Conference Agenda

talk

video

talk

talk

talk

(video recording)

talk

(video recording)

(video recording)

talk

talk

intro

talk

talk

talk

talk

talk

talk

Video Recording

talk

talk

Ecosystem extent information forms the foundation for numerous environmental analyses, serving as the baseline for understanding ecosystem condition, risks, trends, and the effectiveness of conservation and restoration efforts. Accurate data on ecosystem extent is essential for biodiversity assessments, climate change modeling, ecosystem services valuation, and land-use planning. It enables policymakers, scientists, and businesses to identify areas of ecological importance, track habitat loss, and prioritize interventions for protection and restoration.

The Global Ecosystems Atlas will provide this critical information through harmonized, high-resolution maps aligned with the IUCN Global Ecosystem Typology. The Atlas initiative aims to create a trusted, common map of the world’s ecosystems to facilitate consistent and coherent monitoring, reporting, and verification of conservation, sustainable management, restoration goals, and natural capital accounting. This will support users at national, regional, and global levels, including companies’ value chains and investors’ portfolios. At its core, the Atlas is a pioneering geospatial data product developed by integrating existing national and global data on ecosystem extent with new high-resolution Earth observation maps.

By offering a comprehensive and consistent view of global ecosystems distributions, the Atlas will allow users to perform more accurate analyses, inform decision-making processes, and meet reporting requirements under frameworks like the Global Biodiversity Framework (GBF) and the UN System for Environmental-Economic Accounting (SEEA EA).

The Atlas will:

• Integrate existing high-quality ecosystem maps, standardize and harmonize approaches to provide the best available spatial data on ecosystem extent, condition, and risks.
• Identify and fill knowledge gaps on ecosystem extent and condition using the latest Earth observation data, AI/ML technologies, and relevant ecological data.
• Provide tools to support global, regional, and national assessments, reporting, and accounting related to ecosystems.
• Enable businesses to develop coherent nature accounts and assess, report, and verify nature-related risks and key metrics with transparency and consistency.

The Atlas, currently available as a proof-of-concept, will continue to evolve as a collaborative resource for sustainability, risk management, and informed decision-making. It serves as a vital tool for achieving the goals and targets of the Kunming-Montreal Global Biodiversity Framework, ensuring that action is taken where it matters most.

Explore the proof-of-concept at globalecosystemsatlas.org

The Committee on Earth Observation Satellites (CEOS) was created 40 years ago as a way for the world’s civil space agencies to coordinate their activities and exchange ideas to support societal benefit and decision making. Areas of coordination include climate, disasters, capacity building, calibration/validation, and information systems as well as measurements for oceans, atmosphere and land surfaces. However, despite biodiversity’s importance to society and the critical role that Earth Observation (EO) plays in understanding, monitoring, and managing it, CEOS’s engagement with biodiversity has been minimal. To address this gap, CEOS is reaching out to a variety of biodiversity organizations including, among others, the CBD, IPBES, GEO BON and the GEO Ecosystem Atlas, and UNSEEA. The information gained is being used to create a path forward for CEOS and its agencies to better support biodiversity conservation and science and increase the societal impact of EO data.

As the coordinator for the world’s civil space agencies CEOS has tremendous potential to contribute to biodiversity conservation. This is particularly true in the context of numerous forthcoming missions, as sensor technology advances and gets into space, and as other technology such as AI and computing power move forward.

The Convention on Biological Diversity (CBD) calls upon member nations to report on national ecosystem conservation status using metrics on ecosystem extent for terrestrial, freshwater, and marine ecosystems. Similarly, the UN System for Environmental and Economic Accounting (UN SEEA) encourages nations to develop national ecosystem extent accounts. Several of the Sustainable Development Goals (SDGs) also include area-based conservation status metrics which require assessments of ecosystem extent. Both the CBD and UN SEEA processes encourage the use of the IUCN Global Ecosystem Typology (in particular the ecosystem functional groups from the third level of the hierarchy) as a reference classification. Ideally, these national ecosystem extent metrics would be produced by individual nations using a bottom-up, wall-to-wall, fine resolution ecosystem mapping approach. Notably, the GEO Global Ecosystems Atlas initiative has commenced an effort to produce a globally comprehensive ecosystem map as a synthesis and compendium of national ecosystems maps and other relevant ecosystem maps. Commendable progress has been made towards that goal, and a proof-of-concept characterization was presented recently at COP 16 in Cali, Colombia. The GEO Global Ecosystems Atlas initiative is recognized as a multi-year effort which includes a commitment to capacity building. It will be several years before a complete (globally comprehensive) bottom-up draft ecosystems map is available. In the meantime, the question of whether any of the existing top-down, standardized, globally comprehensive ecosystem maps have utility for national ecosystem conservation status reporting is often raised. In this context, the USGS/Esri World Terrestrial Ecosystems are discussed, exploring key dimensions such as mapping approach, source imagery derivation, compatibility with the IUCN GET, currency, spatial resolution, uncertainty, projected future distributions, etc.

For European habitat mapping with EO data and machine learning or deep learning techniques it is a prerequisite to obtain a large amount of in-situ habitat observations across Europe with a high precision and up-to-date. Up to now the training data for EUNIS habitats (level 3) is based on classified plot observations from the European Vegetation Archive (EVA). Although this database is huge in terms of number of vegetation plots (2,6 million) there are three important limitations: 1) Spatial limitation. Not all parts of Europe are evenly covered by plot data. Especially Scandinavia, Eastern Europe and the parts of Spain and Turkey are unrepresented in the database; 2) Temporal limitation. Especially for linking plot observations as ground truth to remotely sensed data, recent data is needed. Only half of the total number of plots (1.3 million) is recorded from the year 2000 and 0.72 million from the year 2010 onwards; and 3) Location uncertainty. The location uncertainty is a major issue in the EVA database. Apart from the fact that there are 343,000 classified plots without an indication of locational accuracy, there are only 183,000 plots with a location uncertainty of 10m or less. Taking only plots into account that have been recorded from the year 2000 the number drops to 115,000.

To fill the gaps in the synoptic observations of EUNIS habitats, the possibility is explored to use combinations of so-called opportunistic species observations. By using GBIF species observation data, extended with species data from EVA, the co-occurrence of species within grids cells of 10 by 10 meter and/or 100 by 100 meter can be used as a proxy for the presence of a EUNIS habitat type. To ease the process of finding optimal thresholds for each EUNIS habitat, an application (called the ‘Eunis Proxy Distribution Viewer’) has been developed, in which we analyse 52.2 million georeferenced plant species records in terms of co-existence of diagnostic, constant and dominant species at grid cells of 10 by 10 m or 100 by 100 m across Europe. The complete method of finding new potential locations of EUNIS habitats at level 3 for training or validation purposes is demonstrated, including the challenges. In the end all good habitat classifications depend on finding sufficient, up-to-date and well-distributed training data.

Earth Observation (EO) has the potential to enhance and accelerate ecosystem accounting within the SEEA EA framework, thereby offering the most economically efficient method for gathering extensive datasets in a standardized format, ensuring both spatial and temporal consistency. The European Space Agency (ESA) project “Pioneering Earth Observation Applications for the Environment – Ecosystem Accounting” (PEOPLE-EA) aimed to study and demonstrate the relevance of Earth Observation (EO) for ecosystem accounting in terrestrial and freshwater ecosystems. Ecosystem accounts are inherently spatial accounts, with the implication that they strongly depend on the availability of spatially explicit datasets. In particular, the development of extent accounts in selected test sites illustrated the changes in extent from one ecosystem type to another over an accounting period. To achieve this, we collected and integrated information from land cover, biodiversity, field surveys, ecological data, and other relevant factors to delineate and classify ecosystems based on their ecological characteristics, processes, and functions. Local, expert knowledge was also integrated in the data collection and validation phases and EUNIS Level 3 maps were generated through AI techniques as the final product. An approach to develop an independent change detection workflow provides a promising perspective, however further work is to be conducted to select the best deep learning network and training dataset to capture the expected transitions in ecosystems. The recently started World Ecosystem Extent Dynamics (WEED) ESA Project, will try make this further step on ecosystem extent mapping and accounting, by developing a global applicable open-source toolbox, leveraging existing datasets and tools while applying creative and novel methods to use EO, and enable users to generate comprehensive maps of the ecosystem extent and the distribution of terrestrial, freshwater and coastal ecosystem types and their temporal variations according to different ecosystem typologies. First approaches and outcomes are herein presented.

Mangroves are critical ecosystems, but data inconsistencies and lack of long-term monitoring hinder effective management in Tanzania. We present a comprehensive analysis of mangrove extent changes from 1990 to 2023, integrating historical and contemporary data sources.

We digitized and preserved unique paper maps from a 1989/1990 forest inventory—the first national-scale assessment of mangroves in mainland Tanzania—and combined these with Landsat, Sentinel-1 and -2 imagery, and training and validation points obtained from field validation using a custom mobile application and manual digitalization from Google Earth, updated with Planet NICFI monthly composites. Using Google Earth Engine (GEE) supervised Random Forest model and an online feedback tool with editable polygon capabilities, we integrated local expertise to iteratively improve the classification of mangrove extent and detect changes, achieving accuracies of 90% (1990) and 94% (2023).

