Remote Sensing

Aerial insects are vital for nature and society. Though methods to observe flying insects have consistently improved in the last decades, insects remain difficult to monitor systematically and consistently over large spatial and temporal scales. Remote sensing with radars has proved to be one of the more effective tools for observation. However, as radars are most sensitive to targets similar in size to the radar wavelength, the detectable sub-group of aerial insects of a certain size range depends on the employed radar. Here, we present a novel method based on data of a zenith-pointing W-band (94 GHz,{lambda} = 0.32 cm) Doppler cloud radar to estimate insect concentration in a vertical profile. Multiple meteorological state-of-the-art algorithms are combined to extract insect signals from the radar data and quantify their abundance from 50 m to 1000 m above the ground. For evaluation, this method is applied to Doppler cloud radar data from a summertime 30 day observation period in central Germany. Results are compared to data from an X-band (9.4 GHz,{lambda} = 3.2 cm) radar in the same region. Aerial insect concentration derived from the W-band radar, which is sensitive to insects in the mm size range, is substantially higher than from the X-band radar, detecting insects in the cm size range. In addition, diel flight timings vary between the different sub-groups of aerial insects observed by the two radar instruments. With its superior sensitivity to smaller insects like aphids, the proposed methodology complements existing entomological radar techniques and contributes to achieving a more complete description of aerial insect activity.

The advent of abundant Earth observation data enables the development of novel predictive methods for forecasting climate impacts on the state and health of terrestrial ecosystems. Here, we target the spatial and temporal variations of land surface reflectance and vegetation greenness, measuring the density of green vegetation and active foliage area, conditioned on current and past climate and the local topography. We train two alternative recurrent deep learning models that rely on convolutional layers for forecasting the spatially resolved deviation of surface reflectance across a heterogeneous landscape from a specified initial state (Baseline Framework). We demonstrate efficiency of the Baseline Framework with respect to training convergence speed. Using data from diverse ecosystems and land cover types across Europe and following a standardized model evaluation framework (EarthNet2021 Challenge), results indicate increased performance in predicting surface greenness during drought events of the models presented here, compared to currently published benchmarks. Our results demonstrate how deep learning methods enable early-warning of vegetation responses to the impacts of climatic extreme events, such as the drought-related loss of green foliage.

In this study, we assess the potential of long time series of Sentinel-1 SAR data to predict forest growing stock volume and evaluate the temporal dynamics of the predictions. The boreal coniferous forests study site is located near the Hyytiala forest station in central Finland and covers an area of 2,500 km2 with nearly 17,000 stands. We considered several prediction approaches (linear, support vector and random forests regression) and fine-tuned them to predict growing stock volume in several evaluation scenarios. The analyses used 96 Sentinel-1 images acquired over three years. Different approaches for aggregating SAR images and choosing feature (predictor) variables were evaluated. Our results demonstrate considerable decrease in RMSEs of growing stock volume as the number of images increases. While prediction accuracy using individual Sentinel-1 images varied from 85 to 91 m3/ha RMSE (relative RMSE 50-53%), RMSE with combined images decreased to 75.6 m3/ha (relative RMSE 44%). Feature extraction and dimension reduction techniques facilitated achieving the near-optimal prediction accuracy using only 8-10 images. When using assemblages of eight consecutive images, the GSV was predicted with the greatest accuracy when initial acquisitions started between September and January. HighlightsO_LITime series of 96 Sentinel-1 images is analysed over study area with 17,762 forest stands. C_LIO_LIRigorous evaluation of tools for SAR feature selection and GSV prediction. C_LIO_LIImproved periodic seasonality using assemblages of consecutive Sentinel-1 images. C_LIO_LIAnalysis of combining images acquired in "frozen" and "dry summer" conditions. C_LIO_LICompetitive estimates using calculation of prediction errors with stand-area weighting. C_LI

