Earth & Environmental Sciences

48 items

Frontiers in Marine Science Sep 28, 2026
Using a threshold of nLw555 = 1.5 mW cm-2 µm-1 sr-1 from MODIS to separate the East China Sea into Clear Water (CW) and Turbid Water (TW), this study reconstructed surface pCO2 for 2003–2023 with a CatBoost model that treats water type as a categorical input alongside multi-spectral satellite variables, achieving independent-test accuracy of R2 = 0.86 and RMSE = 16.55 µatm and cutting TW RMSE by up to about 81% relative to conventional models; the 21-year climatology gives CW 361.2 µatm and TW 415.1 µatm, with TW acting as a weak net source (+0.50 Tg C yr-1) driven by positive non-thermal anomalies (+28.09 µatm), implying conventional models overestimate total ECS carbon uptake by roughly 46% (about 4.04 Tg C yr-1).
Using a threshold of nLw555 = 1.5 mW cm-2 µm-1 sr-1 from MODIS to separate the East China Sea into Clear Water (CW) and Turbid Water (TW), this study reconstructed surface pCO2 for 2003–2023 with a CatBoost model that treats water type as a categorical input alongside multi-spectral satellite variables, achieving independent-test accuracy of R2 = 0.86 and RMSE = 16.55 µatm and cutting TW RMSE by up to about 81% relative to conventional models; the 21-year climatology gives CW 361.2 µatm and TW 415.1 µatm, with TW acting as a weak net source (+0.50 Tg C yr-1) driven by positive non-thermal anomalies (+28.09 µatm), implying conventional models overestimate total ECS carbon uptake by roughly 46% (about 4.04 Tg C yr-1).
Splitting the East China Sea into clear and turbid regimes by nLw555, a CatBoost reconstruction finds the turbid zone is a weak CO2 source and that conventional models overestimate the ECS sink by about 46%
Using a threshold of nLw555 = 1.5 mW cm-2 µm-1 sr-1 from MODIS to separate the East China Sea into Clear Water (CW) and Turbid Water (TW), this study reconstructed surface pCO2 for 2003–2023 with a CatBoost model that treats water type as a categorical input alongside multi-spectral satellite variables, achieving independent-test accuracy of R2 = 0.86 and RMSE = 16.55 µatm and cutting TW RMSE by up to about 81% relative to conventional models; the 21-year climatology gives CW 361.2 µatm and TW 415.1 µatm, with TW acting as a weak net source (+0.50 Tg C yr-1) driven by positive non-thermal anomalies (+28.09 µatm), implying conventional models overestimate total ECS carbon uptake by roughly 46% (about 4.04 Tg C yr-1).
Using a threshold of nLw555 = 1.5 mW cm-2 µm-1 sr-1 from MODIS to separate the East China Sea into Clear Water (CW) and Turbid Water (TW), this study reconstructed surface pCO2 for 2003–2023 with a CatBoost model that treats water type as a categorical input alongside multi-spectral satellite variables, achieving independent-test accuracy of R2 = 0.86 and RMSE = 16.55 µatm and cutting TW RMSE by up to about 81% relative to conventional models; the 21-year climatology gives CW 361.2 µatm and TW 415.1 µatm, with TW acting as a weak net source (+0.50 Tg C yr-1) driven by positive non-thermal anomalies (+28.09 µatm), implying conventional models overestimate total ECS carbon uptake by roughly 46% (about 4.04 Tg C yr-1).
