A global reconstruction of climate‐driven subdecadal water storage variability

Humphrey, Vincent; Gudmundsson, Lukas; Seneviratne, Sonia I.

doi:10.1002/2017gl072564

Cited by 99 publications

(112 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(3) the use of empirical orthogonal function decomposition to reconstruct TWS GRACE data over the Amazon basin by examining the correlation between TWS GRACE and water levels over inter-annual and multi-decadal time periods [54]; (4) the use of statistical models to reconstruct global natural-varying TWS GRACE using rainfall and temperature data [55]; and (5) the use of artificial intelligence to predict the TWS GRACE over a large karst plateau in Southwest China using in situ precipitation, monthly mean temperature, and GLDAS soil moisture [56] and to predict the TWS GRACE over West Africa using rainfall, evaporation, surface temperature, net-precipitation, soil moisture, and climate indices [57].…”

Section: Introductionmentioning

confidence: 99%

Forecasting GRACE Data over the African Watersheds Using Artificial Neural Networks

et al. 2019

View full text Add to dashboard Cite

The GRACE-derived terrestrial water storage (TWS GRACE ) provides measurements of the mass exchange and transport between continents, oceans, and ice sheets. In this study, a statistical approach was used to forecast TWS GRACE data using 10 major African watersheds as test sites. The forecasted TWS GRACE was then used to predict drought events in the examined African watersheds. Using a nonlinear autoregressive with exogenous input (NARX) model, relationships were derived between TWS GRACE data and the controlling and/or related variables (rainfall, temperature, evapotranspiration, and Normalized Difference Vegetation Index). The performance of the model was found to be "very good" (Nash-Sutcliffe (NSE) > 0.75; scaled root mean square error (R*) < 0.5) for 60% of the investigated watersheds, "good" (NSE > 0.65; R* < 0.6) for 10%, and "satisfactory" (NSE > 0.50; R* < 0.7) for the remaining 30% of the watersheds. During the forecasted period, no drought events were predicted over the Niger basin, the termination of the latest (March-October 2015) drought event was observed over the Zambezi basin, and the onset of a drought event (January-March 2016) over the Lake Chad basin was correctly predicted. Adopted methodologies generate continuous and uninterrupted TWS GRACE records, provide predictive tools to address environmental and hydrological problems, and help bridge the current gap between GRACE missions.

show abstract

Section: Introductionmentioning

confidence: 99%

Forecasting GRACE Data over the African Watersheds Using Artificial Neural Networks

et al. 2019

View full text Add to dashboard Cite

show abstract

“…The details on the reconstruction method of climate-driven sub-decadal water storage variability can be found in [48,49]. In brief, first, the effect of one day's precipitation flux to later days can be expressed by an exponential decay filter:…”

Section: Climate-driven Water Storage Variabilitymentioning

confidence: 99%

Groundwater Depletion in the West Liaohe River Basin, China and Its Implications Revealed by GRACE and In Situ Measurements

Zhong

Wang

et al. 2018

Remote Sensing

View full text Add to dashboard Cite

Abstract:The West Liaohe River Basin (WLRB) is one of the most sensitive areas to climate change in China and an important grain production base in the Inner Mongolia Autonomous Region of China. Groundwater depletion in this region is becoming a critical issue. Here, we used the Gravity Recovery and Climate Experiment (GRACE) satellite data and in situ well observations to estimate groundwater storage (GWS) variations and discussed the driving factors of GWS changes in the WLRB. GRACE detects a GWS decline rate of −0.92 ± 0.49 km 3 /yr in the WLRB during 2005-2011, consistent with the estimate from in situ observations (−0.96 ± 0.19 km 3 /yr). This long-term GWS depletion is attributed to reduced precipitation and extensive groundwater overexploitation in the 2000s. Long-term groundwater level observations and reconstructed total water storage variations since 1980 show favorable agreement with precipitation anomalies at interannual timescales, both of which are significantly influenced by the El Niño-Southern Oscillation (ENSO). Generally, the WLRB receives more/less precipitation during the El Niño/La Niña periods. One of the strongest El Niño events on record in 1997-1998 and a subsequent strong La Niña drastically transform the climate of WLRB into a decade-long drought period, and accelerate the groundwater depletion in the WLRB after 1998. This study demonstrates the significance of integrating satellite observations, ground-based measurements, and climatological data for interpreting regional GWS changes from a long-term perspective.

