Faculty Publications
Permanent URI for this communityhttps://idr.nitk.ac.in/handle/123456789/18736
Publications by NITK Faculty
Browse
5 results
Search Results
Item Feature Elimination and Comparative Assessment of Machine Learning Algorithms for Flood Susceptibility Mapping in Kerala, India(Institute of Electrical and Electronics Engineers Inc., 2023) Kundapura, S.; Aditya, B.; Apoorva, K.V.Floods are a catastrophic phenomenon with far-reaching consequences for infrastructure, the economy, and human lives, profoundly impacting regions globally. This study assesses flood susceptibility in four districts of Kerala: Ernakulam, Idukki, Kottayam, and Alappuzha. For the 2018 storm that caused flooding by Cyclone Ockhi, a flood map for the area was produced using Sentinel 1 satellite data in Google Earth Engine environment. The resulting map served as the foundation for further analysis. Based on the literature review, 16 potential flood causative factors were identified and incorporated into spatial maps in the Geographic Information System (GIS) environment. Analysis of the flood dataset was performed using Machine Learning (ML) algorithms, namely, Random Forest (RF), Decision Tree (DT), Gradient Boosting Machine (GBM), and XG Boost (XGB). Grid search was employed to identify the optimal hyperparameters for each algorithm, ensuring improved performance. Recursive Feature Elimination (RFE) was subsequently applied to select the most influential variables, resulting in a refined dataset. The chosen factors' feature importance scores were obtained, which were used to create the flood susceptibility map using the four ML models in a GIS environment. Evaluation metrics such as F1 score, accuracy, precision, recall, and ROC-AUC score were computed for each model, providing insights into the effectiveness of each algorithm in predicting the flood occurrence. The resulting flood susceptibility map for the best-performing ML model visually represents the varying levels of flood risk in the study area. © 2023 IEEE.Item Mapping of 2018 Flood and Estimation of Future Flood Inundation Region for Vembanad Lake System in Kerala, India Using Sentinel-1 SAR Imagery(Springer Science and Business Media Deutschland GmbH, 2024) Kulithalai Shiyam Sundar, K.S.S.; Kundapura, S.Floods have claimed the lives of countless people and caused significant property damage, jeopardizing their livelihoods. The study area is the Vembanad Lake System in Kerala, India has faced severe flooding in 2018 due to torrential rainfall. Considering that Google Earth Engine (GEE) streamlines and simplifies the complex and time-consuming pre-processing of SAR images, this paper evaluates flood inundation mapping using Sentinel-1 SAR data for 2018. The flood inundation zone for the study is calculated using the Land Use Land Cover (LULC) map for 2018 and the forecasted LULC for 2035 and 2050. Hence, the research assesses the areas affected by floods in 2018 and those that may experience flooding of a similar degree in the near future. Thus, the extent of flood inundation during the 2018 floods and the potential flood inundation region for future LULC in 2035 and 2050 are determined. From the analysis, 14.7 km2 of built-up area was inundated during the 2018 floods. The 2018 flood event is used to quantify the flood that may inundate the future LULC in 2035 and 2050; it is found that the flood will affect about 19.87 km2 and 23.32 km2 of the built-up region, respectively. According to the study, the built-up area impacted by the flooding will increase by 34.99% and 58.4% from 2018 to 2035 and 2050, respectively. Examining the flood-prone areas and potential flood-affected areas in the future will be of great use to planners in their efforts to forewarn of an impending tragedy. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.Item Spatio-temporal Dynamics of Land Use Land Cover Changes and Future Prediction Using Geospatial Techniques(Springer, 2022) Abraham, A.; Kundapura, S.Land use land cover (LULC) plays a key role in earth surface processes, and it is important to understand the spatio-temporal dynamics of LULC in an area. The study is carried out in the Meenachil and Manimala basins in Kerala, India, using land change modeller (LCM) to predict future LULC. The Random Forest (RF) classifier is used to classify the LULC in Google Earth Engine (GEE) for the years 1990, 2000, 2008, 2018 and 2021. The overall accuracy obtained for the years 1990, 2000, 2008, 2018 and 2021 is 92.53%, 91.42%, 96.92%, 87.79% and 95.54%, respectively, followed by a Kappa coefficient of 90.67%, 89.27%, 96.12%, 84.55% and 94.39%. LCM is utilised for LULC change detection, the model is validated successfully in predicting the LULC distribution in 2021, and the results were compared with the actual 2021 LULC. The results revealed the expansion of the built-up area and the decline of the agriculture