coherent pixels. In future, assessing different coherence optimization method and comparing with each other is suggested.  

 [1] P.Berardino, et al., “ new algorithm for surface deformation monitoring based on small base line differential SAR interferograms”, IEEE Trans. Geosci. Remote Sens, Vol. 40, no. 11.

[2] V.Navarro-Sanchez et al., “ a contribuition of polarimetry to satellite differential SAR interferometry: increasing the number of pixel candidate, IEEE Geoscience and Remote Sensing Letters, Vol. 7, No. 2, 2010.

[3] S. Cloude et al., “polarimetric optimization in radar interferometry,” Electron let, Vol. 33, No. 13, 1997,

Copahue-Caviahue volcanic complex is located in the province of Neuquen, Argentina, at 37.5°S, 71.1°W. It is constituted by the Caviahue caldera and the Copahue volcano edifice at the border between Argentina and Chile. Copahue volcano is one of the most active volcanoes in Argentina. Historical eruptions have been reported on 1992, 1995 and 2000. Two villages, situated at 5 and 9 km from the crater, are exposed to volcanic hazard. The active crater host an acidic hot lake with ph<1 and temperatures up to 60° Celsius.

Previous works addressing crustal deformation in this area were performed by using ENVISAT, ERS and COSMO-Skymed images acquired between 2002 and 2013 [1]. They show a deflation process centered at the edifice’s northern flank, which acted between 2004 and 2010 at a rate of 2 cm/year. Inverse modeling of these results placed the deformation source at 4 km depth below the volcano, suggesting a relationship with the oscillations of the brittle-ductile boundary where leakage of brines and steams would produce depressurization of the system [1].

After Chilean 8.8 Mw earthquake on February 27 2010, several changes where observed, mainly related with an increment in local seismicity inside Caviahue caldera. Furthermore, two recent eruptions occurred, one of them between Dec 2012 and  Dec 2013, the other between July and December 2014 [2]. SBAS-DinSAR time series show that, since 2011, the volcano inflates at a rate of 12 cm/year until at least middle 2013. Inflation pattern is located exactly in the same region where deflation was previously detected [3] [4].

In this work we analyze the suitability of ALOS2/PALSAR data for characterizing deformation related with the Copahue volcano area. Acquisition frequency in WBD (ScanSAR mode) is about one scene per month, which makes it potentially useful for volcano deformation characterization via time series processing.

We processed a dataset composed by 12 images acquired in WBD mode between November 2014 and October 2015. We were not able to compute coherent interferograms between scenes acquired during 2014 and scenes acquired after February 2015. However, interferograms between scenes acquired after May 2015 show very good degree of interferometric correlation. The issue with the first ones is possibly related with the ending of the ALOS2 commissioning phase, but still we do not have a conclusive argument about the matter which needs further investigation.

Coherent interferograms are still few for computing a deformation time series, but the obtained results with the most recent acquisitions are encouraging.

[1] María Laura Velez, Pablo Euillades, Alberto Caselli, Mauro Blanco, Jose Martínez Díaz. Deformation of Copahue volcano: Inversion of InSAR data using a genetic algorithm. Journal of Volcanology and Geothermal Research 202 (2011) 117–126.

[2] Simthsonian Institution – Global Volcanism Program. http://volcano.si.edu/volcano.cfm?vn=357090. Accessed on October 16th, 2015.

[3] M.L. Velez, P. Euillades, M. Blanco and L. Euillades. Ground Deformation between 2002 and 2013 from InSAR Observations. Copahue Volcano. Active Volcanoes of the World, Springer-Verlag Berlin Heidelberg (2016). DOI 10.1007/978-3-662-48005-2_8

[4] Rayner De Ruyt, Pablo Euillades, Leonardo Euillades. Identificación del Campo de Deformación Cortical del Complejo Volcánico Copahue (CVCC), mediante el Algoritmo DinSAR-SBAS. XXIII Jornadas de Jóvenes Investigadores. Grupo Montevideo. (2015)

Wildfires are among the most serious problems in California (CA) and in European countries with a similar Mediterranean climate, which is characterized by the semi-arid to arid biomes and vegetation types. Although they are a part of the natural cycles related to the regional climate, effective strategic fire management is necessary due to highly variable weather conditions, pervasive drought durations, and significant urban sprawl/growth creating an array of wildland-urban interfaces (WUI) across the extensive built environment. These complexities demand new scientific advances to assess impacts of both natural and anthropogenic factors in a changing climate on wildfire danger to enhance the efficacy of proactive fire management with the utilization of new capabilities from satellite observations of land cover and land use change, including urbanization and ecosystem change.

