The existing gaps (or some missing monthly observations) in the Gravity Recovery and Climate Experiment (GRACE) data limit its use in climate change studies. Data gaps provide an opportunity to reconstruct the time series of GRACE-derived terrestrial water storage (TWS) product or extend it backward to favor climate change assessments. To address this limitation, the use of machine learning models to reconstruct GRACE data is gradually emerging, emphasizing the importance of accurately filling these data gaps. This chapter demonstrates the utility of an integrated machine learning technique that shows faster convergence rates, finer predictions, and more efficient reconstructive properties for non-linear systems. By exemplifying the reconstruction process of TWS using this technique, this chapter discusses how the reconstruction of GRACE data can help improve understanding of the influence of climate variability on terrestrial hydrology.
CITATION STYLE
Ndehedehe, C. (2023). Integrated Machine Learning in Satellite Hydrology. In Springer Climate (Vol. Part F1494, pp. 325–359). Springer Science and Business Media B.V. https://doi.org/10.1007/978-3-031-37727-3_9
Mendeley helps you to discover research relevant for your work.