Floods, earthquakes, landslides, tsunamis and wild fires all have tremendous impact on human societies. Human lives are lost, and significant economic burdens associated with the destructive powers by such disasters cripple societies for long times. To minimize the losses and effectively respond to crisis situations, approaches using today’s state of the art technologies can be deployed to assist already existing procedures and tools.

For exactly those purposes, a consortium of Austrian and Taiwanese partners designed and implemented the QuinJunSAT system that combines very-high resolution remote sensing (i.e., satellite) imagery and information from volunteers performing assigned data-collecting tasks in the field to obtain a quick and accurate assessment of damages in a crisis-hit area. The system implements a multistep approach: 1) damages are detected using change detection approach on pre- and post-event satellite images on a sub-meter level for a first assessment of the situation on the ground (e.g., damaged bridges and roads, collapsed buildings, etc., all geo-located and presented as polygons); 2) based on the first damage assessment, crisis managers task the volunteers who are available and in the vicinity of the area of interest to confirm the damages and to gather additional information (e.g., building height, construction material, etc.); 3) volunteers use QuinJunSAT system’s smartphone apps to gather the requested data in form of, e.g., answers to specific questions and images.; 4) finally, all collected remote sensing and crowdsourcing information is correlated and assessed by crisis managers/responders to create an action plan (e.g., where to distribute resources and response forces).

The system has been successfully deployed in the “921 International Disaster Prevention Drill” in Taiwan in 2018, with a focus on a magnitude 7.0 earthquake. For the demonstration, the crisis response coordination took place at the premises of the Fire Department Central in Hsinchu county, and more than 10 volunteers participated in executing the assigned crowdsourcing tasks in the affected areas of the nearby Zhubei city. Here, we present the results of the demonstration that showed that there is a great potential in the approach, specifically impacting the decision making, response times, allocation of resources, and distribution of response personnel.

ea A - Board 68

In case of natural disaster affecting air traffic, the availability of near real-time observations is essential to warrant and maintain the safety of flight crews and passengers. The aviation and its complex infrastructure requires indeed a reliable and rapid access to standardised and harmonised data, providing key information of natural airborne hazards threating the air traffic management. A global multi-hazard early warning system (EWS) has been developed in the frame of EUNADICS–AV H2020 project (European Natural Airborne Disaster Information and Coordination System for Aviation; http://www.eunadics.eu) with the aim of closing the gap in data availability and enabling all stakeholders in the aviation system to obtain fast, coherent, and consistent information. To achieve this task, improved near real-time (i.e. with maximum delay of data delivery of a few hours) retrievals from satellite, ground-based and in situ instrumentations, are used in order to detect, send notification and generate alert products of volcanic ash and SO2 plumes, desert sandstorms and dust clouds, and aerosols produced from forest fires or radioactive plumes. Based on a review of the user requirements, external reports and recommendations and the inventory of available observational data, our prototype system is designed for triggering model forecasts.

The development of EUNADICS–AV EWS is building on 4 types of existing systems observing continuously volcanic ash and SO2 plumes (1), sandstorms and dust clouds (2), aerosols produced from forest fires (3), and radioactive plume (4), from NRT observational data portals:
1) the first type of existing system concerns the monitoring of volcanic emissions. The Support to Aviation Service (SACS), hosted by BIRA–IASB (http://sacs.aeronomie.be), is the backbone of our EWS. This system initiated by the European Space Agency aims at supporting the Volcanic Ash Advisory Centres, like Toulouse VAAC and London VAAC. Another source of information comes from volcanic observatories such as Icelandic Meteorological Office (IMO) and the National Institute of Geophysics and Volcanology of Italy (INGV), which are equipped with a multitude of instruments. During episodes of volcanic unrest or eruption, observatories provide crucial information about eruptive activity using the Volcano Observatory Notices for Aviation (VONA) messages; see http://wovo.org/aviation-colour-codes.html.
2) The second type of existing system addresses the monitoring of extreme dust events. Outputs from the Copernicus Atmosphere Monitoring Service (CAMS), based on anomalous aerosol detection system, and the WMO Sand and Dust Storm Warning Advisory and Assessment System (SDS–WAS; https://sds-was.aemet.es), is used to complement EUNADICS–AV dusts warning system, that will primarily rely on IASI dust detection and AOD products (from ULB).
3) The third type of system concerns the detection forest fires. Outputs from existing warning infrastructures (e.g., Copernicus EMS – EFFIS, http://effis.jrc.ec.europa.eu; and NASA – FIRMS, https://firms.modaps.eosdis.nasa.gov) and monitoring products (e.g., Fire Radiative Power product based on SEVIRI from the EUMETSAT LSA–SAF; http://landsaf.meteo.pt) are directly considered by our EWS. A relay of such existing notifications is used and combined with EUNADICS–AV products from several platforms/instruments, providing tailored products to interested parties.
4) The fourth type of system addresses the monitoring of nuclear accidents and radionuclide plumes. Most European dose-rate results are available at the EURDEP web site but accessing the site and downloading data requires agreements. A collaboration with the Joint Research Centre of the European Commission to establish easy access to data (including historical data) has been established to proceed radionuclide alert products.

