Collaborating actors

• Department of Soil & Water Conservation Engineering, GBPUAT Pantnagar
• Indian Institute of Technology Delhi (IIT Delhi)
• Dr. Rajendra Prasad Central Agricultural University, Pusa
• Central University of Jharkhand, India
• Assam University, Silchar

Remote sensing can significantly aid in groundwater resource management. Further with the integration of Internet of Things (IoT), the information of groundwater storage and change in groundwater level can be shared through mobile technology to end users, policy makers and also to the government.

Here are a few key steps showing how it can be useful:

1. Mapping and monitoring land use/land cover: Remote sensing helps identify areas of vegetation, agriculture, urbanisation and water bodies, which influence groundwater recharge and extraction.
2. Identifying potential groundwater zones: Satellite imagery, combined with GIS, can analyse geological, hydrological and geomorphological features to locate promising groundwater zones.
3. Monitoring groundwater levels and storage: Missions like Gravity Recovery and Climate Experiment (GRACE) measure changes in Earth's gravity field, enabling estimation of groundwater storage changes over time.
4. Assessing drought and recharge conditions: Remote sensing provides data on precipitation, soil moisture and evapotranspiration, essential for evaluating recharge potential and drought impacts.
5. Supporting sustainable management: Continuous remote sensing data supports long-term planning, policy-making and sustainable groundwater resource development.
6. Integration of remote sensing with IoT: IoT modules can be developed for groundwater level; total groundwater storage; drought level etc and can be sent to end users using mobile technology.

Requirements

Data

• GRACE & GRACE FO satellite data set
• Central Groundwater Board, India
• Water Resources Information System (WRIS) India
• USGS data sets of remote sensing imagery

Software

• QGIS
• Visual MODFLOW flex
• Hydrologic Engineering Center's River Analysis System (HEC-RAS)
• MATLAB
• R software
• ERDAS IMAGINE

Physical

• Workstation as servers lab for developing IoT for groundwater management

Outline steps to a solution

• Worked on GRACE satellite data and used it in field condition to study groundwater anomalies of few cities of India (completed).
• Developed spatio-temporal maps of Standardized Groundwater Index (SGI) (completed).
• Water quality monitoring of water bodies using remote sensing (in progress).
• Water spread mapping and its monitoring, of various water bodies using remote sensing and artificial intelligence (research is in progress).
• Internet of Things (IoT) models which can link groundwater depletion/anomalies information with the end-users (in progress).

Steps to a solution

1. Study area and data acquisition  

• Study area has to be selected for groundwater monitoring and management.  
• The shapefile of the study area has to be downloaded from government websites or can be ordered on request basis. Gridded GRACE products (Level-3) can be used from the Jet propulsion Laboratory (JPL), the National Aeronautics and Space Administration (NASA) to get the monthly water equivalent thickness data.

2. Development of Standardized Groundwater Index (SGI) for understanding the severity of groundwater anomalies or draught

Standardized Groundwater Index (SGI) is a drought indicator which was developed by Bloomfield and Marchant (2013) to quantify groundwater drought. It is used for estimating groundwater level deficit at any time scale which reflects the extreme drought condition of any location. It is similar to the traditional drought index, Standardized Precipitation Index (SPI) and can be calculated on the same basis like SPI. In SGI, groundwater level data is used for measuring drought condition, instead of precipitation data which is used in SPI. Groundwater time series data obtained from ground observation can be appropriately normalized to evaluating the groundwater drought. SGI values can also be analysed by calculating groundwater deviation from the mean groundwater value (Halder et al., 2020). SGI can be given by following formula

where, K is groundwater level of the respective year; M is long term mean groundwater level of 18 years, σ is standard deviation

3. Gravity Recovery and Climate Experiment satellite for groundwater anomalies study

GRACE was launched by NASA on March 17, 2002. It was a joint mission of NASA and German Aerospace Centre (DLR). The two twin satellites of GRACE are monitored to observe the changes in the Earth's gravity field. GRACE satellite, a first remote sensing satellite which provides an efficient and cost-effective way to map Earth’s gravity field and measure the total groundwater storage changes (TWS) with unprecedented accuracy (Yirdaw et al, 2008). GRACE studies the variation in the gravity which are caused due to effects that include: changes due to deep currents in the ocean; runoff and ground water storage on land mass; exchanges between ice sheets or glaciers and the oceans, and variations of mass within the solid Earth. The distance between the twin satellite as they orbit the Earth help in measuring changes in the Earth's gravity field for each month. From these monthly gravity field, time series of regional mass anomalies can be derived using specially designed averaging function. GRACE mission provides an opportunity to directly measure the total groundwater storage changes and with the help of gravity field data of GRACE drought conditions can also be monitored over a region. GRACE satellite has coarse resolution of 300-400 km and provides data in an interval of 30 days. The distance between the two satellite is about 200 km at a starting altitude of about 500 km. The GRACE gridded TWS products (1˚×1˚) from spherical harmonics are provided by the Centre of Research (CSR) at the University of Texas, the Jet Propulsion Laboratory (JPL) and German Research Centre for Geoscience (GFZ). The gridded products estimate the changes in mass in unit of water equivalent thickness (WET).

4. Machine learning algorithms to model data from GRACE with observed data

Using machine learning (ML) for modelling GRACE  satellite data alongside observation datasets (e.g., in-situ hydrological measurements, meteorological data) is a powerful approach to extract spatiotemporal patterns, downscale or predict terrestrial water storage (TWS) anomalies. Machine learning algorithms like artificial neural network, random forest, support vector machines etc can be used to model satellite data with observed data.

5. IoT for groundwater monitoring and sending information to end users

Using IoT for groundwater level monitoring is an effective way to automate the collection, transmission and dissemination of real-time groundwater data to decision-makers, farmers or the public.

Results

Today, no remote sensors can directly monitor groundwater, a combination of surface features anomalies and gravity data obtained by various satellites, allows for optimal groundwater management. Example satellites for monitoring include: GRACE and its Follow-On mission (GRACE-FO) to study groundwater fluctuations, Landsat, Moderate Resolution Imaging Spectroradiometer (MODIS) etc, groundwater management can be done using space technology.

I developed a Standardized Groundwater Index (SGI) for Bihar state of India which proved to be very important to understand the severity of groundwater problems in that region. The spatio-temporal variation of SGI using geographical information systems (see figure 1) was published in the peer reviewed Journal of the Geological Society of India (Kumar and Kumari, 2024).