Spatially accurate estimates of forest above-ground biomass (AGB) are indispensable for carbon-stock accounting and sustainable silviculture. Existing mapping approaches face challenges in densely vegetated Coastal Plain forests because of seasonal optical variability, radar–optical saturation, and limited wall-to-wall structural information. We aimed to (i) develop and evaluate a multisensor, AI-enabled fusion framework for landscape-scale AGB mapping, (ii) quantify the added value of seasonal optical data and photogrammetric canopy-height profiles, and (iii) interpret model drivers using explainable artificial intelligence (AI) to relate predictors to forest structure and composition. We mapped AGB across 10,500 km2 in southeastern North Carolina using wall-to-wall predictors from optical, radar, and photogrammetric sources. Forest Inventory and Analysis plot data (n = 305) were used to train and evaluate an ensemble of gradient-boosted tree models (CatBoost, LightGBM, XGBoost) and a neural network (RealMLP) via cross-validation. Model behavior was interpreted using feature importance and partial dependence analysis. Expanding Sentinel-2 temporal coverage from summer-only to four-season composites improved normalized RMSE by 8.7

Context:High-resolution land-cover maps are essential for ecological monitoring and landscape-level analysis, yet long-term high-resolution time-series products remain scarce. The National Agricultural Imagery Program (NAIP) provides highly valuable aerial imagery, but strong temporal heterogeneity across years, primarily caused by shifts in sensor characteristics, limits the ability to generate consistent multi-year land cover maps. Objectives:This study aims to (1) develop an algorithm capable of producing spatially detailed and temporally coherent 1-m land-cover maps using the state-of-the-art GeoAI/Machine Learning (ML) tools based on time-series NAIP imagery, and (2) address cross-year sensor characteristic shifts without relying on historical training samples. Methods:We designed a dual-track adaptive workflow that applies different strategies to NAIP imagery with different qualities. NAIP imagery collected during 2009-2017 has a higher quality than that collected during 2004-2008. Images from the high-quality years were classified using a foundation model pretrained with U-Net/ResNet-34 and refined with a Segment Anything Model (SAM) for accurate boundary delineation. Images collected in the earlier years were reconstructed using an NLCD-based spatiotemporal bridging and label back-casting pipeline. Accuracy was evaluated across six U.S. counties in North Carolina and Pennsylvania. Results:The algorithm produced stable results across years, yielding overall accuracies of 0.887 (2014), 0.886 (2017), and 0.733 (2004) with Kappa statistics at 0.860, 0.831, and 0.764, respectively. Structure, Water, Wetland, and Cropland exhibited consistently high F1-scores, while performance in low-quality imagery remained coherent despite substantial sensor differences. These findings demonstrate that the algorithm maintains both spatial fidelity and temporal consistency across heterogeneous historical imagery. Conclusions:This study shows using GeoAI/ML tools along with multiple sources of data can effectively produce consistent high-resolution multi-decade land-cover maps based on NAIP imagery. The approach developed in this study provides a scalable solution for generating high-resolution time series land-cover maps across the conterminous USA where NAIP imagery is available, supporting long-term land-change analyses and landscape-level planning.

Accurately monitoring forest canopy height is crucial for sustainable forest management, particularly in southeastern North Carolina, USA, where dense forests and limited accessibility pose substantial challenges. This study presents an explainable machine learning framework that integrates sparse GEDI LiDAR samples with multi-sensor remote sensing data to improve both the accuracy and interpretability of forest canopy height estimation. This framework incorporates multitemporal optical observations from Sentinel-2; C-band backscatter and InSAR coherence from Sentinel-1; quad-polarization L-Band backscatter and polarimetric decompositions from the Uninhabited Aerial Vehicle Synthetic Aperture Radar (UAVSAR); texture features from the National Agriculture Imagery Program (NAIP) aerial photography; and topographic data derived from an airborne LiDAR-based digital elevation model. We evaluated four machine learning algorithms, K-nearest neighbors (KNN), random forest (RF), support vector machine (SVM), and eXtreme gradient boosting (XGB), and found consistent accuracy across all models. Our evaluation highlights our method’s robustness, evidenced by closely matched R2 and RMSE values across models: KNN (R2 of 0.496, RMSE of 5.13 m), RF (R2 of 0.510, RMSE of 5.06 m), SVM (R2 of 0.544, RMSE of 4.88 m), and XGB (R2 of 0.548, RMSE of 4.85 m). The integration of comprehensive feature sets, as opposed to subsets, yielded better results, underscoring the value of using multisource remotely sensed data. Crucially, SHapley Additive exPlanations (SHAP) revealed the multi-seasonal red-edge spectral bands of Sentinel-2 as dominant predictors across models, while volume scattering from UAVSAR emerged as a key driver in tree-based algorithms. This study underscores the complementary nature of multi-sensor data and highlights the interpretability of our models. By offering spatially continuous, high-quality canopy height estimates, this cost-effective, data-driven approach advances large-scale forest management and environmental monitoring, paving the way for improved decision-making and conservation strategies.

