Journal of Urban Environmental Management

Abstract This study addresses the challenge of estimating Biochemical Oxygen Demand over 5 days (BOD₅) in a remote polar lake, where direct measurement is expensive, time-consuming, and often constrained by limited monitoring data. Using more than four decades of historical water quality records from Lake Oulujärvi, Finland, the research developed predictive models based on observations collected from 34 stations across six sub-basins, with particular attention to stratified periods and depth-resolved measurements. The data were first preprocessed through normalization, detection of missing values, removal of invalid observations, and stratification analysis to improve consistency and model reliability. Three machine learning approaches, including Convolutional Neural Networks (CNN), Random Forest (RF), and XGBoost, were then trained to estimate BOD₅ from routinely monitored water quality variables. The selected predictors included dissolved oxygen (DO), pH, turbidity, electrical conductivity, Secchi depth, water temperature, and sampling time, all of which are readily measurable and physically linked to BOD₅ dynamics. Model performance was optimized using grid search and cross-validation. Among the tested models, the CNN achieved the best results, with validation R² values of 0.88–0.89, followed by RF, and XGBoost. Feature importance analysis highlighted DO, pH, turbidity, electrical conductivity, and sampling time as the most influential variables. The findings show that machine learning can provide accurate, cost-effective, and scalable alternatives for BOD₅ estimation in low-accessibility aquatic systems, supporting improved water quality assessment and management in fragile lake ecosystems under environmental change.

Abstract In valley‑mountainous cities, differences in solar radiation and ventilation can modify the classic “hot core” pattern and create heat foci along the margins. This study examines the spatial pattern of the Surface Urban Heat Island (SUHI) in Khorramabad from 16 Mehr to 16 Aban 1404 (7 October to 6 November 2025) and assesses vegetation's role in moderating land surface temperature (LST). LST was derived from Landsat‑8/9 imagery using a single‑channel algorithm with ERA5 atmospheric parameters. NDVI was calculated from Sentinel‑2 images and resampled to 30‑meter resolution. SUHI was computed as the difference between urban LST and the mean LST of peripheral reference areas, redefined with constraints on elevation, slope, and aspect. Results indicated 27.15% of the urban area in the “neutral” class and 4.82% in the “very hot” class, while 50.61% is cold to cool. Very hot hotspots are concentrated in southern margins (airport, 184th brigade, oil depot, barren lands) and the eastern belt (Mehr Housing to Azad University). Cool islands appear around Kio Lake, along Khorramrud River, and in some central neighborhoods. A strong inverse LST‑NDVI relationship (r = –0.82; R² = 0.71; p < 0.001) revealed that each 0.1‑unit increase in NDVI reduces LST by 2.86°C on average. Based on these findings, low‑cost strategies are proposed: protecting intra‑valley green patches, strengthening the green‑blue network along the river and lake, creating green belts in sensitive margins, restricting construction on steep slopes, and preserving valley‑oriented ventilation corridors to alleviate thermal stress.