Search for: All records

Abstract Land surface temperature (LST) is crucial for understanding earth system processes. We expanded the Advanced Baseline Imager Live Imaging of Vegetated Ecosystems (ALIVE) framework to estimate LST in near‐real‐time for both cloudy and clear sky conditions at a five‐minute resolution. We compared two machine learning (ML) models, Long Short‐Term Memory (LSTM) networks and Gradient Boosting Regressor (GBR), using top‐of‐atmosphere observations from the Advanced Baseline Imager (ABI) on the GOES‐16 satellite against observations from hundreds of observation sites for a five‐year period. Long Short‐Term Memory outperformed GBR, especially at coarser resolutions and under challenging conditions, with a clear sky R²of 0.96 (RMSE 2.31K) and a cloudy sky R²of 0.83 (RMSE 4.10K) across CONUS, based on 10‐repeat Leave‐One‐Out Cross‐Validation (LOOCV). GBR maintained high accuracy and ran 5.3 times faster, with only a 0.01–0.02 R²drop. Feature importance revealed infrared bands were key in both models, with LSTM adapting dynamically to atmospheric changes, while GBR utilized more time information in cloudy conditions. A comparative analysis against the physically based ABI_LSTproduct showed strong agreement in winter, particularly under clear sky conditions, while also highlighting the challenges of summer LST estimation due to increased thermal variability. This study underscores the strengths and limitations of data‐driven models for LST estimation and suggests potential pathways for integrating ML models to enhance the accuracy and coverage of LST products. 

The terrestrial carbon cycle varies dynamically on hourly to weekly scales, making it difficult to observe. Geostationary (“weather”) satellites like the Geostationary Environmental Operational Satellite - R Series (GOES-R) deliver near-hemispheric imagery at a ten-minute cadence. The Advanced Baseline Imager (ABI) aboard GOES-R measures visible and near-infrared spectral bands that can be used to estimate land surface properties and carbon dioxide flux. However, GOES-R data are designed for real-time dissemination and are difficult to link with eddy covariance time series of land-atmosphere carbon dioxide exchange. We compiled three-year time series of GOES-R land surface attributes including visible and near-infrared reflectances, land surface temperature (LST), and downwelling shortwave radiation (DSR) at 314 ABI fixed grid pixels containing eddy covariance towers. We demonstrate how to best combine satellite andin-situdatasets and show how ABI attributes useful for ecosystem monitoring vary across space and time. By connecting observation networks that infer rapid changes to the carbon cycle, we can gain a richer understanding of the processes that control it. 

This study investigates high-frequency mapping of downward shortwave radiation (DSR) at the Earth’s surface using the advanced baseline imager (ABI) instrument mounted on Geo- stationary Operational Environmental Satellite—R Series (GOES- R). The existing GOES-R DSR product (DSRABI) offers hourly temporal resolution and spatial resolution of 0.25°. To enhance these resolutions, we explore machine learning (ML) for DSR estimation at the native temporal resolution of GOES-R Level-2 cloud and moisture imagery product (5 min) and its native spatial resolution of 2 km at nadir. We compared four common ML regres- sion models through the leave-one-out cross-validation algorithm for robust model assessment against ground measurements from AmeriFlux and SURFRAD networks. Results show that gradient boosting regression (GBR) achieves the best performance (R2 = 0.916, RMSE = 88.05 W·m−2) with more efficient computation compared to long short-term memory, which exhibited similar performance. DSR estimates from the GBR model through the ABI live imaging of vegetated ecosystems workflow (DSRALIVE) outperform DSRABI across various temporal resolutions and sky conditions. DSRALIVE agreement with ground measurements at SURFRAD networks exhibits high accuracy at high temporal res- olutions (5-min intervals) with R2 exceeding 0.85 and RMSE = 122 W·m−2 . We conclude that GBR offers a promising approach for high-frequency DSR mapping from GOES-R, enabling improved applications for near-real-time monitoring of terrestrial carbon and water fluxes. 