Rising sea levels have driven widespread coastal tree die-off, forming large swaths of standing dead trees known as 'ghost forests'. While reports of coastal forest loss are accumulating, its true severity and factors determining the underlying mortality risks remain poorly understood. Here we mapped over 10 million individual dead trees across the US Atlantic region, a sea level rise hotspot, using deep learning and sub-metre aerial imagery. Our analysis reveals disproportionate and pervasive tree mortality in many previously unrecognized ghost forest hotspots, with over 6 million dead trees concentrated in low-lying (<5 m) forests, primarily driven by salinization rather than flooding alone. Notably, man-made flood obstructions (roads and levees), when present, have reduced forest loss by 40% and 79% respectively, but they can only provide temporary protection against rising seas. These findings highlight the alarming decline in coastal forests and the need for strategic long-term planning to preserve the critical services that coastal ecosystems provide.

Luminescent solar concentrators (LSCs) are large scale sunlight collector and can be used for building-integrated photovoltaics (BIPV). Achieving high-performance LSCs requires fluorophores with broad absorption, high quantum yield and a large Stokes shift. Nevertheless, conventional high-efficiency LSCs typically rely on heavy metal-based quantum dots as fluorophores. Herein, we demonstrated highly efficient LSCs by deeply understanding the polarity of boron-doped carbon nanodots (B-C-dots) in different solvents or polymers, such as polymethyl methacrylate (PMMA) or polystyrene (PS). The hydrothermally synthesized B-C-dots and undoped Cdots exhibit solvent dependent optical properties. Specifically, the emission peak shows a redshift when the Cdots are dispersed in ethanol, hexane and toluene, and the absorption edge shows a redshift with the increase of solvent polarity. For instance, the B-C-dots in toluene exhibit an absorption range of 300-600 nm and an emission peak of 605 nm, while in ethanol the absorption range extends to 300-650 nm and the emission peak is located at 438 nm. Once B-C-dots were dispersed in polymer matrices, the emission peak also shows various due to the different refractive index of polymers. As a proof-of-concept, we fabricated the LSCs (10 x 10 x 0.5 cm3) by selecting different types of polymers. We are able to achieve the LSCs with different colors and the optimized LSC based on B-C-dots/PS exhibits an optical efficiency of 5.48 % under natural sunlight illumination (69 mW/ cm2). Our results pave a simple way to tune the colors and performance of the LSCs to achieve low-cost, highly efficient BIPV applications.

Severe seasonal droughts driven by global climate change significantly alter the cycling of carbon and nutrients in forest ecosystems, while the investigation into the impacts of floor mass and plant roots on soil microbial biomass within the context of recurrent seasonal droughts is still rare. To investigate the environmental determinants governing soil microbial biomass with the escalating severity of seasonal droughts, we conducted a study in a montane subtropical moist evergreen broad-leaved forest in southwestern China from June 2019 to May 2023. The study results revealed that soil microbial biomass, as well as soil moisture, floor mass, and plant roots, showed an apparent single-hump modal within one year. In the comparative analysis of the soil microbial biomass fluctuation amplitudes across control and watered plots, a discernible disparity was observed, indicating significant differences in microbial biomass dynamics between the respective experimental conditions. The pooled data revealed a statistically significant influence of seasonal drought, floor mass, plant roots, and their reciprocal interactions on the soil microbial biomass, highlighting these factors as pivotal determinants of microbial community dynamics. This study elucidates the interactive regulatory mechanisms by which seasonal drought, floor mass, and plant roots collectively modulate soil microbial biomass within tropical and subtropical forests, offering insights into the complex ecological processes governing microbial community dynamics. This interactive regulation might influence the trajectory of plant species and soil microbial communities, facilitating their adaptive development and evolutionary responses.