Here is some examples of key information/parameters provided by our EWS:
- selective detection of volcanic ash (from satellite) and plume height (from radars/observatories)
- SO2 column (with average kernel and error estimate) and plume height (from satellite)
- ground-based measurements of SO2 total column and profiles using spectrometers (from observatories)
- aerosol optical depth – AOD (from satellite)
- lidar AOD/attenuated backscatter/aerosol layer altitude (from observatories and EARLINET networks)
- nuclide concentrations from air sampler networks and external dose rates from radiation monitoring networks (from EURDEP/STUK).

The objective of our alert products is: 1) to enhance situation awareness in case of crisis situation, 2) to trigger atmospheric transport and dispersion models of forecasts/analyses, 3) to facilitate transfer of required relevant data to end-users.
Our global multi-sensors EWS is developed for demonstrating the added-value of integrating NRT harmonised aerosols/radionuclide data observations in forecasts system for cases crises affecting European air space.

Area A - Board 87

The huge and increasing quantity of high temporal and spatial resolution remote sensing data, especially from the European Copernicus Sentinel sensor constellations, calls for a more systemic and capillary integration of remotely sensed information with knowledge of the situation on the ground. This is especially true in the case of environmental hazard, where the need to fruitfully exploit the synoptic information coming from space sensors within remedial and protection actions by local authorities is more pressing. In this direction, the promotion of interactions between remote sensing specialists and local authorities personnel, leading to good practices and operational protocols of interventions, can be considered a viable solution for virtuous environmental management.
We present an example of integration of information from multi-temporal interferometry (MTI) techniques applied to time series of synthetic aperture radar (SAR) data, and in situ measurements over landslides afflicting several small hilltop towns in the Daunia region, Southern Italy. In a first joint project in 2008, a wide-area analysis of persistent scatterers (PS), derived from legacy ESA ERS and ENVISAT SAR image time series, highlighted the presence of ongoing surface displacements over several known landslides. An example is the Pianello landslide, located at the outskirts of the Bovino municipality, where movements were detected in the periods 1995-1999 and 2003-2008, respectively. This prompted local authorities to install boreholes equipped with inclinometers and other sensors on suitable locations. Ground data collected by these sensors during the following years were then compared with more recent PSI data from series of Sentinel-1 images, available from 2014. The integration of wide area MTI surface displacement information, projected onto the sub-vertical satellite line of sight, with pointwise subsurface vector displacement profile measurements in the horizontal plane, collected by inclinometers, as well as other data such as ground water levels measured from piezometers, allows, in several cases, sketching consistent models of a landslide spatial and subsurface structure, leading to a coherent interpretation of remotely sensed and ground measurements. In the above-mentioned case of the Bovino landslide, the limits of the affected area were updated, and the trend and depth of the rupture shear surface was inferred within a coherent picture of the whole mass movement phenomenon. For what concerns the temporal dimension, the integration of ground and satellite measurements time series may allow e.g. to optimize resources to increase the precision of warnings for landslide reactivations. This ongoing collaboration between local authorities and remote sensing specialists is being extended to other town centers in the Daunia area, and could thus represent an example for best practices in environmental management and protection at the regional scale.

Acknowledgements
Part of this research was funded by the Italian Ministry of University and Research (MIUR) in the framework of the “APULIA SPACE” project (PON03PE_00067_6), PON Ricerca e competitività 2007–2013. Copernicus Sentinel-1 data are provided by European Space Agency (ESA). ENVISAT and ERS data provided by ESA Category-1 projects N. 2653 and 5367. Work performed in the framework of the agreement “Convenzione operativa per lo studio del dissesto del suolo nel sub-appennino dauno”, CUP: H38C16000050008.