Land-Use and Land-Cover (LULC) mapping is relevant for many applications, from Earth system and climate modelling to territorial and urban planning. Global LULC products are continuously developing as remote sensing data and methods grow. However, there is still low consistency among LULC products due to low accuracy for some regions and LULC types. Here, we introduce Sentinel2GlobalLULC, a Sentinel-2 RGB image dataset, built from the consensus of 15 global LULC maps available in Google Earth Engine. Sentinel2GlobalLULC v1.1 contains 195572 RGB images organized into 29 global LULC mapping classes. Each image is a tile that has 224 x 224 pixels at 10 x 10 m spatial resolution and was built as a cloud-free composite from all Sentinel-2 images acquired between June 2015 and October 2020. Metadata includes a unique LULC type annotation per image, together with level of consensus, reverse geo-referencing, and global human modification index. Sentinel2GlobalLULC is optimized for the state-of-the-art Deep Learning models to provide a new gate towards building precise and robust global or regional LULC maps.

Developing actionable early detection and warning systems for agricultural stakeholders is crucial to reduce the annual $200B USD losses and environmental impacts associated with crop diseases. Agricultural stakeholders primarily rely on labor-intensive, expensive scouting and molecular testing to detect disease. Spectroscopic imagery (SI) can improve plant disease management by offering decision-makers accurate risk maps derived from Machine Learning (ML) models. However, training and deploying ML requires significant computation and storage capabilities. This challenge will become even greater as global scale data from the forthcoming Surface Biology & Geology (SBG) satellite becomes available. This work presents a cloud-hosted architecture to streamline plant disease detection with SI from NASAs AVIRIS-NG platform, using grapevine leafroll associated virus complex 3 (GLRaV-3) as a model system. Here, we showcase a pipeline for processing SI to produce plant disease detection models and demonstrate that the underlying principles of a cloud-based disease detection system easily accommodate model improvements and shifting data modalities. Our goal is to make the insights derived from SI available to agricultural stakeholders via a platform designed with their needs and values in mind. The key outcome of this work is an innovative, responsive system foundation that can empower agricultural stakeholders to make data-driven plant disease management decisions, while serving as a framework for others pursuing use-inspired application development for agriculture to follow that ensures social impact and reproducibility while preserving stakeholder privacy. Key PointsO_LICloud-based plant disease detection system, easily accommodates newly developed and/or improved models, as well as diverse data modalities. C_LIO_LIEmpower agricultural stakeholders to use hyperspectral data for decision support while preserving stakeholder data privacy. C_LIO_LIOutline framework for researchers interested in designing geospatial/remote sensing applications for agricultural stakeholders to follow. C_LI

Tree detection is a fundamental task in remote sensing for forestry and ecosystem ecology applications. While many individual tree segmentation algorithms have been proposed, the development and testing of these algorithms is typically site specific, with few methods evaluated against data from multiple forest types simultaneously. This makes it difficult to determine the generalization of proposed approaches, and limits tree detection at broad scales. Using data from the National Ecological Observatory Network we extend a recently developed semi-supervised deep learning algorithm to include data from a range of forest types, determine whether information from one forest can be used for tree detection in other forests, and explore the potential for building a universal tree detection algorithm. We find that the deep learning approach works well for overstory tree detection across forest conditions, outperforming conventional LIDAR-only methods in all forest types. Performance was best in open oak woodlands and worst in alpine forests. When models were fit to one forest type and used to predict another, performance generally decreased, with better performance when forests were more similar in structure. However, when models were pretrained on data from other sites and then fine-tuned using a small amount of hand-labeled data from the evaluation site, they performed similarly to local site models. Most importantly, a universal model fit to data from all sites simultaneously performed as well or better than individual models trained for each local site. This result suggests that RGB tree detection models that can be applied to a wide array of forest types at broad scales should be possible.