Our integration of historical data, high-resolution imagery, robust machine learning models, and extensive validation addresses inconsistencies in previous estimates, providing an accurate, reproducible mangrove inventory. This foundation supports planning and conservation strategies, informing mangrove integration into the Tanzania National Forest Inventory. Uniquely, organizations from Tanzania, Germany, and the USA collaborated mainly remotely using online tools, integrating diverse expertise. Models and scripts are openly shared on GitHub, promoting transparency, reproducibility, and enabling future improvement.

Our findings close a fundamental data gap, informing the preparation of the national mangrove management strategy, action plan, and block management plans for mainland Tanzania and Zanzibar, ultimately supporting sustainable conservation of mangroves and the resilience of coastal ecosystems.

Biodiversity Net Gain (BNG) is a relatively new approach which seeks to deliver more sustainability-focused development, by creating or enhancing habitats to secure a net gain in biodiversity following construction. Demonstration of a net gain of 10% became mandatory for most proposed developments in England in February 2024. Land cover maps detailing the spatial distribution of UK Habitat (UKHab) classes are critical components of the baseline BNG assessments completed prior to development. Surveys to collect habitat data must be carried out by trained ecologists in-situ, requiring habitats to be classified according to type, condition and strategic significance. Such surveys are resource-intensive in terms of labour, cost and time, but recent advancements in both machine learning and remote sensing technologies may offer solutions to more rapidly assess BNG. However, existing methodologies for land cover classification may not capture the full complexity of natural habitats, especially for detailed biodiversity assessments. The development of vision-language models (VLMs) has the potential to improve land cover classification, as they enable the integration of visual and textual information. This information increases the understanding of the semantics required to identify and categorise different land cover types. However, few studies have assessed the application of this emerging technology in specific ecological contexts. Responding to this, our work shows that VLMs holds strong promise for automatic detection of land cover by interpreting visual features in the context of descriptive textual data, providing a comprehensive understanding of habitat characteristics. The presentation will show how VLMs can be used with Sentinel-2 and UK National LiDAR data to classify and track changes in UKHab classes. These results contribute to a better understanding of how advanced machine learning methods and open-source remote sensing data can be used to support sustainable development goals.

One of the factors that most limits utilization of EO by decision makers such as conservation managers and planners and those developing policy is the lack of “higher level” products. Because of cost and other factors, most space agencies do not routinely generate many products beyond Level 2 (e.g., Surface Radiance), however, applied biodiversity users need products such as the headline and other indicators of CBD’s Kunming-Montreal Global Biodiversity Framework. This gap between user needs and what is commonly available limits the impact that EO data should have. Bridging that gap is challenging but a federated system of systems approach may enable broader inclusion of organizations that can develop or generate needed products that help to fill that gap. Open, federated systems can grow organically based on well-defined interfaces that enable a range of organizations to participate based on their expertise and resources. This presentation will explore some of the pros and cons of such an approach.

Direct and indirect anthropogenetic activities are affecting global biodiversity, ecosystems functions and services as a whole and in an interconnected manner. Policies have set specific targets to be achieved and minimise these impacts. To measure progress towards these targets, a suite of indicators has been developed based on a combination of in situ data, predictions from models and remote sensing techniques (i.e. Earth Observations). Yet, collecting in situ data is still challenging, especially in remote areas (i.e. tropics, oceans) while predictive models can be uncertain and prone to errors, especially in data deficient areas. These limitations have led to an increased attention and use of satellite remote sensing biodiversity related products to support policy (through indicators) and science (through upscaling in situ observations). Because of the different communities using satellite data and different scales of processes involved, there is currently a disconnection between the different products for land-freshwater-marine ecosystems. Radiometric remote sensing measures the same physical properties across the domains, therefore providing a unifying perspective of the global ecosystem. This paper examines the commonalities across domains and identifies biodiverisity relevant products that offer potential for constructing global satellite derived datasets of biodiversity and environmental (abiotic) drivers through a critical literature review incorporating recent outcomes from projects under the Biodiversity+ Precursors/ESA Flagship action (e.g., EO4Diversity, BiCOME, and BIOMONDO) and ESA Ocean Health (BOOMS). These findings highlight key areas for future research and suggest that further efforts should be invested to enhance the understanding of the biosphere's response to multiple drivers. We highlight the need for global, climate relevant, satellite derived biodiversity and environmental (abiotic) drivers variables datasets across domains (i.e. datasets with long term ambition).

Earth System Science and Earth observation (EO) play a prominent role in the scientific understanding of ecosystems, ecological and biophysical processes. The next generation of biodiversity observing systems and science-based solutions will address the main causes and drivers of biodiversity loss, improving the conservation and restoration of vulnerable ecosystems. However, the massive amount of EO data poses a challenge for scientific applications dedicated to both research and operational use. The Committee on Earth Observation Satellites (CEOS) formed the Ecosystem Extent Task Team (EETT) in 2022 to investigate the use of Earth observation data to support the critical Biodiversity variable of Ecosystem Extent. One objective of this team is to develop data-cube based demonstrators to prepare future biodiversity monitoring systems.

The EETT of CEOS is currently exploring possibilities offered by the combination of cloud infrastructures and standardized protocols for producing essential biodiversity variables. A demonstrator based on free and open software is currently being developed to improve our capacity for ecosystem extent and condition mapping, forest biodiversity mapping and forest degradation and dieback monitoring based on Sentinel-2 time series processing over large scales.

We present results derived from Costa Rican forest ecosystems, leveraging Sentinel-2 data catalogs to generate advanced data cubes that incorporate spectral indices and spectral diversity metrics. The influence of cloud cover on spectral indices composites is analyzed to enhance temporal and spatial consistency. . The spectral diversity maps are subsequently compared with ground-based observations on forest types and species composition, as well as species distribution models.

Urban trees are essential for supporting biodiversity, as they provide habitats for various species and help regulate water storage and temperature, and sequester CO₂ in urban ecosystems.Urban forests have been proposed as a nature-based solution to fight climate change and provide ecosystem services to citizens. Mapping and monitoring urban trees is vital as it facilitates conservation strategies for both flora and fauna, early diagnosis of plant pathogens, and zoning and urban development. However, mapping trees has proved difficult for urban planners since they rely on in situ surveys or community-led projects that may not cover all areas; one such case is London, where the official survey only accounts for ~10% of the estimated 8 million trees in the city. Moreover, the geographic coordinates of trees are surprisingly unreliable due to a lack of precision of measuring devices (e.g. phones or commercial GPS).

We propose a method for calibrating urban tree locations using physical ground sensors as "anchors”. These sensors help reconcile spatial mismatches across various spatial datasets, including high-resolution satellite and aerial imagery and tree surveys collected by city councils or in open-data projects like OSM. These low-power sensors can also collect microclimate and other biodiversity-related data, such as passive acoustic animal activity monitoring, providing a richer picture of tree and urban ecosystem health and enabling high resolution maps not previously possible.

Our ultimate goal is to combine remote sensing information with ground-based measurements to support reliable data that can be used in geographic-based foundation models to help better urban planning strategies around trees that maximise their benefit to humans and nature.

Half of Germany’s land is used for agriculture, making it a crucial factor in the conservation and promotion of biodiversity. Land use, its intensity, and management practices shape biodiversity and consequently affect ecosystem functions and services. Farmland habitat status and quality can serve as proxies for assessing biodiversity, but developing reliable indicators, particularly for evaluating habitat quality, remains challenging. To advance the development of such indicators, a generalized workflow for a farmland habitat biodiversity indicator (FHBI) was proposed by the OECD. Here, agricultural landscape and management diversity is utilized as a proxy for biodiversity. Based on already available data, FHBI can be derived by assigning habitat quality scores of individual land cover classes based on their assumed influence on farmland biodiversity. The aggregated indicator aims at enabling the assessment of the status and trends of farmland habitat quality.

We present the initial results of a pilot study in which we calculate an FHBI for Germany for the period 2017-2023. The FHBI is based on a combination of different datasets derived from satellite data at an aggregation level of 100 ha hexagons. First, we derived pixel-level structural and functional crop diversity based on the frequency and duration of crop sequences and the share of cereal, leave, summer, and winter crops. Second, we derived grassland-use intensity as the number of mowing events detected from satellite time series. Third, we complemented these datasets with maps of small woody features and other perennial land use classes. Fourth, we assigned habitat quality scores ranging from one to five to each pixel in the combined land use (intensity) map. Finally, we calculated the area-weighted average of habitat quality values for each hexagon. Our FHBI map provides a first insight to detect large scale spatial patterns and differences in farmland habitats across Germany.

Soil biodiversity is essential for ecosystem health, driving multiple ecosystem functions and services. However, predicting multi-trophic soil biodiversity remains challenging due to complex environmental interactions and limited observation methods. Our research explores the potential of Earth Observation Foundation Models, using orthophotos with soil, climate, phenology, and landscape tabular data to enhance biodiversity predictions through machine learning (ML).

We collected relative abundance data for 53 trophic groups in the French Alps from the ORCHAMP observatory, which were pre-computed through eDNA metabarcoding and covered categories such as bacteria, collembola, fungi, insects, metazoans, oligochaetes, and protists. We gathered in-situ soil data, CHELSA climate data, COPERNICUS phenology, and THEIA OSO land cover. Additionally, image embeddings were extracted from 20cm resolution IGN Orthophotos using a state-of-the-art self-supervised Dinov2 model pre-trained on satellite imagery.