In the present study, we show effects of large-scale atmospheric circulations on an early start and then extended length of fire seasons. More specifically, we find that multiple climate indices such as North Atlantic Oscillation (NAO) and El Niño-Southern Oscillation (ENSO) are closely related with weather conditions in the southwestern United States region, and they affect local wildfire potential in spring and early summer. This is determined by obtaining a 34-year record of Keetch-Byram Drought Index (KBDI) variability using multi-decadal reanalysis datasets to investigate impacts of multiple climate indices on precipitation, temperature, and the regional wildfire potential characterized by KDBI. Then for an effective fire danger assessment, we successfully test the capability of satellite vegetation indices (VIs) in replicating in-situ live fuel moisture (LFM), which is a critical parameter used by fire agencies to determine fire danger levels in Southern California.  VIs can be derived from data collected with multispectral sensors aboard satellites such as MODIS on Aqua and Terra, VIIRS on Suomi NPP, MERIS on Envisat, and MSI on Sentinel-2. To estimate LFM from satellite data, we identify the relationship between satellite VIs and seasonal/interannual characteristics of in-situ LFM to develop an empirical model function of LFM. Advancing beyond previous studies based on point-wise comparisons, we examine the LFM relationship with VIs averaged over different areal coverage, radially extended by 0.5 km to 25 km from each LFM sampling site in chamise-dominant areas of the chaparral ecosystems.

Moreover, an innovative method, called the Dense Sampling Method based on the patented invention of the Rosette Transform, has enabled the use of data acquired by satellite scatterometers (e.g., QuikSCAT, SeaWinds) to delineate and monitor decadal urban change including lateral sprawl and vertical growth from increasing urbanization across the world in various areas of the built environment including WUI. Together with the satellite-VIs capability for a break-through improvement of the spatial and temporal coverage by more than an order of magnitude compared to current manual methods for LFM sampling, we discuss practical applications of our results to significantly enhance fire danger assessment and prediction.  Beyond California, these results will be directly relevant to European countries in the Mediterranean climate.

Differential interferometry syntethic aperture radar (D-InSAR) is an efficient way to detection and measure the displacement of the Earth's surface. So that using this technology allows for continuous monitoring of small movements of the Earth's surface, with high precision and in a wide range is possible. Because of the wide coverage of satellite images, timeliness and low cost of this technology in them than other methods of field investigation of geologic hazards such as landslide detection, subsidence, earthquakes and volcanic activity is very is common. In this study, the differential interferometric techniques was applied to monitor and determine the amount of displacement of landslide in 70km south of the city Semirom, Isfahan in Iran. For this purpose, the six radar images of ASAR sensor ENVISAT satellite were selected and processed by DInSAR technique. As a result, initial processing, from the images the produced couples, the five couples related to 2010 and 2011 was appropriate further processing. The results of processing of radar images revealed a maximum vertical displacement of the sliding in 2010, from June 24 th 2010 to October 2010 period of 105 days, equivalent to 5 cm, and in 2011, the displacement maximum of the sliding mass of image pairs, September 2011 and October 31, 2011 with a 120-day period, equivalent to 6 cm respectively.

Using DInSAR[1] techniques in the study of earth surface movements such as subsidence, acceptable results can be achieved. In recent years this way was considered by many researchers. The results of this method can be compared to verify the geotechnical methods and Used for interpreting movements occurred. In this paper also to evaluation the process of Damaneh plain subsidence in Iran, from sensor radar data ASAR, 4 images related to the period from 2005 (track 250) and 3 images related to the period 2011-2012 (track 457) was used. These data were processed in software SARSCAPE and was determined the amount and level of subsidence From 5 interferogram obtained. Therefore subsidence rate was obtained in the period of 8 months of 2005, and ranged from 20110712 to 20120108, 7.6 cm and 7 cm, respectively. Subsidence in the northern part of Damaneh plains, Isfahan in Iran has spread greater, which that was adapted with joints and gaps been picked in the field.

