The study proposes Secondary Precipitation Estimate Merging using Machine Learning (SPEM2L) algorithms for merging multiple global precipitation datasets to improve the spatiotemporal rainfall characterization. SPEM2L is applied over the Krishna River Basin (KRB), India for 34 years spanning from 1985 to 2018, using daily measurements from three Secondary Precipitation Products (SPPs). Sixteen Machine Learning Algorithms (MLAs) were applied on three SPPs under four combinations to integrate and test the performance of MLAs for accurately representing the rainfall patterns. The individual SPPs and the integrated products were validated against a gauge-based gridded dataset provided by the Indian Meteorological Department. The validation was applied at different temporal scales and various climatic zones by employing continuous and categorical statistics. Multilayer Perceptron Neural Network with Bayesian Regularization (NBR) algorithm employing three SPPs integration outperformed all other Machine Learning Models (MLMs) and two dataset integration combinations. The merged NBR product exhibited improvements in terms of continuous and categorical statistics at all temporal scales as well as in all climatic zones. Our results indicate that the SPEM2L procedure could be successfully used in any other region or basin that has a poor gauging network or where a single precipitation product performance is ineffective.

Water is an integral aspect of the world necessary for living creatures to thrive. Owing to unplanned urbanization, rapid industrialization, and uncontrollable human intervention, water quality is gradually degrading. This affects not only marine animals but also humans. Thus, the quality of water should be examined regularly. Water quality parameters should be estimated to monitor water quality. In general, water quality parameters are measured by in situ measurements. Although these measurements are accurate, they are costly and do not provide real-time spatial and temporal changes in water quality. To overcome this limitation, water quality parameters can be estimated using machine learning (ML) along with remote sensing (RS) data. A combination of ML and RS data is a powerful approach for the routine assessment of spatial and temporal variations in water quality parameters. In this paper, some articles based on this approach are reviewed. By analyzing the literature, it was found that the integrated use of RS-based geospatial data with ML helps to produce an accurate result. Most of the authors used the regression algorithm in the estimation of the water quality parameters, with a support vector machine (SVM) regression intensively used. The artificial neural network (ANN) algorithm was the most used algorithm of ML in many of the studies. The researchers used multispectral images for their study. By applying ML to RS data, water quality monitoring systems are evolving into real-time artificial intelligence (AI)-enabled models that provide valuable recommendations and insights to support farmers to make decisions and take action in aquaculture.

The study deals with the application of Google Earth Engine (GEE), Landsat data and ensemble-learning methods (ELMs) to map land cover (LC) change over a decade in the Kaski district of Nepal. As Nepal has experienced extensive changes due to natural and anthropogenic activities, monitoring such changes are crucial for understanding relationships and interactions between social and natural phenomena and to promote better decision-making. The main novelty lies in applying the XGBoost classifier for LC mapping over Nepal and monitoring the decadal changes of LC using ELMs. To map the LC change, a yearly cloud-free composite Landsat image was selected for the year 2010 and 2020. Combining the annual normalized difference vegetation index, normalized difference built-up index and modified normalized difference water index, with elevation and slope data from shuttle radar topography mission, supervised classification was performed using a random forest and extreme gradient boosting ELMs. Post classification change detection, validation and accuracy assessment were executed after the preparation of the LC maps. Three evaluation indices, namely overall accuracy (OA), Kappa coefficient, and F1 score from confusion matrix reports, were calculated for all the points used for validation purposes. We have obtained an OA of 0.8792 and 0.875 for RF and 0.8926 and 0.8603 for XGBoost at the 95% confidence level for 2010 and 2020 LC maps, which are better for mountainous terrain. The applied methodology could be significant in utilizing the big earth observation data and overcoming the traditional computational challenges using GEE. In addition, the quantification of changes over time would be helpful for decision-makers to understand current environmental dynamics in the study area.

Land cover (LC) is a crucial parameter for studying environmental phenomena. Cutting-edge technology such as remote sensing (RS) and cloud computing have made LC change mapping efficient. In this study, the LC of Rupandehi District of Nepal were mapped using a Landsat imagery and Random Forest (RF) classifier from 2005 to 2020 using Google Earth Engine (GEE) platform. GEE eases the way in extracting, analyzing, and performing different operations for the earth’s observed data. Land cover classification, Centre of gravity (CoG), and their trajectories for all LC classes: agriculture, built-up, water, forest, and barren area were extracted with five-year intervals, along with their Ecosystem service values (ESV) to understand the load on the ecosystem. We also discussed the aspects and problems of the spatiotemporal analysis of developing regions. It was observed that the built-up areas had been increasing over the years and more centered in between the two major cities. Other agriculture, water, and forest classes had been subjected to fluctuations with barren land in the decreasing trend. This alteration in the area of the LC classes also resulted in varying ESVs for individual land cover and total values for the years. The accuracy for the RF classifier was under substantial agreement for such fragmented LCs. Using LC, CoG, and ESV, the paper discusses the need for spatiotemporal analysis studies in Nepal to overcome the current limitations and later expansion to other regions. Studies such as these help in implementing proper plans and strategies by district administration offices and local governmental bodies to stop the exploitation of resources.

Abstract. Water is a major component in the living ecosystem. As water quality is degrading due to human intervention, continuous monitoring is necessary. One of the indicators is Chlorophyll-a (Chl-a) which indicates algal blooms which are often driven by eutrophication phenomena in freshwater. Lakes should be monitored for Chl-a because Chla-a is related to eutrophication phenomena which are an enrichment of water by nutrients salt. When the environment becomes enriched with nutrients the excessive growth can lead to the death of fish. In this study, the Remote Sensing (RS) and Geographic Information System (GIS) techniques were utilized to determine Chl-a concentration of Phewa Lake of Kaski district. We used Landsat 8 satellite imagery for estimation and mapping of the Chl-a concentration. In-situ measurements from different sample points were taken and used to form a regression model for Chl-a and its concentration over the water body was calculated. The preceding year’s (2016) in situ measurement data of Chl-a concentration at a specific location were assessed with the one evaluated from the regression model thus produced for the succeeding year (2017) using Root Mean Square Error (RMSE) technique. As a result, we concluded that the estimation and mapping of Chl-a of a lake in Nepal can be done with the help of RS and GIS techniques.