Optimisation and interpretation of machine and deep learning models for improved water quality management in Lake Loktak.

Loktak Lake, one of the largest freshwater lakes in Manipur, India, is critical for the eco-hydrology and economy of the region, but faces deteriorating water quality due to urbanisation, anthropogenic activities, and domestic sewage. Addressing the urgent need for effective pollution management, this study aims to assess the lake's water quality status using the water quality index (WQI) and develop advanced machine learning (ML) tools for WQI assessment and ML model interpretation to improve pollution management decision making. The WQI was assessed using entropy-based weighting arithmetic and three ML models - Gradient Boosting Machine (GBM), Random Forest (RF) and Deep Neural Network (DNN) - were optimised using a grid search algorithm in the H2O Application Programming Interface (API). These models were validated by various metrics and interpreted globally and locally via Partial Dependency Plot (PDP), Accumulated Local Effect (ALE) and SHapley Additive exPlanations (SHAP). The results show a WQI range of 72.38-100, with 52.7% of samples categorised as very poor. The RF model outperformed GBM and DNN and showed the highest accuracy and generalisation ability, which is reflected in the superior R2 values (0.97 in training, 0.9 in test) and the lower root mean square error (RMSE). RF's minimal margin of error and reliable feature interpretation contrasted with DNN's larger margin of error and inconsistency, which affected its usefulness for decision making. Turbidity was found to be a critical predictive feature in all models, significantly influencing WQI, with other variables such as pH and temperature also playing an important role. SHAP dependency plots illustrated the direct relationship between key water quality parameters such as turbidity and WQI predictions. The novelty of this study lies in its comprehensive approach to the evaluation and interpretation of ML models for WQI estimation, which provides a nuanced understanding of water quality dynamics in Loktak Lake. By identifying the most effective ML models and key predictive functions, this study provides invaluable insights for water quality management and paves the way for targeted strategies to monitor and improve water quality in this vital freshwater ecosystem.

Assessing the link between land use/land cover changes and land surface temperature in Himalayan urban center, Srinagar.

The rapidly growing urbanization and the consequent land use/land cover (LULC) changes have resulted in unsustainable growth of cities in Indian subcontinent especially in Himalayan region which are highly sensitivity to condition like climate change. Using multi-temporal and multi-spectral satellite datasets, this study analyzes the impact of LULC changes on land surface temperature (LST) in the Himalayan city of Srinagar from 1992 to 2020. For LULC classification, the maximum likelihood classifier technique was utilized, and to extract LST from Landsat 5 (TM) and Landsat 8 (TM) (OLI), spectral radiance was employed. The results show that, among various LULC classes, built-up area has seen a maximum increase of 14% while agriculture has decreased by about 21%. On the whole, Srinagar city has witnessed an increase in LST by 4.5 °C with maximum increase of 5.35 °C especially over marshes and a minimum increase of 4 °C on agriculture landscape. Other LULC categories of built-up, water bodies, and plantation saw an increase in LST by 4.19 °C, 4.47 °C, and 5.07 °C, respectively. The transformation of marshes into built-up saw a maximum increase in LST by 7.18 °C followed by water body to built-up (6.96 °C) and water body to agriculture (6.18 °C) while minimum increase was seen in the conversion of agriculture to marshes by about 2.42 °C followed by agriculture to plantation (3.84 °C) and plantation to marshes (3.86 °C). The findings may be useful to urban planners and policymakers in terms of land use planning and city thermal environment control.

Analysis of peri-urban land use/land cover change and its drivers using geospatial techniques and geographically weighted regression.