We built regression models for each trophic group, leveraging Light Gradient Boosting Machine (LGBM), Random Forest (RF), and Neural Networks (NN). The best R² values reached up to 0.82 for models based solely on tabular data, 0.73 using orthophoto embeddings, and 0.81 combining tabular data with embeddings. The latter often yielded results comparable or slightly lower than using tabular data alone. Principal Component Analysis (PCA) suggests that orthophoto embeddings capture similar, yet less comprehensive, information compared to tabular data. All this highlights the complexity of modeling soil biodiversity, where trophic group-specific characteristics pose distinct challenges.

While models based on tabular data performed best, image-based models offer a viable alternative when traditional data collection is impractical. Our dual-approach strategy - using either comprehensive tabular data or orthophoto embeddings - achieved comparable predictive performance for soil trophic group diversity.

This research demonstrates the potential of Earth Observation data and deep learning models as scalable tools for predicting soil biodiversity and improving our understanding of ecosystem structure, function, and resilience under changing environmental conditions.

Oceanic islands harbour high levels of species endemism and rarity, and are particularly vulnerable to habitat loss, introduction of invasive species, and climate change. Therefore, national, transnational, and global initiatives should explicitly include islands by considering essential biodiversity variables (EBVs) that capture their ecological uniqueness and tailoring protocols to the island context. In BioMonI, we endeavor to build a roadmap for long-term monitoring network adapted to the pressing needs of biodiversity conservation and monitoring on islands of the European Union and beyond. Our working group aims at scaling up island ecosystem monitoring, focusing on multiple EBVs. As such, we will integrate satellite remote sensing data from LIDAR, spectral, and hyperspectral sensors (such as MODIS, Landsat, Sentinel, GEDI and EnMAP) with in situ biodiversity observations.

In 2024, we collected terrestrial laser scans across five major habitats of Tenerife, Spain. We derived structural complexity indices to quantify the occupation of 3D space by the vegetation. Our results demonstrate that these indices can be used to characterize relevant aspects of the structure of different types of habitats, expanding the use of this technology beyond forest ecosystems. We will discuss the future directions of our methodologies to develop island biodiversity monitoring protocols and create an integrated dataset that incorporates measurements from field to space.

The Arctic and Antarctic are changing rapidly through amplified warming at the poles, impacting biodiversity and ecosystem functioning, with major feedbacks to the Earth system. In addition to climate, multiple other drivers of change of biodiversity become stronger, such as increasing industrial activity, tourism, wildfires, plastic pollution, and invasive species. In line with these rapid changes, the planning of the 5th International Polar Year in 2032-33 has started. The 5th IPY provides a unique platform for coordinated interdisciplinary research efforts in the Arctic and Antarctic, involving polar researchers, knowledge holders, rights holders and stakeholders. With this presentation, we aim at initiating discussions on how satellite observations and measurements on the ground could be coordinated and combined for biodiversity assessments in the Arctic and Antarctic. We highlight the need for coordinated measurements, data processing, synthesis and data management in order to close major knowledge gaps and feed results of the polar regions back to global biodiversity efforts and efficiently inform policy makers.

Ecologists suggest that the majority of land animals are either nocturnal or active across both the day and night. It has also been demonstrated that artificial illumination fundamentally impacts animal behavior. However, little information is available about artificial illumination potentially affecting these behavior at large scale factors. In our study, very high resolution multispectral night images from Jilin satellite are used to map artificial light in a rural area of Belgium. The geometric precision and the detection rates of light sources are estimated using a database of public lights and field visits at night. Other artificial light sources, mainly from residential and industrial areas, are also identified to derive “dark frame” connectivity in the study area. In addition, the discrimination of the type of light sources (light emitting diode or sodium vapour light) is measured because it has been hypothesized that the impact on animal behavior was linked with the color temperature. The critical analysis results demonstrate the huge value of multispectral night images for biodiversity studies, but also the need of higher temporal resolution.

In adopting the Kunming-Montreal Global Biodiversity Framework (GBF) and its respective monitoring framework, the Parties to the Convention on Biological Diversity (CBD) committed to establishing national goals and targets for biodiversity and reporting on their progress towards achieving them. The GBF indicators take a pragmatic approach to quantify, monitor and report on the status of biodiversity. At the same time, it is crucial to detect changes in fundamental biodiversity components and attribute those causally to drivers. Repeated, global observations from satellite remote sensing provide a unique opportunity for regularly updated biodiversity products and ultimately Essential Biodiversity Variables (EBVs) i.e., parameters that capture key aspects of biodiversity, to monitor and explain change over time.

Together with workshop participants, we will examine and discuss several questions to incorporate different expertise and perspectives:

1. Which EBVs are relevant for the targets submitted by the EU and member states to the CBD?

2. Reviewing remote sensing biodiversity products and how these can be further developed or strengthened through Earth observation to deliver the required indicators and EBVs?

3. What needs to happen on the science policy interface to support the use of EBVs and improve indicators?

4. How can biodiversity change be detected and attributed to drivers, and how can uncertainty be handled and communicated?

Expected outcomes: The expected outcomes of the workshop includes:

1. A roadmap, setting out what can be achieved from the EO community within the next 5 years to support countries reporting efforts.

2. Shared understanding within the earth observation community on RS biodiversity products and EBVs for monitoring biodiversity changes (including genetics).

Participants should note that the workshop will include a dedicated case on genetic diversity.

talk

Video Recording

Biodiversa+, a multinational partnership co-developed with the European Commission, aims to support biodiversity goals and harmonise monitoring methods across Europe. Globally, natural habitats are increasingly degraded, making new conservation and restoration actions a key priority in the EU's Biodiversity Strategy and Nature Restoration Law. However, inconsistent mapping and monitoring methods hinder effective assessments and conservation planning for valuable habitats. To address this, Biodiversa+ launched the Habitat Pilot.

Remote sensing (RS) offers a cost-effective solution for large-scale habitat monitoring but is underutilised, particularly for high-value habitats, such as those listed under the Habitats Directive. The pilot focuses on testing the applicability of RS methods in two European-wide, threatened habitat types: grasslands and wetlands.

The pilot includes four modules:

1. Review of existing habitat mapping and monitoring methods
2. Application of RS methods to map grasslands and wetlands
3. Use of RS to monitor habitat conditions
4. A synthesis and roadmap for future work

In its initial phase, the pilot involved 11 European countries and reviewed over 40 habitat mapping approaches, evaluating their strengths, weaknesses, and potential for integration into a standardised monitoring framework. Data availability among the partners was also assessed.

The review revealed regional differences in the use of RS technologies, with some areas more advanced and others still more reliant on traditional field methods. Despite these differences, a set of shared RS-based approaches was identified for testing in the subsequent pilot modules.

The pilot is linked with ongoing projects like EU Grassland Watch and integrates new modelling frameworks such as NaturaSat, alongside locally developed methods. The overarching aim is to support knowledge sharing, comparison, testing, and adaptation of methods to pave the way for transnational, harmonised RS-based biodiversity mapping and monitoring.

Satellite remote sensing is playing an increasingly important role for nature conservation agencies by providing spatially explicit and temporally dense data for monitoring and evaluating ecosystems and their use.

From the perspective of a national nature conservation agency, remote sensing methods offer important support in the following areas of application:

• Monitoring of biodiversity, landscapes, as well as the drivers of their change

• Fulfillment of reporting obligations from the local level to the EU, such as

• Enforcement of nature conservation laws

• Monitoring the effectiveness of nature conservation measures

On the basis of the fields of application mentioned, we show how remote sensing is already being used by nature conservation authorities in Germany. Secondly, we outline areas of development and the potentials for the future use of remote sensing for authorities in nature conservation such as for the upcoming activities under the Nature Restoration Law.

We also refer to the fact that remote sensing products are increasingly being used as a basis for ecosystem modeling and nature conservation planning. Therefore, we also aim to consider the future role of remote sensing products as continuous and spatially explicit input data for digital twins.

In addition to technical maturity, organizational and structural prerequisites also play a major role in whether remote sensing can be used successfully for official nature conservation purposes. We hence show which prerequisites should be in place so that remote sensing can support the work of nature conservation authorities in the future.

Overall, satellite remote sensing has great potential to increase efficiency and transparency in official nature conservation by promoting data-based decisions and strengthening accountability to the public.

PEOPLE-ECCO (Enhancing Ecosystems Conservation through Earth Observation Solutions, Capacity Development and Co-design) is a project funded by ESA under the Earth Observation Science for Society (EO4Society) programme. The project answers to critical needs identified by Civil Society Organizations (CSOs) and Non-Governmental Organizations (NGOs) striving to improve evidence-based ecosystem conservation. The project aims to develop and demonstrate innovative Earth Observation (EO)-integrated methods and tools to 1) monitor protected areas conditions and management effectiveness, and 2) identify high-priority areas to be protected. PEOPLE-ECCO follows a co-design and user-centred approach. This means we develop the tools together with conservation CSOs/NGOs and provide tailored capacity development enabling them to integrate these EO methodologies in their operational practices.