Area A - Board 94

Natural disasters can have devastating effects on human societies, including loss of lives, damaged infrastructure, and enormous economic burdens. Multiple large scale events over the last decades offer a stark memorial: the Indian Ocean earthquake and tsunami (2004), hurricane Katrina (2005), the Haiti earthquake (2010) and Tōhoku earthquake and tsunami (2011), however, many smaller incidents that disrupt societal order occur worldwide each year. Such events require decisive decision making, which is based on reliable and timely information about the situation in the affected area. Correct decisions depend on high levels of reliable and accurate information, thus minimizing misplaced spending of resources, optimizing response time and allowing for more location-targeted response. In this paper, we present a novel approach to offer this kind of information to crisis and disaster managers for improving common operation picture and decision making by systematically enriching earth observation (EO) data (i.e., satellite imagery) with information obtained through a targeted crowdsourcing methodology, called “crowdtasking”.

Our approach utilizes very-high resolution (VHR) EO satellite imagery to detect disaster-affected areas. Together with a spatial resolution on a sub-meter level, the satellite’s swath enables coverage of large areas. However, satellite data tells only parts of the story on the ground and critical information is lacking due to, e.g., the viewing angles. Thus, the information obtained from satellites represents a first assessment that needs to be complemented by context-rich data from the ground to offer a more complete picture to crises responders/managers. Thus, satellite imagery initiates (and provides a basis for) data collecting activities on the ground. These activities require a lot of personnel, dispersed over a wide area. To alleviate the stress this puts on the already strained capacities of emergency organizations, citizens near the affected area can be engaged to supplement the resources of disaster response organizations. Citizens acting as volunteers have long been an important factor in crisis and disaster management, often being first on site and providing vital aid in the initial response phase. Volunteers can support decision making by taking over data-collecting tasks (e.g., provide photos from a specific location, confirm or reject assumptions derived from satellite data). Additionally, participants from international aid organizations can also contribute to the crowdsourcing process. Data collected in such a way can complement and bypass some typical EO limitations.

In our research, we evaluate data gathered by volunteers in a geospatial crisis mapping solution, in real-time, thus offering response coordinators improved situational awareness. This approach saves time, optimizes resource allocation and minimizes human and material losses. We designed and implemented a software system, called QuinJunSAT system, based on VHR satellite imagery from historic disaster events, using data from two incidents: firstly, the Central Italian earthquake (August 24, 2016), which caused severe damages in the village of Amatrice and its surroundings. Secondly, we employed data from the typhoon Haiyan (November 3-11, 2013), which almost completely destroyed Tacloban City in the Philippines. Both data sets were used to develop an object-based image analyzing algorithm to initially detect affected regions, which are then shared between the system components to enable access to crisis managers and coordinators. Using this initial, but rapidly gained overview as basis, we provide crisis managers with a system that allows them to assign reconnaissance tasks to volunteers on the ground wherever they need further qualification of satellite data. Our system is based on state of the art information and communication technology and supports querying of quantitative as well as qualitative data: crisis managers can task volunteers on site with providing multimedia content, such as images, or ask specific questions in a structured manner that are adaptable to the current needs of the situation. Volunteers use the system’s smartphone application to locate the tasks they have been assigned in the disaster-affected area and provide the required information. Data gathered from the field is visualized for the crisis manager in real time on an interactive map that fuses both satellite imagery and data gathered on-site, to allow decision making based on current and highly relevant information.
Our system has recently been implemented and successfully demonstrated to a wide audience in a national earthquake drill day in Taiwan under field conditions. After describing the design of our approach, we will offer insights gained and discuss implications of the combined earth observation data and crowdsourcing approach for crisis and disaster management.

Area A - Board 140

Authors:
Anastasiia Shalina (1), Uku Kangur (1), Patrick Joan Thomson (1), Gudrun Vene (2), Peeter L. Nielander

1) University of Tartu, Estonia
2) Tartu Kutsehariduskeskus, Estonia

Contact email: patrickjoanthomson@gmail.com

Introduction:
In the event of a disaster, be it flooding, forest fire, earthquake or terrorism, relief agencies need to get a detailed overview of the area to develop disaster management plans. Satellite imagery can provide a brief overview of the area, but the images often lack context. For example, a satellite image can show flooded areas, but it cannot provide important detailed information such as depth of water or damage to surrounding infrastructure if not visible from above. In the event of a volcanic eruption, the ash cloud blocks the view of the ground below completely. Although satellite images provide lots of data to analyze, they lack context and can’t provide specific details regarding different types of disasters. All of which can be improved by combining data from ground level, where the location can be viewed from a different perspective.

