GRACE Total Water Storage Research Articles

Changes in groundwater levels across China from 2005 to 2016

Groundwater provides perennial, ubiquitous and high-quality freshwater supplies for humanity globally. Difficulties in the collection and collation of groundwater records have hindered our understanding of large-scale groundwater dynamics. This study collected groundwater level depth (GLD) records from 909 groundwater wells across China and the spatiotemporal changes in GLD were investigated in unconfined and confined aquifers from 2005 to 2016 nationally. The GLD distributions were highly heterogeneous, with average (standard deviation) GLDs of 11.9 (20.3) and 21.6 (26.4) m in unconfined and confined aquifers, respectively. From 2005 to 2016, the GLDs decreased significantly over the energy-limited region due to increased precipitation but increased significantly over the water-limited region due to high groundwater abstraction. The GLD changes in confined aquifers were larger than those in unconfined aquifers due to the smaller storativity in confined aquifers. The correlations between the GLDs and GRACE total water storage (TWS) anomalies over the water-limited region were higher than those over the energy-limited region, indicating that groundwater storage contributed more to TWS over arid areas than humid areas. This study highlights the increasing groundwater storage in the energy-limited region and the decreasing groundwater storage in the water-limited region, i.e., enhanced regional differences in groundwater resources. The results improve our understanding of groundwater dynamics in different aquifers and climatic conditions and hence advance robust groundwater management and policy development in China.

Revising precipitation – water storages – vegetation signatures with GRACE-based data assimilation

• Assessing modeled water – vegetation dynamics by observation-based data sets. • Observation-based data derived from GRACE assimilation and remote sensing. • Duration for precipitation refilling water storages is shorter modeled than observed. • Model overestimates water storage amount contributing to vegetation growth. • Insights will help modelers improve model structures; e.g. climate events like ENSO. The availability of freshwater is highly influenced by climate change, extreme climate events and by anthropogenic use. Countries where a large part of the population depends on the agricultural sector, such as South Africa, are strongly affected by changes in climate, which emphasizes that water is an essential source for food production and drinking water. To analyze changes in surface and subsurface water, model simulations and in situ data are commonly used. However, both have limitations, for example, the models rely on potentially erroneous forcing data and insufficient process representations and the in situ data do not represent the larger-scale weather and climate due to spatial and temporal heterogeneity. This can be mitigated by assimilating remote-sensed satellite data into models. In this research, we build a realistic picture of the water and its propagation (measured between peak times) from fluxes as precipitation, to its way through the storages and its impact on vegetation at the 50 km scale by using observation-based data. The observations are derived from MODIS remote-sensing and integrating GRACE total water storage anomaly (TWSA) observations into a hydrological model via data assimilation. Our objective is to identify shortcomings in model simulations by confronting them with the (synthesized) observations. Moreover, we demonstrate the importance of integrating observations into the models. We base these comparisons on signatures or sub-signals (e.g., temporal lags and annual amplitudes) that we derive via regression analysis, principal component analysis, sensitivity analysis and correlation analysis from the synthetic data and the model output. Our main results show that correlations and signatures in real observations are found weaker as compared to what is simulated in the model, e.g. for the contribution of precipitation to groundwater. Lag times between precipitation and surface and groundwater storage peaks are observed to be longer than in the model. The observed propagation of soil water from storages to vegetation is often shorter than in the model, while for groundwater it is longer. We believe our findings will be highly relevant for modelers; the gained knowledge can be used to improve models. In addition, we feel our study underlines the potential of GRACE assimilation into hydrological models.

Recent Changes in Groundwater and Surface Water in Large Pan-Arctic River Basins

Surface and groundwater in large pan-Arctic river basins are changing rapidly. High-quality estimates of these changes are challenging because of the limits on the data quality and time span of satellite observations. Here, the term pan-Arctic river refers to the rivers flowing to the Arctic Ocean basin. In this study, we provide a new evaluation of groundwater storage (GWS) changes in the Lena, Ob, Yenisei, Mackenzie and Yukon River basins from the GRACE total water storage anomaly product, in situ runoff, soil moisture form models and a snow water equivalent product that has been significantly improved. Seasonal Trend decomposition using Loess was utilized to obtain trends in GWS. Changes in surface water (SW) between 1984 and 2019 in these basins were also examined based on the Joint Research Centre Global Surface Water Transition data. Results suggested that there were great GWS losses in the North American river basins, totaling approximately −219 km3, and GWS gains in the Siberian river basins, totaling ~340 km3, during 2002–2017. New seasonal and permanent SWs are the primary contributors to the SW transition, accounting for more than 50% of the area of the changed SW in each basin. Changes in the Arctic hydrological system will be more significant and various in the case of rapid and continuous changes in permafrost.

