1482 papers
This study presents an explainable machine learning framework using XGBoost and Temporal Attention LSTM models to forecast drought severity across six South Australian IBRA subregions with high accuracy, revealing that lagged SPI-12 and rainfall anomalies are dominant predictors and mapping the six-year propagation of the Millennium Drought from the Murray Scroll Belt to the Eyre Mallee.
This study demonstrates that concurrent heat and water deficits, rather than single hazards alone, independently reshape the phenology of China's staple crops and disrupt cropping-system turnaround windows, particularly in southern regions and after 2010, highlighting a critical pathway for climate adaptation policy.
This study reveals that seasonal rainfall predictability, particularly in the South Asian monsoon, is governed not solely by remote oceanic forcing like ENSO but significantly by the persistence and spatial organization of terrestrial memory reservoirs, where soil moisture acts as a primary driver that often mediates oceanic influences and enhances forecast skill when initialized.
This paper proposes a risk-responsive spatial planning framework for the Kullu Valley that integrates downscaled climate projections and hydrodynamic simulations to address the critical exposure of over 8,400 structures to future flood hazards caused by rapid urbanization and climate change, recommending specific adaptation measures such as a re-zoned river corridor buffer and nature-based slope stabilization.
This study compares woody plant assemblages in Seasonal Atlantic Forest remnants and 22–25-year-old restoration sites in southern Brazil, finding that while restoration sites show high floristic similarity driven by planted species, they remain distinct from remnants, highlighting the need to account for historical disturbance in reference selection and acknowledging that full ecological convergence is a long-term process.
This comprehensive critical review synthesizes 312 high-quality studies on machine learning and deep learning for ground deformation analytics, utilizing a novel tripartite framework and bibliometric analysis to identify research gaps in uncertainty quantification and advocate for physics-informed, interpretable models to enhance geohazard mitigation.
This study quantifies the blue water footprint of 25 Iranian hydropower plants from 2004 to 2021, revealing that evaporation losses in arid and semi-arid regions are significantly high and highly variable, thereby underscoring the urgent need for improved reservoir management and climate-integrated strategies to ensure sustainable energy-water interactions.
This paper introduces an object-centered scanning and meshing protocol using industrial-grade laser scanners to generate high-precision 3D digital twins of individual wildland fuel particles, establishing a foundational framework for improving fire behavior, emissions, and ecological modeling through detailed structural characterization.
This study elucidates the microscopic mechanisms and main controlling factors of water huff-n-puff in tight reservoirs under salinity influence, revealing that low-salinity injection enhances oil recovery by synergistically leveraging pressure differential, capillary imbibition, and ion osmosis to mobilize residual oil from low-permeability zones.
Using two decades of satellite data, this study reveals that structural overshoot—characterized by excessive canopy growth followed by stress-induced declines—is intensifying in boreal forests, particularly in the south, thereby eroding ecosystem resilience and undermining long-term carbon uptake potential.