2353 papers
This paper proposes STICEnet, a U-Net-based deep learning model that integrates a Temporal Encoder-Decoder Coordination mechanism with multi-scale spatiotemporal feature encoding and hybrid attention to significantly improve the accuracy of Arctic sea ice concentration forecasting across 1-to-6-month lead times.
This study constructs a future ecological security pattern for the Mu Us Sandy Land by integrating InVEST and groundwater vulnerability assessments with PLUS-based multi-scenario land use simulations for 2030, identifying key ecological sources, corridors, and nodes to guide balanced development and restoration in the Yellow River Basin.
This study utilizes an integrated geospatial and statistical framework to demonstrate that while climate variability (specifically warming and rainfall fluctuations) significantly influences land-use dynamics in Mangochi District, Malawi, the observed substantial expansion of wetland cultivation and contraction of natural ecosystems between 2000 and 2020 are driven by complex, spatially varying interactions where climate acts as a partial, rather than sole, explanatory factor.
This bibliometric synthesis of 18,582 documents from 2005 to 2025 reveals the rapid global expansion of ecological restoration research, highlighting China's quantitative dominance and Brazil's qualitative leadership in tropical ecosystems while identifying key emerging hotspots and calling for strengthened North-South collaboration to align scientific efforts with global climate and biodiversity goals.
This study presents a novel satellite-based framework integrating Sentinel-2 imagery and SWOT observations to reconstruct the three-dimensional evolution of the July 2025 Merzbacher Lake outburst flood, enabling the quantitative estimation of a drained volume of approximately 9.84 million cubic meters and demonstrating a shift from two-dimensional detection to comprehensive stage–area–volume assessment of glacial lake outburst floods.
This study integrates GIS, remote sensing, and machine learning algorithms to map pollution vulnerability in South Africa's Kruger National Park, identifying atmospheric pollutants as key drivers and providing a spatial decision-support framework to advance sustainable ecosystem management and related Sustainable Development Goals.
By analyzing the moist static energy budget, this study reveals that the Mesoamerican Midsummer Drought is driven not by increased atmospheric stability, but by a seasonal shift in circulation efficiency and a specific balance where increased horizontal moisture export is offset by rising surface turbulent fluxes, allowing the atmospheric column to gain energy even as precipitation temporarily declines.
This paper demonstrates that an advection–reaction–diffusion model accurately reproduces the band formation patterns observed in agates, as evidenced by the strong agreement between experimental results and model calculations.
This study reveals that standard gravimetric measurements of atmospheric particulate matter are compromised by a U-shaped variability pattern and significant systematic positive bias caused by strongly bound filter water that survives conditioning, with PM₁ fractions at industrial sites being the most affected, thereby necessitating routine Karl Fischer water analysis to ensure accuracy.
This study demonstrates that local ecological knowledge (LEK) among small-scale fishers in southeastern Brazil is a quantifiable attribute that varies significantly based on experience and species knowledge, playing a fundamental role in enhancing the community's adaptive capacity to socio-environmental changes.