Estimating canopy height and its changes at meter resolution from satellite imagery remains a challenging computer vision task with critical environmental applications. However, the lack of open-access datasets at this resolution hinders the reproducibility and evaluation of models. We introduce Open-Canopy, the first open-access, country-scale benchmark for very high-resolution (1.5 m) canopy height estimation, covering over 87,000 km2across France with 1.5 m panchromatic resolution satellite imagery and aerial LiDAR data. Additionally, we present Open-Canopy-∆, a benchmark for canopy height reduction detection between images from different years at tree level—a difficult task for current computer vision models. We evaluate state-of-the-art architectures on these benchmarks, highlighting significant challenges and opportunities for improvement. Our datasets and code are publicly available at https://github.com/fajwel/Open-Canopy.

Significant efforts have been directed towards adapting self-supervised multimodal learning for Earth observation applications. However, most current methods produce coarse patch-sized embeddings, limiting their effectiveness and integration with other modalities like LiDAR. To close this gap, we present DUNIA, an approach to learn pixel-sized embeddings through cross-modal alignment between images and full-waveform LiDAR data. As the model is trained in a contrastive manner, the embeddings can be directly leveraged in the context of a variety of environmental monitoring tasks in a zero-shot setting. In our experiments, we demonstrate the effectiveness of the embeddings for seven such tasks: canopy height mapping, fractional canopy cover, land cover mapping, tree species identification, plant area index, crop type classification, and per-pixel waveform-based vertical structure mapping. The results show that the embeddings, along with zero-shot classifiers, often outperform specialized supervised models, even in low-data regimes. In the fine-tuning setting, we show strong performances near or better than the state-of-the-art on five out of six tasks.

With the rise in global greenhouse gas emissions, accurate large-scale tree canopy height maps are essential for understanding forest structure, estimating above-ground biomass, and monitoring ecological disruptions. To this end, we present a novel approach to generate large-scale, high-resolution canopy height maps over time. Our model accurately predicts canopy height over multiple years given Sentinel 2 time series satellite data. Using GEDI LiDAR data as the ground truth for training the model, we present the first 10 m resolution temporal canopy height map of the European continent for the period 2019–2022. As part of this product, we also offer a detailed canopy height map for 2020, providing more precise estimates than previous studies. Our pipeline and the resulting temporal height map are publicly available, enabling comprehensive large-scale monitoring of forests and, hence, facilitating future research and ecological analyses. For an interactive viewer, see https://europetreemap.projects.earthengine.app/view/europeheight.

We propose a framework for global-scale canopy height estimation based on satellite data. Our model leverages advanced data preprocessing techniques, resorts to a novel loss function designed to counter geolocation inaccuracies inherent in the ground-truth height measurements, and employs data from the Shuttle Radar Topography Mission to effectively filter out erroneous labels in mountainous regions, enhancing the reliability of our predictions in those areas. A comparison between predictions and ground-truth labels yields an MAE/RMSE of 2.43 / 4.73 (meters) overall and 4.45 / 6.72 (meters) for trees taller than five meters, which depicts a substantial improvement compared to existing global-scale products. The resulting height map as well as the underlying framework will facilitate and enhance ecological analyses at a global scale, including, but not limited to, large-scale forest and biomass monitoring.

This research introduces a new method for predicting changes in corn yield and price, critical for food security. Instead of relying on difficult-to-access pre-harvest production data, our approach uses satellite-derived gross primary production (GPP) data and dimension-reduction techniques to forecast national corn yield and price changes. We conducted case studies in the U.S., Malawi, and South Africa to validate this approach, analyzing predictors from annual GPP variations during peak growing seasons. We applied dimension-reduction strategies, such as spatial averaging and empirical orthogonal functions (EOFs). Additionally, we used deep learning techniques such as autoencoders (AEs) and variational autoencoders (VAEs) to extract meaningful features from the high-dimensional GPP datasets. These features were then used as predictors in yield and price classification models based on generalized linear models (GLMs) and ElasticNet. We also considered a neural network model trained to predict yield and price variations from GPP input data directly. We evaluated the model performances using metrics such as area under the curve (AUC), brier skill score (BSS), and Matthew’s correlation coefficient (MCC). Our results indicate that dimension-reduction techniques based on AEs and VAEs provided better predictive capabilities across all three countries compared to EOF, particularly in Malawi, where the BSS increased from 0.26 with EOF to 0.81 with AE for yield and from −0.004 with EOF to 0.77 with VAE for price. This study demonstrates that integrating open-access satellite data with dimension-reduction techniques can significantly improve crop forecast accuracy, providing an accessible tool to enhance agricultural management and food security.

