The International Charter ‘Space and Major Disasters’ (“Charter”) has been founded in 1999 with three initially participating space agencies: ESA, CNES and CSA (from Europe, France, and Canada). It started its operations in November 2000 with the goal to provide quick access to satellite-based information in cases of major disasters, free of any cost for the user. In 21 years, the Charter has been activated more than 730 times worldwide and covered disasters in 129 countries (as of Nov 2021). The Charter is now composed of 17 agencies worldwide (ABAE, CNES, CNSA, CONAE, CSA, DLR, ESA, EUMETSAT, ISRO, INPE, JAXA, KARI, NOAA, ROSCOSMOS, USGS, UAESA, UKSA).

The Charter covers a large variety of disasters, both natural and man-made. In about 50% of the cases Charter activations are for flood events, including tsunamis. In addition, tropical storms, earthquakes, fires, landslides, volcanic eruptions, ice jams, oil spills, and large industrial accidents are being covered. The Charter is functioning on a best-effort basis.

The Charter illustrates the fact that Earth-observing satellites are able to deliver key information that brings benefit to the definition, planning, implementation and monitoring of disaster relief operations when human life and infrastructure are at stake. Its service is user-driven, i.e. the Charter becomes active after being triggered by an “Authorized User” (AU). Typically, Charter AUs are national disaster management agencies or emergency operations units. Besides these AUs, the Charter can also be triggered by Cooperating Bodies from the UN family in order to make Charter support available for user organisations within the UN, the Red Cross/Red Crescent, as well as countries that do not yet have a Charter AU.

The “Universal Access” policy allows national disaster management agencies, which do not yet have Charter access, to register with the Charter and become an AU after a training. It is also encouraged that countries struck by a disaster bring in their own EO expertise in case of an activation, i.e. Charter activations can be managed by and/or maps generated by in-country capacities. The Charter support is free of charge and based on international collaboration among the Charter member agencies and their humanitarian motivation.

This paper will give an overview of the Charter’s work with a focus on user aspects. Various examples of information products resulting from Charter activations will be shown, with a special focus on products based on synthetic aperture radar (SAR) satellite data.

The idea for the International Charter ‘Space and Major Disasters’ came into being at the third UN space conference UNISPACE III held in Vienna in July 1999. In the face of increasing destruction and damage to life and property caused by natural disasters and conscious of the benefits that space technologies can bring to rescue and relief efforts, the European Space Agency (ESA) and the Centre national d'études spatiales (CNES) set out to establish the text of the Charter, which they themselves signed on 20 June 2000, while inviting other space agencies to do the same. The Canadian Space Agency (CSA) was the first to come onboard and sign the Charter on 19 October 2000. These three founding space agencies then went on to establish the architecture essential for implementing the Charter.

As one of the three founding members of the Charter, CSA will provide in this paper an overview of the Charter’s history, the role played by CSA in its development, and highlight the notable Charter milestones in its evolution. Examples of Charter images that have provided valuable assistance to disaster relief efforts will be presented. The benefits and strengths of the Charter from a Canadian perspective will be examined. The paper will conclude by discussing some of the Charter’s future challenges.

The Sendai Framework for Disaster Risk Reduction (SFDRR) provides an internationally agreed agenda for evidence-based policy with the overall aim to achieve progress in disaster risk reduction (DRR). Main components of the framework are four priorities for action and seven Global Targets to monitor progress in DRR based on thirty-eight indicators. However, the progress of monitoring the set targets is very often obstructed due to a lack of available, accessible, and validated data on disaster-related loss and damage, especially in developing countries. This weakens the accuracy, timeliness, and quality of the SFDRR monitoring process. In the case of floods, which account for the highest number of people affected by hazards, there is a strong need for innovative and appropriate tools for monitoring and reporting impacts, for which Earth Observation (EO) can provide solutions. Previous attempts to address this gap via a geospatial model approach could not be validated due to a lack of in-situ measured loss and damage reference data. The presented research here addresses this gap by developing a geospatial model approach that was validated against reference data provided by partner national institutes from the country of Ecuador. The methodology in this study aimed at combining EO-based information products with additional geospatial data to result in quantitative measures for indicator B-5a of the SFDRR and validate them against the reference data. A semi-automated derivation of flood event characteristics from a full year of Sentinel-1 synthetic aperture radar data was applied to three Ecuadorian focal provinces that best represented the ecological diversity of the country in order to assess flood hazard. An automated thresholding algorithm was applied to delineate flooded areas which yielded flood statistics when applied to Sentinel-1 data for an entire reference year. This assessment was complemented by census and agricultural in situ data to spatially model SFDRR indicator B-5a for the year 2017. The validation procedure involved various steps to produce different models that combined elements of flood exposure and vulnerability to be cross validated with the reference data. A statistical analysis was used to assess the agreement of the models with reference data and their ability to reproduce the reference data. The validation procedure resulted a geospatial model, which integrated also flood vulnerability and has high agreement with the reference data. However, the models were sensitive to different ecological regions.