[1] Differential interferometry Radar 

The SAR (Synthetic Aperture Radar) data are widely used in order to monitoring deformation phenomena impacting the Earth surface and infrastructures. The main remote sensing technique to extract centimeter information from SAR data is the Differential SAR Interferometry (DInSAR), based on the phase information only. Specifically, this technique is based on an appropriate combination of differential interferograms generated by image pairs characterized by a small orbital separation, in order to limit the spatial decorrelation phenomena. However, it is well known that DInSAR technique may suffer for lack of coherence among the considered images. New Earth observation SAR satellite sensors, as COSMO-SkyMed, TerraSAR-X and PAZ, acquire imagery on any point of the Earth with high resolutions, in terms of phase and amplitude value. Thanks to this higher amplitude resolution and to the use of on board dual frequency GPS receivers, which allows to determine the satellite orbit with an accuracy at few centimetres level, the SAR images offer the capability to achieve, in a global reference frame, positioning accuracies in the meter range and even better. The goal of this work is to exploit the slant-range measurements reaching centimetre accuracies using only the amplitude information of SAR images. The leading idea is to measure the distance between the satellite and a well identifiable natural or artificial Persistent Scatterer (PS). To this aim an interferometric repeat-pass stack (N SAR images) and one single same side image acquired on a different orbit are used. Then the positioning accuracy of a PS along the SAR line of sight is evaluated starting from this reference position and the orbital information supplied in the SAR imagery metadata, and accounting both for signal propagation delays and for geophysical effects causing not negligible PS displacements. In order to evaluate the PS displacements, an automatic matching procedure has been implemented to measure the PS image coordinates in all the stack images. In details, it is necessary to recognize and match the same PS across the images of the considered stack with sub-pixel accuracy (millimetres to centimetre level). As mentioned, in order to evaluate the positioning accuracy of a PS along the SAR line of sight it is necessary to correct for the signal propagation delays through the troposphere and ionosphere and to filter out the known geophysical effects that induce periodic and secular ground displacements. The tropospheric delay is close to 2.5 meters in zenith direction at sea level at mid latitude, and the needed correction along the slant path at better than centimetre level can be estimated by GNSS phase observations collected in the area imaged by SAR, using a proper mapping function. The ionospheric delay in X-band is in the order of few centimetres and can be computed by a global model (e.g. Klobuchar). The geophysical effects consist primarily in the solid Earth tides, polar tides, crustal deformations due to ocean loading, which globally can be at the level of half meter and can be corrected using the International Earth Rotation Service Conventions; moreover, geodynamics, crustal deformations due to atmospheric loading, glacial isostatic adjustment and the effect of the seasonal hydrological loading have to be accounted for. The preliminary results, obtained on the Berlin area (Germany) (TerraSAR-X HR-SpotLight ), shown that it is possible achieve a slant-range positioning accuracy with a bias well below 10 cm and a standard deviation of about 3 cm; the results are encouraging for applications of high resolution SAR imagery amplitude data in land and infrastructures monitoring.
 Thanks to the encouraging results this methodology could be conveniently adopted to monitoring deformation phenomena affecting both extended natural areas and localized man-made structures. 

Extreme flooding events are an important risk in many regions around the world. They affect the ecosystems and involve socio-economic consequences for dwellers. There is a real need to upgrade remote sensing techniques to mitigate and better manage natural disasters, and improve citizen's security and wellness.
Remotely sensed imagery has long been used in water resources assessment and it is a powerful source of information to support crisis management. It is available where and when there are no ground-based measurements, and it is especially needed for large-scale views and for the development and deployment of early warning systems.
This presentation shows the use of Deimos Imaging’s DEIMOS-2 very-high-resolution satellite imagery during the severe flood events affecting the Ebro River (Spain) in 2015.
Moreover, the main features of the DEIMOS-2 system are described, highlighting its 24/7 Rush services that have been provided to support crisis management, and that are now available to users all across Europe in the frame of the Copernicus Emergency Service.
During February-March 2015, the levels of the Ebro River raised significantly, causing in numerous occasions the break-up of the banks. Thousands of hectares were inundated causing unprecedented destruction and damage to crops, cattle, infrastructure and houses. Whole villages where affected.
Under these circumstances, satellite imagery acquired by DEIMOS-2 has been crucial to estimate the effects of the flood, allowing the Ebro Hydrographic Confederation (Confederación Hidrográfica del Ebro) to create a vector coverage of the water body throughout the whole Aragon region.
DEIMOS-2 has actively monitored the evolution of the flooding by providing images and value added products. Deimos Imaging 24/7 rush services, including near real time download and delivery, supported the fast damage assessment of the flooding. In addition, a full coverage of the area of interest was acquired during the flood event by using the high temporal resolution of the system.

Tungurahua volcano has been intermittently erupting since 1999, with observed deformation between 2007-2011 using Interferometric Synthetic Aperture Radar (InSAR) from the ALOS satellite of the Japanese Aerospace Exploration Agency.  Our recent analysis during that time period has provided insights into the characteristics of the subsurface, suggesting multiple connected magma chambers underneath the edifice with distinct temporal and spatial behaviors.  However, the previous source models are too simplistic and fail to incorporate realistic physical properties and forces acting within the volcano that would generate the observed deformation.  We use deformation data from InSAR and solve for the optimal deformation source parameters with Finite Element Methods (FEM) by incorporating into the models the material heterogeneities (e.g. temperature, density, elastic, viscoelastic), and stresses (e.g. background stresses, edifice loading, magma chamber overpressure) acting on the volcano.  We attempt to integrate previous multidisciplinary volcanological studies with our results to constrain the subsurface characteristics and understand the volcanic processes generating the observed deformation.

