To obtain seasonable and precise crop yield information with fine resolution is very important for ensuring the food security. However, the quantity and quality of available images and the selection of prediction variables often limit the performance of yield prediction. In our study, the synthesized images of Landsat and MODIS were used to provide remote sensing (RS) variables, which can fill the missing values of Landsat images well and cover the study area completely. The deep learning (DL) was used to combine different vegetation index (VI) with climate data to build wheat yield prediction model in Hebei Province (HB). The results showed that kernel NDVI (kNDVI) and near-infrared reflectance (NIRv) slightly outperform normalized difference vegetation index (NDVI) in yield prediction. And the regression algorithm had a more prominent effect on yield prediction, while the yield prediction model using Long Short-Term Memory (LSTM) outperformed the yield prediction model using Light Gradient Boosting Machine (LGBM). The model combining LSTM algorithm and NIRv had the best prediction effect and relatively stable performance in single year. The optimal model was then used to generate 30 m resolution wheat yield maps in the past 20 years, with higher overall accuracy. In addition, we can define the optimum prediction time at April, which can consider simultaneously the performance and lead time. In general, we expect that this prediction model can provide important information to understand and ensure food security.

Technology Acceptance Model (TAM) is essential to the field of Building Information Modelling (BIM) implementation especially in the construction industry since the government mandate, it has become the key towards this perception as an obstacle of success. While innumerable studies focus on the way contractors, consultants and stakeholder acceptance towards BIM, therefore, it becomes a method for improvisation. However, up until present, limitation have been found for government staffs and perceptions regarding this manner. As to affirm, the government as well as its organization are the backbone of BIM utilization for every country. Research has shown that BIM is also influenced by the one who holds the authority by government mandating its usage strategies, and that representations are generally more effective than appeals from other users. Thus, this study aims and focuses on examines the emerging role of government organizations in the context of perceptions and acceptance towards BIM as well as the readiness of government mandate in facing challenges for Building Information Modelling (BIM) implementation in Malaysia. In this study, an online survey was distributed to government staffs that currently uses BIM across Malaysia. The obtained data were analyzed by statistical tools of IBM SPSS 28 to empirically test the correlation and impact of the proposed variable. The results showed that the TAM for BIM and well-structured government mandate will affect the challenges in BIM implementation. This paper has been organised in the following way. Starts with introduction, literature review, methodology, results and discussions. Finished with conclusions accordingly. Additionally, acceptance of BIM technology would initiate of minimizing the challenges faced by government staffs in Malaysia that involve in BIM. Good obtained values for TAM and government mandate variables in explanatory power, concluded this study. Further research is needed in this area of study to identify well-planned roadmap and proper guideline for government staffs in strengthening the effectiveness of BIM implementation in the future.

Recent advances in land use and land cover (LULC) classification with remote sensing imagery are driven by state-of-the-art models such as Convolutional Neural Networks (CNNs). Advanced CNN architecture like ResNet can enhance overall classification performance by incorporating residual skip connections. The integration of 3D feature extraction and ResNet architecture suggests a potential improvement in classification tasks. This paper explores the potential of the 3D ResNet model for LULC classification, comparing it with baseline approaches (Support Vector Machine, Random Forest, XGBoost, 1D CNN, 3D CNN) and state-of-the-art 3D models (3D VGG, 3D DenseNet) using WorldView-2 satellite imagery. The 3D ResNet-18 model, fine-tuned via transfer learning on multispectral images, demonstrates significant improvements in classification performance over machine learning models. It achieves the highest Overall Accuracy (OA) of 99.66% and Kappa Accuracy (KA) of 99.39% on the primary dataset. Despite having slightly lower performance on the external validation dataset (OA:82.89%, KA:80.05%) than 3D DenseNet, it is highly efficient with processing times of 490.2 minutes and 3.6 minutes for both datasets respectively. McNemar’s test results show 3D ResNet and 3D DenseNet have significant differences in classification performance (p<0.05) against other models consistently for both datasets.

Kathmandu Valley, the capital of Nepal, is located in the seismically active Himalayan belt and has a history of devastating earthquakes causing substantial loss of life and property damage. This study employs Probabilistic Seismic Hazard Analysis (PSHA) using the Foulser-Piggott Attenuation (FPA) model and Travasarou et al. (2003) with R-CRISIS software to calculate Arias intensity in Kathmandu Valley. Historical and recent seismic data within a 500-km radius were analysed, and the earthquake catalogue was declustered and standardized using ZMAP software, a tool developed for the statistical analysis and visualization of earthquake catalogues. Additionally, a Digital Elevation Model (DEM) based topographic analysis was conducted to assess the impact of local topography on seismic site response providing insights into, slope, soil amplification factors, and shear wave velocity across the region. The results reveal Arias intensity values ranging from 0.225 to 0.241 m/s at 2% and 10% probability of exceedance corresponding to 475 and 2475 years, mapped using ArcGIS. The analysis revealed that southwestern Kathmandu and Lalitpur exhibit higher Arias intensity values, while intensity decreases gradually from southwest to northeast. The DEM analysis further revealed that areas with low slopes, particularly in central Kathmandu, have higher soil amplification factors, potentially amplifying seismic waves. The shear wave velocity distribution highlights lower values in sedimentary deposits, indicating increased seismic vulnerability. These findings emphasize the need for effective urban planning and disaster preparedness strategies to mitigate earthquake impacts in Kathmandu Valley.

Urbanisation profoundly alters land-use and land-cover (LULC) patterns, often intensifying urban heat island (UHI) effects and threatening sustainable development. In rapidly growing semi-arid cities such as Kano Metropolis, Nigeria, these dynamics remain underexplored. This study investigates the relationship between LULC transitions and urban thermal dynamics over nearly four decades (1984 - 2023), while also forecasting future changes. Multi-temporal Landsat imagery was classified using the Classification and Regression Tree (CART) algorithm, achieving overall accuracies ranging from 92.1% to 98.5% and Kappa coefficients exceeding 0.85. Vegetation and built-up indices, including the Normalised Difference Vegetation Index (NDVI) and Normalised Difference Built-up Index (NDBI), were analysed alongside Land Surface Temperature (LST) to assess urban heat patterns. Predictive modelling employed the Cellular Automata Artificial Neural Network (CA-ANN) approach, which was validated with an overall accuracy of 92% and a Kappa coefficient of 0.86. Results show that built-up areas expanded from 43.06 km² (2.93%) in 1984 to 381.79 km² (25.95%) in 2023, an almost 800% increase, while bare land declined by 23.2%. Mean LST rose from 39.7°C in 1984 to 41.5°C in 2023, with peak values exceeding 52°C in 2010. A strong positive correlation was found between NDBI and LST (r = 0.57), while NDVI showed a negative correlation with LST (r = -0.32), highlighting the cooling effect of vegetation. Model simulations predict continued urban expansion through 2050, with built-up areas increasing by 6.5% and further intensification of UHI effects if unchecked. These findings emphasise the urgent need for sustainable urban planning, including the preservation of vegetation cover, the development of green infrastructure, and climate-sensitive construction practices. The study offers critical insights for policymakers and urban planners seeking to mitigate thermal stress and foster climate-resilient urban development in sub-Saharan Africa.