Groundwater contamination and vulnerability in urbanized areas are of major concern and need proper attention. Several models including the DRASTIC model are used to evaluate groundwater vulnerability. In the present study, a modified DRASTIC model named as DRASTICA was used, by including anthropogenic influence as a model parameter. The study included an innovative methodology to characterize the anthropogenic influence by using satellite observations of night-lights from human settlements as a proxy and land-use/landcover surrounding the urbanized area in Lucknow, the capital city of the most populous State of Uttar Pradesh in India. Geographical information system was used for spatial integration of different parameter maps. The groundwater vulnerability to pollution indicated that about 0.7 % area is covered under very high vulnerable zone, 24.5 % area under high vulnerable zone, 66.6 % area under moderately vulnerable zone and 8.2 % area under low vulnerable zone. The results were validated using nitrate concentration in ground water. It was shown that the proposed DRASTICA model performed better than conventional DRASTIC model in an urbanized environment. Sensitivity analysis indicated that anthropogenic impact and depth to water table largely influenced the groundwater vulnerability to pollution, thereby signifying that anthropogenic influence has to be addressed precisely in such studies. The modified-DRASTIC/DRASTICA model proposed in this study will help in better categorization of groundwater vulnerable zones to pollution where anthropogenic contamination is high, particularly in and around urban centers.
Machine learning classifiers are being increasingly used nowadays for Land Use and Land Cover (LULC) mapping from remote sensing images. However, arriving at the right choice of classifier requires understanding the main factors influencing their performance. The present study investigated firstly the effect of training sampling design on the classification results obtained by Random Forest (RF) classifier and, secondly, it compared its performance with other machine learning classifiers for LULC mapping using multi-temporal satellite remote sensing data and the Google Earth Engine (GEE) platform. We evaluated the impact of three sampling methods, namely Stratified Equal Random Sampling (SRS(Eq)), Stratified Proportional Random Sampling (SRS(Prop)), and Stratified Systematic Sampling (SSS) upon the classification results obtained by the RF trained LULC model. Our results showed that the SRS(Prop) method favors major classes while achieving good overall accuracy. The SRS(Eq) method provides good class-level accuracies, even for minority classes, whereas the SSS method performs well for areas with large intra-class variability. Toward evaluating the performance of machine learning classifiers, RF outperformed Classification and Regression Trees (CART), Support Vector Machine (SVM), and Relevance Vector Machine (RVM) with a >95% confidence level. The performance of CART and SVM classifiers were found to be similar. RVM achieved good classification results with a limited number of training samples.
The Western Ghats (WG) of India, one of the hottest biodiversity hotspots in the world, has witnessed major land-use and land-cover (LULC) change in recent times. The present research was aimed at studying the patterns of LULC change in WG during 1985-1995-2005, understanding the major drivers that caused such change, and projecting the future (2025) spatial distribution of forest using coupled logistic regression and Markov model. The International Geosphere Biosphere Program (IGBP) classification scheme was mainly followed in LULC characterization and change analysis. The single-step Markov model was used to project the forest demand. The spatial allocation of such forest demand was based on the predicted probabilities derived through logistic regression model. The R statistical package was used to set the allocation rules. The projection model was selected based on Akaike information criterion (AIC) and area under receiver operating characteristic (ROC) curve. The actual and projected areas of forest in 2005 were compared before making projection for 2025. It was observed that forest degradation has reduced from 1985-1995 to 1995-2005. The study obtained important insights about the drivers and their impacts on LULC simulations. To the best of our knowledge, this is the first attempt where projection of future state of forest in entire WG is made based on decadal LULC and socio-economic datasets at the Taluka (sub-district) level.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.