Artificial intelligence for geoscience: Progress, challenges and perspectives

The Innovation, Journal Year: 2024, Volume and Issue: 5(5), P. 100691 - 100691

Published: Aug. 23, 2024

Public summary•What does AI bring to geoscience? has been accelerating and deepening our understanding of Earth Systems in an unprecedented way, including the atmosphere, lithosphere, hydrosphere, cryosphere, biosphere, anthroposphere interactions between spheres.•What are noteworthy challenges As we embrace huge potential geoscience, several arise reliability interpretability, ethical issues, data security, high demand cost.•What is future The synergy traditional principles modern AI-driven techniques holds immense promise will shape trajectory geoscience upcoming years.AbstractThis paper explores evolution geoscientific inquiry, tracing progression from physics-based models data-driven approaches facilitated by significant advancements artificial intelligence (AI) collection techniques. Traditional models, which grounded physical numerical frameworks, provide robust explanations explicitly reconstructing underlying processes. However, their limitations comprehensively capturing Earth's complexities uncertainties pose optimization real-world applicability. In contrast, contemporary particularly those utilizing machine learning (ML) deep (DL), leverage extensive glean insights without requiring exhaustive theoretical knowledge. ML have shown addressing science-related questions. Nevertheless, such as scarcity, computational demands, privacy concerns, "black-box" nature hinder seamless integration into geoscience. methodologies hybrid presents alternative paradigm. These incorporate domain knowledge guide methodologies, demonstrate enhanced efficiency performance with reduced training requirements. This review provides a comprehensive overview research paradigms, emphasizing untapped opportunities at intersection advanced It examines major showcases advances large-scale discusses prospects that landscape outlines dynamic field ripe possibilities, poised unlock new understandings further advance exploration.Graphical abstract

Language: Английский

---

Rainfall‐Driven Extreme Snowmelt Will Increase in the Tianshan and Pamir Regions Under Future Climate Projection

Journal of Geophysical Research Atmospheres, Journal Year: 2025, Volume and Issue: 130(1)

Published: Jan. 2, 2025

Abstract Snowmelt and related extreme events can have profound natural societal impacts. However, the studies on projected changes in snow‐related extremes across Tianshan Mountains (TS) Pamir regions been underexplored. Utilizing regional climate model downscaling bias‐corrected CMIP6 data, this study examined snowmelt water available for runoff (SM ROS , rainfall plus snowmelt) during cold seasons these historical (1994–2014) future (2040–2060) periods under shared socioeconomic pathway (SSP) scenarios (SSP245 SSP585). The results demonstrated that accumulated was to rise by 17.98% 20.36%, whereas SM could increase 26.97% 28.95%, respectively, SSP245 SSP585 scenarios. Despite relatively minimal snowmelt, magnitude of daily maximum (10‐year return level) 28.04 mm expected 15.32% 15.31% scenarios, especially western TS exceeding 26%. Meanwhile, areas with a 50 over 13.5%. A notable its area occupation high intensity highlighted an increased risk rainfall‐driven events. absolute snowfall frequent snow‐rain phase transitions season warming (SSP245: 2.19°C SSP585: 2.22°C) benefits high‐intensity rain‐on‐snow events, leading augmentation. findings emphasize significant role rainfall‐trigger exacerbating climate.

Language: Английский

---

Interpretable machine learning on large samples for supporting runoff estimation in ungauged basins

Journal of Hydrology, Journal Year: 2024, Volume and Issue: 639, P. 131598 - 131598