Estimating the added value of GRACE total water storage and uncertainty quantification in seasonal streamflow forecasting

ABSTRACT Seasonal streamflow prediction is vital for water resources management and disaster mitigation. Multiple linear regression (MLR), as the simplest yet effective statistical model, is widely used in hydrological forecasting. To investigate the added value of incorporating GRACE total water storage anomalies (TWSA) in seasonal streamflow prediction and quantifying the uncertainty in model residuals, this study constructed four MLR models, namely a benchmark model, TWSA-incorporated model, uncertainty quantified model and synthetic model, to predict monthly streamflow in Yarlung Zangbo River, Upper Jinshajiang River, Xiangjiang River and Lanjiang River with lead times from 1 to 11 months. The results show that MLR models perform fairly well, with correlation coefficients greater than 0.8 and Nash-Sutcliffe efficiency coefficients up to 0.73. TWSA contributes moderately to forecast skill, whilst about 15% of the forecast skill is attributed to uncertainty quantification. The synergy of TWSA and uncertainty quantification brings basin-varying contributions, ranging from 5% to 26%.

Multi-decadal assessment of water budget and hydrological extremes in the Tigris-Euphrates Basin using satellites, modeling, and in-situ data

The transboundary Tigris-Euphrates Basin (TEB) is prone to water-scarcity disputes. Water scarcity is related to aridity, climate extremes, limited supplies, upstream reservoir storage, rising water demand, and population growth. Understanding the water budget and storage changes in the basin in relation to hydrological extremes is fundamental to mitigate the drought and flood impacts and the key to efficient water resources management. This study evaluated the water budget related to drought occurrences in the TEB over four decades (1979–2020) based on GRACE/GRACE-FO, and altimetry satellites data, in situ observations, and hydrological modeling using a Bayesian model averaging (BMA) approach. Results show that severe droughts occurred at about decadal timescales with increasing recovery times. Severe and exceptional droughts dominated from (1998 to 2000, 2007 to 2009). Mild to moderate droughts occurred in 1983–1984, 1989–1992, 2011–2013, and 2018. The most severe drought occurred in 2007–2009, with the largest decline (−80 km3) in GRACE total water storage (TWS). Depletion in TWS was dominated by depletion in reservoir storage. In contrast, groundwater (GW) depletion accounted for only 25–30% of TWS decline. Storage depletion was amplified by human intervention (e.g., irrigation and GW abstraction) by at least 50% during drought. Marked recovery in TWS occurred in 2019 and 2020 (totaling ~144 km3 by July 2020, representing ~2× total depletion between 2007 and 2018) in response to regional flooding. Applying the BMA approach to the estimates of water cycle fluxes improved the accuracy and similarity of storage change, but not variability relative to GRACE. In summary, prolonged droughts are the norm rather than the exception in the TEB over the past four decades. The frequency and severity of droughts have substantial implications for water scarcity for countries sharing the TEB and underscore riparian countries' needs to expand their water management portfolio to mitigate drought impacts.

Forecasting terrestrial water storage for drought management in Ethiopia

ABSTRACT Reliable seasonal forecasting of water resources variability may be of great value for agriculture and energy management in Ethiopia. This work aims to develop statistical forecasting of seasonal total water storage (TWS) anomalies in Ethiopia using sea-surface temperature and sea-level pressure indices. Because of the spatial and temporal variability of TWS over the country, Ethiopia is divided into four regions each having similar TWS dynamics. Periods of long-term water deficit observed in GRACE TWS products for the region are found to coincide with periods of meteorological drought. Multiple linear regression is employed to generate seasonal forecasting models for each region. We find that the skill of the resulting models varies from region to region, with R 2 from 0.33 to 0.73 and correlation from 0.27 to 0.77 between predicted and observed values (using leave-one-out cross-validation). The skill of the models is better than the climatology in all regions.

