A report presented by ESA and the European Commission's Joint Research Centre on 2 November 2026 raises a practical question: can we monitor progress in climate adaptation from space? The researchers examined 59 indicators of the global adaptation goal adopted as part of the climate negotiations. Their conclusion is nuanced. Eleven indicators contain physical components that satellites can measure directly; sixteen can receive an indirect contribution; and twenty-one can benefit from a context on risks or environmental conditions. For the others, no specific contribution has been established. No indicator can be fully measured by Earth observation alone.
Measuring a change does not prove adaptation
A satellite can track surface temperature, the extent of a wetland, or the color of water. It cannot, from a single image, establish that a policy has improved the health of inhabitants or that farmers have better weathered a drought. The first result describes a physical state; the second requires linking that state to actions, people, and decisions. The report emphasizes this distinction because the political interest in indicators can lead to the search for a simple answer in an available map.
Let’s take a city that is planting trees to reduce heat. Satellite images can help observe changes in vegetation cover and surface temperature. But they don’t tell us whether residents in the most exposed neighborhoods are actually experiencing cooler air, whether the trees survive for several years, or whether the observed decrease is due to an exceptionally wet season. We need ground-based thermometers, information about maintenance, social data, and a method to compare places before and after the intervention.
How the 59 indicators were ranked
The report summarizes a workshop held from 23 to 25 June 2026 at ESA's Harwell Centre, with participants from scientific, public and private organizations. They evaluated the potential contributions of Earth observation to global indicators. The categories are not ratings assigned to satellites. "Direct" means that a physical component of the indicator can be measured; "indirect" refers to a model or surrogate variable; "contextual" means that the data help to describe the exposure or environment without measuring the indicator itself.
The percentages published by ESA are 18.6% direct contribution for one component, 27.1% indirect, 35.6% contextual, and 18.6% without a specific contribution established. They should not be added together to claim that satellites "account for 81% of the adaptation". This formulation would blur the distinction between measuring, helping to model, and simply providing context. The report explicitly points out that none of the 59 indicators is fully covered by orbital observations alone.
Four avenues close to African needs
Among the initial priorities, the report mentions the environmental quality of water intended for supply, degraded land managed for food and agriculture, ecosystems benefiting from resilience actions, and the threat status of ecosystems. Each of these may be of interest to actors in Cameroon and Central Africa. Satellite data can help identify surface changes, build time series, and guide field visits. They do not replace water sampling or ecological inventories.
For a river near Douala, for example, an image may show a change in turbidity or surrounding vegetation under certain conditions. It does not, by itself, measure all the contaminants that determine whether water is safe to drink. Similarly, a plot that appears greener does not prove that agricultural activity has sustainably increased producers' incomes. The indicator must be defined before selecting the sensor; otherwise, the team may publish the data that is easiest to obtain rather than the data that answers the question.
The continuity of the measures is as important as the map
A repeated series allows us to distinguish a one-off event from a trend. The report therefore requires spatial observations and ground measurements to be maintained over time. A map produced once may launch a project, but it cannot follow an adaptation policy over several years. Spatial resolution, pass frequency, cloud cover and continuity between satellites must be documented. A method that works in a well-equipped city may fail in a municipality where field surveys are lacking.
The authors also call for transparent methods, information on uncertainty, and the integration of data into national and local systems. The report highlights the need for training and funding in countries and municipalities with fewer resources. An open data platform is not enough if no one can verify the results, preserve the series, or integrate them into public decision-making. Adaptation is a social and institutional process as much as it is a matter of pixels.
What RECA could experiment with
A useful demonstration would be to choose a concrete indicator, for example the evolution of a wetland exposed to urbanization, and to publish a complete protocol. This would include the question, the images selected, the dates, the clouds excluded, the field observations, the uncertainty calculation, and the decision that the results can indeed shed light on. The work could be reproduced by a partner university or municipality. A beautiful visualization would be only one step in this demonstration.
The announcement of 2 November thus opens a more serious work programme than the "satellite measures adaptation" formula. The authors show where the orbit really helps, where it provides only an indication, and where other evidence must be sought. It is this boundary, applied to local needs, that can make space data useful for climate policies.
Link people to physical data
An adaptation policy targets populations, services and livelihoods. A series of images may show that a vegetated area has changed, but it does not say who lives there, who has access to water, or who has benefited from an investment. Local surveys, administrative records and field observations remain essential. The ESA-JRC report helps to organise this complementarity: satellites repeatedly describe certain physical phenomena, while social and institutional data allow us to discuss the human consequences. A responsible analysis documents the dates and quality of both types of information.
The case of flooding illustrates this limitation. Sensors can map water after an event, sometimes even when clouds obscure optical instruments. However, to determine whether an adaptation measure has reduced losses, it is necessary to know the exposed buildings, the affected people, the protections installed, and the intensity of the event compared. A map of flooded areas is not sufficient to draw conclusions. It becomes more useful when it is part of a specific question, with observations before and after the measure, and a method to deal with differences between events.
Building a series that stands the test of time
International indicators require comparisons over several years. If images change instruments, resolution, or processing methods, apparent variation may come from the measurement chain rather than the field. The protocol must retain data versions, corrections, and assumptions. For an African observatory, this discipline is as important as initial access to satellites. It enables independent analysis by researchers and prevents a single map from being used as excessive evidence. The next concrete step is to choose a local case, publish the usable data, and honestly explain what the method does not measure.