From Earth Observation Data to Flood Risk Management Services

Abstract

Climate change is making extreme rainfall and flooding more frequent, and small islands such as Bonaire are especially exposed. Infrastructure and economic activity are concentrated in low-lying coastal areas, while the local capacity and data to manage flood risks are limited. The risks became visible during the extreme rainfall of November 2022, when roughly 80 mm fell in a single night, submerging streets and homes, closing the hospital and schools, and sending polluted run-off onto the coral reefs. The economic stakes are high: storms and floods are projected to cause around US$317 million in damage on Bonaire by 2050, and degradation of the reefs that buffer the coast could contract the island’s economy by up to a quarter of its GDP. In January 2026, a court ruled that the Netherlands had not adequately protected the island, placing flood risk higher on the political and legal agenda. Earth Observation (EO) data can help monitor and predict floods, and for the Dutch Caribbean it is already openly available. Raw satellite data, however, cannot be used directly: it must be processed into products that local stakeholders can understand and act on. This study asks how EO data can be translated into entrepreneurial flood risk management services on Bonaire.

The research is a qualitative, exploratory case study built on 21 semi-structured interviews with stakeholders across the local and national government, knowledge institutions, nature organisations, the private sector, and EO companies. The interviews followed a needs-first (Mom Test) logic, probing actual practices and past problems rather than hypothetical demand. Transcripts were analysed through
a hybrid deductive-inductive thematic analysis in ATLAS.ti, using open, axial, and selective coding, which produced seven themes. Two theoretical lenses structure the analysis. The EO value chain, running from data providers through service providers to customers, organises the needs by the level at which they arise. Bryson’s power-interest grid maps stakeholders by their power to act and their interest in flood risk, locating where demand sits and where the power to fund and adopt a service lies.

At the data level, satellite coverage is abundant but insufficient on its own: global datasets are too coarse for a small island, with grids on the order of 100 km, and outputs need in-situ ground validation, historical imagery of past events, and currently missing base registrations before they become reliable.
Subsurface conditions, such as the porous coral dike that carries seawater hundreds of metres inland, are poorly captured by standard EO. At the service provider level, the dominant need is to turn raw data into products that non-technical users can operate, integrating optical imagery, aerial photos, and LiDAR-derived elevation models, and reporting the uncertainty of forecasts and models rather than obscuring it. Matched to these needs, the most suitable modalities are high-resolution optical and LiDAR elevation data for flood mapping and spatial planning, cloud-penetrating SAR for monitoring through the rainy season, and optical time series for mangrove monitoring, the last being the most successful pilot application observed. At the customer level, four needs recurred: restoring water storage through dams, rooien, and saliñas; improving drainage; a faster and more reliable early warning
system, which is currently absent and runs through informal channels; and greater awareness of what EO can do. Most applications fall under flood prevention, with response and recovery still largely unexplored.

The central finding is that the binding constraint is not the availability of satellite data but the translation of that data into accessible, usable products. This service provider role exists on Bonaire only in fragments, through individual projects and pilots, rather than as a continuous, locally embedded service.
Three findings explain why the link is hard to fill. Awareness of EO is low across most user groups, though it varies, with nature organisations and some individuals inside government more aware than others. The main obstacle is financial and organisational rather than technical: services struggle to stay funded once a pilot ends, in a small market that leans heavily on project-based public money, and
the party that would pay is rarely the party that benefits. Finally, adoption depends on governance and trust. The power-interest grid reveals a structural mismatch: the actors with the most interest in flood risk, such as residents, nature organisations, and infrastructure operators, have the least power, while the powerful actors, mainly the Ministry of Infrastructure and Water Management (IenW) and the Ministry of the Interior and Kingdom Relations (BZK), have a distant interest, and real estate developers hold an interest that runs counter to flood management.
EO data can support flood risk management on Bonaire, but the data alone is not enough. It becomes useful only when a service turns it into usable products, funding keeps that service running, and local support ensures it is actually used.

The recommendations are therefore directed at the two groups whose action matters most.
For an entrepreneur, the opportunity lies in the service provider role: turning raw data into an integrated tool that combines elevation, water flows, and infrastructure with local data and knowledge, bundled with education and engagement, and targeting prevention first.
A commercial venture aimed at Bonaire alone is unlikely to be viable; a regional service spanning several islands, which spreads the fixed cost of acquisition and processing across more users, is the more realistic path. Because adoption depends
on trust, such a service would need to work through a local champion, possibly one of the EO-aware individuals already within the OLB.
For the government, the priority is continuity, since the main obstacle is not building a service but keeping it running. This calls on BZK and IenW to move from project-based towards structural funding, comparable to how the Delta Programme secures water safety in the European Netherlands. It also requires closing a governance gap: with neither a province nor a water board, responsibility for flood measures falls between the national government and the OLB, so BZK, as the coordinating ministry
for the special municipalities, should assign a clear owner for such a service rather than leaving it in that gap. At the island level, the OLB can strengthen the basis for permitting itself, where procedures are currently slow because the underlying rules are unclear or absent. Setting clear criteria against which applications are tested, using the zoning-plan revision to assess flood risk at new developments,
and introducing drainage requirements for private developers would give both the zoning plan and EO-based risk information a concrete point of entry into permitting decisions.
The first step, however, is not technical but a matter of building awareness and capacity among the users closest to the problem, and the upcoming zoning-plan revision, the developing climate plan, and the January 2026 court ruling open an institutional window in which to do so.