<p>Droughts are a major disaster in Africa, threatening livelihoods through their influence on crop yields but also by impacting and weakening ecosystems. Modeling the vegetation state can help anticipate and reduce the impact of droughts by predicting the vegetation response over time. Forecasting the state of vegetation is challenging: it depends on complex interactions between the plants and different environmental drivers, which can result in both instantaneous and time-lagged responses, as well as spatial effects. Furthermore, modeling these interactions at the fine resolution of landscape scale can only rely on remote sensing observations, as in-situ measurements are not global and weather models have a coarse grid. With the increasing availability of remote sensing data, deep learning methods are a promising avenue for these spatiotemporal tasks. Here, we introduce both a dataset and a baseline deep neural network, modeling the vegetation response to climate at landscape scale in Africa.</p> <p>EarthNet2021 [1] introduced leveraging self-supervised learning for satellite imagery forecasting based on coarse-scale weather in Europe. Here, we introduce EarthNet2023 with a more narrow focus on drought impacts in Africa. It contains over 45,000 Spatio-temporal minicubes (each 1.28x1.28km) at representative locations over the whole African continent. Alongside Sentinel-2 reflectance, ERA5 weather, and topography, it also contains Sentinel-1 backscatter, soil properties, and a long-term Normalized Difference Vegetation Index (NDVI) climatology based on Landsat. The latter allows evaluating models on vegetation anomalies, thereby including modeling of drought impacts. EarthNet2023 is intended as an open benchmark challenge, allowing multiple research groups to develop their approaches to drought impact modeling in Africa. </p> <p>As a baseline for EarthNet2023, we train a  Convolutional Long Short-Term Memory (ConvLSTM) deep learning model. Previous work has shown it is suitable for spatiotemporal satellite imagery forecasting [2, 3, 4]. The ConvLSTM baseline captures the seasonal evolution of NDVI over a wide range of vegetation types. General spatial patterns are well-captured as well as a first indication of skill during weather extremes is seen, although the accuracy of the predictions is inconsistent, and the confidence in the model is therefore too low. This suggests, with further development, deep learning approaches are promising for modeling vegetation evolution in Africa, potentially even up to the degree to support anticipatory action with drought impact modeling.</p> <p> </p> <p>[1] Requena-Mesa, C., Benson, V., Reichstein, M., Runge, J., & Denzler, J. (2021). EarthNet2021: A large-scale dataset and challenge for Earth surface forecasting as a guided video prediction task. In <em>CVPR 2021</em> (pp. 1132-1142).</p> <p>[2] Diaconu, C. A., Saha, S., Günnemann, S., & Zhu, X. X. (2022). Understanding the Role of Weather Data for Earth Surface Forecasting Using a ConvLSTM-Based Model. In <em>CVPR 2022 </em>(pp. 1362-1371).</p> <p>[3] Kladny, K. R. W., Milanta, M., Mraz, O., Hufkens, K., & Stocker, B. D. (2022). Deep learning for satellite image forecasting of vegetation greenness. <em>bioRxiv</em>.</p> <p>[4] Robin, C., Requena-Mesa, C., Benson, V., Alonso, L., Poehls, J., Carvalhais, N., & Reichstein, M. (2022). Learning to forecast vegetation greenness at fine resolution over Africa with ConvLSTMs. In <em>Tackling Climate Change with Machine Learning: workshop at NeurIPS 2022.</em> </p>

Distribution électronique Cairn.info pour Médecine & Hygiène. © Médecine & Hygiène. Tous droits réservés pour tous pays.La reproduction ou représentation de cet article, notamment par photocopie, n'est autorisée que dans les limites des conditions générales d'utilisation du site ou, le cas échéant, des conditions générales de la licence souscrite par votre établissement. Toute autre reproduction ou représentation, en tout ou partie, sous quelque forme et de quelque manière que ce soit, est interdite sauf accord préalable et écrit de l'éditeur, en dehors des cas prévus par la législation en vigueur en France. Il est précisé que son stockage dans une base de données est également interdit.