International Journal of Data Science and Analytics Sep 28, 2026
The study reformulates vineyard mildew risk prediction as an event-onset warning task—after a minimum disease-free gap, will a new treatment-associated risk event begin within the following 3–7 days?—and, under a chronological 2020–2021/2022/2023 train-validation-test split, compares calendar-only, environmental-only, and combined representations with logistic regression, XGBoost, LSTM, and TCN, finding that calendar-only models were highly competitive (calendar-only LSTM and TCN detected 7 of 8 test events in every run, while a monthly climatological baseline detected 6), that environmental-only models were substantially weaker, that gains from adding environmental variables varied across model families, and that event-onset prediction and conventional daily-status classification show dif
The study reformulates vineyard mildew risk prediction as an event-onset warning task—after a minimum disease-free gap, will a new treatment-associated risk event begin within the following 3–7 days?—and, under a chronological 2020–2021/2022/2023 train-validation-test split, compares calendar-only, environmental-only, and combined representations with logistic regression, XGBoost, LSTM, and TCN, finding that calendar-only models were highly competitive (calendar-only LSTM and TCN detected 7 of 8 test events in every run, while a monthly climatological baseline detected 6), that environmental-only models were substantially weaker, that gains from adding environmental variables varied across model families, and that event-onset prediction and conventional daily-status classification show dif
Calendar-only LSTM and TCN detected 7 of 8 vineyard mildew risk events in 2023, while environmental-only models were substantially weaker
The study reformulates vineyard mildew risk prediction as an event-onset warning task—after a minimum disease-free gap, will a new treatment-associated risk event begin within the following 3–7 days?—and, under a chronological 2020–2021/2022/2023 train-validation-test split, compares calendar-only, environmental-only, and combined representations with logistic regression, XGBoost, LSTM, and TCN, finding that calendar-only models were highly competitive (calendar-only LSTM and TCN detected 7 of 8 test events in every run, while a monthly climatological baseline detected 6), that environmental-only models were substantially weaker, that gains from adding environmental variables varied across model families, and that event-onset prediction and conventional daily-status classification show dif
The study reformulates vineyard mildew risk prediction as an event-onset warning task—after a minimum disease-free gap, will a new treatment-associated risk event begin within the following 3–7 days?—and, under a chronological 2020–2021/2022/2023 train-validation-test split, compares calendar-only, environmental-only, and combined representations with logistic regression, XGBoost, LSTM, and TCN, finding that calendar-only models were highly competitive (calendar-only LSTM and TCN detected 7 of 8 test events in every run, while a monthly climatological baseline detected 6), that environmental-only models were substantially weaker, that gains from adding environmental variables varied across model families, and that event-onset prediction and conventional daily-status classification show dif
arXiv Sep 28, 2026
The authors introduce ARGUS, a bounded retrieval-gated large-language-model pipeline that decomposes a difference-in-differences (DID) study into eleven assumption-implication-evidence dimensions, audits whether the evidence a paper reports adequately supports each identification assumption, and abstains when no relevant evidence can be retrieved; it detects 8 of 11 planted flaws (versus 2 for a keyword baseline), 25 of 33 flaw variants, abstains on roughly 60% of paper-dimension assessments over 26 economics papers for lack of retrievable evidence, and in a five-paper, 55-cell pilot with two reconciled annotators is more severe than the human labels on 25 of the 33 cells it completes, an over-severity a rule fixed before the labels arrived reduces substantially in-sample.
The authors introduce ARGUS, a bounded retrieval-gated large-language-model pipeline that decomposes a difference-in-differences (DID) study into eleven assumption-implication-evidence dimensions, audits whether the evidence a paper reports adequately supports each identification assumption, and abstains when no relevant evidence can be retrieved; it detects 8 of 11 planted flaws (versus 2 for a keyword baseline), 25 of 33 flaw variants, abstains on roughly 60% of paper-dimension assessments over 26 economics papers for lack of retrievable evidence, and in a five-paper, 55-cell pilot with two reconciled annotators is more severe than the human labels on 25 of the 33 cells it completes, an over-severity a rule fixed before the labels arrived reduces substantially in-sample.
ARGUS audits identification assumptions in climate-policy DID studies, detecting 8 of 11 injected flaws and abstaining on about 60% of paper-dimension assessments across 26 economics papers
The authors introduce ARGUS, a bounded retrieval-gated large-language-model pipeline that decomposes a difference-in-differences (DID) study into eleven assumption-implication-evidence dimensions, audits whether the evidence a paper reports adequately supports each identification assumption, and abstains when no relevant evidence can be retrieved; it detects 8 of 11 planted flaws (versus 2 for a keyword baseline), 25 of 33 flaw variants, abstains on roughly 60% of paper-dimension assessments over 26 economics papers for lack of retrievable evidence, and in a five-paper, 55-cell pilot with two reconciled annotators is more severe than the human labels on 25 of the 33 cells it completes, an over-severity a rule fixed before the labels arrived reduces substantially in-sample.
