With climate disasters escalating, ground sensors – such as those used in flood situations reach their limits, while satellites can continue to observe. Through FLOODCARE, Sopra Steria’s CS Group is using space data to help emergency services respond faster and more effectively, explains Adrien Pain, Head of Space Consulting Europe at Sopra Steria’s CS Group.
As natural disasters intensify, civil protection is undergoing a profound transformation. Long based on ground sensors, human observations and local hydrological models, it is now being enriched by a new, strategic source of informational space data.
Beyond that lies a blind spot that satellites can address.
“The satellite makes it possible to cover entire areas, with data going back to more than ten years in the case of Copernicus, and in all weather conditions, day and night,” explains Adrien Pain, Head of Space Consulting Europe at Sopra Steria’s CS Group.
With a resolution that can reach up to 30 centimetres and an image acquisition time that, in favourable cases, can be reduced to between 45 minutes and two hours, space data transforms what was a logistical constraint into an operational advantage.
Rather than competing with existing sensors, Sopra Steria’s CS Group chose to complement them. It achieved this not by operating satellites, but by building the digital chain that transforms spatial imagery into enriched information directly usable by civil protection field operators.
FLOODCARE: Transforming imagery into operational decision-making
FLOODCARE, developed by Sopra Steria’s CS Group within the framework of the France 2030 programme, addresses a very concrete problem faced by those in charge of crisis management.
In a flood situation, a prefect or a head of fire and rescue service a (SDIS - service départemental d’incendie et de secours doesn’t need a raw satellite image. Instead, what is needed is to know whether their teams can reach affected areas, and under what deployment conditions.
FLOODCARE provides four essential pieces of information. It includes an initial rapid outline of flooded areas and a second, and more precise outline to refine the assessment; accessibility to buildings and populations; water depth; as well and a trafficability index.
This final layer translates the image into information that can be used directly for operational decision-making.
“The challenge is not to deliver a raw image, but to indicate to emergency services by what means a flooded area can be reached — light vehicle, reinforced means, or no access at all” says Adrien Pain. This creates a three-level interpretation, that is immediately usable to guide intervention.
Users don’t need to adopt an additional portal to receive this information, either. Instead, it is integrated directly as data layers into CRIMSON, the crisis management platform from Sopra Steria’s CS Group, already used by COGIC (interministerial crisis management operational centre), the COZ (Zonal Operational Centres), and by around 80% of French SDIS organisations.
Integration is also possible within prefectural information systems (SYNAPSE), the NEXSIS crisis management system of the civil security digital agency, but also within geographic information systems such as QGIS or ESRI’s ArcGIS.
The choice is deliberate. It is not to add one more tool, but to enrich environments already mastered by emergency services.
This logic of integration, rather than substitution, constitutes one of the strongest drivers for adoption of the system.
A processing chain designed for continuity of service
FLOODCARE brings together six complementary actors within the framework of specifications jointly produced by CNES and the Ministry of the Interior under the France 2030 programme.
Sopra Steria’s CS Group oversees the overall architecture as well as the trafficability service.
Rapid mapping of flooded areas is entrusted to SERTIT (Service Régional de Traitement d’Image et de Télédétection) and CLS (Collecte Localisation Satellites), mobilised alternately in order to guarantee redundancy and continuity.
Vortex-io deploys monitoring stations under bridges in order to measure water height and detect debris jams, with such accumulations likely to weaken infrastructure. Meanwhile, INRAE (National Research Institute for Agriculture, Food and Environment) contributes flood history and terrain models, while HydroMatters contributes hydrological modelling.
The entire system operates through a coordinated chain designed to deliver information that can be used immediately by decision centres.
This distributed architecture imposes a major technical requirement, that of processing images originating from heterogeneous sensors, whether radar or optical data and from different providers.
And that can be challenging. “
“It is like working with a Nikon, a Canon or a Fujifilm, the rendering varies, and the algorithms must adapt to it,” explains Pain.
In return, this diversity of sources guarantees continuity of service, since: no single provider is capable on its own of determining how the system functions.
Assessing the system in use
The first activations under real conditions made it possible to evaluate the system’s contribution.
During the February 2026 floods, when France experienced its wettest month on record, FLOODCARE was mobilised continuously for more than fifteen days. It covered, nine zones and processed thirty-four images, with delivery times reduced to as little as 47 minutes in the most favourable cases.
Pain says that an initial test had already taken place as early as 2025 with Storm Herminia: “even though the project had not yet been fully launched, with first outputs delivered within a few hours”.
In French territories already heavily instrumented and monitored by the Vigicrues network, satellite data mainly confirmed field observations and made it possible to build a database of flooded extents.
By contrast, in the face of less predictable events, such as a dyke breach or a crisis affecting isolated areas, the satellite’s comprehensive view comes into its own.
It is in situations such as these, when ground sensors are no longer sufficient or become unavailable, that spatial imagery most clearly reveals its operational value.
Space sovereignty: From dependency to the need for autonomy
The war in Ukraine made visible a fragility that few actors had anticipated. Access to satellite imagery remains dependent on commercial providers whose priorities do not necessarily coincide with those of their clients.
“There is a real risk of denial of service,” says Adrien Pain. “A government can requisition a commercial satellite for its own needs, making the resource unavailable to others.
The parallel drawn with the threat of a Starlink shutdown in Ukraine or restrictions on access to satellite imagery in the conflict with Iran gives tangible reality to this risk, which until now had remained largely theoretical.
The European Union is now seeking to reduce this dependency through the EOGS (Earth Observation Governmental Services) programme, which aims to deploy, by 2030, a sovereign capability that is much more responsive.
The ambition is clear; to provide an image of any area within 30 to 45 minutes, thanks to a three-layer architecture combining governmental satellites, member-state constellations and commercial providers.
The issue is not only one of capability; it is also geopolitical – avoiding a situation in which a provider prioritises other clients or other strategic interests.
Sopra Steria’s CS Group already aligns itself with this sovereign approach through an open-source technological approach and a deliberately diversified portfolio of image providers.
“Our role is the digital architecture downstream of the satellite: securing flows, merging and enhancing data, distributing it to field operators,” explains Adrien Pain.
Towards broader use in the service of territorial resilience
The technological building blocks of FLOODCARE are intended to extend beyond the field of flooding alone.
Wildfires, drought, and even certain flooding or landslide phenomena already feature among the avenues being considered.
Monitoring water reserves, already operational across more than 55,000 French bodies of water according to Adrien Pain, “opens new uses: production of drought indices, identification of reserves available for water-scooping operations, or support for planning and territorial governance of water”, he says.
Over the longer term, these services could contribute not only to crisis management, but also to preparedness, territorial planning and the resilience of local authorities facing climate hazards that are expected to multiply.
The institutional target is also expanding.
To date, only the highest national authorities can activate the Copernicus Emergency Management Service (CEMS).
Municipalities and metropolitan authorities remain dependent on a decision that does not belong to them.
Sopra Steria’s CS Group intends to change this within eighteen to twenty-four months.
“The objective is for a local authority to be able to activate the service itself as soon as it has the need, without depending on a higher authority,” says Pain.
By progressively developing a territorial resilience capability rooted in operational uses, Sopra Steria’s CS Group is shaping an approach set to become central as spatial data establishes itself as a critical infrastructure for crisis management.