Prashanth et al. (2025) Unveiling hidden perspectives: Examining COVID-19 impact on non-perennial river flow across ungauged river segments and anthropogenic footprint on the hydrological cycle

Identification

• Journal: Journal of Hydrology Regional Studies
• Year: 2025
• Date: 2025-11-20
• Authors: Thallam Prashanth, Sayantan Ganguly
• DOI: 10.1016/j.ejrh.2025.102959

Research Groups

• Department of Civil Engineering, Indian Institute of Technology Ropar, India

Short Summary

This study investigates the impact of reduced human interference during the COVID-19 pandemic on the rejuvenation potential of the ungauged lower Pennar River, India. It found that the river's rejuvenation potential significantly increased during the lockdown due to decreased groundwater extraction, highlighting human interventions as the primary driver of groundwater drought in the region.

Objective

• Identify the drivers responsible for groundwater drought.
• Identify ungauged river segments influenced by groundwater drought by integrating machine learning, remote sensing, and hydrological drought indices.
• Determine River Rejuvenation Susceptibility (RRS) using the Analytical Hierarchy Process (AHP) and Fuzzy AHP (FAHP) methods.
• Validate RRS using depth-duration curves and Standardized Total Water Storage Anomalies (STWSA) obtained from GRACE satellites.

Study Configuration

• Spatial Scale: Pennar River Basin, India, specifically the ungauged lower Pennar River segments within the YSR Kadapa and SPSR Nellore districts. The basin covers approximately 54,781.06 square kilometers.
• Temporal Scale: Data analysis spans from 2000 to 2023, with specific periods for pre-COVID (2010–2020), during COVID, and post-COVID assessments. Landsat data were used for the period 2001–2023.

Methodology and Data

• Models used:
  • Standardized Precipitation Index (SPI)
  • Standardized Potential Evapotranspiration Index (SPEI)
  • Standardized Groundwater Table Index (SGWTI)
  • Satellite-derived Bathymetry (SDB) models
  • Machine Learning (ML) algorithms: Support Vector Regression (SVR), Random Forest (RF), Decision Trees (DT), eXtreme Gradient Boosting (XG–Boost) regression (RF was identified as the optimal model)
  • Analytical Hierarchy Process (AHP)
  • Fuzzy AHP (FAHP)
  • Spontaneous Rejuvenation Potential Number (SRPN)
  • Water Table Fluctuation (WTF) method
  • Rainfall Infiltration Factor (RIF) method
  • Pearson correlation analysis
  • Kolmogorov-Smirnov (KS) test
  • Hargreaves and Samani (1985) equation for Potential Evapotranspiration (PET)
  • Monte Carlo Simulation (MCS)
  • First-order error propagation algorithms
• Data sources:
  • Satellite-derived bathymetry (SDB)
  • GRACE satellites (for Standardized Total Water Storage Anomalies - STWSA)
  • Landsat-5, 8 Top-Of-Atmosphere (TOA) data
  • ALOS PALSAR Digital Elevation Model (DEM) (12.5 meters spatial resolution)
  • EGM-2008 (Earth Gravitational Model)
  • India Meteorological Department (IMD) (rainfall and temperature data, 0.25° x 0.25° and 1° x 1° resolution respectively)
  • Central Ground Water Board (CGWB) (seasonal and monthly groundwater level data, groundwater extraction data, specific yield, infiltration factor, geological reports)
  • Andhra Pradesh State Groundwater and Water Audit Department (monthly groundwater level data)
  • Central Water Commission (CWC)-Pennar Sub-Division office (measured Surface Water Level (SWL) and streamflow data)
  • HydroRIVER and HydroBASIN websites (river network and basin boundary)
  • Bhukosh website (Geological Society of India) (geology data)
  • NASA Earth Data website (GRACE data, 1° spatial resolution)

Main Results

• A low to moderate correlation (0–0.5) between meteorological drought indices (SPEI, SPI) and the groundwater drought index (SGWTI) indicates that groundwater drought in the Pennar River Basin is predominantly influenced by anthropogenic activities (e.g., pumping) rather than climatic factors.
• Groundwater over-extraction (>70% stage of extraction) was observed in the Kadapa district, while Nellore district showed moderate groundwater usage.
• The Random Forest (RF) model demonstrated optimal performance for estimating water depth from satellite band reflectance.
• A strong correlation (>0.95) between the head difference (Groundwater Table - Surface Water Level, GWT-SWL) and SGWTI in ungauged river segments confirmed the high susceptibility of the lower Pennar River to groundwater drought.
• The River Rejuvenation Susceptibility (RRS) at Chennur (Kadapa) increased from "low susceptible" during pre-COVID (2010–2020) to "highly susceptible" during the COVID-19 lockdown.
• In the Nellore region, RRS increased from "moderately susceptible" (third section) or "low susceptible" (first and second sections) pre-COVID to "highly" or "moderately susceptible" during and post-COVID.
• This increase in RRS during the COVID-19 lockdown was attributed to a significant reduction in groundwater extraction, validated by depth-duration curves and STWSA data from GRACE.
• The lower Pennar River, which had transitioned from intermittent to ephemeral before COVID-19, showed signs of recovery during the lockdown period.
• The overall relative uncertainty in the RRS estimate was 8.08%, with GWT-SWL contributing the most uncertainty (36.89%), followed by SGWTI (21.78%), SRPN (17.05%), SPEI3 (7.75%), and SPI3 (3.85%).

Contributions

• Developed an innovative framework integrating remote sensing and machine learning to analyze groundwater–surface water interactions in ungauged river segments.
• Introduced a novel approach to assess Spontaneous Rejuvenation Potential Number (SRPN) and River Rejuvenation Susceptibility (RRS) in ungauged rivers using Satellite-derived Bathymetry (SDB).
• Quantified the dominant role of anthropogenic activities, particularly groundwater extraction, as the primary driver of groundwater drought in the Pennar River Basin.
• Demonstrated the significant positive impact of reduced human activity during the COVID-19 lockdown on river rejuvenation potential, illustrating a temporary recovery from degradation.
• Provided a robust methodology and framework that can be applied globally to evaluate the ecological resilience and natural health of river systems.

Funding

The authors declare that no funds, grants, or other assistance were received which supported the study presented in this paper.

Citation

@article{Prashanth2025Unveiling,
  author = {Prashanth, Thallam and Ganguly, Sayantan},
  title = {Unveiling hidden perspectives: Examining COVID-19 impact on non-perennial river flow across ungauged river segments and anthropogenic footprint on the hydrological cycle},
  journal = {Journal of Hydrology Regional Studies},
  year = {2025},
  doi = {10.1016/j.ejrh.2025.102959},
  url = {https://doi.org/10.1016/j.ejrh.2025.102959}
}