The authors introduce ARGUS, a bounded retrieval-gated large-language-model pipeline that decomposes a difference-in-differences (DID) study into eleven assumption-implication-evidence dimensions, audits whether the evidence a paper reports adequately supports each identification assumption, and abstains when no relevant evidence can be retrieved; it detects 8 of 11 planted flaws (versus 2 for a keyword baseline), 25 of 33 flaw variants, abstains on roughly 60% of paper-dimension assessments over 26 economics papers for lack of retrievable evidence, and in a five-paper, 55-cell pilot with two reconciled annotators is more severe than the human labels on 25 of the 33 cells it completes, an over-severity a rule fixed before the labels arrived reduces substantially in-sample.
arXiv Sep 28, 2026
The study adapts pretrained Stable Diffusion 3 into an image-only generative backbone with a pruned text stream and patch-wise normalization, conditioning on both photogrammetric DSMs and Pléiades-HR imagery via two ControlNets to refine vertically co-registered DSMs, reducing Dense Urban RMSE from 6.00 m to 3.45 m across eight in-context French cities and from 4.16 m to 2.77 m in the geographically held-out city of Bordeaux.
The study adapts pretrained Stable Diffusion 3 into an image-only generative backbone with a pruned text stream and patch-wise normalization, conditioning on both photogrammetric DSMs and Pléiades-HR imagery via two ControlNets to refine vertically co-registered DSMs, reducing Dense Urban RMSE from 6.00 m to 3.45 m across eight in-context French cities and from 4.16 m to 2.77 m in the geographically held-out city of Bordeaux.
Refining photogrammetric DSMs with a pretrained diffusion model and multimodal conditioning cut Dense Urban RMSE from 6.00 m to 3.45 m in French cities
The study adapts pretrained Stable Diffusion 3 into an image-only generative backbone with a pruned text stream and patch-wise normalization, conditioning on both photogrammetric DSMs and Pléiades-HR imagery via two ControlNets to refine vertically co-registered DSMs, reducing Dense Urban RMSE from 6.00 m to 3.45 m across eight in-context French cities and from 4.16 m to 2.77 m in the geographically held-out city of Bordeaux.
The study adapts pretrained Stable Diffusion 3 into an image-only generative backbone with a pruned text stream and patch-wise normalization, conditioning on both photogrammetric DSMs and Pléiades-HR imagery via two ControlNets to refine vertically co-registered DSMs, reducing Dense Urban RMSE from 6.00 m to 3.45 m across eight in-context French cities and from 4.16 m to 2.77 m in the geographically held-out city of Bordeaux.
JURNAL PESISIR DAN LAUT TROPIS Sep 27, 2026
Using MAWS observations, this study evaluated the bias in BMKG Ina-Flows sea surface temperature forecasts and compared three machine-learning bias-correction algorithms (SVR, LSTM, Bi-LSTM) at Karimun Jawa and Bira, finding a warm bias at both sites, with LSTM best at Karimun Jawa (RMSE 0.207, MAE 0.182, MBE -0.171, a 29.83% RMSE reduction) and Bi-LSTM best at Bira (RMSE 0.218, MAE 0.173, MBE 0.011, a 26.10% RMSE reduction), indicating that correction effectiveness is location-dependent and algorithm choice should rest on local validation.
Using MAWS observations, this study evaluated the bias in BMKG Ina-Flows sea surface temperature forecasts and compared three machine-learning bias-correction algorithms (SVR, LSTM, Bi-LSTM) at Karimun Jawa and Bira, finding a warm bias at both sites, with LSTM best at Karimun Jawa (RMSE 0.207, MAE 0.182, MBE -0.171, a 29.83% RMSE reduction) and Bi-LSTM best at Bira (RMSE 0.218, MAE 0.173, MBE 0.011, a 26.10% RMSE reduction), indicating that correction effectiveness is location-dependent and algorithm choice should rest on local validation.
Machine-learning bias correction cuts Ina-Flows SST forecast error by 29.83% at Karimun Jawa and 26.10% at Bira, with the best algorithm differing by site
Using MAWS observations, this study evaluated the bias in BMKG Ina-Flows sea surface temperature forecasts and compared three machine-learning bias-correction algorithms (SVR, LSTM, Bi-LSTM) at Karimun Jawa and Bira, finding a warm bias at both sites, with LSTM best at Karimun Jawa (RMSE 0.207, MAE 0.182, MBE -0.171, a 29.83% RMSE reduction) and Bi-LSTM best at Bira (RMSE 0.218, MAE 0.173, MBE 0.011, a 26.10% RMSE reduction), indicating that correction effectiveness is location-dependent and algorithm choice should rest on local validation.
