Filling Temporal Gaps Within and Between GRACE and GRACE-FO Records: Advances, Challenges, and Future Opportunities

Abstract:

Terrestrial water storage (TWS) data derived from past Gravity Recovery and Climate Experiment (GRACE; April 2002–June 2017) and current GRACE-Follow On (GRACE-FO; June 2018–present) missions provide insights into mass transport within, and between, different Earth’s systems. In 2017 Earth Science Decadal Survey, a future gravity mission (GRACE-II; 2025) was recommended to provide 30 years of TWS measurements. However, there are currently temporal gaps within GRACE record (20 months) and between GRACE and GRACE-FO missions (11 months), within GRACE-FO record (2 months), and similar gaps could be experienced between GRACE-FO and GRACE-II missions. Temporal gaps in TWS records minimize scientists’ ability to investigate long-term spatiotemporal variability in TWS, increase the level of uncertainty in the spectral analyses of TWS, and obscure the temporal patterns related to episodic events. In this study, we compare the performance of different data-driven models in filling TWS gaps for different hydrologic systems stretched over different climatic, geologic, and hydrologic settings. The investigated hydrologic systems are located in Africa, Australia, North America, and Eurasia and are witnessing both natural and anthropogenic variabilities. Specifically, we used artificial neural networks (ANNs), support vector machines (SVMs), and multiple linear regression (MLR) models to predict TWS data based on the knowledge of relevant datasets/observations such as rainfall, temperature, evapotranspiration, vegetation indices, climate indices. These models were trained, tested, validated, and used to predict TWS data. The performance of the developed models was evaluated using several standard statistical measures. Our preliminary results indicate: (1) ANN models show better performance over the examined systems compared to MLR and SVM models, (2) the performances of ANN, MLR, and SVM models depend mainly on the nature of factors that control TWS in each of the examined hydrologic systems, and (3) higher model performance is achieved when the model input data were further spectrally decomposed. Our research will promote additional and improved use of GRACE products by the scientific community, end-users, and decision makers by providing a continuous uninterrupted TWS record from GRACE and GRACE-FO missions.