The rate of transformation of natural land use land cover (LULC) to the built-up areas is very high in the peri-urban areas of Indian metropolitan cities. Delhi National Capital Region (Delhi NCR) is an inter-state planning region, located in the central part of India. The region has attracted a larger chunk of population by providing better economic opportunities during last few decades. This has resulted in large-scale transformation of the LULC pattern in the region. Thus, this study is intended to analyze and quantify the LULC change and its drivers in the peri-urban areas of Delhi NCR using Landsat datasets. Based on an extensive literature survey, several potential drivers of the LULC change have been analyzed using ordinary least squares (OLS) and geographical weighted regression (GWR) for the Delhi NCR. The results from LULC classification showed that the built-up area has increased from 1.67 to 7.12% of the total area of Delhi NCR during 1990-2018 while other LULC types have declined significantly. The OLS results showed that migration and employment in the tertiary sector are the most important drivers of built-up expansion in the study area. The standard residuals and local R2 results from GWR showed spatial heterogeneity among the coefficients of the explanatory variables throughout the study area. This study can be helpful for the urban policy makers and planners for making better master plan of Delhi NCR and other cities of developing countries.

Modelling built-up land expansion probability using the integrated fuzzy logic and coupling coordination degree model.

The expansion of built-up area is the most noticeable form of urbanization-induced land use/land cover (LULC) change. In the global cities of south, the urban sprawl is increasing rapidly with even higher probabilities of future built-up expansion. These cities are witnessing unsustainable urban growth with no consideration of eco-friendly environmental condition and quality of life due to rapid expansion in built-up area. Indian cities too have been witnessing rapid urban growth and built-up expansion especially in the large metropolitan cities like Delhi. Therefore, the main objective of this study is to model the built-up expansion probabilities in Delhi National Capital Region (Delhi NCR) using remote sensing datasets and an integrated fuzzy logic and coupling coordination degree model (CCDM). For this, initially, the LULC classification was done using random forest (RF) classifier to extract the built-up area. Further, analytical hierarchy process (AHP) integrated fuzzy sets were applied using the extracted built-up area along with a set of economic, demographic, proximity parameters, topographic, and utility services. Five zones of built-up expansion probabilities were made namely very high, high, medium, low and very low. The result shows that the probability of built-up expansion in Delhi NCR is maximum under very high and high probability zones, whereas minimum expansion probabilities come in the very low probability zone for both base year i.e., 2018 and future years. Moreover, between base year and future years, the probability of built-up expansion has increased maximum (5.72%) under the very high zone while it declined by 14.06% in low probability zone. The validation of built-up probability using CCDM shows that the AHP integrated fuzzy logic-based probability model is robust while predicting built-up probability. The results of this study may provide useful insights for the urban planning department and policy makers to mitigate the adverse impacts of built-up expansion. Similar approach may be utilized in the analyzing the built-up urban expansion of other major cities of the world similar geographical conditions.

Application of advanced trend analysis techniques with clustering approach for analysing rainfall trend and identification of homogenous rainfall regions in Delhi metropolitan city.

In the era of global urbanization, the cities across the world are experiencing significant change in the climate pattern. However, analysing the trend and pattern of rainfall over the urban areas has a number of challenges such as availability of long-term data as well as the uneven distribution of rain-gauge stations. In this research, the rainfall regionalization approach has been applied along with the advanced statistical techniques for analysing the trend and pattern of rainfall in the Delhi metropolitan city. Fuzzy C-means and K-means clustering techniques have been applied for the identification of homogeneous rainfall regions while innovative trend analysis (ITA) along with the family of Mann-Kendall (MK) tests has been applied for the trend analysis of rainfall. The result shows that in all rain-gauge stations of Delhi, an increasing trend in rainfall has been recorded during 1991-2018. But the rate of increase was low as the trend slope of ITA and Sen's slope in MK tests are low, which varies between 0.03 and 0.05 and 0.01 and 0.16, respectively. Furthermore, none of the rain-gauge stations have experienced a monotonic trend in rainfall as the null hypothesis has not been rejected (p value > 0.05) for any stations. Furthermore, the study shows that ITA has a better performance than the family of MK tests. The findings of this study may be utilized for the urban flood mitigation and solving other issues related to water resources in Delhi and other cities.

Trend analysis and forecasting of streamflow using random forest in the Punarbhaba River basin.