ViralSAT platform:
We are developing an analytical platform called ViralSAT, that combines social media and satellite images to provide a more comprehensive story of a disaster to relief organizations in order to help them make informed decisions. We have completed a proof of concept of the technology by combining publicly licensed open-source packages to facilitate searching for real-time pictures of disasters. We continuously collect data from additional social media outlets. Furthermore, the images provided by ViralSAT can be used as training and reference data for remote sensing algorithms. We are also in the process of integrating our front end portal with our analytical back end.

ViralSAT can provide additional information to:
Save lives through better initial response plans
Provide faster information analysis thus providing faster response time
Allow for quicker infrastructure damage reports and economic damage estimates

Outlook:
Our development goals for ViralSAT are to:
integrate Artificial Intelligence (AI) functionality to automate event detection.
develop a plug-in that can be used to integrate with imagery portals (ESA, NASA)

Area A - Board 141

The I-REACT project [1] is an operational system for natural hazard risk management taking advantage of the latest technologies, open data, Earth observation (EO), citizen reporting, and social media. This single platform integrates and analyses in real-time all relevant and useful data to manage emergencies, help improve prediction and the management of the response when a natural disaster occurs. Due to climate change, it is expected that extreme weather events will become more frequent and will last longer including flooding, wildfires and heatwaves. I-REACT is the first European-wide platform integrating emergency management data from a variety of sources, including information provided by concerned citizens through social media and crowdsourcing. In this manner actionable information can be provided faster and allow citizens, civil protection services and policymakers to effectively prevent and/or react to the growing disaster threat from natural hazards. At the core of I-REACT is a data management system designed to scale for Big Data analysis, provide easy connections to third-party services including mapping based on EO imagery, a communication system that can handle reporting from both professionals and civilians via a smartphone application, and visualise everything at the command centre. The first European-wide solution able to integrate and analyse in real-time a set of data coming from different sources to help build a more secure and resilient society against natural disasters and supported by the Sentinel EO missions of the Copernicus programme of the European Union. I-REACT has been able to address the following issues based on the latest technologies and specialised models:
• Inability to integrate different emergency systems and visualise the output in a single place especially across borders;
• Lack of tools to take advantage of social media and citizen/volunteer involvement;
• Access and processing of relevant data and information, including EO, to help plan and respond to natural hazards.
The modular architecture is able to adapt to data feeds and present them in a standard format within a control room console. Automated social media tools monitor natural hazard related Tweets in four different languages. The I-REACT mobile application allows professional and volunteer users report on the current situation with two-way communication. Third party services including hazard and risk mapping, and access to open data provide the operational input information to keep users informed 24 hours a day. Furthermore, I-REACT can easily scale to other parts of the world providing the wanted level of integration to help local authorities manage multiple data sources, increase the performance of their response and reduce overall costs.

EO imagery and EO derived data and information are essential to all three emergency phases: prevention, preparedness and response. I-REACT was built as a downstream Copernicus service and therefore integrates a number of those data and services. This presentation will concentrate on flood and wildfire natural hazards among those that I-REACT currently supports. The flood nowcast module can be triggered by an authorised user to produce a localised flood map based on reporting provided by professional responders as well as citizen volunteers using the I-REACT mobile application. The near real-time flood map production can be then coupled with the automated risk mapping service. Risk maps provide vulnerability analysis using current nowcast flood information. A separate module integrates Copernicus Sentinel-1 Synthetic Aperture Radar (SAR) imagery to automatically detect and produce a flood extent map through the establishment of a direct data chain to quickly ingest and process the required imagery.

Wildfire hazards and disasters are also monitored through a variety of techniques. The potential for a wildfire to develop is based on the Fire Weather Index (FWI). This weather-based index coupled with anomalous heat detection from Moderate Resolution Imaging Spectroradiometer (MODIS) and Visible Infrared Imaging Radiometer Suite (VIIRS) imagery provide improved details to the civil protection services. Wildfire nowcasting and forecasting maps are produced from authorised user and citizen volunteers from the field and weather forecasts respectively. Risk maps based on wildfire hazards can also be requested through the I-REACT system. Finally, on-demand Copernicus Sentinel-2 based automated wildfire burned area mapping provides decision makers with post-disaster details. The figure presents a sample map overview of the I-REACT system with some mobile application screens inset.

The I-REACT mobile application allows users to provide vital mapping details for both flood and wildfire hazards. Citizen volunteers through gamification learn about a variety of natural hazards, how to react when in emergency situations and how to provide reports to monitor natural hazards. Through the gamification process, volunteers also learn how to use EO imagery such as those acquired by the Copernicus Sentinels. Please download the I-REACT mobile application now and help fight disasters in Europe [2].

