Accurate vegetation-community classification is essential for ecological monitoring, habitat assessment, and evidence-based environmental management in heterogeneous landscapes. Existing studies often rely on standalone tree ensembles or generic neural networks, although fine-grained ecological classes frequently exhibit overlapping spectral, topographic, and structural characteristics.
The study maps tree species across Denmark using National Forest Inventory plots and Earth Observation data, comparing manually engineered spectral‑temporal features (STF) from Sentinel‑1 and Sentinel‑2 with embeddings from the foundation models TESSERA and AlphaEarth. Random forest, XGBoost, and MLP classifiers were evaluated, with the STF‑based MLP achieving the highest macro F1 scores for pure and mixed stands. The best model was applied nationally to produce a 10 m resolution tree species map, achieving 79.9% area‑adjusted accuracy and released as an open‑access product.
By Alkiviadis Koukos, Spyros Kondylatos, Thomas Nord-Larsen, Lotte Nyborg, Christian T{\o}ttrup, Kenneth Grogan
arXiv:2606. 32023v1 Announce Type: cross Abstract: Forest attributes are essential for national-scale resource monitoring.
By Emilie Vautier, Cl\'ement Mallet, C\'edric Vega
arXiv:2608. 11638v1 Announce Type: new Abstract: Spatially continuous quantification of forest above-ground biomass (AGB) is what makes carbon accounting credible and mitigation strategies actionable.
By Pann Thinzar Seint, Bryan Atwood, Subas Chhatkuli
The study maps tree species across Denmark using National Forest Inventory plots and Earth Observation data, comparing manually engineered spectral‑temporal features (STF) with embeddings from foundation models TESSERA and AlphaEarth. Random forest, XGBoost, and MLP classifiers were evaluated, with the STF‑based MLP achieving the highest macro F1 scores (0.843 for pure stands, 0.653 for mixed stands). The best model was applied nationally to produce a 10 m resolution tree species map, achieving 79.9% area‑adjusted accuracy and released as an open‑access resource.
The paper introduces DWFF‑Net, a Dynamic Weighted Feature Fusion Network designed to improve multi‑scale segmentation for agricultural habitat recognition. It employs a frozen DINOv3 encoder, a data‑level adaptive dynamic weighting strategy, and a decoder with a dynamic weight calculation network and hybrid loss. Experiments on an agricultural habitat dataset show significant gains in mIoU and mF1 over static fusion and several baseline models, especially for tiny features such as scattered trees.
By Kesong Zheng, Zhi Song, Peizhou Li, Shuyi Yao, Tong Li, Yonglin Shen, Zhenxing Bian
arXiv:2606. 26194v1 Announce Type: cross Abstract: Urban tree biomass remains less spatially explicitly quantified than biomass in managed forests because many estimates rely on inventories or coarse products that cannot resolve individual crowns or fine-scale heterogeneity.
By Jose Bermudez (McMaster University, Hamilton, Ontario, Canada), Zilong Zhong (McMaster University, Hamilton, Ontario, Canada), Dominic Cyr (, Environment and Climate Change Canada, Montreal, Quebec, Canada), Camile Sothe (Planet Labs PBC, San Francisco, California, USA), Alemu Gonsamo (McMaster University, Hamilton, Ontario, Canada)
The paper introduces MS-ALS-SPECIES, an open multispectral airborne laser scanning dataset designed for tree species classification. It contains 6,326 segment-level point clouds of individual trees from nine species in southern Finland, collected with two multispectral laser systems—HeliALS and Optech Titan—providing high point densities. The authors also present a crowdsourcing application for efficient field reference data collection and demonstrate the dataset’s utility through new species‑classification analyses building on prior work.
By Matti Hyypp\"a, Klaara Salolahti, Eric Hyypp\"a, Xiaowei Yu, Josef Taher, Leena Matikainen, Matti Lehtom\"aki, Paula Litkey, Teemu Hakala, Harri Kaartinen, Juha Hyypp\"a, Antero Kukko
The study evaluates the use of frozen geospatial foundation embeddings (AlphaEarth) for mapping cultivated versus non‑cultivated land in Maine. Using 192 spatially separated patches and USDA Cropland Data Layer labels, a lightweight classifier achieved 93.7% overall accuracy without fine‑tuning, and a nearest‑class‑centroid rule reached 90.2%. A balanced sample of 60,000 labeled pixels was nearly as effective as the full 8.6 million‑pixel pool, and classifiers trained in one year remained accurate across 2018‑2023. In a blind human validation of 385 points, the AlphaEarth‑plus‑random‑forest map matched 95.3% of the consensus, outperforming the CDL reference (91.7%).
By Mohammad Ammar Mughees, Giovanni Montefoschi, Zhongxin Chen, Maria Antonia Brovelli
A deep learning model (U‑Net‑ID) mapped 6.5 million large trees (crown area ≥ 100 m²) across 78.7 % of the Sierra Nevada Floristic Province using sub‑meter aerial imagery from 2020. The spatial distribution of these trees correlated with elevation, temperature, and precipitation. From 2020 to 2025, Sentinel‑2 time series and BFAST breakpoint analysis revealed that wildfires were the dominant driver of large‑tree mortality, killing 10 % of all large trees, especially during the extreme 2020‑2021 fire seasons.
By Fabien H. Wagner, Dan J. Dixon, Christopher W. Woodall, Mayumi C. M. Hirye, Felipe Saad, Griffin Carter, Ricardo Dalagnol, Lorena Alves, Cynthia Creze, Stephen C. Hagen, Zhihua Liu, Christopher Mihiar, Adugna Mullissa, Le Bienfaiteur Sagang Takougoum, Bryan Shaddy, Yan Yang, Dafeng Zhang, Sassan Saatchi
arXiv:2608. 07801v1 Announce Type: cross Abstract: Precision nitrogen (N) management (PNM) for cotton requires in-season monitoring of crop growth parameters and N status indicators to decide fertilizer timing, placement, and application rates for optimal canopy development and yield.
By Vaishali Swaminathan, Nithya Rajan, J Alex Thomasson, Amrit Shrestha, Karem Meza Capcha, Robert Hardin, Pramod Pokhrel
The paper presents a framework that enhances deep‑learning tree‑cover mapping in New South Wales by fusing multiple imagery sources and normalizing image quality. It introduces an image‑composition technique that removes defects and a prediction‑fusion method that reduces reliance on any single image, together cutting errors by 38.2 % and 53.6 % respectively. Label transfer across diverse imagery further boosts data efficiency, yielding error reductions of 28.1 %–76.2 % and a 13‑fold decrease in performance variability across dates.
By Kal Backman, Jared Wood, Adam Roff