Using MAWS observations, this study evaluated the bias in BMKG Ina-Flows sea surface temperature forecasts and compared three machine-learning bias-correction algorithms (SVR, LSTM, Bi-LSTM) at Karimun Jawa and Bira, finding a warm bias at both sites, with LSTM best at Karimun Jawa (RMSE 0.207, MAE 0.182, MBE -0.171, a 29.83% RMSE reduction) and Bi-LSTM best at Bira (RMSE 0.218, MAE 0.173, MBE 0.011, a 26.10% RMSE reduction), indicating that correction effectiveness is location-dependent and algorithm choice should rest on local validation.
arXiv Sep 26, 2026
The authors retrained AIFS, ECMWF's open-source operational 0.25-degree probabilistic graph-transformer weather model, on satellite-based precipitation observations to produce Laxmi, which improves global probabilistic accuracy by 19%, cuts drizzle overprediction by 33% for amounts below 3 mm per day, raises the global 95th percentile Brier skill score by 57%, and delivers the most accurate 150 mm event-total precipitation forecast in 7 of 10 Indian tropical storms (versus 1 for AIFS and 2 for the leading physical model IFS).
The authors retrained AIFS, ECMWF's open-source operational 0.25-degree probabilistic graph-transformer weather model, on satellite-based precipitation observations to produce Laxmi, which improves global probabilistic accuracy by 19%, cuts drizzle overprediction by 33% for amounts below 3 mm per day, raises the global 95th percentile Brier skill score by 57%, and delivers the most accurate 150 mm event-total precipitation forecast in 7 of 10 Indian tropical storms (versus 1 for AIFS and 2 for the leading physical model IFS).
Retraining AIFS on satellite precipitation observations yields Laxmi, which lifts global probabilistic accuracy by 19% and gives the most accurate 150 mm event-total forecast in 7 of 10 Indian tropical storms
The authors retrained AIFS, ECMWF's open-source operational 0.25-degree probabilistic graph-transformer weather model, on satellite-based precipitation observations to produce Laxmi, which improves global probabilistic accuracy by 19%, cuts drizzle overprediction by 33% for amounts below 3 mm per day, raises the global 95th percentile Brier skill score by 57%, and delivers the most accurate 150 mm event-total precipitation forecast in 7 of 10 Indian tropical storms (versus 1 for AIFS and 2 for the leading physical model IFS).
The authors retrained AIFS, ECMWF's open-source operational 0.25-degree probabilistic graph-transformer weather model, on satellite-based precipitation observations to produce Laxmi, which improves global probabilistic accuracy by 19%, cuts drizzle overprediction by 33% for amounts below 3 mm per day, raises the global 95th percentile Brier skill score by 57%, and delivers the most accurate 150 mm event-total precipitation forecast in 7 of 10 Indian tropical storms (versus 1 for AIFS and 2 for the leading physical model IFS).
Communications Earth & Environment Sep 26, 2026
Using satellite observations of solar-induced fluorescence together with model estimates of water table depth and aridity, and applying causality-guided explainable machine learning, this study quantified the relative roles of groundwater and climatic aridity in shaping the spatial pattern of photosynthesis across the contiguous United States, finding that groundwater's relative importance equals 48% to 101% of aridity's effect on forest photosynthesis, 30% to 58% in savannahs and shrublands, 22% to 42% in grasslands, and 15% to 32% in croplands.
Using satellite observations of solar-induced fluorescence together with model estimates of water table depth and aridity, and applying causality-guided explainable machine learning, this study quantified the relative roles of groundwater and climatic aridity in shaping the spatial pattern of photosynthesis across the contiguous United States, finding that groundwater's relative importance equals 48% to 101% of aridity's effect on forest photosynthesis, 30% to 58% in savannahs and shrublands, 22% to 42% in grasslands, and 15% to 32% in croplands.