The Qinghai–Tibet Plateau is rich in water resources with numerous lakes, rivers, and glaciers, and, as a source of many rivers in Central Asia, it is known as the Asian Water Tower. Under global climate change, it is critical to understand the current influencing factors on surface water area in this region. Although there are numerous studies on surface water mapping, they are still limited by temporal/spatial resolution and record length. Moreover, the complicated topographic condition makes it challenging to map the surface water accurately. Here, we proposed an automatic two-step annual surface water classification framework using long time-series Landsat images and topographic information based on the Google Earth Engine (GEE) platform. The results showed that the producer accuracy (PA) and user accuracy (UA) of the surface water map in the Qinghai–Tibet Plateau in 2020 were 99% and 90%, respectively, and the Kappa coefficient reached 0.87. Our dataset showed high consistency with high-resolution images, indicating that the proposed large-scale water mapping method has great application potential. Furthermore, a new annual surface water area dataset on the Qinghai–Tibet Plateau from 2000 to 2020 was generated, and its relationship with climate, vegetation, permafrost, and glacier factors was explored. We found that the mean surface water area was about 59 481 km2, and there was a significant increasing trend (=322 km2/year,$p < 0.01$) during 2000–2020 in the plateau. Greening, warming, and wetting climate conditions contributed to the increase of surface water area. Active layer thickness and permafrost types may be the most related to the decrease of surface water area. This study provides important information for ecological assessment and protection of the plateau and promotes the implementation of sustainable development goals related to surface water resources.

Global and long-term vegetation optical depth (VOD) dataset are very useful to monitor the dynamics of the vegetation features, climate and environmental changes. In this study, the radar-based global ASCAT (Advanced SCATterometer) IB (INRAE-BORDEAUX) VOD was retrieved using a model which was recently calibrated over Africa. In order to assess the performance of IB VOD, the Saatchi biomass and three other VOD datasets (ASCAT V16, AMSR2 LPRM V5 and VODCA LPRM V6) derived from C-band observations were used in the comparison. The preliminary results show that IB VOD has a promising ability to predict biomass$(\mathrm{R}=0.74,\ \text{RMSE} =44.82\ \text{Mg}\ \text{ha}^{-1})$, which is better than V16 VOD$(\mathrm{R}=0.64,\ \text{RMSE} =51.27\ \text{Mg} \text{ha}^{-1})$and VODCA VOD$(\mathrm{R}=0.72,\ \text{RMSE} =47.14\ \text{Mg}\ \text{ha}^{-1})$. Some retrieval issues for IB VOD were found in boreal regions (e.g., Eastern America, Russia). In the future, we will focus on improving our algorithm in those regions, and produce a global and long-term dataset.

Surface carbon fluxes associated with terrestrial vegetation play a key role in the global carbon cycle. Remote sensing (RS) and land surface models (LSM) have demonstrated to be valuable tools in assessing the gross primary production (GPP). Yet, the seasonal variability of this flux, and timing of the seasonal cycle remain challenging to observe and simulate accurately. Here, the ability of four RS products and two LSM to simulate GPP and its variability was assessed. It was found that the mean seasonal GPP was simulated accurately with RS-based models, but that it failed to capture the seasonal variability and timing of the seasonal cycle. In contrast, the LSMs demonstrated their ability to simulate seasonal anomalies in GPP, but had trouble to simulate the associated phenology.

Vegetation Optical Depth (VOD) plays an increasingly important role in studying global carbon, water and energy transformation [1], [2]. This study explores the performance of the X-MEB (X-band microwave emission of the biosphere) model at global scale. Similar to the L-MEB model, the X-MEB model, built by INRAE (Institut national de recherche pour l'agriculture, l'alimentation et l'environnement) Bordeaux, aims to retrieve VOD (referred to as IB X-VOD) at X-band. To avoid the ill-posed problem caused by retrieving two parameters of interest (soil moisture (SM) and VOD) from mono-angular and dual-polarized observations (AMSR2), which are strongly correlated, we used the ERA5 SM product as an input to the X-MEB inversion. At a first step, we produced global IB X-VOD in year 2015 using the parameters (soil roughness and effective scattering albedo) calibrated in the African continent and evaluated the retrieved X-VOD with three vegetation parameters including Above-Ground Biomass (AGB), Leaf Area Index (LAI) and Normalized Difference Vegetation Index (NDVI). The evaluation results indicate X-MEB model has a great potential for global VOD retrievals from AMSR2 satellite data.