[1] I-REACT – http://www.i-react.eu/
[2] I-REACT mobile app for the citizen – Android: search for I-REACT on the Play Store (https://play.google.com/); iOS: search for I-REACT on the App Store.

Area A - Board 142

Over the past 35 years there has been almost a fivefold increase in the number of recorded natural disaster events. This includes those from droughts, floods, epidemics, extremes of weather and temperature, landslides and mass movements, wildfires, tsunamis, volcanic activity and earthquakes. Whilst many of these disasters are directly due to geological processes or cascading processes (e.g. volcanic activity, earthquakes) a significant number result from non-geological process; for example heavy rainfall may lead to landsliding. Data from Munich Re shows that whilst the occurrence of natural disasters is increasing, the contribution to the total number of those disasters with a geological basis (i.e. geohazards or geotechnical perils) is proportionally greater. In parallel, the estimated economic damage from disasters is also increasing, with the poor and disadvantaged being hardest hit. International recognition of the situation is highlighted in the Sendai Framework for Disaster Risk Reduction. Established in 2015, its aim is the “substantial reduction of disaster risk and losses in lives, livelihoods and health and in the economic, physical, social, cultural and environmental assets of persons, businesses, communities and countries.” The response to disasters via the timely use of satellite imagery helps to meet these aims.

BGS has utilised satellite imagery in its response to several disasters in recent times, including Thailand (2004), Montserrat (2008), Japan (2011), Serbia (2014), Ecuador (2015), Nepal (2015), Mexico (2017), Sierra Leone 2017, Philippines (2018). This presentation will focus on the natural disasters that the British Geological Survey (BGS) has responded to over the last 5 years. It will highlight the important role that evolving satellite imagery combined with geological knowledge, expertise and experience has in the production of useful, timely products and advice for disaster response. Case studies will be presented on landslide inventories and hazard advice for Nepal following the earthquake in 2015, Sierra Leone following heavy rain in 2017 and for the Philippines following the Typhoon in 2018. A case study for volcanic event in Montserrat (2008) will be presented also to highlight a particular application of EO data.

Satellite data and imagery have a wide applicability for the full disaster lifecycle, including response. By their nature satellites capture consistent data, of differing spatial and spectral resolutions over large spatial areas with no risk to human life. These characteristics along with the complimentary nature of optical and radar data improve what can be monitored over difficult terrain, at times when access is made harder by the disaster, in many cases satellite imagery is the only timely source of data for emergency response. Recognising these benefits, many satellite imagery providers are making their imagery freely available for disaster response through a variety of mechanisms. A prime example, the International Charter for Space and Major Disasters, established in 2000, not only forms a collaborative framework for the combination of earth observation data from different providers but also the value adding work carried out by organisations such as BGS to rapidly create information products required by relief teams on the ground.

We will also discuss the research opportunities that these disasters have presented, for example the opportunity to study Tsunami sediments in the weeks following the Japanese Tohoku Tsunami in 2011 has enabled better understanding of tectonically induced Tsunamis and therefore better modelling and hence planning and preparedness.

Earth Observation enables improved natural disaster preparedness and response; a key focus of the BGS Overseas Development Assistance (ODA) programme. The Global Geological Risk activities are building upon the disaster response activities to strengthen multi hazard resilience through integrating citizen science, innovative technologies, and understanding of environmental processes, hazards and impacts. Within this framework, BGS are actively working on the establishment of disaster response systems and protocols to ensure the maximum useful information can be extracted from the imagery and presented to user in a manner, which is easy to digest and use. These include 3D visualisation systems populated with worldwide data facilitating the interpretation of a disasters effects and the presentation of such to users.

Area A - Board 143

The changing geological landscape under changing climate and resulting from the tectonic setting of the alluvial plains, obscured the buried landforms such that today it is difficult to reconstruct the original river courses. In many cases the buried landforms such as paleochannel are investigated using detailed field surveys to reconstruct the underground structures. However, these techniques are time-consuming and required high costs. In this work, an innovative methodology for the automatic deformation pattern interpretation of time series obtained using A-DInSAR techniques is presented, in order to support geological mapping. The Venetian-Friulian coastal Plain (NE Italy) was selected for this study as test site, it is characterized by a general long-term subsidence (average rate 0.4-0.5 mm/yr), mainly related to crustal flexuring and sediment compaction including much stronger values recently measured in specific areas with peculiar subsoil conditions, such as buried incised river valley formed in the Late Glacial and completely filled in the last millennia with softer sediments. The methodology was applied to interferometric data derived by the PS-InSAR processing of ERS-1/2, Envisat, COSMO Sky-Med and Sentinel-1 SAR data covering the periods 1992-2000, 2003-2010, 2012-2016 and 2014-2017. The results give insight about the capability of the proposed methodology to use the ground motion areas changes from 1992 to 2017 induced by natural and man-induced processes such as sediment compaction, land subsidence due to load of new buildings, and seasonal ground motion due to groundwater level variations to detect priority areas where to address further in situ investigations to verify the presence of localized buried landforms.