class in these basins. The study then utilised LCM to predict future LULC up to the year 2050 at decadal intervals. The predicted future LULC maps revealed the drastic expansion of built-up; these basins might witness in the coming decades. The built area from 1990 to 2050 is expected to increase to 100.88 km2 and 60.75 km2 in Meenachil and Manimala basins, respectively. The agriculture area showed a decrease from 861.7 to 728.29 km2 in Meenachil and 743.5–676.89 km2 in Manimala basin. The outcome of the study showed the transformation of the considered land cover classes due to developmental activities in the region. The outcomes of the study can be considered as suitable inputs to land use planners for effective land use planning and management. © 2022, Indian Society of Remote Sensing.Item Spatiotemporal variation in the water quality of Vembanad Lake, Kerala, India: a remote sensing approach(Springer Science and Business Media Deutschland GmbH, 2023) Kulithalai Shiyam Sundar, K.S.S.; Kundapura, S.Water quality is one of the essential parameters of environmental monitoring; even a slight variation in its characteristics may significantly influence the ecosystem. The water quality of Vembanad Lake is affected by anthropogenic effects such as industrial effluents and tourism. The optical parameters representing water quality, such as diffuse attenuation (Kd), turbidity, suspended particulate matter (SPM), and chlorophyll-a (Chl-a), are considered in this study to evaluate the water quality of Vembanad Lake, Kerala, India. As this lake is regarded as of ecological importance by the Ramsar Convention and has faced severe concerns over recent years, there was a substantial change in the water quality during the lockdowns of the COVID-19 pandemic. This research is aimed at examining the change in water quality using optical data from Sentinel-2 satellites in the ACOLITE processing software from 2016 to 2021. The analyses showed a 2.5% decrease in the values of Kd, whereas SPM and turbidity show a reduction of about 4.3% from the year 2016 to 2021. The flood and the COVID lockdown had an impact on the improvement in the quality of water from 2018 to 2021. The findings indicated that the reduction in industrial activities and tourism had a more significant effect on the improvement in the water quality of the lake. There was no substantial change in the Chl-a until 2020, whereas an average decrease of 12% in Chl-a values was observed throughout 2021. This decrease can be attributed to the reduction in the lake’s hydrological residence time (HRT). Thus, these findings will be a valuable reference to help the government and non-government organizations (NGO) during strategic planning. © 2023, The Author(s), under exclusive licence to Springer Nature Switzerland AG.Item Spatial Mapping of Flood Susceptibility Using Decision Tree-Based Machine Learning Models for the Vembanad Lake System in Kerala, India(American Society of Civil Engineers (ASCE), 2023) Kulithalai Shiyam Sundar, P.; Kundapura, S.Floods have claimed the lives of countless people and caused significant property damage in many countries, putting their livelihoods in the jeopardy. The Vembanad lake system (VLS) in Kerala, India, has faced adverse mishappening during 2018, 2019, and 2021 floods in the state due to torrential rainfall. The goal of this research is to construct effective decision tree-based machine learning models such as adaptive boosting (AdaBoost), random forest (RF), gradient boosting machines (GBMs), and extreme gradient boosting (XGBoost) for integrating data, processing, and generating flood susceptibility maps. There are 18 conditioning parameters considered, which include seven categories and 11 numerical data. These seven categorical data were converted to numerical data, bringing the total amount of input data to 61. The recursive feature elimination (RFE) was utilized as the feature selection technique, and a total of 22 layers were chosen to feed into the machine learning models to generate the flood susceptibility maps. The efficiencies of the models were evaluated using receiver operating characteristic (ROC)-area under the ROC curve (AUC), F1 score, accuracy, and kappa. According to the results, the performance of all four models demonstrated their practical application; however, XGBoost fared well in terms of the model's metrics. For the testing data set, the ROC-AUC values of XGBoost, GBM, and AdaBoost are 0.90, whereas it was 0.89 for RF. The accuracy varied significantly among the four models, with XGBoost scoring 0.92, followed by GBM (0.88), RF (0.87), and AdaBoost (0.87). As a result, this map may be utilized for early mitigation actions during future floods, as well as for land-use planners and emergency managers, assisting in the reduction of flood risk in regions prone to this hazard. © 2023 American Society of Civil Engineers.