Research Projects

The Ganga River Basin is known for its extensive agricultural activities, and it supports millions of people for their livelihood. With the increasing pressure of the large food demands of India, agricultural fertilizer use has enormously increased in the basin, leading to nitrate contamination of aquifers. This environmental issue of increasing nitrate pollution has raised serious concerns for several government bodies, including the Central Ground Water Board (CGWB) and the Public Health Departments (PHEDs). It is therefore important to understand the migration route of nitrate contamination in the aquifers. To monitor nitrate transport and design preventive strategies, groundwater models are often employed in research studies. In several studies, the MODFLOW-groundwater flow model and the MT3DMS-contaminant transport model are coupled to simulate the nitrate transport in groundwater. Conventionally, the MODFLOW model takes groundwater data (head) observed in the dug and bore wells through a water level meter as input, and MT3DMS takes nitrate concentration in the groundwater samples as input. The initial input data for both these models is collected through field visits and laboratory experiments. However, field data is more often collected three to four times to define the starting head and concentrations in the model, and calibrate and validate the model with this data. The field data is collected at coarse time resolution, i.e., after several weeks and months, due to several reasons, including funding availability, remote site locations, labor-intensive data collection, and time-consuming field visits. Therefore, models calibrated and validated with limited field data may not reproduce the actual time-series plot of head and nitrate concentration at several locations in the study area. In other words, this conventional modeling technique is not completely reliable to numerically understand or monitor the nitrate transport in real-world groundwater systems. To overcome this shortcoming, Internet of Things (IoT)-enabled sensors offer great advantages in real-time groundwater and nitrate monitoring in modern times. However, IoT sensors are normally expensive and installed at limited places to monitor the groundwater quality of an aquifer. An integrated technique of combining IoT sensors and groundwater models can improve the model predictions and reduce the overall monitoring cost. Therefore, the proposed research study aims to develop a novel technique that uses real-time IoT sensor data in groundwater flow and nitrate transport modeling. This work will use real-time IoT data to calibrate and validate the MODFLOW and MT3DMS models. The MODFLOW parameters, including aquifer hydraulic conductivity and recharge rate, will be calibrated by linking the IoT sensor-based water level data with the PEST model for parameter estimation. The transport parameters, including longitudinal, transverse, and vertical dispersivities, will be calibrated through simulation-optimization (S/O) model coupling MT3DMS and particle swarm optimization (PSO). The coupled MT3DMS-PSO model will minimize the difference between the IoT sensor data and simulated nitrate concentration. The proposed work will demonstrate the field application of integrated modeling technology combining IoT sensors and groundwater models for real-time nitrate monitoring in the lower Ganga River Basin. The selected study area includes the four administrative blocks, namely Dubrajpur, Suri-I, Suri-II, and Mohammad Bazar, of Birbhum district of West Bengal. The preliminary investigation at 66 sampling locations confirms the nitrate concentration exceeding the WHO permissible limit of 45 mg/L within these blocks, with several points indicating nitrate concentration around 275 mg/l. The proposed IoT sensor-based groundwater modeling can be a benchmark study in the cost-effective real-time monitoring of nitrate transport at a regional scale.