Area A - Board 145

We present first full resolution state-wide multi-temporal InSAR analysis of the whole Slovakian territory exploiting a significantly enhanced spatio-temporal coverage and free data availability of the Sentinel-1 interferometric wide swath (IWS) SAR imaging.

All Sentinel-1 IWS SLC data covering Slovakia (three ascending and three descending orbit stacks), and acquired over past 4 years since the beginning of the mission, are utilized within our automatized processing chain. It is based on the current and open-source Sentinels Application Platform (SNAP) and modified Stanford method for Persistent Scatterers (StaMPS), their functionally combined and optimised within fully parallelised python processing framework.

Additionally, incorporating this open-source approach allowed us to test its feasibility, addressing the unfavourable InSAR conditions of Slovakian sparsely urbanized and mountainous environments as well as its dynamic and spatially variable troposphere, which especially influences the quality of InSAR observations over large distances.

Outcome of the analysis – millions of identified Persistent Scatterers (PS) with known displacement time series over monitored time period serve as a “natural geodetic network” to monitor kinematic behaviour of a whole cities up to an individual structures.

Several significant ground motion phenomena have been recognized, most notably: undermining subsidence areas, active (or recently re-activated) landslides over inhabited villages, instability of transport infrastructure (e.g. railways and highways), seasonal deformation over strategically important waterworks, or slow mountain rockslides endangering the operation of a cable car at a ski centre.

These findings are supported by ground-truth validation with conventional geodetic techniques (i.e. levelling, terrestrial and GNSS surveys). Respecting the relative nature of InSAR measurements as well as its line-of-sight geometry, comparison is carried out forming spatio-temporal double-differences. For several sites this proves the technique’s capability to capture millimetre scale movements.

Considering the precision comparable to ground-based geodetic techniques, current 6-day repeatability and low-cost merit of Sentinel-1 multi-temporal InSAR, our findings emphasize its operational potential to monitor natural hazards at the country-scale level.

Area A - Board 149

Taiwan Island, resulting from oblique collision between Philippine sea plate and Eurasian plate converging at a rate of about 8 cm per year, is one of the most active tectonic region 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 in several domains of Earth Sciences, in particular for earthquake cycle study and earthquake hazard assessment, for subsidence and landslide monitoring, and also to better understand the kinematics and mechanics of mountain building.
Taiwan benefits from a remarkable GNSS network. However, due to a complex geodynamical context and high strain rate, the pattern of deformation is not well solved by GNSS. In complement, INSAR shows its contribution with respect to GNSS, as it allows to dramatically increase the spatial information.
Combining SAR dataset provided by ALOS-1, ALOS-2 and Sentinel-1 enables to generate consistent time series and dense maps of ground displacements by InSAR on the whole island over different periods. This possibility of long time series of observations is particularly interesting for earthquake cycle study.
Our processing uses NSBAS interferometric chain (Doin et al., 2015), based on a SBAS approach, that includes several corrections applied before unwrapping, in particular correction of atmospheric delays predicted from the global atmospheric re-analysis ERA-Interim model, and local DEM error correction. These corrections are of particular importance as they reduce the variance of the phase across regions with high topographic gradients, like the Central Range in Taiwan, hence facilitating unwrapping step.
Using the full archive of ALOS-1 PALSAR images, a first complete deformation map of Taiwan has been derived over the period 2007-2011. Our InSAR results offer an unprecedented continuous view of deformation field of the entire Island. For instance, in the Central Range, the LOS velocity map shows a clear pattern of deformation, consistent with a rapid uplift (cm/y) of the Central Range South of the island. This uplift, already partially documented by GPS and leveling, is clearly mapped here and seems to show an overall continuity. In southwestern Taiwan, the InSAR LOS velocity map provides a good coverage in the foothills area, revealing several localized areas of interseismic deformation that were overlooked in GNSS, and that can be correlated with tectonic structures. Among them, is the 15 km-long Lungchuan anticline, showing relative surface displacement toward satellite by several cm/year. Those observations, combined with a geological study and field survey (Le Beon et al., 2017), suggest the existence of a back-thrust fault that reaches the surface on western side of Lungchuan ridge and roots on the ~4 km deep Tainan detachment. This structure has also been activated during 2010 Mw 6.3 Jia-Shian Earthquake and the Meinong earthquake (02/05/2016, Mw6.4).
A time series analysis can also be conducted on 2014-2018 period with Sentinel-1 data. Since end of 2014, we benefit from S1 SAR images, acquired in C-band, thus less favorable than PALSAR L-band on Taiwan Island. However, this drawback is balanced thanks to the high frequency of image acquisitions (12 days on Taiwan with S1-A and S1-B). We can also take advantage of the 2 different geometries of acquisition (both ascending and descending) to derive horizontal and vertical components of the deformation.
The combination of ALOS and Sentinel-1 InSAR results, in addition to their high density of measure, covers different time periods and gives the opportunity to investigate temporal evolution of the deformation. Some areas, in particular in SW Taiwan, show changes in the tectonic deformation pattern, thus revealing transient behavior of some structures.
Those observations can also be completed on several areas with previous ERS and Envisat INSAR results, offering an unique monitoring of more than 20 years.

