Sylhet Agricultural University

Soil load-bearing capacity (SLBC) is a critical determinant of sustainable soil management and the efficient operation of agricultural machinery, yet continuous field-based assessment is constrained by high costs and labor demands. This study presents a novel, spatially explicit remote sensing–machine learning (RS–ML) framework for mapping SLBC across croplands in northeastern Bangladesh, particularly in haor regions. Field observations from 127 agricultural sites were collected using a Dynamic Cone Penetrometer (DCP), with penetration-per-blow values converted to DCP indices and subsequently transformed into California Bearing Ratio (CBR) and SLBC. A predictor set, including soil properties, topographic derivatives, and spectral indices, was preprocessed in Google Earth Engine and harmonized into raster layers at 250 m resolution. Extreme Gradient Boosting (XGBoost), Gradient Boosting Machine (GBM), and Random Forest (RF) models were trained using hyperparameter tuning, 10-fold cross-validation, and a 70/30 train–test split. During 10-fold cross-validation, XGBoost (MSE 3.03–43.57, RMSE 1.74–6.60, MAE 1.49–3.92, R² 0.567–0.955) and GBM (MSE 2.81–46.60, RMSE 1.68–6.83, MAE 1.32–3.75, R² 0.531–0.958) showed better predictive performance for SLBC than RF (MSE 6.59–60.91, RMSE 2.57–7.80, MAE 2.14–4.44,R² 0.230–0.901). However, RF achieved lower error on the test dataset, while the Wilcoxon signed-rank test showed no statistically significant differences among the models. Bulk Density was identified as the most influential factor shaping spatial variability in soil load-bearing capacity (SLBC). Higher SLBC in Sylhet Sadar and lower values in Tahirpur and Bishwambarpur were observed. Limitations include relatively homogeneous soil conditions, limited field coverage, lack of seasonal variability, and restricted model extrapolation beyond training data. Despite these constraints, the proposed ML–RS framework provides spatially explicit SLBC predictions that can inform data-driven decision-making and support mechanized farming in wetland-prone croplands of northeastern Bangladesh and similar agroecosystems. This graphical abstract presents a spatially explicit, multi-feature-based machine learning–remote sensing (ML–RS) framework for mapping Soil Load Bearing Capacity (SLBC) across persistent croplands in northeastern Bangladesh. Field-based SLBC measurements were collected from 127 locations using a Dynamic Cone Penetrometer (DCP). Penetration-per-blow values were converted into DCP indices and subsequently transformed into California Bearing Ratio (CBR) and SLBC using established empirical relationships, which served as the model target. A comprehensive suite of multi-source remote sensing predictors was assembled, including soil intrinsic properties (bulk density, sand, silt, and clay), hydro-thermal variables (soil moisture and soil temperature), topographic derivatives (slope, aspect, curvature, and topographic wetness index), and vegetation- and water-related spectral indices (NDVI and NDWI). All predictor layers were harmonized to 250 m spatial resolution across croplands using integrated workflows in Google Earth Engine, ArcGIS, and Python. Extreme Gradient Boosting (XGBoost), Gradient Boosting Machine (GBM), and Random Forest (RF) models were trained with hyperparameter tuning, 10-fold cross-validation, and a 70/30 train–test split. XGBoost and GBM showed stronger and faster-converging performance during cross-validation and learning-curve analysis than RF. However, RF achieved lower error on the independent test dataset, while the Wilcoxon signed-rank test indicated no statistically significant differences among the models. Bulk Density was the most influential factor shaping SLBC, with higher values in Sylhet Sadar and lower values in Tahirpur and Bishwambarpur. With larger data coverage, incorporation of seasonal data, machinery interaction, and hybrid models, the ML–RS framework could become more scalable and provide more rigorous SLBC predictions. A spatial, multi-feature ML–RS framework mapped SLBC across northeastern Bangladesh croplands. Model performance varied between cross-validation and test evaluation, but differences were not statistically significant. SLBC was higher in Sylhet Sadar, comparatively lower in Bishwambarpur, and Tahirpur. Bulk density was the dominant positive driver of SLBC. Larger datasets, seasonal effects, machinery interaction, and hybrid models may improve predictions.