INVESTIGATION OF THE CORRELATION BETWEEN GRACE TWS AND SOIL MOISTURE IN SARAKHS CATCHMENT

Abstract. Drought is a gradual phenomenon that may intensify with time. In fact, drought is the result of interactions between natural and human activities and usually appears in the following ways: reduction in water resources, accelerating desertification, the vegetation cover change and etc. Monitoring and predicting drought is essential for sustainable water management and mitigating socio-economic damages. The GRACE satellite mission provided monthly Earth’s gravity field anomalies, which can be processed to obtain changes in Total Water Storage (TWS) content. TWS is a sum total of various compartments of water near the surface of the Earth, such as soil moisture, surface water, snow, etc. In this study, we analyze GRACE TWS and the modelled soil moisture (SM) over Sarakhs catchment in north-east of Iran, between 2003–2016. We find a time lag of 19 months between GRACE TWS and soil moisture, where both TWS and soil moisture content show negative trend over Sarakhs catchment. Our findings are similar to other reports on the water scarcity in this region.

River Discharge Estimation based on Satellite Water Extent and Topography: An Application over the Amazon

Abstract River discharge (RD) estimates are necessary for many applications, including water management, flood risk, and water cycle studies. Satellite-derived long-term GIEMS-D3 surface water extent (SWE) maps and HydroSHEDS data, at 90-m resolution, are here used to estimate several hydrological quantities at a monthly time scale over a few selected locations within the Amazon basin. Two methods are first presented to derive the water level (WL): the “hypsometric curve” and the “histogram cutoff” approaches at an 18 km × 18 km resolution. The obtained WL values are interpolated over the whole water mask using a bilinear interpolation. The two methods give similar results and validation with altimetry is satisfactory, with a correlation ranging from 0.72 to 0.89 in the seven considered stations over three rivers (i.e., Wingu, Negro, and Solimoes Rivers). River width (RW) and water volume change (WVC) are also estimated. WVC is evaluated with GRACE total water storage change, and correlations range from 0.77 to 0.88. A neural network (NN) statistical model is then used to estimate the RD based on four predictors (SWE, WL, WVC, and RW) and on in situ RD measurements. Results compare well to in situ measurements with a correlation of about 0.97 for the raw data (and 0.84 for the anomalies). The presented methodologies show the potential of historical satellite data (the combination of SWE with topography) to help estimate RD. Our study focuses here on a large river in the Amazon basin at a monthly scale; additional analyses would be required for other rivers, including smaller ones, in different environments, and at higher temporal scale.

Total Water Storage in Small Islands Using Multi-satellite Products: The case of Cyprus

The world’s longest international undersea water pipeline has recently been laid from southern Turkey to Northern Cyprus to address the water scarcity problems in the Cyprus Island. When completed, the project will add 19.8 million gallons of water annually for drinking and irrigation. Moreover, it will spur further development in the Turkish area divided from Greek community for four decades. Under such circumstances, satellite remote sensing provides a unique tool to evaluate the water policy for the entire island. The objective of this study is, therefore, to examine the potential of satellite-based remote sensing hydrologic models covering a small-scale Mediterranean island, which is under drought conditions. Satellite-based measurements such as GRACE (total water storage), TRMM (precipitation) and MODIS (evapotranspiration) data over a 1°–1° grid together with related information from global hydrological model, specifically WGHM and GLDAS, were collected for this purpose.

Exploring hydro-meteorological drought patterns over the Greater Horn of Africa (1979–2014) using remote sensing and reanalysis products