Area A - Board 150

The ASTER (Advanced Spaceborne Thermal Emission and Reflection Radiometer) is an optical imager operating since 1999 on NASA’s Terra platform. ASTER is a joint project between Japan’s METI agency, and U.S. NASA. ASTER acquires -525 images per day in 9 spectral bands in the visible, near infrared, and thermal infrared wavelength regions, with a spatial resolution of 15 to 90 m per pixel. An off-nadir visible stereo band allows creation of digital elevation models.
Early in the mission, the ASTER Science Team realized there would be a need for urgent data acquisitions, and rapid data turn-around. As part of mission operations, the US ASTER project and NASA’s Land Processes DAAC developed an expedited data tasking and processing pipeline to respond quickly in events of natural disasters, national emergencies, humanitarian crises, etc. A system was put in place that allows scheduling of data acquisitions up to 36 hours in advance, and provides 2-4 hour processing time after scenes are acquired. This tasking process is made available to emergency responders, including the US Geological Survey’s emergency response team, who coordinates requests from a number of US Federal agencies. Processed data are archived on a website with ftp access, and requesters are sent notification that the data are available. In a typical year, we respond to over 350 requests for urgent acquisitions and were able to provide data to more than 95% of users.
The International Charter (Space and Major Disasters) has been designed to provide a unified system of space data acquisition and delivery of products to those affected by natural and man-made disasters. The Charter operates on a volunteer, best-efforts basis by the member organizations. In the United States, the U.S. Geological Survey and the National Oceanic and Atmospheric Administration are Authorized Users. NASA responds to internal requests from these two agencies to acquire data from its resources. ASTER is one of the instruments frequently asked to provide data on behalf of the US Authorized Users
In this presentation, examples will be shown of the uses of ASTER data and DEMs during episodes of natural disasters. The principal events we respond to are: 1) Flooding from hurricanes (Hurricane Katrina), snow and ice melt (Midwestern US), and tsunamis (Sumatra, Tohoku Earthquake); 2) Fires: forest fires all over the world; 3) Earthquakes: Szechuan and Haiti earthquakes, and resultant landslides; 4) Volcanic eruptions: Kamchatka volcanoes and aircraft hazards; SO2 plume detection.
Though not christened as an operational instrument, given the constraints dictated by its design as a science research and applications tool, ASTER has nevertheless functioned in a quasi-operational mode,.
Work done by Abrams was performed at the Jet Propulsion Laboratory/California Institute of Technology under contract with the National Aeronautics and Space Administration.

