Synergizing google earth engine and earth observations for potential impact of land use/ land cover on air quality

Results in Engineering, Journal Year: 2024, Volume and Issue: 22, P. 102039 - 102039

Published: March 24, 2024

Changes in land use and cover are imperative drivers of climate change urbanization. The conversion modifies the physical thermal characteristics surface also has an impact on air quality. This study aims to assess potential Land Use Cover (LULC) quality Gujarat state, India for 6 years, 2018, 2020, 2023. Six land-use types, water bodies, forest, agricultural land, built-up, barren scrubland obtained from Landsat 8 product processed GEE, where LULC each category was estimated. analysis findings indicated that variations pollution response exhibit distinct differences across different regions, influenced by natural factors or human activities like deforestation Over years 2018–2023, seems consistently decrease area, but urban areas saw exponential growth. combined percentage forest area slightly decreased 61.08% 60.7%, while spread increased 4.07% 5.13%. bare 29.59% 27.56%, mainly due urbanization converting soil into built-up areas. Sentinel-5P satellite data used estimate atmospheric i.e., carbon mono-oxide (CO), Nitrogen dioxide (NO2), Methane (CH4), Sulfur Dioxide (SO2), formaldehyde (HCHO). past decade, a significant portion transitioned vegetation western region, rapidly expanded eastern central-western parts.

Language: Английский

---

Artificial intelligence to predict soil temperatures by development of novel model

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: April 30, 2024

Abstract Soil temperatures at both surface and various depths are important in changing environments to understand the biological, chemical, physical properties of soil. This is essential reaching food sustainability. However, most developing regions across globe face difficulty establishing solid data measurements records due poor instrumentation many other unavoidable reasons such as natural disasters like droughts, floods, cyclones. Therefore, an accurate prediction model would fix these difficulties. Uzbekistan one countries that concerned about climate change its arid climate. for first time, this research presents integrated predict soil temperature levels 10 cm depth based on climatic factors Nukus, Uzbekistan. Eight machine learning models were trained order best-performing widely used performance indicators. Long Short-Term Memory (LSTM) performed predictions depth. More importantly, developed here can with measured predicted levels. The without any ground measurements. be effectively planning applications sustainability production areas

Language: Английский

---

Comparative analysis of different rainfall prediction models: A case study of Aligarh City, India

Results in Engineering, Journal Year: 2024, Volume and Issue: 22, P. 102093 - 102093

Published: April 5, 2024

This research paper delves into creating and comparing rainfall prediction models, employing diverse machine learning algorithms, including Logistic Regression, Decision Tree Classifier, Multi-Layer Perceptron classifier (neural network), Random Forest. The study aims not only to predict patterns but also evaluate the performance of each model through metrics such as Accuracy, Cohen's kappa coefficient, Receiver Operating Characteristic (ROC) curve analysis. Additionally, relevance predictors employed in is thoroughly assessed. results extensive experimentation analysis reveal that Regression (Accuracy = 82.80 %, ROC 82.45 Kappa 65.05 %) Neural Network 82.59 81.94 64.40 has emerged most promising approach, achieving highest percentage accuracy, metrics; among models considered. outcome underscores effectiveness architectures capturing intricate relationships within data.

Language: Английский

---

Explainable artificial intelligence to estimate the Sri Lankan (Ceylon) Tea crop yield

Smart Agricultural Technology, Journal Year: 2025, Volume and Issue: unknown, P. 100999 - 100999

Published: May 1, 2025

Language: Английский

---

Predictive modeling of rice milling degree for three typical Chinese rice varieties using interpretative machine learning methods

Journal of Food Science, Journal Year: 2024, Volume and Issue: 89(10), P. 6553 - 6574

Published: Sept. 1, 2024

Brown rice over-milling causes high economic and nutrient loss. The degree of milling (DOM) detection prediction remain a challenge for moderate processing. In this study, self-established grain image acquisition platform was built. Degree bran layer remaining (DOR) datasets is established with capturing processing (grain color, texture, shape features extraction). mapping relationship between DOR the DOM in-depth analyzed. Rice typical machine learning deep models are established. results indicate that optimized Catboost model can be cross-validation grid search method, best accuracy improving from 84.28% to 91.24%, achieving precision 91.31%, recall 90.89%, F1-score 91.07%. Shapley additive explanations analysis indicates feature affect accuracy, importance: color > texture shape. YCbCr-Cb_ske GLCM-Contrast make most significant contribution quality prediction. importance provides theoretical practical guidance model. PRACTICAL APPLICATION: valuable process in application. paper, methods provide an automated, nondestructive, cost-effective way predict rice. study may serve as reference methods, retaining nutrition, reducing broken yield.

