Improving disaster resilience is a key global challenge that unites different stakeholders across the planet. Between 1998 and 2017, around 4.5 billion people have been affected by disasters, with catastrophes such as the Australian bushfires of 2019-2020 killing 1.25 billion animals and affecting biodiversity to an unprecedented extent. Furthermore, in 2018 alone a sample of 63 countries have reported $17.5 billion in direct economic loss, with agricultural losses reaching $13 billion. Specialists estimate that real economic losses amount to hundreds of billions of dollars – in 2019 the figure reached $232 billion. The grim reality represented in these figures becomes even more critical when considering that both natural and human induced disasters such as floods, droughts, fires, extreme weather events and geo-hazards, are predicted to increase in frequency and intensity due to climate change, unplanned and rapid urbanization, poverty, and other factors.

Building resilience is a key factor in addressing this reality. In that regard, Earth Observation (EO) plays an important role as it allows for real- and quasi-real-time monitoring, forecasting and assessment of disasters and their impact. The advancement of both the EO sector and the EO-enabling technologies such as more efficient data gathering, processing, or analysing, make EO-based solutions more effective in supporting disaster resilience, and give rise to new market trends associated with significant opportunities for business growth.

Enabled by technological developments, EO has greatly increased its role in security and emergency response in the last 10-15 years. This is perhaps best exemplified in satellite-powered global real-time information services such as the Copernicus Emergency Management Service (EMS). Today, EO has a critical role in discovery, monitoring, and assessment of crises and emergencies, identifying affected areas, mapping infrastructures and their damages, evaluating how many people are at risk, and designing mitigation measures. Dr. Haris Kontoes, Research Director in the Institute for Astronomy and Astrophysics Space Applications and Remote Sensing of the National Observatory of Athens (NOA, leader of the e-shape’s Disasters Resilience showcase ),Coordinator of the BEYOND Center of Excellence, emphasises that “Earth Observation goes beyond monitoring and assessment of disasters”. Indeed, sophisticated simulation modelling that incorporates satellite data enables a variety of new added value services such as predicting bushfires in Australia and preventing large-scale damage in the future.

This capability can be nicely understood when considering floods, the most widespread hazard in Europe with around 85% of all European civil protection measures addressing this issue. For flood hazard and risk assessment, the images from Copernicus Sentinel-1A and -1B satellites are processed and classified in an automated way, combined with in-situ data and population density information. Integration of different data inputs, classification through trained algorithms and simulation modelling enable not only predicting the hazard, but also more efficient rescue and recovery efforts and fast and precise damage assessment. For instance, the flood maps created by the researchers of the Luxembourg Institute of Science and Technology (LIST), processed using the algorithm the scientists had developed, were used in the aftermath of the hurricanes Harvey and Irma that ravaged the United States in the end of summer 2017. Similarly, EO informs a better and more efficient crisis response and recovery in the case of other disasters.

The increasing role of EO in disaster resilience building is enabled by technology and data availability and quality. Copernicus Sentinels-1, -2, and -5p with their regular revisit times provide vast amounts of high-resolution data (SAR, multispectral) which can then be “translated” into actionable information thanks to advanced algorithms. As pointed out by Dr. Kontoes, “the current exponential growth in the amount of data brings great advantages, but also challenges for ensuring capacity and capabilities needed for processing it”. In numbers, NOA/BEYOND receives 80.000 images per day covering the whole globe and in a seasonal estimate for one region in Greece processes 1.000 terabytes of information. According to global estimations, less than 5% of available data is currently being used and less than 10% of disasters are mapped. Therefore, technologies and solutions such as Cloud Computing, data cubes, Cloud repositories, or machine learning are crucial. For instance, Artificial Intelligence (AI) is used in crisis management for applications such as the TSAR AI platform which automates analysing and annotating disaster maps.

By improving disaster resilience, EO has a role in achieving specific goals agreed by the United Nations and stated in major political frameworks – most notably the Sendai Framework for Disaster Risk Reduction and the Sustainable Development Goals of the 2030 Agenda ( SDGs: 2-zero hunger, 11-sustainable cities and communities, as well as 1-no poverty, and 13-climate action). Collaboration between international, but also national and regional authorities and EO actors is identified as an important pre-requisite for efficient resilience building. For example, institutions such as the National Observatory Athens partner up with both regional authorities and Copernicus for research and development of services enabling real-time crisis support for floods, fires, earthquakes, etc. Dr. Kontoes stresses the urgency of action needed from all involved parties: “We all know the specific targets we need to reach by 2030. We have the necessary data; we have the capacity. If we want to be successful in achieving these targets, we have to be chasing them today.”

