Using Copernicus data to monitor economic development through the analysis of changes in land use
A proof-of-concept (PoC) activity explored how Copernicus data can be used to monitor economic development by analysing changes in land use and detecting new construction activities.
The ECB’s work and the exploration of Copernicus data
In line with its mandate to maintain price stability, the European Central Bank (ECB) conducts extensive macroeconomic analysis to assess economic developments and risks to the inflation outlook. This is supported by a robust framework of data, advanced statistical methods and an extensive suite of macroeconomic models.
The ECB strives to continuously feed its analyses and models with high quality data in terms of comprehensiveness, timeliness, and granularity, while operating in a context of evolving technologies and increase availability of data sources and non-traditional data. The development of the Copernicus Data Space Ecosystem has significantly enhanced the public access to and processing of Earth Observation (EO) data (e.g. satellite imagery, temperature, atmospheric, and radar data), while ensuring privacy by capturing environmental and geographical features at resolutions that do not allow the identification of individuals or personal activities. This innovation has unlocked opportunities for economists and researchers to obtain near real-time, highly granular information potentially relevant for economic analysis as a complement to traditional statistics.
To explore the potential of EO data in supporting the monitoring of economic development, a team of ECB staff designed a use case that was then developed as a proof-of-concept (PoC), facilitated by EUSPA. This initiative demonstrated the potential ability of EO data to detect early-stage changes in land use and the beginning of new construction projects with remarkable precision.
Andrej Ceglar (ECB Climate Change Centre) declared "Earth observation turns the planet into a real-time dataset - revealing shifts in land-use patterns as they emerge. This holds great potential to monitor economic development in real-time as well as detect the early impacts of natural hazards on communities and infrastructure."
Monitoring economic development from space
Unlike official macroeconomic statistics, that rely on surveys and national accounts and are processed and published with a delay, EO data enables continuous observation of physical changes on the ground, potentially providing near real-time insights on specific aspects of the economy.
After exploring a few alternative options that could be achieved with Copernicus data and within the timeframe of the PoC, it was decided to focus on the analysis of construction activity and urban expansion. From a technical perspective, construction activity can be identified with satellite imagery at a relatively early stage. From an economic perspective, an increase in construction activity can locally signal higher investment levels, job creation, and/or stronger demand for housing or commercial real estate.
Proving the utility of EO data through the PoC: detecting new construction activity
For the scope of the PoC, a representative area in Romania was selected. This region was chosen due to its notable growth over the past 10–15 years, particularly in infrastructure and real estate, as evidenced by both official statistics and anecdotal observations. The aim of the PoC was to assess whether EO data could provide, on a small scale, comparable insights into such development trends. Specifically, a filter (i.e. a “mask layer”) has been applied to the satellite images to identify areas that had transitioned from predominantly vegetated land to bare ground and then remained persistently bare over an extended period. This pattern of shift from vegetation to long-term bare ground was interpreted as indicative of land clearance for construction purposes.
The activity used a series of Copernicus Sentinel-2 images to analyse the area of interest. These images were trimmed to match the boundaries used by the Copernicus Land Monitoring Service (CLMS) and Urban Atlas (UA). To detect bare ground, a vegetation index was applied to highlight areas where plant growth was absent. This allowed to construct “bare ground maps” for each satellite acquisition. By combining these maps, it became possible to identify newly emerging bare ground and monitor how long specific areas remained in that state. Areas that exhibited prolonged periods of bare ground were correlated with construction activity. Subsequently, individual ex-post checks were conducted to validate the accuracy of this identification approach.
Oleg Reichmann and Lorenzo Ferrante (ECB) confirmed "the proof-of-concept (PoC) demonstrated the significant potential of Earth Observation (EO) data to make our analyses more granular and timelier, and to offer new insights by complementing traditional approaches. While many challenges remain before alternative data (like EO data) can be seamlessly integrated into our daily workflows, the PoC was a first step in that direction."
Looking Ahead
Through the lens of this PoC, the ECB could testify the potential of EO Copernicus data for monitoring local economic developments in near real-time. Looking ahead, the ECB will further investigate whether EO data could be incorporated into their workflows by refining the techniques developed within this PoC and scaling them up to fit larger geographical areas and/or additional applications.
Media note: This feature can be republished without charge provided the European Union Agency for the Space Programme (EUSPA) is acknowledged as the source at the top or the bottom of the story. You must request permission before you use any of the photographs on the site. If you republish, we would be grateful if you could link back to the EUSPA website.