An Overview of Machine Learning and Deep Learning Applications in Earth Sciences in 2024: Achievements and Perspectives DOI
Mikhail Krinitskiy

Moscow University Physics Bulletin, Год журнала: 2024, Номер 79(S2), С. S739 - S749

Опубликована: Дек. 1, 2024

Язык: Английский

Predicting urban Heat Island in European cities: A comparative study of GRU, DNN, and ANN models using urban morphological variables DOI Creative Commons
Alireza Attarhay Tehrani, Omid Veisi,

Kambiz kia

и другие.

Urban Climate, Год журнала: 2024, Номер 56, С. 102061 - 102061

Опубликована: Июль 1, 2024

Continued urbanization, along with anthropogenic global warming, has and will increase land surface temperature air anomalies in urban areas when compared to their rural surroundings, leading Urban Heat Islands (UHI). UHI poses environmental health risks, affecting both psychological physiological aspects of human health. Thus, using a deep learning approach that considers morphological variables, this study predicts intensity 69 European cities from 2007 2021 projects impacts for 2050 2080. The research employs Artificial Neural Networks, Deep Gated Recurrent Units, combining high-resolution 3D models data analyze trends. results indicate strong associations between form, weather patterns, intensity, highlighting the need customized planning policy measures reduce foster sustainable settings. This enhances understanding dynamics serves as valuable tool planners policymakers address challenges climate change, pollution, ultimately aiding improvement outcomes building energy consumption. Moreover, methodology effectively demonstrates ability GRU link its scores projections, offering crucial insights into potential impacts.

Язык: Английский

Процитировано

19

A review of urban heat island mapping approaches with a special emphasis on the Indian region DOI

N Renugadevi,

Manu Mehta,

G. T.

и другие.

Environmental Monitoring and Assessment, Год журнала: 2025, Номер 197(4)

Опубликована: Март 8, 2025

Язык: Английский

Процитировано

1

Modelling urban heat island effects: a global analysis of 216 cities using machine learning techniques DOI Creative Commons

Glenn Kong,

Yanni Zhao,

Jonathan Corcoran

и другие.

Computational Urban Science, Год журнала: 2025, Номер 5(1)

Опубликована: Апрель 8, 2025

Abstract Urban areas globally have become home to over half of the world's population, leading intensification urban heat island (UHI) effect, where cities experience higher temperatures than their rural counterparts. The current study develops a new model predicting UHI intensity for 216 across all climate zones both Global North and South using machine learning techniques, focusing on years 2019 2023. Utilising novel dataset, integrating climate, economic, land use data from worldwide, model, trained Support Vector Regression (SVR), demonstrates mean absolute error (MAE) 0.86 °C. Results reveal that wind speed significantly mitigates intensity, while in temperate climates exhibit more pronounced effects compared those located within tropical climbs. Additionally, results show crucial role coastal proximity reducing find no significant differences between South. Findings offer important empirical actionable insights alongside robust tool planners policymakers measure, map, monitor contributing development liveable sustainable environments.

Язык: Английский

Процитировано

1

Urban heat and pollution island in the Moscow megacity: Urban environmental compartments and their interactions DOI

Nikolay Kasimov,

Sergey Chalov,

Natalia Chubarova

и другие.

Urban Climate, Год журнала: 2024, Номер 55, С. 101972 - 101972

Опубликована: Май 1, 2024

Язык: Английский

Процитировано

5

Exploring intra-urban thermal stress vulnerability within 15-minute city concept: Example of heat waves 2021 in Moscow DOI
Natalia Shartova,

E. A. Mironova,

Mikhail Varentsov

и другие.

Sustainable Cities and Society, Год журнала: 2024, Номер 114, С. 105729 - 105729

Опубликована: Авг. 4, 2024

Язык: Английский

Процитировано

5

Enhancing machine learning performance in cardiac surgery ICU: Hyperparameter optimization with metaheuristic algorithm DOI Creative Commons
Ali Bahrami, Morteza Rakhshaninejad, Rouzbeh Ghousi

и другие.