Streamflow rate changes due to damming are hydro-ecologically sensitive in present and future times. Very less studies have done an investigation of the damming effect on the streamflow along with future forecasting, which can be the solution for the existing problems. Therefore, this study aims to use the Pettitt test as well as standard normal homogeneity test (SNHT) to discover trends in streamflow with the future situation in the Punarbhaba River in Indo-Bangladesh from 1978 to 2017. Trend was spotted using Mann-Kendall test, Spearman's rank correlation approach, innovative trend analysis, and a linear regression model. The current work additionally uses advanced machine learning techniques like random forest (RF) to estimate flow regimes using historical time series data. 1992 appears to be a yard mark in this continuum of time series datasets, indicating a significant transformation in the streamflow regime. The MK test as well as Spearman's rho was used to find a significant negative trend for the average (-0.57), maximum (-0.62), and minimum (-0.48) flow regimes. The consistency of the flow regime has been losing consistency, and the variability of flow regime has increased from 2.1 to 6.7% of the average water level, 1.5 to 6.5% of the maximum streamflow, and 3.1 to 5.8% of the minimum streamflow in the post-change point phase. The forecast trend using random forest for streamflow up to 2030 are negative for all four seasons with a flow volume likely to be reduced by 0.67% to-5.23%. Annual and monthly streamflows revealed very negative tendencies, according to the conclusions of unique trend analysis. Flow declination of this magnitude impacts downstream habitat and environment. According to future estimates, the seasonal flow will decrease. Furthermore, the outcome of this research will give a wealth of data for river management and other places with comparable environment.

Monitoring drought pattern for pre- and post-monsoon seasons in a semi-arid region of western part of India.

Drought has become a regular phenomenon in the western semi-arid regions of India, where severe drought occurs once in 8-9 years. Therefore, two drought indices, namely temperature condition index (TCI) and vegetation condition index (VCI), were prepared from using Landsat datasets to appraise and monitor of drought pattern for the pre- and post-monsoon seasons between 1996 and 2016 in the Latur district, the north-western part of India. Additionally, the average frequency layers (AFL) of all drought and land use indices were prepared to analyse the correlation between them. The results show a substantial increase in the area under high, very high and severe drought classes both pre- and post-monsoon seasons during the study period. The highest increase was noticed from the high drought class from 2532.45 to 4792.49 sq. km and 1559.84 to 3342.32 sq. km for pre- and post-monsoon season, respectively, based on the TCI and 1269.81 to 1787.77 sq. km in very high drought class for the post-monsoon season using the VCI. The correlation analysis showed that there exists a significant relationship between the land use indices and drought indices. However, the spatial pattern of correlation was heterogeneous for both pre- and post-monsoon seasons. The results of this research can help in the drought management and mitigation planning in the study area. In addition, a similar approach may be applied to analyse drought patterns in other places with similar geographic characteristics as both VCI and TCI are cost-effective and less time-consuming methods and produce reliable outcomes.

Groundwater potential mapping of Tawi River basin of Jammu District, India, using geospatial techniques.

The demand for water is increasing around the world due to population growth, urbanization, industrialization, etc., which is making groundwater a vital natural resource for meeting the growing demand for water. According to the central groundwater report, Jammu district has adequate groundwater potential (GWP) and comes under the safe category. However, the GWP has not been fully utilized, thereby leading to a water shortage in the district. Therefore, this study has been designed to examine the GWP zones in the Tawi River basin of Jammu district using geospatial techniques. For this, several GWP conditioning parameters, such as elevation, slope, geology, geomorphology, rainfall, soil, land use/land cover, topographic wetness index (TWI), drainage density, lineament density, roughness, and curvature, were utilized. The analytical hierarchy process (AHP) technique was used to evaluate the weights of the selected criteria after a pair-wise comparison of each criterion with the rest of the criteria. The result showed that the high GWP zone accounts for 21.98% of the area, the moderate zone covers an area of 40.54%, the low GWP area accounts for about 34.91%, and only 2.57% of the area lies under the very low GWP zone. The validation of GWP zones using 25 monitoring wells showed an accuracy of 80% in GWP modeling. The findings of this study may be utilized in meeting the rising demand for water in the region.