PEOPLE-ECCO commenced in October 2024 and will run for two years. In this presentation we will outline our overall approach which consists of two interacting parts: a user-focused part dedicated to user engagement, requirement consolidation and capacity development, and a technology-focused part focussing on EO-integrated methods and tools testing, development and demonstration. A central role is reserved for six NGOs/CSOs active in conservation actions with an interest in taking up EO solutions. These “Early Adopters” will jointly contribute to the development of actionable and relevant EO-integrated methods and tools. The Early Adopters in PEOPLE-ECCO (African Parks, Bulgarian Society for the Protection of Birds, Lebanon Reforestation Initiative, IUCN Vietnam, Prince Edwards Island Watershed Alliance and Reef Check Malaysia) are distributed over four continents, and the ecosystems they jointly manage cover a range of terrestrial and aquatic ecosystems.

Outputs of PEOPLE-ECCO aim to contribute to the EU Biodiversity Strategy for 2030 and the Kunming-Montreal Global Biodiversity Framework (GBF), especially to GBF Target 3 (Conserve 30 percent of land, water and seas) and Target 20 (Strengthen Capacity-Building, Technology Transfer, and Scientific and Technical Cooperation for Biodiversity).

The Horizon Europe project Nature FIRST, Forensic Intelligence and Remote Sensing Technologies for Nature Conservation, is generating different tools to support biodiversity monitoring and human-wildlife conflict (HWC) prevention.

Using Satellite Remote Sensing (SRS) technologies with a collaborative approach, Nature FIRST demonstrated the generation of a habitat mapping model in a given territory, which integrates the knowledge of their key actors. This results in semi-automatic, efficient, affordable and easy to update habitat distribution maps (EUNIS, Habitats of community interest), along with an automatic change detection system through available Copernicus data. The Habitat mapping model approach is intended to be applicable to protected area management, to monitor the conservation status of habitats and their dynamics. It also supports the establishment of conservation objectives, along with action and monitoring plans by the entities responsible for biodiversity in the territory.

In this context, the integration of data and information is key. The Nature FIRST system, based on the Sensing Clues platform, makes use of semantic knowledge graphs, which links species, habitat and Natura 2000 site data, together with SRS data. This framework supports additional applications for biodiversity management, such as predictive species movement, habitat suitability maps, and digital twins for monitoring and predicting HWC.

We showcase the practical outcomes of Nature FIRST, i) the creation of habitat mapping models on the territories of Bulgaria, Romania, Spain and Ukraine; ii) An associated habitat change detection system; iii) how the organisation of SRS and in situ data has allowed us to generate a predictive model of brown bear movements, their habitat suitability maps and a digital twin to monitor and predict conflicts, the Human-Bear Conflict Radar.

Maintaining functional ecosystems under anthropogenic pressures requires understanding cumulative impacts on habitat suitability and connectivity to support species conservation. We propose an integrative framework for identifying and preserving functionally connected habitats, utilizing computational tools that enhance conservation planning. This approach begins by modeling effective connectivity through three main steps: (1) estimating habitat permeability, (2) quantifying ecological distances, and (3) calculating effective connectivity for each species.

The approach then scales effective connectivity to the landscape level through the concept of “functional habitat,” linking niche suitability in environmental space with connectivity in geographic space to assess cumulative impacts across landscapes for conservation planning. The framework combines geographic information science, ecological niche modeling, and network science to model species movement across complex landscapes. Applied through scenario analysis to hydropower development impacts in Norway, this framework revealed extensive habitat loss due to fragmentation. The development of the ConScape library enable rapid, high-resolution assessment of connectivity and habitat functionality, facilitating data-driven conservation.

Finally, a sensitivity analysis developed within this framework identifies priority areas for conservation by examining the effects of local landscape changes. In Southern Norway, this analysis suggested that strategically placed wildlife overpasses could achieve a fourfold increase in connected habitat. Together, these methodologies support sustainable landscape management through scenario analysis, spatial prioritization, and mitigation strategies.

The German Federal Ministry for the Environment, Nature Conservation, Nuclear Safety and Consumer Protection has launched the Natural Climate Protection Action Program and commissioned the German Space Agency at DLR to implement measure 8.9 "Tapping the potential of remote sensing for natural climate protection". Started on 01.01.2025, the aim of the EO4ANK-project is to set up the EO4ANK-portal, including a modular EO toolbox, together with partners from science and industry and in close consultation with representatives of the German authorities, who will be the main users. The EO4ANK-Portal will support German authorities at federal, state and local level in implementing the measures from the action program and provide tools for an efficient environmental and nature conservation monitoring. Therefore, a total of 18 tools from the areas of peatlands, floodplains, forests, wilderness, soils and urban areas will be developed and made operationally available on the portal (e.g. heat islands in cities, determination of greenhouse gas emissions from peatlands and their reduction through rewetting, overflow areas etc.). The first tools should be operational by the end of 2026. It is important to provide the portal without any follow-up costs, which is why the toolbox is largely based on Copernicus data. In addition to the development and implementation of the necessary technical solutions, user training also plays a key role. The tools developed must be integrated into the operational working environment of the authorities and users must be trained accordingly, which is why numerous learning materials are produced and made available on the portal.

Habitat destruction, fragmentation, and degradation via human-induced land-cover and land-use change are the predominant drivers of biodiversity loss and are the most significant threats to chimpanzee survival. Conservation practitioners and decision-makers must understand and monitor the relative condition of chimpanzees and other forest and woodland habitats, the threats they face, and how this changes over time to plan and implement cost-effective conservation strategies and measure success. Recent developments in remote sensing and cloud computing such as NASA’s OPERA Land Surface Disturbance Alerts provide near-real-time access to vegetation cover loss intelligence from Harmonized Landsat and Sentinel-2 (HLS) scenes. It provides updates on vegetation cover, and disturbance, and estimates confidence every 2-4 days at 30-meter resolution across the globe. Decision-makers could potentially move from simply documenting the forests already lost toward faster action to stop illegal activities on the ground, slowing and preventing deforestation before it happens. However, to realize this potential, local decision-makers need easy-to-use, cost-effective, and practical solutions to connect and access relevant information and tools. There is an urgent need to find innovative ways to convert these near-real-time EO data into actionable information, meaningful and useful to support specific decision-making processes and build local capacities to access and use these products to drive action and impact. In this presentation, we will discuss the feasibility of OPERA data combined with Planetary Variables from Planet to support local communities and governments to monitor and manage chimpanzee habitats in private, village, district, and national protected areas in Tanzania and Uganda. We will then share ongoing efforts to integrate OPERA alerts into an existing Decision Support System to monitor habitats and threats and inform conservation strategies, and actions and measure success as part of national chimpanzee action plans in Tanzania and Uganda.

The EU Habitats Directive mandates the protection and monitoring of wetland habitats within Natura 2000 sites. However, comprehensive and timely assessment of wetland conservation status remains challenging. The reporting under article 17 of the Habitats directive is missing the detailed, spatially explicit information required for accurate assessment of wetland habitats conservation status, and in particular indicators of degradation.

This initiative, developed in collaboration with the European Commission's DG Environment (DG ENV) and the European Environment Agency (EEA), aims to design an operational geospatial information system to monitor critical wetlands, detect degradation, and assess conservation status within Natura 2000 sites.

Leveraging the Knowledge Centre on Earth Observation's (KCEO) policy-focused value chain and Deep Dive assessment methodology, we translate specific policy needs into technical requirements for Earth Observation (EO) products. We analyze the fitness-for-purpose of existing products and services, evaluating gaps, and provide recommendations to support the EU's commitment to biodiversity protection.

Our approach extends beyond assessment to prototype a Policy-driven Service for monitoring wetlands on selected areas. Ongoing and planned key activities include:

• Characterizing various European wetland habitats, their ecological functioning, and main pressures leading to degradation.

• Determining appropriate indicators for selected habitats and the relevant EO products, prioritizing wetland types based on current degradation levels (per Article 17 of the Habitats Directive), relevance beyond the Directive, and biodiversity value.

• Designing advanced spatial and temporal analysis tools for policymakers and conservation managers integrating cutting-edge EO technologies with ground-truth data and modelling.

This project will enhance our understanding of wetland dynamics and support more effective implementation of EU environmental policies, including the Biodiversity Strategy 2030 and the Nature Restoration Law. The insights and methodologies developed through this project will serve as the foundation for implementing a comprehensive web-based platform for monitoring all wetlands across the EU.

Wetlands are critical biodiversity hotspots, supporting 40% of the world’s plants and animals (https://doi.org/10.1017/S1464793105006950), and are important for storing water, reducing the impacts of droughts and flood events, recharging groundwater, improving water quality, and improving human well-being. Wetland ecosystems vary considerably across the globe, including vast boreal peatland complexes at higher latitudes, and seasonal prairie potholes in lower latitude grasslands. Detailed, reliable, up-to-date inventories of these wetlands is key to accurately monitoring and understanding changes due to natural or anthropogenic factors.

The Alberta Biodiversity Monitoring Institute (ABMI) brings together open-source Earth observation datasets from Sentinel-1/2, machine learning, and Google’s Earth Engine platform to support this crucial knowledge need. In 2021, we published a novel province-wide, temporally consistent, publicly accessible wetland inventory of Alberta bogs, fens, marshes, swamps, and open water. The dataset contains >3 million wetland polygons, producing overall accuracies of 80% or more. When combined with ABMI’s human footprint data, it reveals the dominant influences of agriculture, forestry, urban and industrial development on Alberta wetlands.