Cotopaxi volcano entered a new eruptive phase during August 2015 after nearly 73 years of eruptive quiescence.  We use the COSMO-SkyMed satellites from the Italian Space Agency, which acquired data over Cotopaxi during 2014-2015, to search for indications of volcanic deformation covering the new eruptive phase.  Preliminary analysis of interferograms do not show evidence of significant deformation related to the 2015 eruptions, but we can not exclude that there was ground deformation of less than ~3 cm.  We attempt to make an InSAR time series analysis over Cotopaxi volcano to search for indications of subtle deformation signals related to the eruptions.

Vegetation fires significantly affect and shape Mediterranean ecosystems with the strongest events occurring during the hot and dry summers months and in Southern countries (e.g., Portugal, Spain, Southern France, Italy, Greece). Mapping the areas affected by fires is the first step for the assessment of the relevant impacts on the vegetation and the atmosphere. However, systematic monitoring at the regional scale is feasible only by relying on satellite data. Satellite data acquired by optical sensors have long been used for monitoring wildfires and mapping burned areas by relying on the effects of fire on vegetation health and the straightforward interpretation of the burn spectral signature. However, the exclusive use of optical data poses some limitation for operative applications due to cloud cover, which prevents from observing the Earth surface and the spectral overlap between burned areas and low albedo surfaces. In this framework, the combined use of SAR (Synthetic Aperture Radar) and optical images could enhance monitoring capabilities. Due to its dual-pol capability and short revisiting time, ESA Sentinel-1 mission [1] offers a unique opportunity for monitoring fire events and the implementation of dedicated methodologies addressed to burned area mapping. Previous studies conducted in Mediterranean regions and focusing on the identification of burned areas from SAR images have highlighted the numerous variables that could influence the radar response pertaining to the occurrence of a fire; among them wavelength and operative polarization, vegetation structure and conditions, topography, dielectric properties of soil and vegetation as well as environmental conditions such as rainfall events. In particular, the availability of ESA Sentinel-1 A (S1A) data for over one year opens to the possibility of systematically investigating burned areas in Mediterranean vegetation through the analysis of C-band radar signature.

In this work, we preliminary analyze C-band backscatter as a function of operative conditions (e.g. local incidence angle, polarization) as well as vegetation and fire characteristics. These issues are addressed by processing multi-temporal dataset of S1A images over different study sites in southern Europe, covering heterogeneous land cover and environmental conditions. The relevance of the information conveyed by SAR images for burned area mapping is discussed, and inherent limitations are critically addressed also with reference to available optical observations. These analyses are also addressed to the integration of S1A data into a fuzzy burned area mapping algorithm, which was originally developed in [2], and further developed (for optical and microwave data integration) and tested with ENVISAT C-band ASAR data and Landsat Optical data over Portugal [3]-[4]. The proposed algorithm relying on fuzzy set theory represents a multi-source approach to burned area mapping, thus offering a highly flexible framework for the integration of SAR and optical data. Preliminary tests have been carried out by jointly using Landsat-8 OLI and Sentinel-1 SAR data; in addition, the proposed approach is flexible and can be suitably  adapted to  jointly use Sentinel 1 and Sentinel 2 datasets.

[1]     F. De Zan, A. M. Guarnieri, “TOPSAR: Terrain Observation by Progressive Scans”, IEEE Trans. on Geoscience and Remote Sensing, vol. 44, no.9, pp. 2352–2360, 2006.

[2]     Stroppiana, D.; Bordogna, G.; Carrara, P.; Boschetti, M.; Boschetti, L.; Brivio, P.A. “A method for extracting burned areas from Landsat TM/ETM+ images by soft aggregation of multiple Spectral Indices and a region growing algorithm”, ISPRS J. Photogramm. 2012, 69, 88–102.

[3]     D. Stroppiana, R. Azar, F. Calò, A. Pepe, P. Imperatore, M. Boschetti, J. M. N. Silva, P.A. Brivio, and R. Lanari, “Remote Sensing of Burned Area: A Fuzzy-Based Framework for Joint Processing of Optical and Microwave Data” Proc. of IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015, Milan, Italy, July 2015.

[4]     D. Stroppiana, R. Azar, F. Calò, A. Pepe, P. Imperatore, M. Boschetti, J. M. N. Silva, P.A. Brivio, and R. Lanari, “Integration of optical and SAR data for burned area mapping in Mediterranean regions”, Remote Sens.vol.7, no. 2, pp.1320-1345, Jan. 2015.