<p>Wood density is an emergent property resultant of tree growth strategies modulated by local edapho-climatic and stand development conditions. It is associated with the biomechanical support of trees and hydraulic conductivity or safety, directly and indirectly influencing a range of ecological processes, including, among others, tree growth, tree resistance to disturbances, and mortality. Tree wood density is also crucial for assessing vegetation carbon stocks by supporting the link between a volumetric retrieval and a mass estimate. Earlier studies based on tree-level wood density measurements have reported significant relationships between wood density, environmental conditions, and tree growth strategies. However, these were either regionally focused or suffering from data availability, lacking a representative large-scale and spatially explicit representation of factors influencing tree wood density. This study collects and collates information from several sources to construct a global database of 28,822 tree-level wood density measurements alongside with a wide set of climate, soils, topography, and Earth observation covariates to support the development of statistical models for wood density. The dataset, consisting of more than 3,000 global covariates, is used for training different machine learning models, including random forest model (RF), light gradient boosting model (LGBM), extreme gradient boosting model (XGBoost), and bagged trees models. The experimental design considers six cross-validation approaches: either random 5-fold; according to two sets of climate classifications, land cover types, ecozones, or latitudinal ranges. Model performances are assessed with the coefficient of determination (R<sup>2</sup>) and Root-mean-square errors (RMSE) when predicting an independent test subset of wood density. The top ten models show a prominent performance (R<sup>2</sup> > 0.67 and RMSE < 0.09), and their ensemble mean, and standard deviation are considered the best estimation and uncertainty in wood density predictions, respectively. Systematic underestimation biases are observed around the low northern latitudes (0º-20ºN), primarily due to the lack of wood density measurements. Further analysis of sources of uncertainties and their quantification support the generation of a global quantitative and spatially explicit representation of wood density. The ecological interpretation and quantitative assessment of global wood density, and associated uncertainties aim to contribute to improving predictions of vegetation biomass and inferring ecosystem resistance under current and future climate scenarios.</p>

<p>The biosphere displays high heterogeneity at landscape-scale. Vegetation modelers struggle to represent this variability in process-based models because global observations of micrometeorology and plant traits are not available at such fine granularity. However, remote sensing data is available: the Sentinel 2 satellites with a 10m resolution capture aspects of localized vegetation dynamics. The EarthNet challenge (EarthNet2021, [1]) aims at predicting satellite imagery conditioned on coarse-scale weather data. Multiple research groups approached this challenge with deep learning [2,3,4]. Here, we evaluate how well these satellite image models simulate the vegetation response to climate, where the vegetation status is approximated by the NDVI vegetation index.</p> <p>Achieving the new vegetation-centric evaluation requires three steps. First, we update the original EarthNet2021 dataset to be suitable for vegetation modeling: EarthNet2021x includes improved georeferencing, a land cover map, and a more effective cloud mask. Second, we introduce the interpretable evaluation metric VegetationScore: the Nash Sutcliffe model efficiency (NSE) of NDVI predictions over clear-sky observations per vegetated pixel aggregated through normalization to dataset level. The ground truth NDVI time series achieves a VegetationScore of 1, the target period mean NDVI a VegetationScore of 0. Third, we assess the skill of two deep neural networks with the VegetationScore: ConvLSTM [2,3], which combines convolutions and recurrency, and EarthFormer [4], a Transformer adaptation for Earth science problems. </p> <p>Both models significantly outperform the persistence baseline. They do not display systematic biases and generally catch spatial patterns. Yet, both neural networks achieve a negative VegetationScore. Only in about 20% of vegetated pixels, the deep learning models do beat a hypothetical model predicting the true target period mean NDVI. This is partly because models largely underestimate the temporal variability. However, the target variability may partially be inflated by the noisy nature of the observed NDVI. Additionally, increasing uncertainty for longer lead times decreases scores: the mean RMSE in the first 25 days is 50% lower than between 75 and 100 days lead time. In general, consistent with the EarthNet2021 leaderboard, the EarthFormer outperforms the ConvLSTM. With EarthNet2021x, a more narrow perspective to the EarthNet challenge is introduced. Modeling localized vegetation response is a task that requires careful adjustments of off-the-shelf computer vision architectures for them to excel. The resulting specialized approaches can then be used to advance our understanding of the complex interactions between vegetation and climate.</p> <p><br /><br /></p> <p> [1] Requena-Mesa, Benson, Reichstein, Runge and Denzler. EarthNet2021: A large-scale dataset and challenge for Earth surface forecasting as a guided video prediction task. CVPR Workshops, 2021.</p> <p> [2] Diaconu, Saha, Günnemann and Zhu. Understanding the Role of Weather Data for Earth Surface Forecasting Using a ConvLSTM-Based Model. CVPR Workshops, 2022.</p> <p> [3] Kladny, Milanta, Mraz, Hufkens and Stocker. Deep learning for satellite image forecasting of vegetation greenness. bioRxiv, 2022.</p> <p> [4] Gao, Shi, Wang, Zhu, Wang, Li and Yeung. Earthformer: Exploring Space-Time Transformers for Earth System Forecasting. NeurIPS, 2022.</p>