Abstract Restoring and preserving the world's forests are promising natural pathways to mitigate some aspects of climate change. In addition to regulating atmospheric carbon dioxide concentrations, forests modify surface and near‐surface air temperatures through biophysical processes. In the eastern United States (EUS), widespread reforestation during the 20th century coincided with an anomalous lack of warming, raising questions about reforestation's contribution to local cooling and climate mitigation. Using new cross‐scale approaches and multiple independent sources of data, we uncovered links between reforestation and the response of both surface and air temperature in the EUS. Ground‐ and satellite‐based observations showed that EUS forests cool the land surface by 1–2°C annually compared to nearby grasslands and croplands, with the strongest cooling effect during midday in the growing season, when cooling is 2–5°C. Young forests (20–40 years) have the strongest cooling effect on surface temperature. Surface cooling extends to the near‐surface air, with forests reducing midday air temperature by up to 1°C compared to nearby non‐forests. Analyses of historical land cover and air temperature trends showed that the cooling benefits of reforestation extend across the landscape. Locations surrounded by reforestation were up to 1°C cooler than neighboring locations that did not undergo land cover change, and areas dominated by regrowing forests were associated with cooling temperature trends in much of the EUS. Our work indicates reforestation contributed to the historically slow pace of warming in the EUS, underscoring reforestation's potential as a local climate adaptation strategy in temperate regions. 

Abstract Climate change is intensifying the hydrologic cycle and altering ecosystem function, including water flux to the atmosphere through evapotranspiration (ET). ET is made up of evaporation (E) via non‐stomatal surfaces, and transpiration (T) through plant stomata which are impacted by global changes in different ways. E and T are difficult to measure independently at the ecosystem scale, especially across multiple sites that represent different land use and land management strategies. To address this gap in understanding, we applied flux variance similarity (FVS) to quantify how E and T differ across 13 different ecosystems measured using eddy covariance in a 10 × 10 km area from the CHEESEHEAD19 experiment in northern Wisconsin, USA. The study sites included eight forests with a large deciduous broadleaf component, three evergreen needleleaf forests, and two wetlands. Average T/ET for the study period averaged nearly 52% in forested sites and 45% in wetlands, with larger values after excluding periods following rain events when evaporation from canopy interception may be expected. A dominance analysis revealed that environmental variables explained on average 69% of the variance of half‐hourly T, which decreased from summer to autumn. Deciduous and evergreen forests showed similar E trajectories over time despite differences in vegetation phenology, and vapor pressure deficit explained some 13% of the variance E in wetlands but only 5% or less in forests. Retrieval of E and T within a dense network of flux towers lends confidence that FVS is a promising approach for comparing ecosystem hydrology across multiple sites to improve our process‐based understanding of ecosystem water fluxes. 

Water use efficiency (WUE) is a critical ecosystem function and a key indicator of vegetation responses to drought, yet its temporal trajectories and underlying drivers during drought propagation remain insufficiently understood. Here, we examined the trajectories, interdependencies and drivers of multidimensional WUE metrics and their components (gross primary production (GPP), evapotranspiration, transpiration (T), and canopy conductance (Gc)) using a conceptual drought propagation framework. We found that even though the carbon assimilation efficiency per stomata increases during drought, the canopy‐level WUE (represented by transpiration WUE (TWUE)) declines, indicating that stomatal regulation operates primarily at the leaf level and cannot offset the drought‐induced reduction in WUE at the canopy scale. A stronger dependence on T and TWUE indicates that the water–carbon trade‐off relationship of vegetation more inclines toward water transport than carbon assimilation. Gc fails to prevent the sharp decline in GPP during drought and has limited capacity to suppress T, as reflected by the reduction magnitude and the threshold (the turning point at which a component shifts from a normal to drought‐responsive state). The primary drivers of the water–carbon relationship under drought propagation include vapor pressure deficit and hydraulic traits. Among plant functional types, grasslands show the strongest water–carbon fluxes in response to drought, whereas evergreen broadleaf forests exhibit the weakest response. These findings refine our comprehensive understanding of multidimensional ecosystem functional dynamics under drought propagation and enlighten how the physiological response of vegetation to drought affects the carbon and water cycles.