Language: Английский

---

Drivers of PM10 Retention by Black Locust Post-Mining Restoration Plantations

Atmosphere, Journal Year: 2025, Volume and Issue: 16(5), P. 555 - 555

Published: May 7, 2025

Atmospheric pollution due to an increased particulate matter (PM) concentration imposes a threat for human health. This is particularly true regions with intensive industrial activity and nature-based solutions, such as tree plantations, are adopted mitigate the phenomenon. Here, we report on case of lignite complex western Macedonia (LCWM), largest in Greece, where extensive Robinia pseudoacacia L. plantations have been established during last 40 years post-mining reclamation, but their PM retention capacity controlling parameters not assessed date. Thus, 2021 growth season (May October), determined PM10 capture by leaves sampled twice per month, across four 10-m long transects, each consisting five trees, at three different heights along canopy. During same period, also measured leaf area index (LAI) collected climatic data, well data production belt conveyors system, main polluting source site. We estimated that plantations’ foliage captures average c. 42.85 μg cm−2 developed robust linear model describes basis, function production, LAI (a proxy seasonal changes area), distance from emitting source, wind speed height within crown. The accuracy estimates performance were tested bootstrap cross-validate resampling technique. spring early summer following increase LAI, its peak August October was controlled highest elevated energy demands. Moreover, facilitated speed, it higher lower part trees’ On contrary, load decreased increasing conveyor system frontline plantations. Our findings support positive role R. heavily polluted areas, mines provide estimation based basic environmental drivers characteristics which could be helpful planning future management.

Language: Английский

---

Forecasting of time-dependent scour depth based on bagging and boosting machine learning approaches

Journal of Hydroinformatics, Journal Year: 2024, Volume and Issue: 26(8), P. 1906 - 1928

Published: July 17, 2024

ABSTRACT Forecasting the time-dependent scour depth (dst) is very important for protection of bridge structures. Since result a complicated interaction between structure, sediment, and flow velocity, empirical equations cannot guarantee an advanced accuracy, although they would preserve merit being straightforward physically inspiring. In this article, we propose three ensemble machine learning methods to forecast at piers: extreme gradient boosting regressor (XGBR), random forest (RFR), extra trees (ETR). These models predict given time, dst, based on following main variables: median grain size, d50, sediment gradation, σg, approach U, y, pier diameter Dp, time t. A total 555 data points from different studies have been taken research work. The results indicate that all proposed precisely estimate depth. However, XGBR method performs better than other with R = 0.97, NSE 0.93, AI 0.98, CRMSE 0.09 testing stage. Sensitivity analysis exhibits highly influenced by scale.

Language: Английский

---

Landscape Metrics as Ecological Indicators for PM10 Prediction in European Cities

Land, Journal Year: 2024, Volume and Issue: 13(12), P. 2245 - 2245

Published: Dec. 21, 2024

Despite significant progress in recent decades, air pollution remains the leading environmental cause of premature death Europe. Urban populations are particularly exposed to high concentrations pollutants, such as particulate matter smaller than 10 µm (PM10). Understanding spatiotemporal variations PM10 is essential for developing effective control strategies. This study aimed enhance prediction models by integrating landscape metrics ecological indicators into our previous models, assessing their significance monthly average concentrations, and analyzing correlations with across European urban landscapes during heating (cold) non-heating (warm) seasons. In research, we only calculated proportion land uses (PLANDs), but according current research hypothesis, have a impact on quality. Therefore, expanded independent variables incorporating that capture compositional heterogeneity, including Shannon diversity index (SHDI), well reflect configurational heterogeneity landscapes, Mean Patch Area (MPA) Shape Index (SHI). Considering data from 1216 quality (AQ) stations, applied Random Forest model using cross-validation discover patterns complex relationships. Climatological factors, temperature, wind speed, precipitation, mean sea level pressure, emerged key predictors, season when temperature increased 5.80% 22.46% at 3 km. Landscape metrics, SHDI, MPA, SHI, were significantly related concentration. The SHDI was negatively correlated levels, suggesting heterogeneous could help mitigate pollution. Our enhanced achieved an R² 0.58 1000 m buffer zone 0.66 3000 zone, underscoring utility these improving predictions. findings suggest complexity, patch sizes, more fragmented associated sources built-up areas, along larger evenly distributed green spaces, can contribute reduction

Language: Английский

---