Aqueous zinc-ion batteries (AZIBs) have emerged as highly promising options for large-scale energy storage systems due to their cost-effectiveness, substantial energy capacity, and improved safety features. However, the Zn anode faces challenges such as self-corrosion and dendrite formation, which limit its practical use in AZIB applications. In this work, a simple blade-coating method was used to successfully coat poly (vinylidene fluoride–hexafluoro propylene) (PVDF-HFP) on the Zn anode. The coated Zn anode (P-Zn) displayed a stable cycling performance (700 h) at 1 mA cm−2 current density in the symmetric cell. In addition, the full cell using MnO2 as the cathode and P-Zn as the anode retained almost full capacity even after 1400 cycles at 2C, far outperforming the full cell using the unmodified Zn anode with only 50% capacity retention after 600 cycles. In situ optical observations of Zn deposition demonstrate that the special organic coating significantly enhances the uniform deposition of Zn2+, thus effectively mitigating corrosion and hydrogen evolution. Density Functional Theory (DFT) calculations show that the PVDF-HFP coating effectively narrows the adsorption energy gap between the P-Zn (002) and (101) planes, leading to the homogeneous deposition of Zn2+ with fewer Zn dendrites. A simple and feasible strategy for designing ultra-stable AZIBs by coating an organic protective layer on the Zn surface is provided by this work.

Lakes provide important water resources and many essential ecosystem services. Some of Earth's largest lakes recently reached record-low levels, suggesting increasing threats from climate change and anthropogenic activities. Yet, continuous monitoring of lake levels is challenging at a global scale due to the sparse in situ gauging network and the limited spatial or temporal coverage of satellite altimeters. A few pioneering studies used water areas and hypsometric curves to reconstruct water levels but suffered from large uncertainties due to the lack of high-quality hypsometry data. Here, we propose a novel proxy-based method to reconstruct multi-decadal water levels from 1992 to 2018 for both large and small lakes using Landsat images and ICESat (2003-2009) and recently launched ICESat-2 (2018+) laser altimeters. Using the new method, we evaluate reconstructed levels of 342 lakes worldwide, with sizes ranging from 1 to 81,844 km2. Reconstructed water levels have a median root-mean-square error (RMSE) of 0.66 m, equivalent to 57% of the standard deviation of monthly level variability. Compared with two recently reconstructed water level data sets, the proposed method reduces the median RMSE by 27%-32%. The improvement is attributable to the new method's robust construction of high-quality hypsometry, with a median R2 value of 0.92. Most reconstructed water level time series have a bi-monthly or higher frequency. Given that ICESat-2 and Landsat can observe hundreds of thousands of water bodies, this method can be applied to conduct an improved global inventory of time-varying lake levels and thus inform water resource management more broadly than existing methods. Landsat images and laser altimeters were leveraged to reconstruct multi-decadal lake levels of both large and small lakes Reconstructed water levels were validated against observed levels on 342 global lakes with a median error of 0.66 m Most of the reconstructed lake level time series have a bi-monthly or higher frequency

Maize and wheat are two major staple foods that collectively contribute two-thirds of the world's grain supply. The extensive use of nitrogen (N) fertilizers during the cultivation of both crops leads to significant losses of reactive nitrogen (Nr) into the environment. Here, using machine learning algorithms, we generate high-resolution maps of crop-specific soil Nr losses based on global field measurements. We estimate that global annual soil Nr losses from the use of synthetic N fertilizer in 2020, including direct emissions of nitrous oxide (N2O), nitric oxide (NO), ammonia (NH3), N leaching and run-off, amount to 0.18, 1.62, 0.09, 1.47 and 1.10 million tonnes N for maize, and 0.12, 1.33, 0.07, 1.21 and 0.95 million tonnes N for wheat, respectively. The annual indirect N2O emissions induced by synthetic N fertilizer use from these soil Nr losses are estimated to be 45,000 and 37,000 tonnes for maize and wheat, respectively, with hydrologic pathways playing a predominant role. Enhancing N use efficiency up to 60% for regions below this value can achieve a total soil Nr loss mitigation potential of 4.00 million tonnes per year for the two crops, thereby reducing indirect N2O emissions by 49%. Our results contribute to constrain global N budgets from the use of fertilizer in agriculture, which then can help to improve projections of nitrogen cycle-climate feedbacks using modelling approaches.