This validated geospatial model approach, is - to the best of the authors’ knowledge - the first attempt to validate geospatially measured Sendai indicators against reference data. The derivation of open source information products was conducted in close collaboration with the National Service for Risk and Emergency Management of the Government of Ecuador, the Ecuadorian Ministry of Agriculture, the Ecuadorian National Institute for Statistics and Census, the Ecuadorian country office of the United Nations Development Program, and the United Nations Office for Disaster Risk Reduction. It is thereby assured that the development and validation of this methodology is in line with the Ecuadorian national and the international approach of implementing the SFDRR. The information products were shared with partner institutions through a “training-of-trainers” capacity building workshop which enabled the sustainable transfer of results and learning outcomes for Sendai Monitoring.

Consequently, this validated geospatial model approach provides an opportunity to support countries without information on disaster-related loss and damage in monitoring indicator B-5a of the SFDRR. Due to its retrospective ability to assess loss and damage, a baseline measure of the indicator can be derived as a reference for monitoring progress. This approach also successfully integrates characteristics of vulnerability into the UNDRR methodology to better capture the heterogeneous nature of flood impacts. Future research should seek the application and modification of the developed and validated model for additional Sendai indicators and targets, and predominantly explore solutions to overcome the sensitivity of the models for different ecological regions.

Satellite imagery has great potential for monitoring and understanding the behaviour of volcanoes, especially at the 45% of potentially active volcanoes without ground-based monitoring. Here, we present results from Committee on Earth Observation Satellites’ Volcano Demonstrator. Our primary goal is to increase the uptake of satellite imagery, and especially Interferometric Synthetic Aperture Radar (InSAR), for volcano research and monitoring. We aim to support the use of satellite data for disaster risk reduction by providing otherwise restricted civilian data to volcano observatories, and by supporting capacity building. The Volcano Demonstrator developed from a 4-year Pilot (2013-2017) to demonstrate the usefulness of satellite data for monitoring large numbers of volcanoes. This pilot focussed on Latin American volcanoes and detected unrest not observed by ground based networks, as well as feeding into volcanic observatory decisions about alert levels and sensor deployments. Satellite measurements of deformation contribute both our knowledge of sub-volcanic magmatic zones, and to volcano monitoring, especially when used in combination with satellite measurements of thermal anomalies and volcanic gases.

Since 2019, the Volcano Demonstrator has targeted volcanoes that present the greatest risks and potential for useful remote sensing measurements in Latin America, SE Asia and Africa. In some cases, we have provided a route for scientists at volcano observatories to access CosmoSkyMed, TerraSAR-X and Pleiades imagery, while in others we have contributed to data processing and interpretation. We have responded to unrest and eruptions at volcanoes including in St Vincent, Guatemala, Peru, Chile, Indonesia, DRC and the Canary Islands. Research based on CEOS demonstrator data has included novel retrievals of topographic change during eruptions (Fuego 2018, St Vincent 2021, Nevados de Chillán) and forensic analysis of recent eruptions with high spatial resolution imagery (Agung 2017, Anak Krakatoa 2018). In addition, we coordinate the acquisitions, and where possible, the analysis, of ‘baseline’ high resolution radar images at active volcanoes to complement open datasets such as Sentinel-1. From our experience in the CEOS pilot and demonstrator programs, we believe that integrating multiple SAR platforms is critical for volcano monitoring because it maximises temporal coverage and because some volcanic events can only be observed at specific radar wavelengths, geometries or repeat times. Integration of these observations with satellite measurements of gas, thermal anomalies and the measurements of ground-based monitoring networks are critical for interpretation of these data.
The CEOS volcano demonstrator links volcano observatories around the world with experts in satellite remote sensing and space agencies. Our long-term aim is to make this project sustainable and to demonstrate the necessity of inclusive, international coordination of satellite imagery acquisition for volcanology.

