This is a Preprint and has not been peer reviewed. This is version 1 of this Preprint.
Satellite-Based Assessment of Terrestrial Water Storage Change and Groundwater Stress in Zhob District, Balochistan
Downloads
Authors
Abstract
Groundwater underpins domestic supply and irrigated agriculture across much of Balochistan, Pakistan, and satellite-based subsidence studies have documented severe groundwater-related stress around Quetta, Gulistan, Pishin, Mastung, and Kalat districts. Zhob District, in northern Balochistan, is comparably dependent on tubewell and karez irrigation but has not been covered by any published satellite-based subsidence or storage-trend study identified in this review. This study assembles three independent, freely available satellite datasets — NASA GRACE/GRACE-FO terrestrial water storage (TWS) anomalies (2002–2024), Sentinel-1 SAR VV backscatter amplitude (2014–2026), and Sentinel-2 NDVI/NDWI surface indices (2016–2025) — processed within Google Earth Engine, and subjects them to an independent data audit and a battery of trend statistics (ordinary least squares, Theil–Sen, Mann-Kendall, and seasonal Mann-Kendall) rather than relying on a single period-mean comparison.
Careful audit of the four underlying satellite time series found that the GRACE-derived "depletion rate" reported in early working calculations (-0.23 cm/year) was obtained by dividing a period-mean difference by an assumed year gap rather than from a validated trend estimator; formal trend tests on the full GRACE record (OLS, Theil–Sen, Mann-Kendall) are not statistically significant (p > 0.5). The period-mean difference itself (-2.4 to -2.5 cm between 2002–2016 and 2017–2024) is robust to alternative period boundaries and is consistent in direction with independent, verified hydrogeological literature on groundwater decline in upland Balochistan, but should be read as evidence of a regional TWS shift consistent with, not proof of, groundwater depletion in Zhob specifically, since GRACE integrates groundwater, soil moisture, and surface water at ~300 km resolution. Sentinel-1 amplitude shows no statistically detectable trend (Theil–Sen 95% CI spans zero), consistent with its known insensitivity to subsidence-scale deformation. Sentinel-2 indices, examined with seasonal Mann-Kendall to control for the strong annual cycle, show small but statistically significant trends that a simple period-mean comparison alone would have missed: NDVI increasing (greening, p < 0.001) and NDWI decreasing (drying, p < 0.001) over 2016–2025, a pattern discussed as a hypothesis-level finding possibly consistent with agricultural intensification rather than reduced irrigation.
Independent, newly verified literature — including an ADB (2017) SWAT-based water balance for the Zhob River Basin and pre-InSAR GPS subsidence work in Quetta (Khan et al., 2013) — is incorporated to correct an overstated novelty claim and to provide basin-scale hydrological context. The paper concludes that current evidence is consistent with, but does not independently prove, groundwater-related stress specific to Zhob District, and identifies phase-based InSAR processing, local well validation, and quantitative rainfall/soil-moisture decomposition (not performed here owing to data-access constraints) as the necessary next steps.
DOI
https://doi.org/10.31223/X54R4R
Subjects
Earth Sciences, Environmental Sciences
Keywords
terrestrial water storage; GRACE/GRACE-FO; Sentinel-1; Sentinel-2; NDVI; NDWI; Google Earth Engine; Balochistan; Zhob District; groundwater stress; data audit; trend statistics
Dates
Published: 2026-08-20 13:21
Last Updated: 2026-08-20 13:21
License
CC BY Attribution 4.0 International
Additional Metadata
Conflict of interest statement:
None
Data Availability:
The four satellite-derived time series (GRACE LWE thickness, Sentinel-1 VV backscatter, Sentinel-2 NDVI and NDWI) analyzed in this study were exported from Google Earth Engine and are available from the corresponding author upon reasonable request. The Google Earth Engine processing scripts are available from the corresponding author.
Metrics
Views: 37
Downloads: 2
There are no comments or no comments have been made public for this article.