Two sound level maps currently exist for the contiguous United States. One was developed by the National Park Service (NPS) using machine learning methods and sound pressure level monitoring data, and the other by the Bureau of Transportation Statistics (BTS) using transportation noise models of roadway, aviation, and rail sources. Developed for different purposes, each has distinct strengths and weaknesses. This study aimed to compare the two models, develop a hybrid model integrating both, and evaluate its performance against field measurements. Linear regression with data from 378 NPS field sites was used to relate the NPS L50 metric to Leq. A positive association was observed, and the resulting regression equation was used to convert L50 to Leq. Comparing BTS 2018 and 2020 with the converted NPS model, we found strong correlation and small bias between BTS years (Pearsons r = 0.90, Spearmans rho = 0.88, bias = 0.3 dBA), but larger differences between BTS and NPS, with BTS levels on average [~]6 dBA higher. A hybrid model was created by filling censored BTS areas with converted NPS Leq values. Evaluation against 708 NPS measurements and 757 metropolitan measurements showed good performance (bias = 0.4 dBA, MAE = 5.0 dBA for NPS; bias = -0.5 dBA, MAE = 3.8 dBA for metropolitan sites). Using the hybrid model, we estimated that [~]36.4 million people (11.1% of the U.S. population) are exposed above 55 dB Leq. The hybrid model provides a resource to inform noise-related environmental health research, policy, and planning.

High-resolution monitoring of forest structure and productivity is essential for effective natural resource management. However, monitoring approaches such as field-based forest inventories or extensive lidar campaigns are costly, time-intensive, and spatially limited. Therefore, inexpensive and accessible methods are needed. SatCHM (Satellite Canopy Height Model) was developed to be an accessible and open-source tool for researchers, allowing for site-specific and temporally flexible predictions of canopy height with limited computational resources. SatCHM requires four inputs: panchromatic satellite imagery, solar and sensor angle metadata of satellite imagery, digital elevation models (DEMs), and lidar-produced CHMs for an area of interest. After SatCHM pre-processes inputs, data is loaded into a collection of convolutional neural networks (CNNs) for image-to-image regression. This ensemble cooperates to yield high-resolution predictions (up to 0.5-meter) of three-dimensional tree structure with discernible tree crowns across a broader defined area of interest. After calculating the mean absolute error for each prediction output, the median of these mean absolute errors was 6.06 meters.

O_LIThe exodus of swallows from communal nighttime roosts is often visible as an expanding ring-shaped pattern in weather radar data. The WSR-88D network operated by the National Weather Service archives more than 25 years of data across 143 stations in the contiguous US. However, access to information about the roosting behavior of swallows is limited by the cost of manual annotation of these scans. C_LIO_LIWe develop an AI system to detect and track swallow roosts in weather radar data. Our model is based on the Faster R-CNN architecture and is customized to incorporate multiple spatial and temporal channels in volumetric radar scans using novel adaptor layers. We systematically study the impact of network architecture and input representation for this task. We incorporate our detection outputs into an AI-assisted system with an interface for human screening to collect research-grade data about roosting behavior. We deploy the system to collect information from 12 radar stations in the Great Lakes region of the US spanning 21 years. C_LIO_LIThe addition of temporal information improves roost detection performance from 47.0% mean average precision to 54.7%. Temporal information helps the model recognize the expanding pattern of roosts and filter false positives due to rain and static structures. Our system allowed the annotation of 15,628 roost signatures with 64,620 single-frame detections in 612,786 radar scans with 183.6 total hours of human screening, or 1.08 seconds per radar scan. C_LIO_LIOur AI-assisted system provides research-quality roost data with far less human effort than manual annotation of radar scans. The data contains critical information about the phenology and population trends of swallows and martins, a declining group of aerial insectivores. Our successful deployment to collect historical data for 8% of the radar stations in the contiguous US lays the groundwork for continentscale analysis of swallow roosts, and provides a starting point for analysis of other family-specific phenomena in weather radar, such as bat roosts and mayfly hatches. C_LI

The measurement of light received at the cornea of the eye is a paramount consideration for the understanding of the relation between environmental illumination and the non-image-forming effects of light. The field of view (FOV) at the cornea is less than a full hemisphere, because it is partially occluded by human facial morphology. The International Commission on Illumination (CIE) has defined a standard model of human FOV. A suitably designed physical occluder attached to the sensor (of a light meter) has been proposed as a means of incorporating the effect of human FOV when taking measurements. Similarly, when using simulation to predict light received at the cornea, a geometrical description of the occluder at the eye point(s) can be added to the 3D model of the scene. The first occluder model proposed to represent CIE human FOV was enumerated in terms of: the CIE definition; the radius of the occluder; and, the radius of the light sensor disc. We present a simpler model based only on the CIE definition and the occluder radius. Both models were tested using a virtual goniophotometer. Various sensor response functions describing the spatial sensitivity across the sensor disc, including several we characterized through laboratory measurements, were included in the test. For all functions considered, the performance of the simpler occluder model was equivalent to or better than the model first proposed.