Excessive withdrawal of underground fluid particularly ground water is one of the most important factor of land subsidence formation that can cause extremely expensive damages to concrete buildings, roads, rail tracks, pipelines, well casing and can also cause flood. A necessary step to perform a proper analysis of the land subsidence is to obtain accurate measurements of the actual amount of subsidence at certain intervals. Nowadays the Interferometry Synthetic Aperture Radar (InSAR) ability to detect the surface deformation in millimeter exactness has been demonstrated in many countries of the world. This study used satellite InSAR technique and Persistent Scatterer Interferometry (PSI) method to measure ground motion over Port-Harcourt city from February 2007 to May 2009. The research was initiated as a pilot project to bring out the beauty, precision and the reliability of InSAR technique over conventional methods in ground deformation monitoring. This work presents an accurate and ground motion result for Port-Harcourt city. The measurement points delivered are derived from an analysis of 18 descending pass from Synthetic Aperture Radar (SAR) images obtained from the processing of ENVISAT radar satellite data with the Stable Point Network (SPN) Interferonmetric software. The analysis of the result from the study area revealed that more than 52.83% of the Port-Harcourt city is undergoing low subsidence velocity of between -2.5 to -5mm/year, while 20.66% of the area is experiencing high subsidence velocity of values of more than -5mm/year. 26.36% indicates a stable ground of velocity values between -2.5 to 2.5 mm/year. 0.12% of the study area revealed a low uplift velocity of between 2.5 to 5mm/year and 0.02% of the area indicates a high uplift with a velocity value of more than 5mm/year. No points are obtained in heavily vegetated region or in areas in descending mode layover. This displacement values are given relatively to a reference point located in the city center.

This work presents an integration of various remote sensing datasets and a geotechnical modelling to perform a morphological reconstruction of a volcanic area, Mount Etna, where rapid changes can occur.

In January 2011, a sequence of short and violent eruptions started at the “New Southeast” crater (NSEC) of Etna, leading to the formation of a cinder cone that has is presently the highest peak of the volcano.. This cone is located at the top of the steep slopes of Valle del Bove and it represents an element of concern because of its potential destructive evolution.

A multi-sensor and multi-temporal dataset has been gathered to investigate the morphological changes of the summit area of the volcano, above ~3000m, especially around that new crater. WorldView and Pleiades stereopairs have been used to calculate high resolution Digital Elevation Models and orthoimages at different epochs between 2011 and 2014. COSMO-SkyMed interferometric pairs provide additional information.

Finally, a probabilistic model has been developed to assess the stability of the NSEC, quantify the possible unstable volumes and define the covered area from the a potential collapse. The modeling has been implemented by introducing strength parameters from volcanoclastic-like materials.

From a geomatic point of view, our approach may constitute an advance in the monitoring capability of inaccessible areas through the combination of various high resolution optical and SAR interferometry space images.

In the future data from other missions could boost the capabilities of our method, such as those acquired by Sentinel 1, active from April 2014.

Our probabilistic model, once further validated, could contribute to define a methodology to evaluate instabilities induced by mechanisms influencing the load of the magma column at the base of a cinder cones that, in case of sufficient pressure, may be destabilized and the gravitational load of the cones themselves.

In relation to hazard assessment issues, such investigations contribute to the understanding of the causes that can determine a collapse of even greater proportions and to the definition of risk mitigation actions for downslope human activities.  

Finally, in a large and high volcano in continuous and persistent activity like Mount Etna, it is not always easy to conduct monitoring activity from the ground, and with a persistent hazard, especially near the summit. Therefore the approach of multi-sensor assimilation observations gathering various EO data is promised to becoming increasingly used in the future also considering a deeper assimilation with ground based and UAV observations.

The Gulf of Corinth (Greece) has been long identified as a site of major importance due to its intense and high variety of past geophysical activity. It has one of the highest seismicity rates in the Euro-Mediterranean region occurred from a variety of fault mechanisms.  The rifting mechanism observed is crucial for the stability of the region as it can lead to submarine slope failures, tsunamis and surface landslides.

Landslides are recognized as natural hazards having a major social–economic impact and represent a significant risk for citizens as well as infrastructures. During the last years, a significant increase of landslide occurrences has been recorded globally as a result of increased urbanization and development, continued deforestation and severe meteorological events. In Greece, this increase exceeds the 20% of the totally recorded cases during the last fifteen years.

Landslide phenomena constitute a major geological hazard in Greece and especially in the western part of the country as a result of anthropogenic activities, growing urbanization and uncontrolled land – use. More frequent triggering events and increased susceptibility of the ground surface to instabilities as consequence of climate change impacts (continued deforestation mainly due to the devastating forest wildfires and extreme meteorological events) have also increased the landslide risk. Monitoring and acquiring in near-real time the dynamic kinetic situation of certain regions of high risk, using remote sensed data, is of paramount importance for public authorities dealing with civil protection, natural hazards and large infrastructures.