Causality-guided explainable machine learning shows groundwater accounts for 48% to 101% of aridity's effect on forest photosynthesis across the contiguous United States
Using satellite observations of solar-induced fluorescence together with model estimates of water table depth and aridity, and applying causality-guided explainable machine learning, this study quantified the relative roles of groundwater and climatic aridity in shaping the spatial pattern of photosynthesis across the contiguous United States, finding that groundwater's relative importance equals 48% to 101% of aridity's effect on forest photosynthesis, 30% to 58% in savannahs and shrublands, 22% to 42% in grasslands, and 15% to 32% in croplands.
Using satellite observations of solar-induced fluorescence together with model estimates of water table depth and aridity, and applying causality-guided explainable machine learning, this study quantified the relative roles of groundwater and climatic aridity in shaping the spatial pattern of photosynthesis across the contiguous United States, finding that groundwater's relative importance equals 48% to 101% of aridity's effect on forest photosynthesis, 30% to 58% in savannahs and shrublands, 22% to 42% in grasslands, and 15% to 32% in croplands.
Mining of Mineral Deposits Sep 26, 2026
In the Wadi Umm Eish El-Zarqa area of Egypt, this study for the first time combines machine learning with EnMAP hyperspectral data, training Random Forest (RF) and Support Vector Machine with Radial Basis Function (SVM-RBF) models on pixel spectra from three known gold mining sites and expanding the training set with a 0.05 spectral probability tolerance, while using ALOS-PALSAR DEM to extract drainage networks linking upstream bedrock sources to downstream placers; ground-truth validation showed SVM-RBF had slightly higher overall accuracy (91.8% vs 89.
In the Wadi Umm Eish El-Zarqa area of Egypt, this study for the first time combines machine learning with EnMAP hyperspectral data, training Random Forest (RF) and Support Vector Machine with Radial Basis Function (SVM-RBF) models on pixel spectra from three known gold mining sites and expanding the training set with a 0.05 spectral probability tolerance, while using ALOS-PALSAR DEM to extract drainage networks linking upstream bedrock sources to downstream placers; ground-truth validation showed SVM-RBF had slightly higher overall accuracy (91.8% vs 89.
Random Forest and SVM-RBF target gold at Wadi Umm Eish El-Zarqa, Egypt using EnMAP hyperspectral data, with RF achieving perfect precision and 37% more high-confidence targets
In the Wadi Umm Eish El-Zarqa area of Egypt, this study for the first time combines machine learning with EnMAP hyperspectral data, training Random Forest (RF) and Support Vector Machine with Radial Basis Function (SVM-RBF) models on pixel spectra from three known gold mining sites and expanding the training set with a 0.05 spectral probability tolerance, while using ALOS-PALSAR DEM to extract drainage networks linking upstream bedrock sources to downstream placers; ground-truth validation showed SVM-RBF had slightly higher overall accuracy (91.8% vs 89.
In the Wadi Umm Eish El-Zarqa area of Egypt, this study for the first time combines machine learning with EnMAP hyperspectral data, training Random Forest (RF) and Support Vector Machine with Radial Basis Function (SVM-RBF) models on pixel spectra from three known gold mining sites and expanding the training set with a 0.05 spectral probability tolerance, while using ALOS-PALSAR DEM to extract drainage networks linking upstream bedrock sources to downstream placers; ground-truth validation showed SVM-RBF had slightly higher overall accuracy (91.8% vs 89.