On August 14th, 2021, a 7.2 magnitude earthquake hit the southern peninsula of Haiti. Two days later, Grace, a tropical storm, struck the three departments of Nippes, South & Grand’Anse. These combined events, and all aftershocks caused many landslides, especially along the Enriquillo-Plantain Garden fault. Following these disasters, emergency management services have been triggered such as the International Charter Space and Major Disasters and the Copernicus Emergency Management Service Rapid Mapping.

The third CEOS Recovery Observatory demonstrator was activated on September 6th, 2021 on behalf of the European Union, United Nations Development Programme (UNDP), and the World Bank, in charge of the Post Disaster Needs Assessment (PDNA). The priority was to estimate the number of landslides and their location throughout the peninsula. Taking into account the size of the affected area, the difficulty in accessing this region, it was decided that the best way to obtain and provide this information was through satellite imagery. Earth Observation provides an overall and factual view of the situation.

Many contributors assembled to provide a rich package of information in a rush and best effort mode. Automatic landslide extraction layers realized by NASA and BGC Engineering Inc. were provided. ICube-SERTIT as RO liaison officer, but also as a value-added producer and expert in Earth Observation aggregated, validated, and completed these data over the whole southern peninsula. Landslides were extracted from optical satellite imagery. The International Charter “Space and Major Disasters” provided a Pléiades-HR image over the national Reserve of Macaya Park located in one of the most affected areas. In order to cover the whole peninsula, Sentinel-2 and Landsat-8 data were used.

The Haitian National Center of Geospatial Information (CNIGS) shared several reference datasets (land use / land cover maps, Digital Elevation Model…) which enabled the elaboration of statistical information concerning the landslides. Nearly 7.000 ha of landslides were mapped within the southern peninsula, especially affecting wood and bushy areas, in mountainous agro-ecological zones.

This landslide dataset and its related statistics helped the PDNA team to complete and precise their report. It represents also the principal input data to the second RO phase’s products, which ends at the end of February 2022. To support the reconstruction phase, RO will provide products aimed at user needs concerning land cover and the hydrological changes induced by these disasters. Furthermore, the computation of a soil erosion susceptibility index is planned to identify new risks in the areas that need to recover.