In the current study we have investigated the ground deformations by means of multitemporal interferometry, caused by creeping faults and landslides in the area of South Gulf of Corinth. Specifically, the landslides of Platanos and Panagopoula, Karia-Souli, the Psathopyrgos fault and two man-made ground deformations were studied. The Psathopyrgos fault zone, which is considered to be a presently active structure, acts as a transfer zone between the Corinth and Patras rift.

In order to support our findings and enrich our knowledge in the temporal evolution of the studied phenomena we have used a big volume of SAR acquisitions from four satellite SAR sensors, with different characteristics.  Specifically, we have used archived ASAR/ENVISAT acquisitions to assess the sensor detecting and measuring capabilities for the specific landslides and very high resolution archived RADARSAT-2 over the test site of Karia - Souli. In parallel, very high resolution new SAR acquisitions of TERRASAR-X, RADARSAT-2 and COSMO-SkyMed, over the test sites (Panagopoula and Platanos) were used to produce mutitemporal interferometry deformation maps.

Land subsidence due to ground water overexploitation is one of the most significant geo-hazards in many areas of the world. It is also an ongoing problem for many groundwater basins in Iran causing varied forms of ground failure including sinkholes, rural building collapse, fissures in agricultural lands and damage to infrastructures. Monitoring the spatial and temporal pattern of land subsidence is essential for better subsidence hazard mitigation in the country. In this study, different SAR data are utilized and processed to map subsidence area in Marvdasht plain located in Fars province, south of Iran. The plain is among the largest producers of wheat in the country. Due to lack of precipitation and rainfall in recent years the plain has been dependent heavily on groundwater in order to provide the required water for irrigation of agricultural lands. We use more than 40 SAR archive from Envisat, ALOS PALSAR, TerraSAR-X (TSX), ALOS-2 and Sentinel-1 and process the data using multi-temporal technique of Small Baseline Subset (SBAS) to map the deformation area. The results and subsidence maps are presented and discussed in full paper and are analyzed with the information from head declines and subsurface geology to assess rheological properties of the aquifer system in the region.   

The most of water reservoirs located in mountainous regions are potentially affected by a number of geohazards. The most significant among them are landslides which activity might decrease reservoir’s volumetric capacity, damage roads, buildings and transmission infrastructure. Moreover, landslides could be harmful for the dams. Historical records, as well as numerous previous studies, show that unexpectedly triggered and rapidly moving landslides, and debris flow may initiate devastating local tidal waves.

Monitoring of all slopes surrounding reservoirs by conventional in-situ techniques (e.g., geodetic surveying, GNSS, borehole instrumentation, etc.) is technically challenging and expensive task. Thus, remote sensing techniques and more specifically the SAR interferometry, are an obvious alternative for monitoring the entire water reservoir and surroundings. Nevertheless, most of reservoirs are located in mountainous areas. This can be often a limiting and challenging factor for the interferometric analysis. Mainly due to the low coherence due to very rough topography, sparse urbanization, dense vegetation, forest cover and relatively long winter season with the snow cover.

In order to give a give a quantitative and qualitative assessment of the applicability of InSAR for this specific application, a number of validation campaigns on pre-selected test sites have been performed. Considering the above mentioned challenges in the context of InSAR applicability, two landslides were selected as the optimal candidate sites. One in Poland (around Roznow Reservoir in Nowy Sacz area), and another one in the area of reservoir in the Three Gorges in China. Both reservoirs are located in mountainous areas and surrounded by the slopes that are affected by landslides. Moreover, both sites are equipped with in-situ monitoring systems, and these data are used for both validation and fusion of/with InSAR results.

In this study, we applied a two-step approach. In the initial step, SAR Interferometry (DInSAR) technique is used to obtain a preliminary results on the areas with the high displacement rates. This initial step serves as a “localization tool”, where areas of strong deformation are localized both in terms of its location and magnitude of the signal. Because of the short revisit time TerraSAR-X data is used in this initial step. In the second step initial DInSAR results are utilized as a-priori information and extended with additional X-band (TerraSAR-X) data analysis including SpotLight and Staring SpotLight modes, and analysis of the archives of historical C-band (ERS 1/2 and Envisat) and L-band (ALOS) SAR data. In the second analysis step, multitemporal interferometric techniques for time series analysis to estimate the velocity of the landslides is used. The key research questions of this study is to explore the applicability of archived SAR data (ERS-1/2, Envisat, ALOS PALSAR) and their synergy with TerraSAR-X and in-situ terrestrial measurements to improve and validate detectability of deformation signal in InSAR challenging mountainous regions.

The results from InSAR analyses for both areas were examined against landslide inventory maps and the obtained deformation velocities are compared with in-situ monitoring data and laser scanning data (terrestrial and airborne). It was found that the landslide activity reported by InSAR match very well with the mapping inventory and it allows to fine-tune areas of contemporary activity.