Recent, advanced mapping efforts in collaboration with Ducks Unlimited Canada and the Government of Alberta combined newer machine learning approaches, additional field and Earth observation datasets, and recent lidar acquisitions in two contrasting boreal areas, a parkland and a prairie pilot area. The new wetland inventories met provincial wetland mapping standards at the upland-wetland level, the class level (i.e. bog, fen, marsh, swamp and open water) and form level (i.e. open, shrubby, treed). These approaches are already used elsewhere to support groundwater dependent ecosystem mapping in Alberta’s northern oil sands region.

Complementary work at the ABMI is capturing lentic surface water dynamics. The goal is to deliver regularly updated hydroperiod information for long-term monitoring that reflects the state of Alberta’s wetland and freshwater shoreline habitats, which are sensitive to climate changes and human pressures.

Wetlands are the most threatened realm in South Africa, similar to the findings of the global assessment of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (IPBES) in 2019. Wetlands in the predominantly temperate and arid climatic regions that dominate the South African landscape, are small, narrow and mostly palustrine (vegetated). Continuous work is underway to improve the representativity of wetlands, while monitoring of their integrity remains challenging. The availability of Sentinel-1 and -2 images have revolutionised the capability of mapping wetland biodiversity in South Africa, and tracking changes in their extent over time. Case studies will be presented with examples of both the lacustrine and palustrine wetland biomes, including: (a) biodiversity mapping and phenological variation in lacustrine wetlands; (b) tracking changes in the extent of estuarine and freshwater ecosystem functional groups; (c) the importance of the Africa land cover for assessing river ecosystem types and their ecological condition; and (d) monitoring of essential biodiversity variables such as above-ground biomass (i.a., for teal carbon), soil moisture as well as the hydrological regime and phenology metrics. These outputs have contributed to the capabilities of refined reporting to the Sustainable Development Goal 6.6.1a; the reporting of changes in ecosystems to target 1, 2 and 3 of the Kunming-Montreal Global Biodiversity Framework reporting in 2030, and also Red Listing of Ecosystems.

Eco-Patterns is an Innovate UK project that aims to develop a standardised methodology to assess peatland condition remotely. Peatlands contain more carbon than all other UK vegetation combined, however, 80% of peatlands are degraded. Degraded peatlands actively release carbon and impact water quality and flood control. Eco-Patterns is led by Gentian Ltd in collaboration with the University of East London, BSG Ecology, and the IUCN UK Peatland Programme. Our approach combines high-resolution multispectral imagery (<1 m) with advanced deep learning models to identify and classify the spatial and spectral patterns that characterise peatland health. Rather than focusing solely on species, Eco-Patterns analyses habitat “fingerprints”—texture patterns created by species assemblages and structural features unique to these ecosystems. This method provides a comprehensive way to remotely assess peatland condition, offering the potential to underpin emerging market standards like the Peatland Code. Project validation partners include the West Midlands Combined Authority, the National Trust, Natural Resources Wales, NatureScot, and Northern Ireland Water, who are providing ground data and testing sites across the UK.

The European Space Agency (ESA) activity “Biodiversity+ Precursors” is a contribution to the joint EC-ESA Earth System Science Initiative to advance ESS and its response to the global challenges. The Precursor BIOMONDO was focused on biodiversity in freshwater ecosystems. Based on analysis of relevant sources for scientific and policy priorities, the main knowledge gaps and challenges in biodiversity monitoring were compared to possibilities from Earth Observation (EO). These findings were the basis for the development of innovative integrated earth science solutions (Pilots) that integrates EO based products, biodiversity modelling (GLOBIO and Delft3 model suites) and in situ data using advanced data science and information technology. The three pilots were focused on eutrophication, heat waves and river fragmentation, and its effect on biodiversity. The generated products were also implemented in a BIOMONDO Biodiversity data cube. In addition, time series of the cube’d data were analysed using Machine Learning (ML) technique and integrated Thematic Ecosystem Change Indices (TECI), e.g., water quality and lake water temperature evolution, were deduced and analysed. Validation of the integrated products was a key task within BIOMONDO, and several biodiversity and policy experts have been consulted. They were also provided access to the novel EO products in the cube via API or the implemented data viewer, a tool for visualisation and easy access to products and data.

Freshwater biodiversity faces challenges worldwide. One approach to describing its status is through the study of freshwater phenology, which is listed as an Environmental Biodiversity Variable (EBV). As phytoplankton is a central component of a lentic ecosystem, monitoring the phenology of it can be highly relevant in relation to freshwater biodiversity. While terrestrial phenology based on EO is advanced significantly, the study of phenology in lake ecosystems is in its early stages.

Phenological shifts of phytoplankton can be derived from chlorophyll-a concentrations, which are effectively measured using Earth observation techniques. Our method utilizes time series analysis to detect seasonal variations in phytoplankton blooms, identifying key characteristics such as the timing of bloom peaks, the duration of blooms, and their spatial distribution. Working with Copernicus satellite products, the data enables the observation of phytoplankton phenology across the whole waterbody, making it possible to detect the spatial distribution of individual bloom events and providing insights into these events at both spatial and temporal scales.

This method is currently being developed as part of the OBSGESSION EU Horizon Europe project and tested for the study sites in Sweden and Finland. We are aiming to scale our method to make it applicable to different lake types.

In the context of climate change and increasing water scarcity, lakes serve as water reservoirs and are supporting services, such as agricultural irrigation or maintaining discharge during low-flow periods. Their presence in a catchment impacts downstream ecosystem and biodiversity by altering water, sediment, nutrients and pollutants cycles. Moreover, increasing temperatures, declining water levels and nutrients fluxes are the principal drivers of eutrophication, threatening water quality and biodiversity in both lakes and downstream ecosystems. Monitoring these water bodies is essential for assessing eutrophication risk and informing management solutions, yet less than 1% of lakes in France are monitored by public authorities and most of the time with few data at the temporal scale.

In this study, we focused on thousands of reservoirs within the Adour-Garonne basin (South-Western France). We developed a methodology that combined machine-learning models to predict (1) nitrates and phosphorus inputs into lakes, and (2) lake chlorophyll-A dynamics and trends, from various environmental drivers such as meteorological data, land-use, land management data or lake characteristics. The training data for these basin-wide models were derived from field observations (nutrients) and Sentinel-2 images (Chlorophyll-A and turbidity). The Sentinel-2 images were analyzed for all reservoirs in the Adour-Garonne basin with surface areas exceeding 10,000 m² from 2018 to 2023, as part of the SCO XTREMQUALITY project. The first results indicate promising model performances, with good accuracy for chlorophyll-A prediction in lakes. Results help characterize eutrophication status and trends in thousands of various sized lakes and untangle relationships between eutrophication and driving factors, mainly land use and lake characteristics. Limitations and potential improvements in satellite image processing will also be discussed. These insights allow for the identification of priority lakes for enhanced monitoring or tailored management strategies, aiming to mitigate eutrophication impacts and preserve biodiversity in vulnerable aquatic ecosystems.

The hydrological cycle is critical for Earth system stability, involving intricate coupled processes and feedbacks tied closely to terrestrial ecosystems. Changes in key hydrological functions can have significant impact on both ecological and social systems, affecting biodiversity, crop yields, and ecosystem structure and function. Through the spatial connectivity of the water cycle, the effect of these changes may be felt from the local to the continental scale. Anthropogenic pressures, such as deforestation and land-use change, have led to a reduced capacity of ecosystems to recover from external perturbations, or resilience loss, in regions that are closely coupled to the water cycle, but the reciprocal impact of changes to terrestrial ecosystems on the resilience of hydrological functions remains an open question.

Here, we use remotely sensed data on soil moisture (SMOS), evapotranspiration (GLEAM), and precipitation (SSMI/S), and employ an early warning signal-based detection of the resilience of these key hydrological variables at the global scale. In doing so, we aim to present a first assessment of global-scale water resilience, and a characterisation of regions vulnerable to abrupt changes, or close to sensitive thresholds related to the stability of the hydro-climatic cycle. We compare our findings to assessments of resilience loss in terrestrial ecosystem variables, and assess the key driving variables to contribute to a holistic understanding of resilience in the terrestrial freshwater cycle.

The ambitions of the EU Green New Deal (e.g. ‘nature as a solution’, ‘building a bioeconomy’) as well as recent legislation (e.g. the Nature Restoration Regulation, the ecosystem accounting module under Regulation 691/2011) require much better data on biodiversity and ecosystems than currently available (in terms of spatial and thematic accuracy).

The ambitions of the EU Green New Deal (e.g. ‘nature as a solution’, ‘building a bioeconomy’) as well as recent legislation (e.g. the Nature Restoration Regulation, the ecosystem accounting module under Regulation 691/2011) require much better data on biodiversity and ecosystems than currently available (in terms of spatial and thematic accuracy).

The EU Copernicus program provides important data sets for monitoring the environment. Work on behalf of the European Environment Agency, the European Space Agency, in various (EU) research projects etc. has explored options for using satellite data in support of ecosystem and nature monitoring. However, converting research outcomes into operational Copernicus products for ecosystem monitoring is challenging and resource intensive.