Published: July 5, 2024

Language: Английский

---

Large-scale flood modeling and forecasting with FloodCast

Water Research, Journal Year: 2024, Volume and Issue: 264, P. 122162 - 122162

Published: July 27, 2024

Large-scale hydrodynamic models generally rely on fixed-resolution spatial grids and model parameters as well incurring a high computational cost. This limits their ability to accurately forecast flood crests issue time-critical hazard warnings. In this work, we build fast, stable, accurate, resolution-invariant, geometry-adaptive modeling forecasting framework that can perform at large scales, namely FloodCast. The comprises two main modules: multi-satellite observation modeling. the module, real-time unsupervised change detection method rainfall processing analysis tool are proposed harness full potential of observations in large-scale prediction. physics-informed neural solver (GeoPINS) is introduced, benefiting from absence requirement for training data networks (PINNs) featuring resolution-invariant architecture with Fourier operators. To adapt complex river geometries, reformulate PINNs space. GeoPINS demonstrates impressive performance popular partial differential equations across regular irregular domains. Building upon GeoPINS, propose sequence-to-sequence handle long-term temporal series extensive domains employs learning hard-encoding boundary conditions. Next, establish benchmark dataset 2022 Pakistan using widely accepted finite difference numerical solution assess various simulation methods. Finally, validate three dimensions - inundation range, depth, transferability spatiotemporal downscaling utilizing SAR-based data, traditional benchmarks, concurrent optical remote sensing images. Traditional hydrodynamics exhibit exceptional agreement during water levels, while comparative assessments depth show outperforms hydrodynamics, smaller errors. experimental results demonstrate enables high-precision, an average MAPE 14.93 % Mean Absolute Error (MAE) 0.0610 m 14-day simulations facilitating reliable precipitation data.

Language: Английский

---

A novel insight on input variable and time lag selection in daily streamflow forecasting using deep learning models

Environmental Modelling & Software, Journal Year: 2024, Volume and Issue: 179, P. 106126 - 106126

Published: June 25, 2024

Language: Английский

---

Advanced streamflow forecasting for Central European Rivers: The Cutting-Edge Kolmogorov-Arnold networks compared to Transformers

Journal of Hydrology, Journal Year: 2024, Volume and Issue: unknown, P. 132175 - 132175

Published: Oct. 1, 2024

Language: Английский

---

Climatic Controls on Hydrological Landslide Triggering in the Northern Himalayas

Water Resources Research, Journal Year: 2025, Volume and Issue: 61(2)

Published: Feb. 1, 2025

Abstract Hydrologically‐induced landslides are ubiquitous natural hazards in the Himalayas, posing severe threat to human life and infrastructure. Yet, landslide assessment Himalayas is extremely challenging partly due complex drastically changing climate conditions. Here we establish a mechanistic hydromechanical modeling framework that incorporates impacts of key water fluxes stocks on triggering risk evolution mountain systems, accounting for potential change conditions period 1991–2100. In drainage basin largest river northern Himalayas– Yarlung Zangbo River Basin (YZRB), estimate rainfall, glacier/snow melt permafrost thaw contribute ∼38.4%, 28.8%, 32.8% landslides, respectively, 1991–2019. Future will likely exacerbate primarily increasing whereas contribution decreases owing deglaciation snow cover loss. The total Gross Domestic Productivity projected increase continuously throughout 21st century, while population shows general declining trend. results yield novel insights into climatic controls provide useful guidance disaster management resilience building under future Himalayas.

Language: Английский

---

Research on machine learning hybrid framework by coupling grid-based runoff generation model and runoff process vectorization for flood forecasting

Journal of Environmental Management, Journal Year: 2024, Volume and Issue: 364, P. 121466 - 121466

Published: June 12, 2024

Language: Английский

---

Integrated risk analysis for urban flooding under changing climates

Results in Engineering, Journal Year: 2024, Volume and Issue: unknown, P. 103243 - 103243

Published: Oct. 1, 2024

Language: Английский

---

Artificial intelligence for life sciences: A comprehensive guide and future trends

The Innovation Life, Journal Year: 2024, Volume and Issue: unknown, P. 100105 - 100105

Published: Jan. 1, 2024

<p>Artificial intelligence has had a profound impact on life sciences. This review discusses the application, challenges, and future development directions of artificial in various branches sciences, including zoology, plant science, microbiology, biochemistry, molecular biology, cell developmental genetics, neuroscience, psychology, pharmacology, clinical medicine, biomaterials, ecology, environmental science. It elaborates important roles aspects such as behavior monitoring, population dynamic prediction, microorganism identification, disease detection. At same time, it points out challenges faced by application data quality, black-box problems, ethical concerns. The are prospected from technological innovation interdisciplinary cooperation. integration Bio-Technologies (BT) Information-Technologies (IT) will transform biomedical research into AI for Science paradigm.</p>

Language: Английский

---