Reconstruction of GRACE Mass Change Time Series Using a Bayesian Framework.

Gravity Recovery and Climate Experiment and its Follow On (GRACE (-FO)) missions have resulted in a paradigm shift in understanding the temporal changes in the Earth's gravity field and its drivers. To provide continuous observations to the user community, missing monthly solutions within and between GRACE (-FO) missions (33 solutions) need to be imputed. Here, we modeled GRACE (-FO) data (196 solutions) between 04/2002-04/2021 to infer missing solutions and derive uncertainties in the existing and missing observations using Bayesian inference. First, we parametrized the GRACE (-FO) time series using an additive generative model comprising long-term variability (secular trend + interannual to decadal variations), annual, and semi-annual cycles. Informative priors for each component were used and Markov Chain Monte Carlo (MCMC) was applied to generate 2,000 samples for each component to quantify the posterior distributions. Second, we reconstructed the new data (229 solutions) by joining medians of posterior distributions of all components and adding back the residuals to secure the variability of the original data. Results show that the reconstructed solutions explain 99% of the variability of the original data at the basin scale and 78% at the one-degree grid scale. The results outperform other reconstructed data in terms of accuracy relative to land surface modeling. Our data-driven approach relies only on GRACE (-FO) observations and provides a total uncertainty over GRACE (-FO) data from the data-generation process perspective. Moreover, the predictive posterior distribution can be potentially used for "nowcasting" in GRACE (-FO) near-real-time applications (e.g., data assimilations), which minimize the current mission data latency (40-60 days).

+50 Years of Terrestrial Hydroclimatic Variability in Africa's Transboundary Waters.

GRACE Terrestrial Water Storage (TWS) provides unique and unprecedented perspectives about freshwater availability and change globally. However, GRACE-TWS records are relatively short for long-term hydroclimatic variability studies, dating back to April 2002. In this paper, we made use of Noah Land Surface Model (LSM), and El Niño-Southern Oscillation (ENSO) data in an autoregressive model with exogenous variables (ARX) to reconstruct a 66-year record of TWS for nine major transboundary river basins (TRBs) in Africa. Model performance was evaluated using standard indicators, including the Nash Sutcliffe Efficiency criteria, cumulative density frequency, standardized residuals plots, and model uncertainty bounds. Temporally, the reconstruction results were evaluated for trend, cycles, and mode of variability against ancillary data from the WaterGAP Model (WGHM-TWS) and GPCC-based precipitation anomalies. The temporal pattern reveals good agreement between the reconstructed TWS, WGHM-TWS, and GPCC, (p-value < 0.0001). The reconstructed TWS suggests a significant declining trend across the northern and central TRBs since 1951, while the southern basins show an insignificant trend. The mode of variability analysis indicates short storage periodicity of four to sixteen-month in the northern basins, while strong intra-annual variability in the central and southern basins. The long-term TWS records provide additional support to Africa's water resources research on hydroclimatic variability and change in shared transboundary water basins.

Investigation of potential sea level rise impact on the Nile Delta, Egypt using digital elevation models.

In this study, the future impact of Sea Level Rise (SLR) on the Nile Delta region in Egypt is assessed by evaluating the elevations of two freely available Digital Elevation Models (DEMs): the SRTM and the ASTER-GDEM-V2. The SLR is a significant worldwide dilemma that has been triggered by recent climatic changes. In Egypt, the Nile Delta is projected to face SLR of 1 m by the end of the 21th century. In order to provide a more accurate assessment of the future SLR impact on Nile Delta's land and population, this study corrected the DEM's elevations by using linear regression model with ground elevations from GPS survey. The information for the land cover types and future population numbers were derived from the Moderate Resolution Imaging Spectroradiometer (MODIS) land cover and the Gridded Population of the Worlds (GPWv3) datasets respectively. The DEM's vertical accuracies were assessed using GPS measurements and the uncertainty analysis revealed that the SRTM-DEM has positive bias of 2.5 m, while the ASTER-GDEM-V2 showed a positive bias of 0.8 m. The future inundated land cover areas and the affected population were illustrated based on two SLR scenarios of 0.5 m and 1 m. The SRTM DEM data indicated that 1 m SLR will affect about 3900 km(2) of cropland, 1280 km(2) of vegetation, 205 km(2) of wetland, 146 km(2) of urban areas and cause more than 6 million people to lose their houses. The overall vulnerability assessment using ASTER-GDEM-V2 indicated that the influence of SLR will be intense and confined along the coastal areas. For instance, the data indicated that 1 m SLR will inundate about 580 Km(2) (6%) of the total land cover areas and approximately 887 thousand people will be relocated. Accordingly, the uncertainty analysis of the DEM's elevations revealed that the ASTER-GDEM-V2 dataset product was considered the best to determine the future impact of SLR on the Nile Delta region.