The US NEXRAD radar network has monitored the aerosphere over the US and its territories continuously since the 1990s and archived nearly 300 million radar volume scans. These data contain a wealth of information about the movements of birds, bats, and insects. Historically, this biological information was difficult to access due to the amount of data and challenges in analyzing it. In the last 15 years, fueled by computational and methodological advances, large-scale aeroecology research has blossomed. However, comprehensive analyses of the NEXRAD archive remain very costly. We collected measurements of biological activity from every volume scan in the NEXRAD archive--nearly 300 million data files total--to assemble a dataset of aerial biomass over the US from 1995 to 2025. The core data are vertical profiles, which summarize biological activity at different heights above the radar station for each volume scan. We also provide time series data products that aggregate vertical profiles to point measurements at radar stations across time. These data products can support a range of aeroecology analyses at significantly reduced effort.

Individual tree structure plays a key role in forest monitoring, biomass estimation, and ecological assessment. However, ground-based remote sensing methods such as terrestrial and mobile laser scanning frequently produce incomplete point clouds due to occlusion, particularly in the upper canopy. This limits the accuracy of derived structural metrics such as tree height or crown volume. In this study, we present a novel deep learning-based method to reconstruct the outer crown shape of coniferous trees from incomplete point clouds. Instead of completing the full tree structure, we focus on predicting the alpha-shape of the crown, enabling a more efficient and generalizable approach for structural reconstruction. We train a geometry-aware transformer model (AdaPoinTr) on synthetically generated partial tree crowns and evaluate its performance across three independent datasets encompassing different forest types and acquisition conditions. The model consistently improved crown shape similarity metrics and reduced height estimation errors compared to using partial data alone (reduced bias from -11% to -3.5%). Our results demonstrate that this shape-based strategy enables the extraction of key tree-level parameters from incomplete data, offering a practical solution for gaining improved 3D forest structural information from cost-sensitive or logistically constrained forest monitoring acquisitions.

The acquisition of phenological events play an integral part in investigating the effects of climate change on forest dynamics and assessing the potential risk involved with the early onset of young leaves. Large scale mapping of forest phenological timing using earth observation data, could facilitate a better understanding of phenological processes due to an added spatial component. The translation of traditional phenological ground observation data into reliable ground truthing for the purpose of the training and validation of Earth Observation (EO) mapping applications is a challenge. In this study, we explored the possibility of predicting high resolution phenological phase data for European beech (Fagus sylvatica) with the use of Unmanned Aerial Vehicle (UAV)-based multispectral indices and machine learning. Using a comprehensive feature selection process, we were able to identify the most effective sensors, vegetations indices, training data partitions, and machine learning models for phenological phase prediction. The best performing model that generalised well over various sites was the model utilising the Green Chromatic Coordinate (GCC) and Generalized Addictive Model (GAM) boosting. The GCC training data was derived from the radiometrically calibrated visual bands from a multispectral sensor and predicted using uncalibrated RGB sensor data. The final GCC/GAM boosting model was capable in predicting phenological phases on unseen datasets within a RMSE threshold of 0.5. This research shows the potential of the interoperability among common UAV-mounted sensors in particular the utility of readily available low cost RGB sensors. Considerable limitations were however discovered with indices implementing the near-infrared (NIR) band due to oversaturation. Future work involves adapting models to facilitate the ICP Forests phenological flushing stages.