Arctic-boreal wetlands, important ecosystems for biodiversity and ecological services, are experiencing hydrological changes including permafrost thaw, earlier snowmelt, and increased wildfire susceptibility. These changes are affecting wetland productivity, species diversity, and biogeochemical cycles. However, given the diverse forms and structures of wetland vegetation communities, traditional wetland maps generated from lower spatial and spectral resolution satellite imagery lack community-level vegetation classification and miss spatially complex patterns. In this study, we built a cloud-based workflow to map wetland vegetation community of the Peace-Athabasca Delta (PAD), Canada, by leveraging high-resolution (5-m) airborne multi-sensor datasets, namely NASA's Airborne Visible/Infrared Imaging Spectrometer-Next Generation (AVIRIS-NG) and Uninhabited Aerial Vehicle Synthetic Aperture Radar (UAVSAR), and a historical LiDAR archive. Validation of our classifications using ground references indicates that classifications derived from AVIRIS-NG have higher accuracies (≥87.9%) than either UAVSAR (65.6%) or LiDAR (75.9%) for mapping wetland vegetation communities. We also show improved classification accuracy when combining information from multiple sensors. In particular, incorporating AVIRIS-NG and UAVSAR datasets substantially reduced omission errors of wet graminoid and wet shrub classes from 29.6% to 20.5% and from 10.8% to 7.5%, respectively. Combining AVIRIS-NG and LiDAR datasets further improves overall accuracy (+2.2%) for most classifications, especially emergent vegetation, wet graminoid, and wet shrub. The best performing model, using features derived from all three sensors, achieved an overall accuracy of 93.5%. The framework established here can be used to leverage extensive airborne AVIRIS-NG and UAVSAR datasets collected across Alaska and northwest Canada to understand the spatial distribution of Arctic-Boreal wetland vegetation communities.

Redox-active functional groups in dissolved organic matter (DOM) are crucial for microbial electron transfer and methane emissions. However, the extent of aquatic DOM redox properties across northern high-latitude lakes and their relationships with DOM composition have not been thoroughly described. We quantified electron donating capacity (EDC) and electron accepting capacity (EAC) in lake DOM from Canada to Alaska and assessed their relationships with parameters from absorbance, fluorescence, and ultrahigh resolution mass spectrometry (FT-ICR MS) analyses. EDC and EAC are strongly tied to aromaticity and negatively related to aliphaticity and protein-like content. Redox-active formulae spanned a range of aromaticity, including highly unsaturated phenolic formulae, and correlated negatively with many aliphatic N and S-containing formulae. This distribution illustrates the compositional diversity of redox-sensitive functional groups and their sensitivity to ecosystem properties such as local hydrology and residence time. Finally, we developed a reducing index (RI) to predict EDC in aquatic DOM from FT-ICR MS spectra and assessed its robustness using riverine DOM. As the hydrology of the northern high-latitudes continues to change, we expect differences in the quantity and partitioning of EDC and EAC within these lakes, which have implications for local water quality and methane emissions.

The use of ground control points (GCPs) for georeferencing is the most common strategy in unmanned aerial vehicle (UAV) photogrammetry, but at the same time their collection represents the most time-consuming and expensive part of UAV campaigns. Recently, deep learning has been rapidly developed in the field of small object detection. In this letter, to automatically extract coordinates information of ground control points (GCPs) by detecting GCP-markers in UAV images, we propose a solution that uses a deep learning-based architecture, YOLOv5-OBB, combined with a confidence threshold filtering algorithm and an optimal ranking algorithm. We applied our proposed method to a dataset collected by DJI Phantom 4 Pro drone and obtained good detection performance with the mean Average Precision (AP) of 0.832 and the highest AP of 0.982 for the cross-type GCP-markers. The proposed method can be a promising tool for future implementation of the end-to-end aerial triangulation process.