This workshop reviews the key success factors for a successful operational implementation of ecosystem monitoring with satellite data. It has a particular focus on the components that need to be paired with modern satellite technology: habitat-level in situ data as well as stable operational infrastructure and expert capacity for developing and maintaining regular monitoring products.

The workshop will review current experience with developing ecosystem extent data sets in the European Union, present an overview of available and needed in situ data and engage participants in a discussion on how to overcome current bottlenecks and constraints in developing successful ecosystem monitoring products in an EU context.

Expected outcomes: The workshop outcomes include a better understanding of possibilities and limitations for using satellite data sets for ecosystem monitoring and a set of proposals for developing ecosystem monitoring products in an EU context.

Objectives of the workshop:

- Raise attention for existing EU investment gap in making satellite approaches effective

- Highlight the critical data gap on biodiversity in situ data

- Discuss EU policy priorities for biodiversity and ecosystem monitoring

- Review need for increasing institutional capacity for regular application ready data sets

talk

talk

talk

talk

talk

talk

Video recording

Video Recording

talk

As international conservation efforts aim to address biodiversity loss driven by climate change and anthropogenic disturbance, comprehensive information on species distributions which when assessed over a specific temporal scope represent that species distribution essential biodiversity variable (SD EBV) – has become a potential key resource for policymakers and stakeholders. Currently, much of this data is publicly accessible through expert range maps (e.g., IUCN Red List) and species occurrence records (e.g., GBIF). However, these kinds of data are fundamentally incomplete and subject to a variety of biases, thus completing such an information base that addresses all species of a taxon at spatial resolutions relevant to actionable conservation plans demands a highly scalable approach to species distribution modeling (SDM). One central pillar for temporally specific, spatially contiguous and global predictions is the integration of a range of remote sensing and remote sensing-supported climate products, collected at different spatial scales. In support of a range of information products and downstream indicator users, Map of Life has been advancing 1km2 global SDMs for a diverse array of taxa from vertebrates to invertebrates, common to rare species, and generalists to specialists, all with their associated challenges. In this presentation, we will share insights from our progress in producing, curating, and validating global SD EBVs, and also discuss the integration of novel data streams, such as trait-based data to apply target group background sampling, to enhance predictions. We will also address critical remaining challenges, including data gaps, computational limitations, and the complexities of generating SDMs for multiple taxa, and outline key next steps toward establishing a comprehensive, robust set of global SD EBVs that support global biodiversity measurement and conservation.

In the context of ever-increasing human impacts and accelerating climate warming, we need to better understand and predict species occurrences and abundances over space and time. In recent years, new types of Species Distribution Models (SDMs) based on deep learning (deep-SDMs) have been successfully applied for occurrence prediction from satellite remote sensing data. It has been shown that deep-SDMs outperform conventional SDMs for occurrence prediction and their architecture is very promising for solving abundance prediction challenges. However, deep-SDMs require millions of observations to be trained and cannot therefore be trained to predict abundance. Indeed, due to acquisition difficulties, abundance datasets are considerably smaller than presence-only datasets.

Here we overcome this limitation by using the transfer learning from occurrence deep-SDM to abundance deep-SDM with the underlying hypothesis that the neural network layers of a previously trained model with presence-only data can capture general information and patterns that can be reused for abundance predictions. As an example, we used coastal fish in the Mediterranean Sea. We assessed the extent to which deep-SDM trained on only 406 fish counts can predict the abundance of fish species by taking advantage of transfer learning from a deep-SDM trained on 62,000 presence-only.

We show that this approach significantly improves the abundance prediction performance of deep-SDM, with average gains of 35% (based on the D2 Absolute Log Error score). This allows deep-SDMs to be more efficient than conventional SDMs, with an average gain of 20%. These gains are mainly linked to a better prediction of the abundance of rare species. This ability of deep-SDM to extract relevant information predicting species abundance from presence-only data is a new and unexpected result. This finding paves the way towards a more general use of deep learning to predict species abundance and biodiversity patterns, especially for rare species.

While policymakers are committed to a 30% global protection target by 2030, including the ocean, our knowledge of marine species distributions remains limited compared to terrestrial species. This gap is a major barrier to science-based decision-making in the field of marine conservation.

The high cost of data acquisition is partly responsible for the situation. But a significant lever would be to use the sparse data we have more efficiently. Indeed, current species distribution models (SDMs), though robust, are simplistic in terms of environment-species interactions: they often rely only on long-term climate averages. This seriously hinders the discovery of dynamic processes, such as seasonal migrations or adaptation to environmental change, and therefore limits our knowledge of how marine species use space.

Recently, computational ecologists have successfully designed and tested new types of SDMs based on deep learning to model plant species distributions in the terrestrial realm. They treat the environmental landscape as a multi-layered image and use convolutional neural networks to extract relevant geographical features and predict associated species, with promising results on data-poor species using knowledge transfer.

Our work adapts this approach to the open ocean, in particular by taking into account the high variability of environmental conditions. In an initial trial, we predicted relative presence probabilities for 38 marine taxa at a global scale using 18 satellite-derived environmental variables, achieving 89% Top-3 accuracy. This work provided valuable insights on data curation, variable importance and hyperparameter fine-tuning.

This approach provides ways to enhance our knowledge of data-poor marine regions or species, and to deepen our understanding of the dynamic impact of environmental conditions. It paves the way for a diversity of use cases ranging from other marine environments to predictions of future species distributions. This allows more comprehensive biodiversity mapping, which is a significant step towards well-designed ecosystem protection measures.

Species distributions are one of the fundamental units in biogeography and conservation and a key Essential Biodiversity Variable. Species distribution models (SDMs) are popular tools that characterize a species distribution by statistically relating occurrence data with environmental variables, remote sensing products, and other habitat variables. SDMs have become increasingly sophisticated, but they are unsuitable for data-deficient species; 30% of known species have insufficient data to characterize geographic distributions appropriately. Since many analyses rely on robust SDM outputs, data-deficient species are often left out, biasing scientific and conservation efforts.

Recently, ecology has seen unprecedented growth in the amount and variety of data collected, including occurrence data, phylogenetic data, and fine-resolution remote sensing products. We integrate these data sources and present a novel modeling framework that extends SDMs to allow data-deficient species to borrow strength from data-rich species. Specifically, we demonstrate how evolutionary history, WorldClim, EarthENV, and MODIS products can inform the distributions of data-deficient species. We apply our modeling framework to the tropical clade of South American hummingbirds, where SDMs are often hampered by a need for more data, even in well-studied taxons such as birds.

The results of our analysis include up to 40% improvement in the Area Under the Curve for over 75% of the species. Species that showed little to no improvement lacked a recently diverged sister species, indicating that this method works best when species’ pairs are recently diverged. We quantify the improvement of our model and produce novel richness maps. We suggest these maps are our best current understanding of South American species’ distributions. This work represents a concrete way forward for SDMs to integrate phylogenetic information and remote sensing products. By improving data-deficient distribution estimates, we will develop more robust species distribution-related EBVs and better understand how our biodiversity is distributed across geographic space.

Preserving species' genetic diversity is crucial for maintaining ecosystem functions and services in the face of global change. However, to preserve it effectively, we first need to monitor genetic diversity efficiently. While DNA sequencing remains the gold standard for assessing genetic diversity, it is often too expensive and time-consuming for routine and large-scale monitoring.

The Genes from Space project offers a complementary approach: using Earth Observations (EO) for large-scale monitoring of species' genetic diversity. Changes in species' genetic diversity often reflect local population declines driven by habitat changes that can be readily detected via EO—such as deforestation, ecosystems response to shifting climate and other biotic and abiotic conditions. Therefore, tracking habitat changes with EO allows for large-scale and routine monitoring of genetic diversity using EO-based indicators.

In collaboration with the BON in a Box platform, we have developed a tool that enables monitoring of genetic diversity across a range of ecosystems using EO data. The tool is versatile, allowing users to integrate their own species distribution or habitat change data, or to retrieve such data from publicly available databases (e.g., GBIF, Global Forest Watch, or global land cover maps). Based on these inputs, the tool produces genetic diversity indicators consistent with the global biodiversity conventions.

This free and open-source tool is designed to accommodate a range of users. For example, biodiversity conservation practitioners can access the tool via a user-friendly website interface to generate indices for specific ecosystems. For advanced users with programming expertise, the tool can be run locally, and they are encouraged to contribute to its development by adding new workflows.

The Genes from Space monitoring tool provides a scalable, accessible solution for monitoring genetic diversity across large spatial scales, serving as an early-warning system to direct and optimize in situ and DNA-based assessments where needed.

We present a framework to monitor biodiversity by calculating ecological niche models over time with a temporal series of remote sensing products. We have implemented this methodology in the Natural Park of Montesinho (Northeast Portugal) through the MontObEO research project. The framework estimates species vulnerability by analysing trends (Mann-Kendall test) over time (2001-2023) of the habitat suitability index from a set of ecological niche models (Maxent) calculated with a time series of remote sensing variables (MODIS sensor). Positive trends are associated with increases in habitat suitability; negative trends with decreases in habitat suitability. All procedures (e.g. gathering the satellite data, calculating the MaxEnt models, and analysing the habitat suitability trends) are performed in Google Earth Engine (GEE). We considered five taxonomic groups: vascular flora, amphibians, reptiles, birds, and mammals. We analysed habitat suitability trends for each species, taxonomic group, functional group, and potential species richness over time. We created an R package and a GEE App, where users can use our framework easily and efficiently. We built a spectral library for some vascular flora key species in Montesinho. Our framework is an effective monitoring methodology as it can be adapted to any study area at different spatial and temporal resolutions. This work is funded by Centro de Investigação em Ciências Geo-Espaciais, reference UIDB/00190/2020, funded by COMPETE 2020 and FCT, Portugal.