Vegetation optical depth (VOD) is used to parameterize microwave extinction effects within the vegetation layer. Many studies have showed VOD presents interesting features for applications in ecology, water and carbon cycles, and VOD is only marginally impacted by signal disturbances and artefacts from atmospheric, cloud and sun illumination effects. As soil moisture (and not VOD) has generally been the main factor of interest in retrieval studies from microwave observations, there is room for improvement in the retrieved VOD products. In this context, INRAE Bordeaux recently developed alternate VOD products from the SMOS, AMSR2 and ASCAT sensors, by addressing specifically the ill-posed problem of retrieving both SM and VOD from observations which may be strongly cross-correlated. Promising results were obtained particularly in terms of spatial correlation of these alternate VOD indices with biomass.

Decade years of efforts on the retrieval of soil moisture based on radiative transfer model have largely improved the accuracy of soil moisture (SM). This paper focus on the other parameter, namely vegetation optical depth (VOD). We retrieved X-band VOD from AMSR-E and AMSR2 observations by inverting the L-MEB model (Wigneron et al. 2007 [1]) at X-band, considering that SM was known. As SM input to the L-MEB inversion we used the ECMWF SM product. This step avoids correlation between VOD and SM retrievals from the mono-angular AMSR-E observations. In a first step we evaluated the retrieved VOD with the Copernicus Global Land Service (CGLS) LAI. The evaluation results indicate our model has a great potential for VOD retrievals from AMSR-E/2 satellite data.

Tropical aboveground carbon changes during 2010–2016 were estimated by a newly developed vegetation optical depth (VOD) product retrieved from the low-frequency L-band (1.4 GHz) passive microwave observations from the Soil Moisture and Ocean salinity (SMOS) satellite. The aboveground carbon changes estimated by VOD in the tropical region during 2010–2016 indicate the tropical region acts as a net carbon source of 111 Tg C yr-1 during 2010–2016. The declines in tropical aboveground carbon were found mainly in eastern America, African drylands and Indonesia.

The existing land data assimilation projects (NLDAS, GLDAS, ELDAS) do not include interactive vegetation land surface models, which limits the use of remote sensing data. The analysed variable in LDAS is soil moisture, only, and there is a need to account for vegetation biomass to monitor the biosphere vegetation-atmosphere CO/sub 2/ exchange. The geoland Integrated Project (2004-2007) co-funded by the European Commission, aims at addressing European and global environment issues, based on the use of remote sensing data. The carbon observatory of geoland, hereinafter referred to as "geoland/Carbon", provides a pre-operational global carbon accounting system, dealing with the impact of weather and climate variability on ecosystems fluxes and carbon stocks, on daily to seasonal and inter-annual time scales. The solution chosen in geoland/Carbon is to merge the LDAS approach and the interactive vegetation models, by making two communities work together: the meteorologists involved in ELDAS and the carbon modellers. In particular, we investigate the relationship between weather and climate variability and terrestrial CO/sub 2/ fluxes. The integration of in situ meteorological measurements and different satellite remote sensing sources of information are made by using assimilation techniques. In order to integrate the existing approaches and to deliver an assessment based on independent modelling results, two land surface models are used: 1) an operational scheme (ECMWF) used in numerical weather forecast models, modified to describe an interactive vegetation (based on ISBA-A-gs, Meteo-France); 2) a carbon-water-energy land surface scheme, fitted with carbon dynamics in biomass and soil pools, and with ecosystem dynamics (LSCE). The assimilation system can be run at the global scale with both carbon models. The assimilated output fields are checked against global observations of different nature, such as eddy covariance networks, long term ecological time series (e.g. IGBP transects), forest and soil carbon inventories, or satellite products that were not used at first in the assimilation procedure. At the end of this project, ECMWF is able to propose a single near-operational system based on components of the two approaches.