Hydrogen is increasingly needed as a clean energy source with high energy density. Hydrogen evolution reaction (HER) via electrocatalysis is a viable way to produce clean hydrogen. The key to achieving active HER is to select highly-active electrocatalysts. Although Pt-based noble metal catalysts have excellent performance, they are expensive and earth-abundant. Compared to noble metals, cobalt chalcogenide has been used for efficient HER. Here, we first demonstrated a porous material based on C-dots/cobalt chalcogenide nanofibers as an effective electrocatalyst for HER. Co3O4/10 wt% C-dots nanofibers-based electrode reached an overpotential of 383 mV at the current density of 10 mA & BULL;cm  2 with Tafel slope of 93.4 mV & BULL;dec  1 in 1 M KOH, which is much better than bare Co3O4 (758 mV at 10 mA & BULL;cm  2, Tafel slope is 111.3 mV & BULL;dec  1). In addition to the excellent HER performance, the Co3O4/C-dots nanofibers performed very well in terms of stability after a 10-hour operation, and the overpotential was even reduced by 6.7%. With the further sulphuration, the Co3S4/C-dots nanofibers exhibited an overpotential of 280.2 mV at the current density of 10 mA & BULL;cm  2. This work indicates that the C-dots can be used as an efficient additive for cobalt chalcogenide based electrocatalysts.

Abstract In nearly all reservoirs, storage capacity is steadily lost due to trapping and accumulation of sediment. Despite critical importance to freshwater supplies, reservoir sedimentation rates are poorly understood due to sparse bathymetry survey data and challenges in modeling sedimentation sequestration. Here, we proposed a novel approach to estimate reservoir sedimentation rates and storage capacity losses using high‐resolution Sentinel‐2 satellites and daily in situ water levels. Validated on eight reservoirs across the central and western United States, the estimated reservoir bathymetry and sedimentation rates have a mean error of 4.08% and 0.05% yr−1, respectively. Estimated storage capacity losses to sediment vary among reservoirs, which overall agrees with the pattern from survey data. We also demonstrated the potential applications of the proposed approach to ungauged reservoirs by combining Sentinel‐2 with sub‐monthly water levels from recent satellite altimeters.

This paper reports the use of multi-spot melt strategy coupled with smaller layer thickness to additively manufacture Y2O3/γ-TiAl nanocomposite. In contrast to the hatch melt, the multi-spot melt strategy results in a lower fraction of γ and B2-phase, a smaller lamellar spacing of 208±72 nm with straight interface between α2/γ, and a more uniform distribution of nanoparticles with a finer size of 90±38 nm. Twins can form in the equiaxed γ grains and within γ lamellae; this applies to both the multi-spot and hatch melt samples. Twins within the γ lamellae can propagate across the twin interface but terminate at the γ/α2 interface. A good combination of 556±11 MPa (tensile strength) and 17.0±3.1% (ductility) at 800 °C with 16.5±0.3 MPa√m (room-temperature fracture toughness) is achieved in the as-built condition. Quantitative microscopy confirms a homogeneous microstructure within the X-Y plane for the multi-spot sample, whilst the use of smaller layer thickness helps to reduce the microstructure degradation due to thermal cycling. In terms of the Y2O3 nanoparticles, both the rod-like Y2O3 with monoclinic and the near spherical Y2O3 with the cubic crystal system are identified using transmission electron microscopy (TEM). High-resolution TEM reveals that the Y2O3/TiAl interface is clean and free of interfacial reactions, exhibiting either a semi-coherent or coherent type interface, suggesting a strong interfacial bonding.

Alkaline water electrolysis is a safe and efficient method for producing hydrogen that is favored by the industry. At present, single-atom catalysts have attracted extensive attention due to their extremely high atom utilization. However, studies on the single-atom site catalytic mechanism of the Volmer step, the rate-determining step of the cathodic HER reaction, are lacking. Here, we used ZIF-8 to obtain double-layer N-doped porous carbon nanocages, and successfully achieved the co-loading of Pt single atoms and their clusters (NPCN–Pt). The turnover frequency (TOF) and mass activity of NPCN–Pt are 5 and 7.2 times higher than those of commercial Pt/C, respectively, and the former exhibits significantly better stability than Pt/C. Theoretical calculations show that Pt clusters have a lower activation energy barrier for water molecules compared to Pt single atoms, which favors the Volmer step. The ΔGH* of Pt single-atom sites is smaller than that of Pt clusters, which is more conducive to the release of H2. The study shows that Pt clusters and single atoms can synergistically catalyze the HER reaction, providing new ideas for the development of efficient HER catalysts.