GIScience & Remote Sensing Sep 26, 2026
The study proposes an interactive evaluation framework in which foundation model agents incrementally explore partially observable 20×20 grid-based symbolic maps of roads, intersections, and POIs, then probes spatial understanding with direction judgment, distance estimation, proximity judgment, POI density recognition, and path planning; by systematically varying exploration strategies, memory representations, and reasoning prompts, it finds that exploration has limited impact on final reasoning accuracy, that memory representation (especially node-sequence and graph memory) is central, that structured memory and advanced prompts repair reasoning failures through explicit spatial reconstruction, and that spatial reasoning performance saturates across model versions and scales beyond a cap
The study proposes an interactive evaluation framework in which foundation model agents incrementally explore partially observable 20×20 grid-based symbolic maps of roads, intersections, and POIs, then probes spatial understanding with direction judgment, distance estimation, proximity judgment, POI density recognition, and path planning; by systematically varying exploration strategies, memory representations, and reasoning prompts, it finds that exploration has limited impact on final reasoning accuracy, that memory representation (especially node-sequence and graph memory) is central, that structured memory and advanced prompts repair reasoning failures through explicit spatial reconstruction, and that spatial reasoning performance saturates across model versions and scales beyond a cap
After step-by-step exploration of 20×20 symbolic maps, node-sequence memory lifts GPT-5.2 total accuracy from 43.89% to 77.78%, while further model versions and parameter scale add little spatial reasoning
The study proposes an interactive evaluation framework in which foundation model agents incrementally explore partially observable 20×20 grid-based symbolic maps of roads, intersections, and POIs, then probes spatial understanding with direction judgment, distance estimation, proximity judgment, POI density recognition, and path planning; by systematically varying exploration strategies, memory representations, and reasoning prompts, it finds that exploration has limited impact on final reasoning accuracy, that memory representation (especially node-sequence and graph memory) is central, that structured memory and advanced prompts repair reasoning failures through explicit spatial reconstruction, and that spatial reasoning performance saturates across model versions and scales beyond a cap
The study proposes an interactive evaluation framework in which foundation model agents incrementally explore partially observable 20×20 grid-based symbolic maps of roads, intersections, and POIs, then probes spatial understanding with direction judgment, distance estimation, proximity judgment, POI density recognition, and path planning; by systematically varying exploration strategies, memory representations, and reasoning prompts, it finds that exploration has limited impact on final reasoning accuracy, that memory representation (especially node-sequence and graph memory) is central, that structured memory and advanced prompts repair reasoning failures through explicit spatial reconstruction, and that spatial reasoning performance saturates across model versions and scales beyond a cap
The University of Ha’il Journal of Science (UOHJS) Sep 26, 2026
Using a qualitative literature review plus multi-case comparison of three operating projects above 100 MW — Hornsdale Power Reserve (100 MW/129 MWh), Gateway Energy Storage (250 MW) and Victorian Big Battery (300 MW) — the study analyzes how battery energy storage systems (BESS) provide frequency regulation, peak shaving, renewable firming, voltage support and black start in smart grids, and identifies drivers such as public–private partnerships, FCAS and arbitrage revenue and green-bank financing, alongside barriers such as thermal-runaway fires, cooling-system failure, siting disputes, missing regulatory standards and high initial capital cost.
Using a qualitative literature review plus multi-case comparison of three operating projects above 100 MW — Hornsdale Power Reserve (100 MW/129 MWh), Gateway Energy Storage (250 MW) and Victorian Big Battery (300 MW) — the study analyzes how battery energy storage systems (BESS) provide frequency regulation, peak shaving, renewable firming, voltage support and black start in smart grids, and identifies drivers such as public–private partnerships, FCAS and arbitrage revenue and green-bank financing, alongside barriers such as thermal-runaway fires, cooling-system failure, siting disputes, missing regulatory standards and high initial capital cost.
Three 100 MW-class battery storage case studies show frequency-regulation revenue and policy financing drive deployment, while thermal-runaway fires and missing standards hold it back
Using a qualitative literature review plus multi-case comparison of three operating projects above 100 MW — Hornsdale Power Reserve (100 MW/129 MWh), Gateway Energy Storage (250 MW) and Victorian Big Battery (300 MW) — the study analyzes how battery energy storage systems (BESS) provide frequency regulation, peak shaving, renewable firming, voltage support and black start in smart grids, and identifies drivers such as public–private partnerships, FCAS and arbitrage revenue and green-bank financing, alongside barriers such as thermal-runaway fires, cooling-system failure, siting disputes, missing regulatory standards and high initial capital cost.
Using a qualitative literature review plus multi-case comparison of three operating projects above 100 MW — Hornsdale Power Reserve (100 MW/129 MWh), Gateway Energy Storage (250 MW) and Victorian Big Battery (300 MW) — the study analyzes how battery energy storage systems (BESS) provide frequency regulation, peak shaving, renewable firming, voltage support and black start in smart grids, and identifies drivers such as public–private partnerships, FCAS and arbitrage revenue and green-bank financing, alongside barriers such as thermal-runaway fires, cooling-system failure, siting disputes, missing regulatory standards and high initial capital cost.

Page 3 · showing 10