Road traffic accidents represent a global public safety crisis, necessitating advanced computational tools for accurate injury severity prediction and effective decision support. This study evaluates high-performing ensemble machine learning models, including AdaBoost, XGBoost, LightGBM, HistGBRT, CatBoost, Gradient Boosting, NGBoost, and Random Forest, using a comprehensive National Highway Traffic Safety Administration (NHTSA) dataset from 2018 to 2022. While all models demonstrated exceptional predictive accuracy, with HistGBRT achieving the highest overall accuracy of 92.26%, a defining achievement of this work is the perfect classification (100% precision and recall) of fatal injuries across all ensemble architectures. To bridge the gap between predictive performance and actionable intelligence, this research integrates SHapley Additive exPlanations (SHAP) to provide both global insights into dataset-wide risk factors and local, instance-specific rationales for individual crash events. The global analysis identified ethnicity, airbag deployment, and harmful event type as primary drivers of injury severity, while local force and waterfall plots revealed the precise "push and pull" of variables for specific incidents. The results offer a robust, interpretable framework for stakeholders tasked with improving traffic safety and mitigating crash-related harm.

This paper presents methods and results that combine multiple carbon storage and above-ground biomass datasets using a simple decision tree approach. The resulting dataset combines the positive attributes of the multiple input datasets in order to have global, high-resolution extent while utilizing the best statistical methods where possible. Visual inspection shows very different spatial configurations of carbon storage between the results and the input data, suggesting that combination of methods can improve estimates. The summation of the decision tree result was 336.55 petagrams while the summation of a dataset based on the IPCC Tier 1 method was 502.38 petagrams (49.27% higher than the results from the decision tree).

Hyperspectral cameras are a key enabling technology in precision agriculture, biodiversity monitoring, and ecological research. Consequently, these applications are fuelling a growing demand for devices that are suited to widespread deployment in such environments. Current hyperspectral cameras, however, require significant investment in post-processing, and rarely allow for live-capture assessments. Here, we introduce a novel hyperspectral camera that combines live spectral data and high-resolution imagery. This camera is suitable for integration with robotics and automated monitoring systems. We explore the utility of this camera for applications including chlorophyll detection and live display of spectral indices relating to plant health. We discuss the performance of this novel technology and associated hyperspectral analysis methods to support an ecological study of grassland habitats at Wytham Woods, UK.

Geospatial modelling can give fundamental insights in the biogeography of life, providing key information about the living world in current and future climate scenarios. Emerging statistical and machine learning approaches can help us to generate new levels of predictive accuracy in exploring the spatial patterns in ecological and biophysical processes. Although these statistical models cannot necessarily represent the essential mechanistic insights that are needed to understand global biogeochemical processes under ever-changing environmental conditions, they can provide unparalleled predictive insights that can be useful for exploring the variation in biophysical processes across space. As such, these emerging tools can be a valuable approach to complement existing mechanistic approaches as we aim to understand the biogeography of Earths ecosystems. Here, we present a comprehensive methodology that efficiently handles large datasets to produce global predictions. This mapping pipeline can be used to generate quantitative, spatially explicit predictions, with a particular emphasis on spatially-explicit insights into the evaluation of model uncertainties and inaccuracies.

Statewide tracking of urban tree canopy change is essential for evaluating progress toward policy targets, but detecting real change requires both high-resolution mapping and rigorous uncertainty estimation. We produced a four-year canopy cover time series for all California census-designated places using 60-cm NAIP aerial imagery and a U-Net deep learning model trained with semi-automated LiDAR-derived labels and manually annotated tiles. Canopy cover and change were estimated using stratified, error-adjusted area estimation, enabling comparisons across years. Statewide canopy cover showed a modest negative trend from 2016 to 2022 (Sens slope: -0.60% per year), but confidence intervals included zero across all groups and climate zones, indicating that trends were not statistically distinguishable from no change. Urban canopy cover was consistently lower than non-urban canopy by approximately six percentage points, and canopy cover was highest in the Northern California Coast and lowest in the Southwest Desert. Residential parcels accounted for 55-56% of canopy within incorporated urban areas across all years, indicating that statewide canopy increase goals will require engagement with private landowners. Error adjustment substantially altered canopy estimates relative to raw pixel-count totals, with direct implications for AB 2251 canopy tracking where baselines and targets drawn from unadjusted maps may not reflect true canopy extent. This open-source workflow is transferable to future NAIP acquisition years and other U.S. states, providing a scalable framework for long-term urban forest monitoring.

