Comparison of Data‐Driven Techniques to Reconstruct (1992–2002) and Predict (2017–2018) GRACE‐Like Gridded Total Water Storage Changes Using Climate Inputs

Abstract

The Gravity Recovery and Climate Experiment (GRACE) mission ended its operation in October 2017, and the GRACE Follow-On mission was launched only in May 2018, leading to approximately 1 year of data gap. Given that GRACE-type observations are exclusively providing direct estimates of total water storage change (TWSC), it would be very important to bridge the gap between these two missions. Furthermore, for many climate-related applications, it is also desirable to reconstruct TWSC prior to the GRACE period. In this study, we aim at comparing different data-driven methods and identifying the more robust alternatives for predicting GRACE-like gridded TWSC during the gap and reconstructing them to 1992 using climate inputs. To this end, we first develop a methodological framework to compare different methods such as the multiple linear regression (MLR), artificial neural network (ANN), and autoregressive exogenous (ARX) approaches. Second, metrics are developed to measure the robustness of the predictions. Finally, gridded TWSC within 26 regions are predicted and reconstructed using the identified methods. Test computations suggest that the correlation of predicted TWSC maps with observed ones is more than 0.3 higher than TWSC simulated by hydrological models, at the grid scale of 1°r esolution. Furthermore, the reconstructed TWSC correctly reproduce the El Nino-Southern Oscillation (ENSO) signals. In general, while MLR does not perform best in the training process, it is more robust and could thus be a viable approach both for filling the GRACE gap and for reconstructing long-period TWSC fields globally when combined with statistical decomposition techniques.

Reconstructing GRACE-type time-variable gravity from the Swarm satellites

“…Ref. 19 ) have developed pre-GRACE era reconstructions in this way, but so far this method has been limited to specific compartments (e.g. only hydrological storage variations) and often to basin averages or continental areas.…”

Monitoring Droughts From GRACE

“…The above mentioned issues with GRACE products are currently being investigated by researchers, for example, (a) spatial downscaling of GRACE by assimilating it with other hydrological observations to produce higher spatial resolution TWS products (Zaitchik et al, 2008;Miro and Famiglietti, 2018). (b) With the launch of GRACE Follow On mission, we are expecting another decade long GRACE data, and several efforts are going on to reconstruct GRACE TWS for filling the data gaps and reconstructing TWS prior to 2002 (Humphrey and Gudmundsson, 2019;Li W. et al, 2019;Li et al, 2020), and (c) novel daily and weekly GRACE products have been developed and shown to detect short-lived extreme events, such as floods (Kvas et al, 2019). Therefore, ongoing improvements will make GRACE even more effective for drought research.…”

Reconstruction of GRACE Total Water Storage Through Automated Machine Learning

“…Li et al. (2020) presented a framework for reconstructing GRACE‐like TWSA using only climate forcing data and climate indices. Instead of predicting the total TWSA, their framework reconstructs three detrended TWSA components (i.e., seasonal, interannual, and residual) separately.…”

“…They showed that all three regression models perform reasonably well in humid and low-intensity irrigated areas, but vary significantly in dry and/or intensively irrigated regions. Li et al (2020) presented a framework for reconstructing GRACE-like TWSA using only climate forcing data and climate indices. Instead of predicting the total TWSA, their framework reconstructs three detrended TWSA components (i.e., seasonal, interannual, and residual) separately.…”