Acknowledgments: This research is supported by ESA Dragon 3 Project (ID10606), DLR research project for the scientific utilization of Staring Spotlight (GEO2477), and PGI-NRI Landslide Counteracting System project (SOPO) funded by National Fund for Environmental Protection and Water Management of Polish Ministry of Environment

To answer a social debate related to environmental issues of shale gas exploitation and to check whether any of ongoing fracturing activities in Poland introduced any measurable surface deformations, Polish Geological Institute - National Research Institute (PGI-NRI) initiated the demonstration project that aimed to measure and monitor any of such potential terrain surface deformations.

For the project purposes 3 test areas surrounding the hydraulic fracturing sites has been selected. For each of the sites monitoring infrastructure has been designed and installed. In order to detect a sub-millimeter displacement the monitoring system consisting of geodetic (leveling and GNSS) benchmarks, and specially designed the corner reflectors (CR) for InSAR were deployed. This geodetic monitoring network was also strengthened by additional leveling and GNSS benchmark has been installed outside of expected potential deformation zone. In terms of SAR, data from TerraSAR-X and Sentinel-1 are being acquired. Also the historical data analysis (ERS-1/2 and Envisat ASAR) was to performed to determine whether any long-term, or previous terrain surface deformations occurred within the area of interest. It is expected that the historical data analysis will greatly improve a model of seasonal changes.

This is a long term study. It is foreseen that the terrestrial measurements in combination with CR-interferometry will reveal and precisely locate any potential deformations and allow their validation. The results of first 1.5 years of analysis indicate that the study areas can be mainly considered stable. However, extremely dry summer of 2015 and high temperatures influenced the results in a form of measurable seasonal effects.

We will report on the current status of the project, and elaborate on the initial analysis based on 1.5 years data record.

The Permian salt resources in Poland are related to 4 thick rock salt complexes with extent to almost two thirds of the territory of Poland, There are 15 documented large salt deposits both of stratiform and diapir types. The salt diapirs due to their shallow depth to the earth's surface are subjected to the mining exploitation and also considered as potential structures for storage of hydrocarbons in salt caverns. This second application require accurate information regarding their geological stability. However, the areas of salt tectonics are commonly associated with glaciotectonic deformation in Quaternary sediment cover. It is still an open question whether the salt diapirs remain active in Pleistocene and whether there are any evidences on contemporary occurring deformations.

To answer these general questions Polish Geological Institute – National Research Institute (PGI-NRI) started the project to determine the magnitude and extent of the recent dynamics of the terrain surface of the area of salt tectonics in Poland.  The study utilize SAR interferometry based on ESA data archives, TerraSAR-X and Sentinel-1 acquisitions. For the areas of specific interest, especially these exhibiting deformation and / or containing sensitive infrastructure  a detailed monitoring studies including corner reflector, high-resolution data acquisitions and geodetic measurements will be performed.

In this contribution we will elaborate on a design of the monitoring system and results of one of the validation case studies. Namely,  the  area of Wapno city. In the wider area of Wapno catastrophic deformation occurred after salt mine collapse that took place in 1977. Due to still unstable hydrogeological balance the area is subjected to subsidence up to 7.5 mm/yr, as  recorded from 1997 to 1995. Since 2007 the subsidence is also associated with sinkholes that significantly increase the hazard to urban areas.

To monitor the terrain deformations in Wapno area the existing network for in-situ measurements was supplemented by PGI-NRI by 10 leveling benchmark and 7 corner reflectors located in key areas. Since July 2015 TerraSAR-X data of Wapno area are systematically acquired including both ascending and descending passes.

Specifically in this contribution we will, to our best effort, summarize, in an itemize like manner, on all specific challenges needed to be addressed for monitoring by fast deforming areas by InSAR. Also, we will report on first results of historical data processing, and present results of the initial analysis of monitoring of the Wapno area.

ERS1/2 and Sentinel satellite images (AO358 ESA/ESRIN Project) were used for research of the earthquake hazard areas of Vrancea and Crimea. The main objects are: 1) geomorphology and displacement of the earth surface surfaces; 2) modelling of the Earth structure using the developed gravity tomography method with the geoid data [Greku R, Gozhik P., Litvinov V., Usenko V, Greku T..  Atlas of the Antarctic Deep Structure with the Gravimetric Tomography: Kiev.- 2009. -67 p.- ISBN 978-966-02-4937-0].

This method includes the solution of the following tasks: determination of harmonic density of anomalous disturbing masses by the geoid spherical functions; determination of the relationship between degrees of harmonic series expansion of the geoid topography and depths of disturbing layers of the Earth; creation and displaying of the structured model of dense inhomogeneities for the studied region.

Vertical cross-sections and maps of the lateral distribution of dense anomalies have been created for the area of the Crimean earthquakes, the hypocenter of which is in the Black Sea.

