Search

Terrestrial Water Storage (TWS) changes have been estimated at basin to continental scales from gravity variations using data from the Gravity Recovery and Climate Experiment (GRACE) satellites since 2002. The relatively low spatial resolution (∼300 km) of GRACE observations has been a main limitation in such studies. Various data processing strategies, including mascons, forward modeling, and constrained linear deconvolution (CLD), have been employed to address this limitation. Here we develop a revised CLD method to obtain a TWS estimate that combines GRACE observations with much higher spatial resolution land surface models. The revised CLD constrains model estimates to agree with GRACE TWS when smoothed. As an example, we apply the method to obtain a high spatial resolution TWS estimate in Australia. We assess the accuracy of the approach using synthetic GRACE data. Plain Language Summary: The estimation of terrestrial water storage (TWS) changes using gravity recovery and climate experiment (GRACE) satellites suffers from low spatial resolution, making it challenging to interpret local‐scale mass changes. In this study, we improved the sparse resolution of GRACE observations by incorporating high‐resolution land surface models (LSM) that provides detailed hydrological information. Through synthetic experiments, we confirmed the accuracy of our estimations in regional‐ and local‐scale. When applied to real GRACE data, our new TWS estimations show better spatial resolution compared to conventional GRACE products. Further, our estimations consistently yield reliable results although different LSM were used. Key Points: High‐resolution terrestrial water storage was estimated by combining gravity recovery and climate experiment and land surface modelsOur new estimates reduced both land‐ocean and inter‐basin leakages simultaneously [ABSTRACT FROM AUTHOR]

Farm-level seasonal irrigation water usage is often highly variable across time and space in an irrigation district. Identifying the driving factors of this variation can help researchers and managers understand the underlying efficiency of water usage, identify the sources of water waste and develop best irrigation practices to facilitate the development of more efficient and sustainable irrigated cropping systems. This study explored driving factors for the seasonal irrigation water usage across an irrigation district in south-eastern Australia, using extensive climate, soil, cropping and water use data from 312 farms (a total of 1099 annual crop water use records). It focused on three irrigated crops (corn/maize, cotton and rice) for 2011–2019. Factors considered included climate, soils, irrigation practices and water allocation, from which key driving factors and their effects on the seasonal irrigation water usage were identified. A Bayesian hierarchical model averaging approach was developed to determine key predictors and estimate seasonal irrigation water usage. The Bayesian hierarchical modelling framework allowed seasonal irrigation water usage to be estimated at all crop-farm-year observation points simultaneously while also enabling different key predictors to be selected for different crop types. Results showed that the seasonal irrigation water usage was mainly driven by irrigation practices and soil, but the water use of different crop types was driven by different subsets of irrigation and soil predictors. Modelling was undertaken in natural log space and the model showed reasonable accuracy in estimating the seasonal irrigation water usage (R2 = 0.62). Our model is considered useful in 1) identifying drivers of variation of the seasonal irrigation water usage across time and space, and 2) suggesting potential adjustments to irrigation practices and quantifying their benefits to reduce irrigation water usage. • Factors driving seasonal irrigation usage were explored in an irrigation district. • A Bayesian hierarchical model (BHM) was developed for 1099 crop water use records. • Seasonal irrigation usage was mainly influenced by soil and irrigation practices. • The BHM showed good accuracy in estimating seasonal irrigation usage (R2 = 0.62). [ABSTRACT FROM AUTHOR]