Spatio-temporal patterns of hydrological droughts over the Greater Horn of Africa (GHA) are explored based on total water storage (TWS) changes derived from time-variable gravity field solutions of Gravity Recovery And Climate Experiment (GRACE, 2002–2014), together with those simulated by Modern Retrospective Analysis for Research Application (MERRA, 1980–2014). These hydrological extremes are then related to meteorological drought events estimated from observed monthly precipitation products of Global Precipitation Climatology Center (GPCC, 1979–2010) and Tropical Rainfall Measuring Mission (TRMM, 1998–2014). The major focus of this contribution lies on the application of spatial Independent Component Analysis (sICA) to extract distinguished regions with similar rainfall and TWS with similar overall trend and seasonality. Rainfall and TWS are used to estimate Standard Precipitation Indices (SPIs) and Total Storage Deficit Indices (TSDIs), respectively that are employed to characterize frequency and intensity of hydro-meteorological droughts over GHA. Significant positive (negative) changes in monthly rainfall over Ethiopia (Sudan) between 2002 and 2010 leading to a significant increase in TWS over the central GHA region were noted in both MERRA and GRACE TWS (2002–2014). However, these trends were completely reversed in the long-term (1980–2010) records of rainfall (GPCC) and TWS (MERRA). The four independent hydrological sub-regions extracted based on the sICA (i.e., Lake Victoria Basin, Ethiopia-Sudanese border, South Sudan, and Tanzania) indicated fairly distinct temporal patterns that matched reasonably well between precipitation and TWS changes. While meteorological droughts were found to be consistent with most previous studies in all sub-regions, their impacts are clearly observed in the TWS changes resulting in multiple years of extreme hydrological droughts. Correlations between SPI and TSDI were found to be significant over Lake Victoria Basin, South Sudan, and Tanzania. The low correlations between SPI and TSDI over Ethiopia are likely related to inconsistency between TWS and precipitation signals. Further, we found that hydrological droughts in these regions were significantly associated with Indian Ocean Dipole (IOD) events while El Niño Southern Oscillation (ENSO) plays a secondary role.

Deriving scaling factors using a global hydrological model to restore GRACE total water storage changes for China's Yangtze River Basin

This study used a global hydrological model (GHM), PCR-GLOBWB, which simulates surface water storage changes, natural and human induced groundwater storage changes, and the interactions between surface water and subsurface water, to generate scaling factors by mimicking low-pass filtering of GRACE signals. Signal losses in GRACE data were subsequently restored by the scaling factors from PCR-GLOBWB. Results indicate greater spatial heterogeneity in scaling factor from PCR-GLOBWB and CLM4.0 than that from GLDAS-1 Noah due to comprehensive simulation of surface and subsurface water storage changes for PCR-GLOBWB and CLM4.0. Filtered GRACE total water storage (TWS) changes applied with PCR-GLOBWB scaling factors show closer agreement with water budget estimates of TWS changes than those with scaling factors from other land surface models (LSMs) in China's Yangtze River basin. Results of this study develop a further understanding of the behavior of scaling factors from different LSMs or GHMs over hydrologically complex basins, and could be valuable in providing more accurate TWS changes for hydrological applications (e.g., monitoring drought and groundwater storage depletion) over regions where human-induced interactions between surface water and subsurface water are intensive.

Monitoring the water balance of Lake Victoria, East Africa, from space

Summary Using satellite gravimetric and altimetric data, we examine trends in water storage and lake levels of multiple lakes in the Great Rift Valley region of East Africa for the years 2003–2008. GRACE total water storage estimates reveal that water storage declined in much of East Africa, by as much as 60 mm year , while altimetric data show that lake levels in some large lakes dropped by as much as 1–2 m. The largest declines occurred in Lake Victoria, the Earth’s second largest freshwater body. Because the discharge from the outlet of Lake Victoria is used to generate hydroelectric power, the role of human management in the lake’s decline has been questioned. By comparing catchment water storage trends to lake level trends, we confirm that climatic forcing explains only about 50decline. This analysis provides an independent means of assessing the relative impacts of climate and human management on the water balance of Lake Victoria that does not depend on observations of dam discharge, which may not be publically available. In the second part of the study, the individual components of the lake water balance are estimated. Satellite estimates of changes in lake level, precipitation, and evaporation are used with observed lake discharge to develop a parameterization for estimating subsurface inflows due to changes in groundwater storage estimated from satellite gravimetry. At seasonal timescales, this approach provides closure to Lake Victoria’s water balance to within 17 mm month . The third part of this study uses the water balance of a downstream water body, Lake Kyoga, to estimate the outflow from Lake Victoria remotely. Because Lake Kyoga is roughly 20 times smaller in area than Lake Victoria, its water balance is strongly influenced by inflow from Lake Victoria. Lake Kyoga has been shown to act as a linear reservoir, where its outflow is proportional to the height of the lake. This model can be used with satellite altimetric lake levels to estimate a time series of Lake Victoria discharge with an rms error of about 134 m 3 s .