To address the needs of the Haitian community in the south-west of the country involved in recovery and rehabilitation after the impact of Hurricane Matthew in October 2016, the Committee on Earth Observation Satellites (CEOS) triggered the 4 year-long Recovery Observatory (RO) pilot project, led by the National Center for Geo-spatial Information (CNIGS) with technical support from the French National Centre for Space Studies (CNES) [www.recovery-observatory.org].
During the RO pilot, the Italian Space Agency (ASI) contributed with scientific research aiming to develop – jointly with the RO management and technical team from CNES and CNIGS – a workflow which encompassed: (i) tasking of satellite high resolution Synthetic Aperture Radar (SAR) data for regular observations of the priority areas defined by the Haitian users; (ii) image processing, also testing computing and algorithm resources that could ensure future sustainability; (iii) generation of value-added geohazard products providing information on terrain motion and change detection; and (iv) in-situ validation. In addition, a basic SAR image course was given to the URGEO master in Université d’État d’Haiti (UEH). The motivation to focus on SAR data (to complement more consolidated techniques based on optical satellite imagery) was provided by the Haitian partners’ expressed need to approach the domain of SAR remote sensing and interferometry (InSAR) that, on one side, bring well-known advantages for land applications and disaster risk reduction in tropical regions but, on the other, require computing facilities, training and capacity building to be feasibly used as sources of geospatial information.
The RO pilot was successfully completed at the beginning of 2021. A final workshop was held by involving all the space agencies, national champions and users who collaborated to demonstrate the benefits of better integrating information from satellite imagery and their derived products into the post-Matthew reconstruction and recovery process. That event provided the opportunity to reflect collectively on the lessons learnt and the legacy of the RO pilot experience.
The present paper aims to contribute to the key objectives of ESA Living Planet Symposium session D1.05 “International Collaboration to better understand risks using satellite EO (GEO, CEOS, etc.)”, by:
- sharing the technical achievements and challenges in the use of repeated SAR data from high revisit sensors (e.g. Sentinel-1) and on-demand acquisitions from high resolution sensors (e.g. COSMO-SkyMed) for terrain motion and land surface change applications;
- highlighting the role that the collaboration with users and stakeholders can play to add value to SAR-based scientific products.
For the purposes of a wide-area regional analysis, Sentinel-1 data were processed and interferometric products were generated using ESA’s Geohazards Exploitation Platform (GEP) [Cigna et al. 2020]. In particular, we will showcase the value of accessing such infrastructure and its hosted processing routines, discuss the impact of possible external constraints that may limit the exploitation by users (e.g. skill gap, limited internet connectivity), and outline possible actions towards an effective use of this resource (e.g. dedicated training).
In parallel, a bespoke campaign to monitor three priority areas defined by the Haitian users – i.e. Jérémie, Camp Perrin and Carrière Arniquet – was undertaken with ASI’s COSMO-SkyMed constellation using the Enhanced SpotLight mode, at 1-m spatial resolution, 16 days site revisit, in both ascending and descending modes, from December 2017 to December 2020 until the RO completion. These data were used to generate maps that allowed the identification of different categories of surface changes including:
(a) environmental, located along the estuarine section of the Grand’Anse River south of Jérémie and mixed with anthropogenic activities mostly related to quarrying and unregulated waste disposal [De Giorgi et al., 2021];
(b) geological, along the rock cliffs north-west of Jérémie where susceptibility of local lithologies to fracturing, toppling and lateral spreading may be worsened by the impact of hurricanes and storms, thus causing potential risks to small villages and isolated dwellings.
(c) urban, within the outskirts of Jérémie due to reconstruction, as well as new constructions, in areas where the Sentinel-1 InSAR analyses highlighted ground motions;
(d) rural, due to landslides to be distinguished by similar signals associated with agricultural practices along the slopes in the Camp Perrin.
Each of the above categories was validated based on ground-truth data collected during a technical mission that was jointly carried out by ASI, CNIGS, CNES, ICube-SERTIT and Bureau des Mines et de l'Energie d'Haïti (BME). Lessons learnt will be discussed in detail, in order to outline some recommendations on how to effectively integrate a range of SAR observations and products and pave the way for their embedding into the decision making process for recovery and resilience building.
In this regard, the discussion will also encompass the analysis of the feedback received by Haitian stakeholders (e.g. the Civil Protection, mayors of the municipalities affected by the hurricane Matthew, the Comité Interministériel d'Aménagement du Territoire) during the dedicated workshops that were held in Jérémie and Port-au-Prince to present these SAR and InSAR derived products, as well as from the group discussion at the final workshop. Among the feedback:
- the integration between the Sentinel-1 InSAR ground motion products and the COSMO-SkyMed-based urbanization map was positively assessed for purposes of urban planning, as a satellite evidence-base to highlight areas where cascading hazards may be triggered and thus new urbanization is not recommended;
- in light of the proven benefits of InSAR and SAR change detection techniques, there is the need for capacity building not only to transfer knowledge, but also to create a technical capability (also through the involvement of the local university) to exploit the RO pilot technical legacy after the project completion.

References:
- Cigna, F.; Tapete, D.; Danzeglocke, J.; Bally, P.; Cuccu, R.; Papadopoulou, T.; Caumont, H.; Collet, A.; de Boissezon, H.; Eddy, A.; Piard, B.E. Supporting Recovery after 2016 Hurricane Matthew in Haiti With Big SAR Data Processing in the Geohazards Exploitation Platform (GEP). Proceedings of 2020 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Waikoloa, HI, USA, 26 September–2 October 2020; pp. 6867–6870. https://doi.org/10.1109/IGARSS39084.2020.9323231
- De Giorgi, A.; Solarna, D.; Moser, G.; Tapete, D.; Cigna, F.; Boni, G.; Rudari, R.; Serpico, S.B.; Pisani, A.R.; Montuori, A.; Zoffoli, S. Monitoring the Recovery after 2016 Hurricane Matthew in Haiti via Markovian Multitemporal Region-Based Modeling. Remote Sensing 2021, 13 (17), 3509. https://doi.org/10.3390/rs13173509