The EO industry and its SMEs must build their maturity and convince the governments and private companies to use the solutions proposed by the EO community in the pursuit of these goals. In addition, there is a need to develop services evaluating the efficiency of the proposed solutions. “We need to be able to understand to what extent a particular service has addressed a specific target of the SDGs already today, in 5 years from now, and in 10 years from now, which is the timeframe by which the SDGs have to be achieved,” says Dr. Kontoes.

In recent years, the EO sector has extended the circle of the stakeholders involved in disaster resilience building, to include not only the civil society, civil protection authorities, or government institutes, but also private actors representing commercial sectors and different industries. In that regard, the know-how of modelling and simulations developed at the highest TRL levels can be used to propose services targeting markets which benefit the most from EO-powered risk assessment and forecasting. New stakeholders include insurance companies, the construction industry, energy companies, the aviation industry, and others.

The pilot projects developed within e-shape serve as concrete examples to that effect as they have been conceived with the vision of an “umbrella” of services helping authorities, but also enabling a commercial application. More specifically, within Pilot 6.1 (EO4D_ASH – EO Data for Detection, Discrimination & Distribution (4D) of Volcanic ash), meteorological modelling combining in-situ data and remote EO data can provide volcanic ash spread estimations to commercial airlines. Pilot 6.2 (GEOSS for Disasters in Urban Environment) enables more precise predictions for high impact weather events in urban and peri urban environment and provides decision-making support to Civil Protection Agencies, hydro-meteorological predictions agencies, and disaster risk reduction institutions. Pilot 6.3 (Assessing Geo-hazard vulnerability of Cities & Critical Infrastructures) allows detecting small-scale ground deformation and forecast damages to pipeline networks, informing energy companies on which parts of their underground network need inspection to avoid leakage damaging for both the company and the general society. Pilot 6.4 (ReSAgri – Resilient & Sustainable ecosystems including Agriculture & food) integrates an advanced hazard forecasting system with EO based assimilation processes and provides risk assessment at parcel scale to insurance companies, agriculture agencies, farming cooperatives and even individual farmers.

The insurance sector can be singled out as one of the sectors that can the most directly benefit from both the technology, as well as the vast expertise in risk evaluation and damage assessment that can be provided by EO companies. EO helps building the competitiveness and viability of the insurance sector by increasing the accuracy of risk assessment and reducing the costs of damage assessment. For instance, CyStellar uses machine learning enabled algorithms to analyse multispectral, high resolution images provided by Copernicus Sentinel 2b satellites. The images inform about the state of the roof, presence of solar panels (increasing risk of accidents, damages, fire), state of other nearby buildings, any vegetation (increasing risk of fire), and other risk increasing or mitigating factors. This information is coupled with other data from the insurance company and third parties to provide a more informed assessment of the rate of the insurance policy. Furthermore, EO can enable expanding insurance business to areas where no viable risk assessment data has been previously available. For example, Geodata for Inclusive Finance and Food (G4IFF) initiative can provide both the insurance and the smallholder farmer crucial information on weather, drought, and soil moisture, thus providing access to financial services for farmers in the most rural areas.

Despite its strong demonstrated capability to help in disaster resilience building, the EO sector has many technical challenges to face. This includes issues such as constant calibration and validation, including complex weather influences such as strong wind, adopting systematic and automatic mapping of large areas to improve the reliability of models, or improving mapping exercises for complex areas such as urban spaces. These technical or scientific challenges are tackled and improved together with the development of the upstream sector. According to Dr. Kontoes “the main challenge that needs to be addressed at this stage, is the upscaling of these solutions and engagement of the user communities.” An important barrier towards a more widespread development of market applications is the need for capacity building to enable and spread the use of these high-level services. In other words, users need to be trained and scientific information transformed into information accessible by the user. In addition, knowledge of the specific needs of sectors can be improved to tailor solutions that are perhaps less complex, but therefore target these needs more precisely. Alongside this capacity building, constant re-evaluation of skills or gaps in skills is necessary to maintain the competitiveness and advancement of the sector.