Regular counts of walruses (Odobenus rosmarus) across their pan-Arctic range are necessary to determine accurate population trends, and in turn understand how current rapid changes in their habitat, such as sea ice loss, are impacting them. However, surveying a region as vast and remote as the Arctic with vessels or aircraft is a formidable logistical challenge, limiting the frequency and spatial coverage of field surveys. An alternative methodology involving very high-resolution (VHR) satellite imagery has proven to be a useful tool to detect walruses, but the feasibility of accurately counting individuals has not been addressed. Here, we compare walrus counts obtained from a VHR WorldView-3 satellite image, with a simultaneous ground count obtained using a remotely piloted aircraft system (RPAS). We estimated the accuracy of the walrus counts depending on 1) the spatial resolution of the VHR satellite imagery, providing the same WorldView-3 image to assessors at three different spatial resolutions (i.e., 50, 30, and 15 cm per pixel) and 2) the level of expertise of the assessors (experts vs a mixed level of experience – representative of citizen scientists). This latter aspect of the study is important to the efficiency and outcomes of the global assessment programme because there are citizen science campaigns inviting the public to count walruses in VHR satellite imagery. There were 73 walruses in our RPAS “control” image. Our results show that walruses were under-counted in VHR satellite imagery at all spatial resolutions, and across all levels of assessor expertise. Counts from the VHR satellite imagery with 30 cm spatial resolution were the most accurate, and least variable across levels of expertise. This was a successful first attempt at validating VHR counts with near-simultaneous, in situ, data. But further assessments are required for walrus aggregations with different densities and configurations, on different substrates.

Monitoring vulnerable wandering albatross (Diomedea exulans) populations presents significant challenges due to their remote nesting locations, making traditional ground or aerial surveys costly, infrequent, and often incomplete. However, with advancements in geospatial remote-sensing technologies, citizen science is emerging as a valuable tool for generating accurate, georeferenced wildlife data. In this study, we conducted the first citizen science campaign aimed at counting wandering albatrosses in South Georgia, utilising 31 cm resolution satellite imagery. The campaign spanned 24 breeding areas with imagery captured between 2015 and 2022. Volunteers were tasked with identifying presumed albatrosses in 150 m x 150 m image chips (with 5 m overlap), each reviewed by a minimum of seven unique users. Over the course of the campaign, 639 citizen scientists classified a total of 11,839 image chips, covering 154 km². Our results show a strong positive correlation (r = 0.98, df = 16, P < 0.001) between adjusted ground counts and satellite-based estimates, with deviations ranging from 4.5% to 30.9% for colonies containing more than 100 breeding pairs. This study demonstrates the accuracy and effectiveness of crowdsourcing as a reliable method for long-term monitoring of wandering albatross populations and highlights the potential to expand this approach to other seabird species and breeding sites, offering a scalable solution for wildlife monitoring in remote regions.

Geomagnetic navigation as an animal migratory strategy has been studied across several taxa, but how animals, mainly long-distance migrants (i.e. birds), use the geomagnetic field on their journeys is still relatively unknown. The Earth’s magnetic field varies across both space and time. The variability across temporal scales, which are relevant for animal navigation (at seconds to days), mostly comes from solar activity and may affect the potential animal choice of direction during navigation. To date, ecologists have not been able to study how these short-term geomagnetic field variations affect navigation because of the lack of reliable geomagnetic data (Deutschlander 2014). This, however, is important since it has been demonstrated that animals can sense minor geomagnetic field differences (Beason 1987; Semm and Beason 1990). This talk will introduce a tool called MagGeo that will help ecologists obtain detailed geomagnetic data at the location and time of the passing animal. I will describe the technical process of implementing a spatio-temporal data fusion method (Benitez-Paez 2021) to link wildlife tracking data with geomagnetic data provided by Swarm Constellation (from the European Space Agency - ESA). Our tool, MagGeo, is available as free and open-source software (FOSS) and uses a set of Jupyter notebooks to let users interact with the process.

Ecosystem resilience represents the capacity to withstand and recover from perturbations. Resilience is a fundamental functional property of forests, especially in view of increasing anthropogenic and climate pressures. The focus of recent large-scale resilience studies has been on the relationship between resilience and climate, with little exploration on the relationship between forest resilience and its diversity, upon which management practices can have an impact. In this study, the sensitivity of European forest resilience to structural diversity is quantified using remotely-sensed data. Two established resilience indicators are extracted from MODIS derived kNDVI time series, and forest structural diversity is accounted for by horizontal, vertical and combined horizontal and vertical metrics derived from NASA’s GEDI instrument. A Random Forest model is leveraged to isolate the interplay between resilience and structural diversity and to disentangle confounding environmental variables such as background climatic conditions. The study finds that European forests with a higher level of structural diversity are systematically associated with higher resilience levels. Importantly, diversity in canopy complexity is more important for resilient forests than variability in canopy height, and this relationship is consistent under increasing temperature patterns. This suggests that forest management promoting forest heterogeneity and especially canopy complexity has the potential to offset the decline in forest resilience associated with climate warming.

Monitoring Biodiversity Change is essential for planning and tracking the effectiveness of conservation initiatives aligned with international agreements.

Satellite remote sensing enables monitoring Biodiversity at scales relevant to conservation by providing continuous and repeatable ecosystem observations. Current frameworks primarily focus on mapping binary metrics, like changes in forest extent, which alone misses the impact of global (climate, air pollution) and small-scale degradation on the integrity and resilience of forests’ biodiversity.

This work introduces a monitoring framework to directly track Biodiversity Integrity Change, where Biodiversity Integrity is defined by the degree to which a forest composition, structure and function fall within a dynamic range of reference states that account for seasonal phenology and multi-annual resilience to past stressors such as drought.

Our approach uses Artificial Intelligence (AI) models that integrate multi-sensor satellite time series, including Imaging Spectroscopy, RADAR and LiDAR, capturing changes in composition, structure and function. These AI models analyze spatiotemporal patterns to understand seasonal and multi-annual variability at multiple spatial and temporal scales, and it pinpoints to forests that are deviating from expected phenological, structural or functional reference states.

Our AI-driven analyses enhance existing forest extent monitoring systems, by directly observing Biodiversity Change (structure, composition and function) within and between ecosystems, which is essential to plan and monitor progress towards achieving area-based conservation targets, as well as to understand the main threads to Biodiversity Integrity (e.g. land use conversion, climate change and extremes, air pollution).

We present preliminary findings from tracking changes in forests across the US Pacific region (California, Oregon, and Washington) from 2015 to 2024. We evaluate the effectiveness of various AI models, including novel models developed by our team that integrate Landsat and Sentinel-1 data, as well as Foundation Models (NASA’s Prithvi and PRESTO) which leverage multi-sensor satellite time-series to analyze spatiotemporal patterns.

Tropical rainforests face significant challenges, with nearly 40% of the remaining areas considered disturbed or degraded. This degradation has profound implications for both climate change mitigation and biodiversity conservation efforts. The carbon emissions resulting from tropical forest degradation are substantial, sometimes even surpassing those from deforestation in certain regions, though estimates vary widely. Beyond carbon concerns, degradation-induced changes in forest structure can significantly impact ecosystem function and integrity. These alterations can lead to reduced biodiversity and diminished ecological services provided by healthy rainforests. Recent advancements in satellite remote sensing technology have made it possible to detect even minor disturbance events in tropical forests. However, the full impact of these disturbances remains poorly understood, highlighting a critical gap in our knowledge of forest ecosystem dynamics. The Kunming-Montreal biodiversity framework, while ambitious, is hampered by a lack of high-quality indicators for ecosystem integrity. This deficiency makes it challenging to effectively monitor and assess the impacts of human-induced degradation on forest ecosystems. To address this issue, a novel approach has been developed to evaluate the impacts of disturbance events on tropical forests. This method compares near-coincident GEDI (Global Ecosystem Dynamics Investigation) shots that happen to sample forest carbon and structure before and after disturbance events detected by other remote sensing systems, providing valuable insights into changes. We show how this technique can be applied across the wet tropics to assess the impacts of various types of disturbances on forest ecosystems. By quantifying these effects, we aim to better understand the consequences of degradation and inform more effective conservation and restoration strategies for tropical rainforests

Conservation of forest ecosystems is essential for maintaining biodiversity and ecosystem services. This study leverages Earth Observation (EO) data to address global biodiversity monitoring challenges in the face of increasing natural and anthropogenic disturbances. Focusing on the Ticino Park, a temperate mixed forest in northern Italy, we investigated the impact of drought—an escalating stressor on Earth system functioning—by analysing Sentinel-2 imagery from 2017 to 2022, particularly during the severe drought of 2022.