InP quantum dots (QDs) are promising building blocks for use in solar technologies because of their low intrinsic toxicity, narrow bandgap, large absorption coefficient, and low-cost solution synthesis. However, the high surface trap density of InP QDs reduces their energy conversion efficiency and degrades their long-term stability. Encapsulating InP QDs into a wider bandgap shell is desirable to eliminate surface traps and improve optoelectronic properties. Here, we report the synthesis of "giant" InP/ZnSe core/shell QDs with tunable ZnSe shell thickness to investigate the effect of the shell thickness on the optoelectronic properties and the photoelectrochemical (PEC) performance for hydrogen generation. The optical results demonstrate that ZnSe shell growth (0.9-2.8 nm) facilitates the delocalization of electrons and holes into the shell region. The ZnSe shell simultaneously acts as a passivation layer to protect the surface of InP QDs and as a spatial tunneling barrier to extract photoexcited electrons and holes. Thus, engineering the ZnSe shell thickness is crucial for the photoexcited electrons and hole transfer dynamics to tune the optoelectronic properties of "giant" InP/ZnSe core/shell QDs. We obtained an outstanding photocurrent density of 6.2 mA cm-1 for an optimal ZnSe shell thickness of 1.6 nm, which is 288% higher than the values achieved from bare InP QD-based PEC cells. Understanding the effect of shell thickness on surface passivation and carrier dynamics offers fundamental insights into the suitable design and realization of eco-friendly InP-based "giant" core/shell QDs toward improving device performance.

Northern high‐latitude lakes are hotspots for cycling dissolved organic carbon (DOC) inputs from allochthonous sources to the atmosphere. However, the spatial distribution of lake dissolved organic matter (DOM) is largely unknown across Arctic‐boreal regions with respect to the surrounding landscape. We expand on regional studies of northern high‐latitude DOM composition by integrating DOC concentrations, optical properties, and molecular‐level characterization from lakes spanning the Canadian Taiga to the Alaskan Tundra. Lakes were sampled during the summer from July to early September to capture the growing season. DOM became more optically processed and molecular‐level aromaticity increased northward across the Canadian Shield to the southern Arctic and from interior Alaska to the Tundra, suggesting relatively greater DOM incorporation from allochthonous sources. Using water isotopes (δ 18 O‐H 2 O), we report a weak overall trend of increasing DOC and decreasing aromaticity in lakes that were hydrologically isolated from the landscape and enriched in δ 18 O‐H 2 O, while within‐region trends were stronger and varied depending on the landscape. Finally, DOC correlated weakly with chromophoric dissolved organic matter (CDOM) across the study sites, suggesting that autochthonous and photobleached DOM were a major component of the DOC in these regions; however, some of the northernmost and wetland‐dominated lakes followed pan‐Arctic riverine DOC‐CDOM relationships, indicating strong contributions from allochthonous inputs. As many lakes across the North American Arctic are experiencing changes in temperature and precipitation, we expect the proportions of allochthonous and autochthonous DOM to respond with aquatic optical browning with greater landscape connectivity and more internally produced DOM in hydrologically isolated lakes.

Capacitive deionization (CDI) is an emerging water desalination technology for removing different ionic species from water, which is based on electric charge compensation by these charged species. CDI is becoming popular because it is more energy‐efficient and cost‐effective than other technologies, such as reverse osmosis and distillation, specifically in dealing with brackish water having low or moderate salt concentrations. Over the past decade, the CDI research field has witnessed significant advances in the used electrode materials, cell architectures, and associated mechanisms for desalination applications. This review article first discusses ion storage/removal mechanisms in carbon and Faradaic materials aided by advanced in situ analysis techniques and computations. It then summarizes research progress toward electrode materials in terms of structure, surface chemistry, and composition. More still, it discusses CDI cell architectures by highlighting their different cell design concepts. Finally, current challenges and future research directions are summarized to provide guidelines for future CDI research.