Area A - Board 151

Stakeholders and user communities have brought to the fore the requirements for an effective emergency response capacity for Disaster Risk Management (DRM), encompassing priority theme areas such as floods, seismic hazards, landslides, subsidence and volcanoes. To address these emerging needs, the ESA study on “Disaster Risk Reduction using Innovative Data Exploitation Methods and Space Assets” has analysed the enhanced use of space technologies and assets in the DRM domain, with specific focus on Disaster Risk Reduction (DRR), in particular dealing with hazard mitigation/prevention/preparedness.
One of the objectives of this study is to assess the contribution of current and firmly-planned in-orbit space assets to the broad DRM activity framework (with focus on DRR), to identify gaps in the fulfilment of the user needs, and to analyse new types of dedicated missions (gap-fillers) that could be designed by ESA and by the international space community to meet the user needs.
Two hazard types have been analysed in details in the study, namely floods and volcanoes, and the corresponding user requirements have been specified leveraging the output of user consultations.
The DRR contributions enabled by existing space assets encompass primarily Earth Observation (EO) satellites and in particular ESA missions, including data archives and the current and future Sentinel missions, as well as European national EO missions featuring SAR and optical acquisition capabilities (e.g. Cosmo-SkyMed, Pleiades, TerraSAR, SPOT 6-7, DEIMOS-1, DEIMOS-2, etc.) and other relevant non-European missions (e.g. WorldView, Landsat, DMC 2&3, RADARSAT Constellation Mission, etc.).
The key performance parameters driving the space segment design, the ground segment and the operations concepts for DRR scenarios include revisit time, coverage, data latency (for emergency response), data availability and data quality. Good revisit time and spatial resolution performances and long-term continuity of SAR and optical acquisitions emerge as key enablers for effective DRR applications, which require frequent observations over high-risk areas.
A candidate mission concept to fill the detected gaps for the Volcano Hazard Case has been proposed, namely a constellation of optical satellites, which fulfils the requirement of 1-week revisit time with optical instruments at 5-m spatial resolution for accurate mapping. Space assets embarking optical payloads with the required resolution should be added to the existing or firmly-planned ones.
To exploit the synergies with the Copernicus Sentinels, the candidate mission concept to fill the identified gaps for the Flood Hazard Case consists of a constellation of Sentinel-1-like satellites flying in convoy with Sentinel-1A and Sentinel-1B. These Sentinel-1-like satellites will embark simpler C-band SAR instruments than the one on-board Sentinel-1, with spatial resolution of 50-100 m. Eight satellites are needed to guarantee 1-day maximum revisit time.
The results of the study highlight that federated satellite systems emerge as a promising gap-filling solution to guarantee long-term continuity of SAR and optical acquisitions, so as to build time series of high-quality data and data archives that are paramount for DRR applications. Federated systems with in-orbit missions could be set up launching a tailored version of already-flying satellites, but embarking simpler instruments that cover specifically the spectral bands identified for DRR. Synergy of the acquisitions performed by federated satellite missions could also be conducive to fulfilling revisit time requirements that are not met by existing EO missions.

Area A - Board 152

Flood events are one of the main emergencies that European countries have had to face in the last years. The number of devastating floods has more than doubled in Europe since 1980 and flood disasters often have a broad extent and a high frequency. Earth Observation techniques have become a key tool for flood prevention and monitoring, and for damage assessment. SAR and optical images are frequently employed for flood mapping and monitoring purpose, being the fastest means of data collection for pre- and post event disaster analyses.
The Copernicus Earth Observation programme managed by the European Commission has been designed to provide accurate, timely and easily accessible information to improve the management of the environment and to map a wide range of emergency situations. Sentinel satellites provide open data in all-weather conditions, day and night images and multispectral and radar data at different spatial resolution.

The present study offers the analysis of the flood event that affected Emilia-Romagna region (Italy) in December 2017. Between the 8th and the 12th of December, various meteorological phenomena hit the region. Between the 10th and 11th exceptional rains (300-500 mm recorded in 36 hours) caused the increase of the water level in the Po's Emilian tributaries and in the Reno river; in some sections the water level exceeded the historical highs. Along the Enza river, the flood wave caused severe damages to the embankment resulting in widespread flooding in the built-up area of Lentigione (RE).

The purpose of this study is the detection of the flooded areas caused by this extraordinary event, by means of Sentinel multitemporal data. Both optical and radar images, acquired by the Sentinel 1 and 2 missions, have been processed for mapping and for monitoring the affected area during and after the emergency.
Firstly, the areas along the Enza river, inundated immediately after the overflows, have been studied by radar imagery, refining the analysis by optical data for obtaining the total affected zones. The resulting flood maps were compared with the grading maps produced by the Copernicus EMS in order to evaluate the performance of adopted approach.
In a second phase, the study has been focused on the most affected zones extending the analysis to a longer period in order to evaluate the water retention phenomenon for the recovery monitoring.
In May 2018, 6 months after the event, a field survey has been performed on the most affected areas, identified in the monitoring map to assess the recovery level.

The obtained results and the in situ campaign highlighted the effectiveness of the satellite remote sensing in the identification of exposed areas where the mitigation actions should be concentrated.
The findings of this work can be useful not only for the emergency management purpose, but also for supporting the development and the protection (insurance sector) of the local economy, especially with regard to the agricultural sector.

Area A - Board 153