Woody plant encroachment (WPE)--a phenomenon similar to species invasion--is shifting many grasslands and savannas into shrub and evergreen-dominated ecosystems. Tracking WPE is difficult because shrubs and small trees are much smaller than the coarse resolution of common remote sensing platforms (> 10 m2) and the impassibility of encroaching woody thickets slows ground-based approaches. Many agencies have been investing in fine resolution (< 2 m2) remote sensing through programs such as the United States Department of Agriculture (USDA) National Agriculture Imagery Program (NAIP) and the National Ecological Observatory Network (NEON). Both use low-flying planes and provide data to end users in an easy-to-use format at large spatial extents. By removing entry barriers, these publicly available open-source programs could increase the accessibility and extent of remote sensing. We compared two common methods of machine learning classification of land cover (random forests and support vector machines) factorially crossed with these two freely available remote sensing platforms to determine if we could quickly and accurately develop remote sensing of major vegetation types in a tallgrass prairie landscape undergoing encroachment by shrubs and trees. Our work took place at Konza Prairie Biological Station--a landscape scale experiment that results in a wide range of land cover types. All models had very high overall classification accuracy (>90%), with the NEON-based models a few percent more accurate than NAIP. A model using both inputs had the highest accuracy. However, the accuracies of NAIP and NEON models differed for woody vegetation: compared to NEON, NAIP accuracy was, 82-93% compared to 94-98% for shrubs, 72-92% compared to 93-98% for deciduous trees, and 52-78% compared to 83-86% for evergreen trees (specifically Juniperus virginiana). NEON-based models relied on canopy height (LiDAR) to make classifications, whereas the several bands of light make similar contributions to accuracy in the NAIP models. Finally, we found that both machine learning approaches had similar accuracy, but random forests ran substantially faster. We conclude that with large training datasets, publicly available aerial imagery and similar products (e.g., UAVs, micro-satellites) can produce fine-scale, high-accuracy remote sensing of WPE in this region with low up-front costs.

Semi-arid grasslands and other ecosystems combine green and senescent leaves featuring different biochemical and optical properties, as well as functional traits. Knowing how these properties vary is necessary to understand the functioning of these ecosystems. However, differences between green and senescent leaves are not considered in recent models representing radiative transfer, heat, water and CO2 exchange such as the Soil-Canopy Observation of Photosynthesis and Energy fluxes (SCOPE). Neglecting the contribution of senescent leaves to the optical and thermal signal of vegetation limits the possibilities to use remote sensing information for studying these ecosystems; as well as neglecting their lack of photosynthetic activity increases uncertainty in the representation of ecosystem fluxes. In this manuscript we present senSCOPE as a step towards a more realistic representation of mixed green and senescent canopies. senSCOPE is a modified version of SCOPE model that describes a canopy combining green and senescent leaves with different properties and function. The model relies on the same numerical solutions than SCOPE, but exploits the linear nature of the scattering coefficients to combine optical properties of both types of leaf. Photosynthesis and transpiration only take place in green leaves; and different green and senescent leaf temperatures are used to close the energy balance. Radiative transfer of sun-induced fluorescence (SIF) and absorptance changes induced by the xanthophyll cycle action are also simulated. senSCOPE is evaluated against SCOPE both using synthetic simulations, forward simulations based on observations in a Mediterranean tree-grass ecosystem, and inverting dataset of ground measurements of reflectance factors, SIF, thermal radiance and gross primary production on a heterogeneous and partly senescent Mediterranean grassland. Results show that senSCOPE outputs vary quite linearly with the fraction of green leaf area, whereas SCOPE does not respond linearly to the effective leaf properties, calculated as the weighted average of green and senescent leaf parameters. Inversion results and pattern-oriented model evaluation show that senSCOPE improves the estimation of some parameters, especially chlorophyll content, with respect SCOPE retrievals during the dry season. Nonetheless, inaccurate knowledge of the optical properties of senescent matter still complicates model inversion. senSCOPE brings new opportunities for the monitoring of canopies mixing green and senescent leaves, and for improving the characterization of the optical properties of senescent material.