Blue color shows the less dense anomalies. Within this area are marked in red the still less dense masses that are distributed with breaks at depths of 8 km and 13-15km. Such gaps are linked, obviously, with the delamination process due to compression by the denser masses of the East European platform and Anatolia. That is, the less dense body surrounded by more dense masses may tend to emerge to the surface. It may be accompanied, in turn, by an earthquake.

A similar situation was observed in the region of the Vrancea seismogenetic body at the depth of 170-180 km.

Analysis of the earth surface displacement under the influence of the Crimean earthquakes was carried out using the ERS1/2 Tandem images and the DInSAR technology. Displacement of the earth surface during the period 04.05.1995-01.10.2010 on the Crimean Peninsula is observed within two areas (Figure 2). They are oriented parallel to the length of the Crimean mountains southwest - northeast. The shift up to the 5 cm in direction to the northwest is observed in the southern part of the Crimea near the epicenter of the earthquake. It is direction of the radio emission on the right side of the satellite. To the north of the Crimean Mountains the displacement is reduced up to 0 cm. The area of displacement further to the west in the steppe part of Crimea is re-formed up to 5-6 cm.

Transport networks across Europe face significant challenges in monitoring and predicting ground deformation along their transport infrastructure.  Incidents related to landslides and subsidence on road and rail transport systems can have a negative impact on society and local business communities.  Therefore, it is in the interest of owners and operators of transport infrastructure to understand and manage their exposure to geological hazards to minimise their impact.  Live Land, an ESA ARTES 20 IAP funded development project led by CGG, aims to assist the transport networks, initially in Scotland but with the potential to expand further into Europe, by providing improved intelligence to facilitate the proactive management of landslide and ground deformation events across transport networks. 

Current monitoring practices provide reactive rather than proactive information on landslide and ground deformation events across a transport network, and primarily consist of in-situ sensor technologies and site visits.  Live Land is driven by user needs and aims to provide a three tiered system of information provided by experts within their field; from earth observation satellites, in-situ low cost GNSS units, and forecast modelling technologies. This aims to provide up to date and detailed information on the areas surrounding road and rail networks to support transport owners and operators with their hazard and asset management systems. 

The long term societal and economic benefits of reduced disruptions and improved commercial performance of the transport network will have an impact not only on the transport companies themselves in terms of reduced fines and closure of the network, but also on the day to day lives of the general public.

In line with its data and information policy and with the available services and products, the Copernicus programme supplies free, full and open access to data and products to the users. Depending on their needs, users can obtain these data either from the Copernicus services or directly from the Copernicus space component.

In addition to the data produced by the Sentinels satellites, Copernicus users can also have access, under certain conditions, to the data produced by other satellite missions (Contributing Missions).

Processing of open data in a service-oriented infrastructure is already covered by interfaces defined within the Open Geospatial Consortium (OGC).

In line with the Big Data Management and Openness concepts, e-GEOS proposition includes new algorithms for the NRT processing automatization of Sentinel-1 (a central asset to Europe’s Copernicus programme), worldwide. This capability will also combine the use of commercial and Open Data exploitation for long time historical data series, including COSMO-SkyMed SAR data.

The following key applications and operational services will benefit from the proposed automated Image Processing and Data Fusion system:

Oil Spill and Ship Detection services, with improved frequency and features classification capabilities

Automated integration with ancillary data, such as AIS, weather and met-oceanographic data

Seeps detection

Fisheries monitoring in open oceans and support to natural resources sustainable exploitation

Wind and Wave fields extraction

The proposed approach is of paramount importance to Maritime and Oceanographic applications, to face the challenge of very wide areas surveillance in the shortest time possible, both to monitor marine environment and to detect potential threats associated to illegal activities at sea.

SAR data time series are extremely useful to provide systematic water bodies extent and floods observations worldwide and to improve the overall flood detection accuracy being able, for instance, to distinguish seasonal floods from exceptional floods or to increase the flood detection rate in urban and vegetated areas.

The combined use of booth systematic (e.g. Sentinel-1) and non systematic (e.g. COSMO-SkyMed) SAR satellite missions in a fully automated fashion within an appropriate exploitation environment (e.g. as in the concept of the ESA Thematic Exploitation Platforms) is a key element in order to ensure:

a proper revisit time of special interest targets (e.g. river basins exposed to highest losses)

suitable processing and delivery performances in terms of quality and timeliness to provide actionable information.

In addition to systematic flood monitoring, it is possible to derive flood statistical indicators that can be further assimilated into small and large scale global and local hydraulic models both as calibration and as validation data.

Derived Value Adding products and information layers will be accessible thanks to the GIS integration and Data Visualization innovative functions developed by e-GEOS.