PLoS ONE, Год журнала: 2025, Номер 20(2), С. e0311250 - e0311250

Опубликована: Фев. 10, 2025

The healthcare industry is generating a massive volume of data, promising potential goldmine information that can be extracted through machine learning (ML) techniques. Intensive Care Unit (ICU) stands out as focal point within hospitals and provides rich source data for informative analyses. This study examines the cardiac surgery ICU, where vital topic patient ventilation takes center stage. In other words, ventilator-supported breathing fundamental need limited availability ventilators in has become significant issue. A crucial consideration professionals ICU prioritizing patients who require immediately. To address this issue, we developed prediction model using four ML deep (DL) models-LDA, CatBoost, Artificial Neural Networks (ANN), XGBoost-that are combined an ensemble model. We utilized Simulated Annealing (SA) Genetic Algorithm (GA) to tune hyperparameters models constructing ensemble. results showed our approach enhanced sensitivity tuned 85.84%, which better than without hyperparameter tuning those achieved AutoML improvement performance underscores effectiveness hybrid among patients.

Язык: Английский

Процитировано

0

Examining the Spatiotemporal Dynamics of Urban Heat Island and Its Impact on Air Pollution in Thailand DOI Creative Commons
Veeranun Songsom,

Pawarit Jaruk,

Thongchai Suteerasak

и другие.

Environmental Challenges, Год журнала: 2025, Номер unknown, С. 101120 - 101120

Опубликована: Март 1, 2025

Язык: Английский

Процитировано

0

A machine learning approach for the efficient estimation of ground-level air temperature in urban areas DOI

Iñigo Delgado-Enales,

Joshua Lizundia-Loyola,

Patricia Molina-Costa

и другие.

Urban Climate, Год журнала: 2025, Номер 61, С. 102415 - 102415

Опубликована: Апрель 24, 2025

Язык: Английский

Процитировано

0

Testing the Performance of Large-Scale Atmospheric Indices in Estimating Precipitation in the Danube Basin DOI Creative Commons
Constantin Mares,

Venera Dobrică,

Ileana Mares

и другие.

Atmosphere, Год журнала: 2025, Номер 16(6), С. 667 - 667

Опубликована: Июнь 1, 2025

The objective of this study was to analyse the influence two large-scale climate indices on precipitation in Danube basin, both separately and combination. evolution hydroclimatic regime area is particular importance but has received limited attention. One for these data well-known North Atlantic Oscillation (NAOI) index, which been used numerous investigations; aim using index determine its various variables many regions globe. other Greenland–Balkan (GBOI), demonstrated have a greater Southeastern Europe compared NAOI. First, through different bivariate methods, such as estimating wavelet total coherence (WTC) time–frequency domain applying partial (PWC), performance GBOI contributing basin with that NAOI winter season. Then, by relatively simple multivariate methods multiple linear regression (MLR) variant thereof called ridge (RR), notable results were obtained regarding prediction overall training period 90 years (1901–1990), testing 30 (1991–2020). Nash–Sutcliffe (NS) criterion varied between 0.65 0.94, depending preprocessing approach applied input data, proving statistical modelling season powerful modern deep learning methods.

Язык: Английский

Процитировано

0

Deep Learning-Based Prediction of Urban Heat Island Intensity in European Cities Using Urban Morphological Features DOI
Alireza Attarhay Tehrani, Omid Veisi,

Kambiz Kia

и другие.

Опубликована: Янв. 1, 2024

Continued urbanization, along with anthropogenic global warming, has and will increase land surface temperature air anomalies in urban areas when compared to their rural surroundings worldwide, leading a phenomenon called Urban Heat Islands (UHI). This study aims predict the intensity of UHI European cities by using novel deep-learning methodology regarding morphological variables. The research utilizes Artificial Neural Networks, Deep Gated Recurrent Unit examine trends across 69 merging high-resolution 3D models environmental data. methodology's strength is shown its capacity establish connection between scores (GRU) projections UHI, providing valuable information on possible health consequences. results indicate strong associations form, weather patterns, highlighting need for customized planning policy measures reduce impacts foster sustainable settings. Additionally, GRU can accurately forecast 92\% R2 scores. work enhances overall comprehension dynamics. It provides helpful instrument planners policymakers tackle difficulties posed climate change, pollution.

Язык: Английский

Процитировано

2