Heatwaves are increasingly impacting rural communities in Malaysia, with disproportionate effects on youth due to limited adaptive capacity and a lack of targeted interventions. This study presents a data-driven, participatory heatwave co-adaptation mapping initiative involving thirty Form Three students from a rural high school in Segamat, Johor. The structured, interactive workshop comprised five core activities: emoji-based reflection and gamification exercises to elicit emotional responses and visualize the climate change process; a Mini Talk with interactive materials and quizzes on local climate trends and adaptation options; and hands-on group dotmocracy mapping sessions to identify cold and thermal hotspots, assess vulnerabilities, and propose adaptation measures. Analysis of the school environment revealed a long-term warming trend, with daily Tmax increasing by 0.022 °C per year (≈ 0.22 °C per decade; t = 46.86, p < 2 × 10⁻¹⁶) from 1950 to 2022. March recorded the highest frequency of extreme heat days (> 35 °C) with 160 days (7.07

A range of cropping patterns is practiced in Bangladesh, which is likely to alter soil physical, physico-chemical, and hydraulic properties. To investigate the effects of different cropping patterns on soil physical, physico-chemical, and hydraulic properties, soil samples were collected from an 8-year long-term field experiment conducted in acidic soil during January–February 2024 at Sylhet, Bangladesh. The experiment followed a completely randomized design (CRD) and the tested patterns were: Fallow- Fallow–Rice (Transplanted (T.)-aman, Oryza sativa L.) (F1), Legume (Phaseolus vulgaris L.)–Fallow–Rice (T. aman) (F2), and Brinjal (Solanum melongena L.)–Fallow–Fallow (F3), which are the predominant cropping patterns and all maintained continuously for eight years. The results revealed that bulk density was significantly (p = 0.03) lower (1.11 g cm⁻³) in F2 compared with F1 (1.25 g cm⁻³) and F3 (1.26 g cm⁻³). Soil organic carbon differed significantly among cropping patterns (p < 0.0001), with the following order: F2 (2.29

ABSTRACT The sustainable replacement of conventional fish meal (CFM) remains a critical challenge for aquaculture, given its rising cost and limited supply. This study evaluated the nutritional composition, physical properties, safety, and economic feasibility of suckermouth catfish (Hypostomus plecostomus) meal (SMC) as a potential alternative protein source and as a management strategy for controlling this invasive species in the Buriganga river. Proximate composition, amino acid profile, fatty acid composition, selected physical properties, and contaminant levels (heavy metals, pesticide residues, dyes, and antibiotics) were analyzed using standard laboratory procedures. Proximate analysis revealed slightly lower crude protein in SMC than in CFM, but comparable lipid and ash contents. Physical property assessments indicated favourable bulk density, water stability, and solubility index, supporting its suitability for pelleted feed formulations. Amino acid profiling highlighted high levels of glutamic acid, leucine, and lysine, while fatty acid analysis showed the dominance of palmitic and oleic acids, though with relatively low levels of EPA and DHA. Heavy metal concentrations were within international safety limits, while trace pesticide residues were detected. Economic analysis indicated substantially lower production cost compared to CFM. However, pesticide residues and malachite green were detected, reflecting persistent contamination of the Buriganga river and underscoring potential food safety risks. The findings suggest that SMC may serve as a partial fishmeal substitute under controlled conditions, subject to further validation through feeding trials and safety monitoring. This study provides the first comprehensive evaluation of SMC from Bangladesh and highlights its promise and limitations as a sustainable aquafeed ingredient.

The development processes of soil are influenced by temperature and precipitation; moreover, changes in soil properties are influenced by land use factors. This study aimed to identify the pattern of changes in climate over 40 years (from 1985–2024) and, concomitantly, the spatial and temporal variabilities in the soil properties of Sylhet, a humid subtropical region of Bangladesh. Soil samples were taken from 8 unions of the Sylhet sadar upazila during 2024 and tested in the laboratory to determine the present status of different soil parameters. The temporal changes in soil properties were assessed by comparing the laboratory-analysed data with soil resource development institute-reported soil data from 2007. The annual maximum temperature during the last 20 years of the study (2005–2024) increased by 0.42°C on average, and the hottest year was 2023. The total annual precipitation sharply decreased by 762.34 mm compared with that in the first 20 years (1985–2004), with continuous below-normal precipitation events (-1270.96 mm to -104.32 mm) from 2018–2023 and high interannual variability. In addition, this study revealed greater spatial variability in soils than temporal variability, which was influenced mainly by inherent soil characteristics such as texture and pH and assumed land management practices. Moderate temporal variability in soil organic carbon and other nutrient availability might be associated with changing climates. These findings on spatial and temporal changes in soil properties are expected to be helpful for guiding future soil and crop management in humid subtropical regions in a better way under changing climates.