Search

In the face of rapid urbanization in Saudi Arabia, understanding the impact of landscape changes on land surface temperature (LST) is crucial for sustainable urban planning. This study assesses the influence of landscape morphology on LST and predicts future LST changes. A multi-temporal land use and land cover (LULC) analysis using random forest (RF) quantified urban expansion and its ecological impacts. Accuracy assessment using the kappa coefficient illustrated the precision of the classification techniques. Morphological Spatial Pattern Analysis (MSPA), developed in Python, analyzed the structural evolution of urban areas, vegetation and exposed rocks. Polynomial regression models established the relationship between landscape morphology and LST and predicted future temperature trends. Results for Asir show a significant improvement in accuracy of the LULC models over three decades, with overall accuracy increasing from 89.99% in 1990 to 91.72% in 2020. The bootstrapping trend analysis showed an urban expansion with a positive slope of 9.27 and a decline in water bodies with a negative slope of −0.03. The MSPA analysis reflected a significant urban expansion, with the core area growing from 45.19 km2 in 2001 to 230.33 km2 in 2021. The vegetated areas showed resilience and dynamic connectivity despite slight reduction and fragmentation. The polynomial regression predicted an increase in future average LST by 2030, with urban core areas reaching 58.47 °C, vegetated cores 42.19 °C and exposed rock cores 55.79 °C. These results highlight the link between urban expansion and LST rise and make the case for integrating green infrastructure and cooling strategies into urban development.

Soil erosion poses a significant threat to sustainable land management and agricultural productivity. Addressing this issue requires advanced predictive models that can accurately identify areas at risk and inform soil conservation strategies. This study focuses on the development and interpretation of well-optimized deep learning (DL) models to predict soil erosion probability, aiming to enhance decision-making in land management. Utilizing the Revised Universal Soil Loss Equation (RUSLE) in conjunction with ground-truthing, we identified critical erosion-prone areas. To predict soil erosion probability, we employed Bayesian optimization to fine-tune Deep Neural Network (DNN), Convolutional Neural Network (CNN), Fully Connected Neural Network (FCNN), and DNN-CNN hybrid models. These DL models were verified using a set of metrics. SHAP value analysis as a means of explainable artificial intelligence (XAI) was used to interpret these DL models for better decision-making. RUSLE estimations and ground truthing highlight that Soil erosion rates in the northeastern and northwestern regions are nearing the highest observed at 25 tonnes per hectare annually, largely due to steep slopes and limited vegetation. In contrast, the southern and southeastern areas have lower erosion rates, due to denser vegetation and gentler slopes. Deep learning models, optimized using Bayesian methods, demonstrate high performance in spatially modeling soil erosion probability. The DNN model achieved an accuracy of 0.93, a precision of 0.92, and an F1-score of 0.94, identifying 222.73 sq. km as highly susceptible to erosion, which indicates its strong ability to detect true erosion events. The CNN model identified 49.68% of the study area (503.30 sq. km) as high-risk, with an accuracy of 0.90 and a precision of 0.91. The FCNN model showed a balanced risk distribution, indicating 37.04% of the land (375.25 sq. km) as very low risk and 36.43% (369.11 sq. km) as very high risk, with an accuracy of 0.91. The DNN-CNN hybrid model highlighted 41.58% of the area (421.20 sq. km) as high risk, demonstrating its effectiveness in capturing spatial patterns of erosion susceptibility. SHAP value analysis indicates that land use and soil type (LULC and K-factor) are crucial in erosion predictions, with LULC having a significant predictive influence in the DNN model. These insights facilitate the prioritization of soil conservation measures, enabling decision-makers to focus on the most impactful factors for mitigating soil erosion.

We aim to explore the seasonal influences of meteorological factors on COVID-19 era over two distinct locations in Bangladesh using a generalized linear model (GLM) and wavelet analysis. GLM model findings show that summer humidity drives COVID-19 transmission to coastal and inland locations. During the summer in the coastal area, a 1 °C earth’s skin temperature increase causes a 41.9% increase in COVID (95% CL 86.32%-2.54%) transmission compared to inland. Relative humidity was recorded as the highest at 73.97% (95% CL, 99.3%, and 48.63%) for the coastal region, while wind speed and precipitation reduced confirmed cases by −38.62% and −22.15%, respectively. Wavelet analysis showed that coastal meteorological parameters were more coherent with COVID-19 than inland ones. The outcomes of this study are consistent with subtropical climate regions. Seasonality and climatic similarity should address to estimate COVID-19 trends. High societal concern and strong public health measures may decrease meteorological effect on COVID-19.