This study developed a rapid rice yield estimation workflow and customized yield prediction model by integrating remote sensing and meteorological data with machine learning (ML). Several issues need to be addressed while developing a crop yield estimation model, including data quality issues, data processing issues, selecting a suitable machine learning model that can learn from few available time-series data, and understanding the non-linear relationship between historical crop yield and remote sensing and meteorological factors. This study applied a series of data processing techniques and a customized ML model to improve the accuracy of crop yield estimation at the district level in Nepal. It was found that remote sensing-derived NDVI product alone was not sufficient for accurate estimation of crop yield. After incorporating other meteorological variables into the ML models, estimation accuracy improved dramatically. Along with NDVI, the meteorological variables of rainfall, soil moisture, and evapotranspiration also exhibited a strong association with rice yield. This study also found that stacking multiple tree-based regression models together could achieve better accuracy than benchmark linear regression or standalone ML models. Due to the unique and distinct physio-geographical setting of each district, a variation in estimation accuracy from district to district could be observed. Our data processing and ML model workflow achieved an average of 92% accuracy of yield estimation with RMSE 328.06 kg/ha and MAE 317.21 kg/ha. This methodological workflow can be replicated in other study areas and the results can help the local authorities and stakeholders understand the factors affecting crop yields as well as estimating crop yield before harvesting season to ensure food security and sustainability.

Environmental degradation can be predicted and managed in a sustainable manner by the perodic analysis of the Land Use/Land Cover (LULC) change pattern, which not only helps to revitalize the environment but also helps to improve future land-use policies. With the Rohingya influx in 2017, the Kutupalong Mega Camp area in Bangladesh is at a severe risk of environmental degradation as the area is experiencing remarkable LULC change. The aim of this research is to illustrate the LULC change in the Kutupalong Mega Camp before and after the refugee influx, as well as its impact on the surrounding environment because of this change. The spatial and temporal variation of the LULC is analyzed from the classified multi-temporal Landsat images for years-2015, 2018, and 2021. The study reveals gradual decrease in forest cover of the area, which is replaced by the increasing human settlements. The study found an inverse relation between the refugee influx and the vegetation cover, where a positive relation to the bare land and settlement exists. The area experienced about ten times increase in human settlements during 2015–2021, which resulted deforestation of surrounding forest cover. Between 2015 to 2021, 74 % of forest cover of the studied area has been cleaned up for newer settlements, with an increase of wetland to meet the needs of increasing refugee population which has made the scenario worse. We also noticed an increase of Land Surface Temperature (LST) within a short period, where the average temperature increase rate is 0.06% during 2015–2018 and 0.01% during 2018–2021. The ecosystem, wild-habitat, and the thermal environment has been disturbed to a great extent due to this drastic change of forest cover mostly by the increasing anthropogenic activities in this area. The study represents the present scenario in comparison to its natural setting just a few years ago, and may serve as a guidance for the concerned authorities and international humanitarian organizations to develop a sustainable, comprehensive, and environment-friendly land management plan in order to protect the surrounding forest-ecology as well as the humanitarian works.