show abstract

“…Unlike physically based models, pure data‐driven methods (black box models) seek to establish a regression model between climate forcings (e.g., precipitation and temperature) and GRACE TWS (Humphrey et al, ; Long et al, ; Seyoum & Milewski, ) or between TWS and its various components (Miro & Famiglietti, ; A. Y. Sun, ; Zhang et al, ).…”

Section: Introductionmentioning

confidence: 99%

Combining Physically Based Modeling and Deep Learning for Fusing GRACE Satellite Data: Can We Learn From Mismatch?

Sun

Scanlon

Zhang

et al. 2019

Water Resources Research

158

View full text Add to dashboard Cite

Global hydrological and land surface models are increasingly used for tracking terrestrial total water storage (TWS) dynamics, but the utility of existing models is hampered by conceptual and/or data uncertainties related to various underrepresented and unrepresented processes, such as groundwater storage. The gravity recovery and climate experiment (GRACE) satellite mission provided a valuable independent data source for tracking TWS at regional and continental scales. Strong interests exist in fusing GRACE data into global hydrological models to improve their predictive performance. Here we develop and apply deep convolutional neural network (CNN) models to learn the spatiotemporal patterns of mismatch between TWS anomalies (TWSA) derived from GRACE and those simulated by NOAH, a widely used land surface model. Once trained, our CNN models can be used to correct the NOAH-simulated TWSA without requiring GRACE data, potentially filling the data gap between GRACE and its follow-on mission, GRACE-FO. Our methodology is demonstrated over India, which has experienced significant groundwater depletion in recent decades that is nevertheless not being captured by the NOAH model. Results show that the CNN models significantly improve the match with GRACE TWSA, achieving a country-average correlation coefficient of 0.94 and Nash-Sutcliff efficient of 0.87, or 14% and 52% improvement, respectively, over the original NOAH TWSA. At the local scale, the learned mismatch pattern correlates well with the observed in situ groundwater storage anomaly data for most parts of India, suggesting that deep learning models effectively compensate for the missing groundwater component in NOAH for this study region.Plain Language Summary Global hydrological models are increasingly being used to assess water availability and sea level rise. Deficiencies in the conceptualization and parameterization in these models may introduce significant uncertainty in model predictions. GRACE satellite senses total water storage at the regional/continental scales. In this study, we applied deep learning to learn the spatial and temporal patterns of mismatch or residual between model simulation and GRACE observations. This hybrid learning approach leverages strengths of data science and hypothesis-driven physical modeling. We show, through three different types of convolution neural network-based deep learning models, that deep learning is a viable approach for improving model-GRACE match. The method can also be used to fill in data gaps between GRACE missions. Key Points: • Strong interests exist in fusing GRACE satellite TWS data into global hydrological models to improve their predictive performance • We train CNN deep learning models to learn the mismatch between TWS anomalies simulated by a land surface model and that observed by GRACE • Results show deep learning models significantly improved the predictive skills of land surface model by compensating for missing components Supporting Information: • Supporting Information S1

show abstract

A global reconstruction of climate‐driven subdecadal water storage variability

Cited by 99 publications

References 58 publications

Forecasting GRACE Data over the African Watersheds Using Artificial Neural Networks

Forecasting GRACE Data over the African Watersheds Using Artificial Neural Networks

Groundwater Depletion in the West Liaohe River Basin, China and Its Implications Revealed by GRACE and In Situ Measurements

Combining Physically Based Modeling and Deep Learning for Fusing GRACE Satellite Data: Can We Learn From Mismatch?

Contact Info

Product

Resources

About