To enhance the detection of plant water stress, we conducted direct and continuous monitoring of functional traits that indicate tree health and structural status in relation to drought conditions. Specifically, we derived high-temporal-resolution time series of leaf area index (LAI), canopy chlorophyll content (CCC), and canopy water content (CWC) from Sentinel 2. We also analysed forest environmental characteristics and species composition to assess their influence on physiological responses and corresponding spectral changes observed by EO satellites.

Our results showed strong correlations between Sentinel 2 -derived plant traits and ground measurements, with CCC having the highest correlation with ground data (r² = 0.82, nRMSE = 13.56%) and LAI closely following (r² = 0.75, nRMSE = 11.49%). Daily standardized anomaly (DSA) analysis highlighted significant variations linked to forest types, showing that pine and black cherry experienced the greatest stress, while hygrophilic species such as black alder and chestnut were less affected. The DSA maps provided spatial and temporal patterns related to drought-induced vegetation stress in the Ticino forests in 2022.

Our workflow provides quantitative insights into forest functional states at high spatial resolution (10 m), crucial for effective management and conservation measures. These findings highlight the importance of understanding species-specific responses to drought for improved forest monitoring and management strategies in response to climate-induced challenges.

Foliar functional traits are dynamic plant properties that vary across space and time, serving as principal tools for monitoring plant physiology and terrestrial ecosystem processes. Phenolics are the most crucial secondary metabolites that play key roles in plant defence against biotic and abiotic stressors, leaf decomposition, as well as consequent influence on nutrient cycling and soil microbial composition. However, spatially continuous information on canopy phenolic remains poorly characterized at the landscape level. Current and proposed spaceborne imaging spectrometers offer unique opportunities to map foliar phenolics quantitatively through space and time. Our recent work (Xie et al, 2024) demonstrated that foliar phenolics can be accurately estimated across temperate tree species using leaf spectroscopy. In this study, we leveraged imaging spectroscopy data from PRecursore IperSpettrale della Missione Applicativa (PRISMA) mission to predict and map foliar phenolic variations at canopy scale in a mixed European temperate forest. Two data-driven approaches, namely partial least square regression and Gaussian processes regression, were applied to link lab-measured phenolic concentration with PRISMA plot-level spectra (400–2400 nm). The performance statistics indicated reasonable precision and accuracy of the model results. Maps derived from the best-performing model (based on cross-validated nRMSE) provided a wall-to-wall assessment of canopy phenolics, capturing both inter and intra-species variations across the landscape. Further, we compared the phenol map with the distribution of leaf mass per area and canopy nitrogen. The results indicated that the synergy patterns across the three functional traits were consistent with the known leaf economic spectrum. These findings highlight the potential of spaceborne imaging spectroscopy to characterize spatial and temporal dynamics of ecologically important plant phenolics. Our study also paves the way for improved global monitoring of ecosystem integrity and plant responses to environmental stress and climate change, particularly with the anticipated launch of hyperspectral missions like ESA’s CHIME and NASA’s SBG.

Accurate in-situ biodiversity estimates are crucial for effective conservation strategies and rely on the integration of satellite remote sensing (SRS) data with on-the-ground measurements. Recent advancements in SRS technology enable high-resolution, near real-time observations of land use-land cover (LULC) changes. Model-based indicators, such as the Biodiversity Intactness Index (BII) and GLOBIO, are designed to translate such dynamics into estimated changes in biodiversity intactness of ecosystems. However, existing indicators are subject to some important limitations, including lack of evaluation of predictive performance against observed data, reliance on a relatively small fraction of available biodiversity data, and not integrating potentially important SRS data products.

In this project, we particularly address the lack of model performance testing, deep diving into different evaluation strategies for large-scale intactness models. We explore several cross-validation approaches, from standard random sampling to spatial, environmental, and cross-study alternatives. Using these approaches, we estimate the predictive performance of the BII model for relative species abundance. While the in-sample accuracy is high, predictive capabilities do not generalize to unseen, out-of-sample data, which is driven by the structure of the model. To improve generalization, we develop a Bayesian hierarchical model pipeline, with a hierarchical structure based on biogeographical entities. This model also includes a richer set of environmental predictors. While the Bayesian model performs significantly better in standard cross-validation, it struggles considerably when train-test splits are done across spatial, environmental and study dimensions.

These results highlight that the prospect of building good broad-scale predictive models is currently very challenging due to data limitations. This especially concerns the lack of at-scale, representative biodiversity inventories for many parts of the world and many taxonomic groups. We outline some potential paths forward to improve predictive models in this space, including environmental DNA for large-scale sampling and the need for more historical, high-resolution SRS products.

Here, we pioneer the use of multi-sensor Earth Observation (EO) data and insect in situ data collated from various "big data" platforms (iNaturalist, GBIF, and GenBank) to develop a framework for measuring insect-based biodiversity intactness patterns across Africa. The insect taxa used in this framework are sensitive to ecological changes stemming from unsustainable farming practices, urbanization, and logging. Insect diversity patterns have proven to be valuable indicators of overall ecosystem biodiversity intactness. Compared to megafauna, insects occur in all climate zones and occupy diverse micro-habitats, making them excellent predictors of ecosystem intactness at spatially explicit scales, even over larger regions.

The UN Convention on Biological Diversity (CBD) and its technical working group for the post-2020 framework have called for unbiased (i.e., accurate), measurable, and scalable frameworks and indicators for biodiversity. These frameworks should ideally account for localized drivers of biodiversity loss, support the estimation of planetary boundaries, and assess ecosystems' capacity to deliver services (such as pollination by insects). The UN Kunming-Montreal Global Biodiversity Framework likewise emphasizes the need to connect biodiversity loss with ecosystem services and focuses specifically on the integrity of agro-ecological landscapes.

We estimated biodiversity intactness as the ratio between the actual (or currently observed) insect diversity (o) and the historic or potential estimated insect diversity (p). Predictors included spectral features from 10-20m Sentinel-2 satellite data, 1-km WorldClim climate variables, 25-m tree heights from the Global Ecosystem Dynamics Investigation sensor, and 1-km human footprint data. Pixel-based biodiversity intactness predictions could be aggregated at the country level or across conservation priority corridors. Across Africa, high insect-based biodiversity intactness was observed in natural tropical forests, montane "sky islands," wetlands, islands in Lake Victoria, and arid countries such as Namibia. The framework can be adapted to focus on locally threatened or endemic insect species by analyzing individual species within the assemblage. The indicator values remain stable across diverse climate zones, and pixel-level data can be spatially aggregated to support country-level reporting mechanisms.

Landscape change and habitat fragmentation are recognised drivers of biodiversity change, but properly isolating and assessing their impacts can be challenging without appropriate data and statistical techniques. Indeed, the spatial scales at which they occur are often confounded by other stressors such as anthropogenic pressures or climate change. Here, we combine remotely sensed land cover time series with a recent technique from causal inference, synthetic controls [1,2], to test the impact of landscape modification on French breeding bird diversity metrics (Temporal Monitoring of Common Birds, STOC programme, 2001-2019). This method requires time series of treated and untreated units and a date of treatment assignment. This date is detected from annual land cover products [3]. By constructing appropriate controls, variations in STOC metrics can finally be attributed to landscape changes.

In parallel, foundation models trained on remotely sensed imagery (RSFMs) offer unprecedented predictive accuracy and generalisation power for downstream tasks such as biodiversity metric estimation, without requiring tailored training and precise understanding of the ecological processes at play. Therefore, a second objective is to analyse RSFM predictions based on annual Landsat imagery mosaics (RGB, NIR, SWIR bands) centered on altered landscape plots identified by synthetic controls: Do the deep learning models detect and rely on the same structural changes that have disentangled effects on biodiversity metrics, or do they miss these elements? The results of this cross-analysis between causal effect estimation on the one hand, and interpretation of deep learning predictions on the other, has the potential to increase understanding, confidence, or possibly caution in the adoption of foundation models for biodiversity modelling.

[1] Abadie, Alberto. Journal of economic literature 59.2 (2021): 391-425.

[2] Fick, Stephen E., et al. Ecological Applications 31.3 (2021): e02264.

[3] Zhang, Xiao, et al. Earth System Science Data 16.3 (2024): 1353-1381.

Conserving biodiversity is a global priority that urgently requires effective decision-making. Open, Operational Biodiversity Data Products (OOBDPs) that deliver information to decision-makers at the appropriate spatiotemporal scale are critical to informing conservation policy and action across governments, corporations, and local communities. However, efficiently scaling spatiotemporal biodiversity data products beyond the few areas/periods/variables with comprehensive in situ data, while maintaining scientific integrity, is challenging, as uncertainties inherent in these products negatively scale with in situ data coverage and quality.

Biodiversity is intrinsically local and context-specific, making the integration of in situ data and remote sensing essential for reducing these uncertainties. Despite significant advancements, current practices often fail to explicitly address or propagate uncertainty, limiting the utility and trust in these products among decision-makers.

This workshop will explore how to conceptualize, quantify, communicate, and reduce uncertainty in biodiversity data. By discussing intensive field campaigns, data integration, and emerging technologies, participants will work collaboratively to identify best practices, address key gaps, and provide actionable recommendations for improving biodiversity data products, optimizing resource allocation, and enhancing decision-making processes.