BirdWatch - a Copernicus-based service for the improvement of habitat suitability of farmland birds via satellite-enabled monitoring, evaluation and optimisation of CAP greening measures

Informations projet

BirdWatch

N° de convention de subvention: 101082634

DOI

10.3030/101082634

Projet clôturé

Date de signature de la CE 9 Decembre 2022

Date de début 1 Février 2023

Date de fin 31 Janvier 2026

Financé au titre de

• Digital, Industry and Space

Coût total

€ 2 581 641,25

Contribution de l’UE

€ 2 248 116,00

Coordonné par

LUP-LUFTBILD UMWELT PLANUNG GMBH
Germany

Periodic Reporting for period 1 - BirdWatch (BirdWatch - a Copernicus-based service for the improvement of habitat suitability of farmland birds via satellite-enabled monitoring, evaluation and optimisation of CAP greening measures)

Période du rapport: 2023-02-01 au 2024-07-31

Résumé du contexte et des objectifs généraux du projet

BirdWatch develops a monitoring tool for farmland bird habitat suitability, based on rigorous species distribution modelling and EO-based geospatial data to derive habitat suitability at the scale of an individual farmland parcel up to landscape and national scale.
On a local scale, BirdWatch will provide the users with the information on the agri-environmental measures (AEM), which apply in their region, and contrast them with capabilities to implement them, including budgetary and operational constraints, to identify viable pathways for habitat improvement. To determine the most appropriate choice of AEMs with respect of the available ressources, BirdWatch will optimise for the benefits of both, bird species and farmers, thus providing them with a decision support for financial and operational planning. On a regional scale, supervisory institutions, including environmental and paying agencies, will have a tool to compare the implemented AEMs against indicators, which reflect the actual state of habitat suitability of the farmer's parcels. On a national scale, policy-makers will be able to evaluate the efficiency of their guidelines by being able to evaluate the impact of the chosen AEMs. Thus, they will have access to data-driven indicators which can support decision-making for future policy guidelines.
BirdWatch's main internal components will consist of three main elements: a) a rigorous species distribution modelling framework which translates habitat requirements into a probability of occurrence; b) a large set of satellite-based geospatial indicators, chosen to reflect both the habitat and the agricultural setting; c) a powerful land-use allocation algorithm, which is able to identify the optimal pathway under a set of given constraints.
Access to service will be given via a web-based application.
With the BirdWatch service, we aim to make a signifant contribution to the goals of the European Green Deal, especially supporting the Farm to Fork and the EU's biodiversity strategy for 2030. BirdWatch will provide a monitoring service on the state of habitat suitability of the EU's farmland and help to evaluate current policies as well as their implementations and to identify necessary new guidelines to improve the health of our ecosystems.

Travail effectué depuis le début du projet jusqu’à la fin de la période considérée dans le rapport et principaux résultats atteints jusqu’à présent

The first iteration of the underlying workflow to establish the evaluation and monitoring of farmland bird habitat suitability has been established and includes the identification of the target environmental input parameters for the evaluation of habitat suitability, together with data constraints and necessary steps to prepare these parameters as inputs for SDM.
It also includes the setup of a bird observation data preparation workflow for running the SDM as well as the training and testing of the first habitat models based on the target environmental parameters.
Both the workflows for environmental parameter input preparation and bird data pre-processing are being used to establish the first habitat suitability models for each test region.
The optimisation models are strongly interlinked with the modalities under which the habitat suitability is being evaluated. The optimisation workflow is based on the principles of spatial allocation optimisation thus allowing the setup of dummy models which later on can be filled with proper input data. These dummy models were established by the end of last year to test the optimisation logic and computational constraints. This was followed by the identification of target areas and optimisation scenarios for optimisation in each of the test regions, based on existing or planned bird species protection programs, regional needs for environmental improvements (i.e. agriculturally intensive landscapes) and implemented policies (i.e. the presence of financial support for AEMs). Presently, the output of the first habitat models is being adapted to serve as an input for the first optimisation scenarios for Flanders.
The technical and operational requirements of the web-based platform were collected and will be revisited continuously throughout the project. A minimum set of platform requirements were derived via stakeholder questionnaires, discussions with stakeholders as well as in our technical and stakeholder meetings. A central component of the platform is the backend database which has been implemented already. It serves as the connecting node between the evaluation/monitoring and the optimisation services but will also store user and administration data once the platform has been launched officially. The first implementation of the platform has been finalised this summer and will serve as input for internal and user testing.

Progrès au-delà de l’état des connaissances et impact potentiel prévu (y compris l’impact socio-économique et les conséquences sociétales plus larges du projet jusqu’à présent)

1. BirdWatch will enable monitoring farmland bird habitat suitability using EO data.
BirdWatch automates the analyses of the structural, phenological make-up of parcels and evaluates them in terms of bird-specific habitat criteria. Satellite-based geoinformation for comprehensive habitat suitability monitoring is currently not available at the scope foreseen in BirdWatch. BirdWatch, it is possible to move from a point-based farmland bird biodiversity index to obtaining a 2D distribution of the spatial variance of biodiversity and its changes over time.
2. Multitemporal, multi-modal EO data and EO data derivatives as input for SDM
The service fuses data from different sensors (multispectral and radar) to create biophysical indicators and textural measures to turn them into an input for SDM. Due to the repeated acquisition of satellite data, the monitoring of changes in the parameters relevant for habitat suitability is being made possible. Over time this will contribute to the understanding of the drivers of biodiversity loss as well as the effectiveness of measures. Additionally, the EO-data supplied by BirdWatch would not only serve bird species monitoring but can also be used for other species.
3. A habitat optimisation algorithm to trade-off biological and economic consequences of AEMs
A spatially explicit land-use allocation algorithm is being developed which analyses the biological and economic consequences of different AEMs. By considering bird preferences and stakeholder specific requirements, this algorithm defines the best land use for a specific parcel to maximise the number of breeding bird species present in the specific region. By providing evidence-based, algorithmically-derived recommendations, this will become a novel approach to derive the optimum AEMs.
4. Evidence-based evaluation of biodiversity and AEMs on different scales
As BirdWatch evaluates the current state of an area of interest, identifies its changes over time and provides recommendations for the most appropriate AEMs, the platform can become an integral, data-and evidence-based monitoring tool, e.g. for the administrations and policy-makers. Regarding biodiversity policies, policy-makers still base decisions on subjective input. With BirdWatch, argumentation, claims and decisions can be substantiated by actual measures.