Quantifying the amount of precipitation and its uncertainty is a challenging task all over the world, particularly over the African continent, where rain gauge (RG) networks are poorly distributed. In recent decades, several satellite remote sensing (SRS)‐based precipitation products have become available with reasonable spatial and temporal resolutions to be applied in hydrological and climate studies. However, uncertainties of these products over Africa are largely unknown. In this study, the generalized ‘three‐cornered‐hat’ (TCH) method is applied to estimate uncertainties of gridded precipitation products over the entire African continent, without being dependent to the choice of a reference dataset. Six widely used SRS‐based precipitation products (at monthly scales) were evaluated over the entire continent during the period of 2003–2010. The TCH results are further compared to those of the classical evaluation using the Global Precipitation Climatology Centre (GPCC) over entire Africa, as well as to the RG observations over the Greater Horn of Africa (GHA). Overall, for the study period (2003–2010), the TCH results indicate that the RG‐merged products contain smaller error amplitudes compared to the satellite‐only products, consistent with the GPCC‐based evaluation. A multiple comparison procedure ranking, which was applied based on signal‐to‐noise ratios (SNRs), indicated that PERSIANN contains the highest SNR and thus suitable over most of Africa, followed by ARCv2, TRMM, CMORPH, TAMSAT, and GSMaP. To extract the main spatio‐temporal variability of rainfall over Africa, complex empirical orthogonal function technique was applied, from which the extracted patterns of GPCC, TRMM, PERSIANN, and ARCv2 were found to be similar but different from those of TAMSAT, CMORPH, and GSMaP. Finally, the TCH and RG‐based validation methods were found to provide similar evaluations for the SRS‐only products (CMORPH and GSMaP) over GHA, with CMORPH emerging to be the most suitable product, consistent with previous studies.

The estimation of large‐scale evapotranspiration (ET) is complex, and typically relies on the outputs of land surface models (LSMs) or remote sensing observations. However, over some regions of Africa, inconsistencies exist between different estimations of ET fluxes, which should be investigated. In this study, we evaluate and combine different ET estimates from moderate‐resolution imaging spectroradiometer (MODIS), Global Land Data Assimilation System (GLDAS) and terrestrial water budget (TWB) approaches over the Volta Basin, West Africa. ET estimates from water balance equation are obtained as residuals from monthly terrestrial water‐storage (TWS) changes derived from Gravity Recovery and Climate Experiment (GRACE), Tropical Rainfall Measurement Mission (TRMM)'s rainfall data, and in situ discharge from Akosombo Dam (Ghana). An averaged estimation of ET time series is derived from all the ET estimations under study, while taking into account their uncertainties. The resulting ensemble‐averaged ET was then used to assess each of the individual ET estimates. Overall, out of the seven investigated ET estimates (two from the water balance approach of which one considers water storage using GRACE‐derived TWS and the other ignoring it, four from GLDAS and one from MODIS), only MODIS (28.12 mm month^–1), GLDAS–NOAH (32.74 mm month^–1) and TWB (32.84 mm month^–1) were found to represent the range of variability close to the computed averaged reference ET (30.25 mm month^–1). ET estimations inferred from MODIS were also found to represent relatively lower magnitude of uncertainties, that is, 3.99 mm month^–1 over the Volta Basin (cf. 7.06 and 18.85 mm month^–1 for GLDAS‐NOAH and TWB‐based ET estimations, respectively).

Brazil has recently experienced one of its worst droughts in the last 80 years, with wide-ranging consequences for water supply restrictions, energy rationing, and agricultural losses. Northeast and Southeast Brazil, which share the São Francisco River basin (SFRB), have experienced serious precipitation reduction since 2011. We used terrestrial water-storage (TWS) fields, inverted from the Gravity Recovery and Climate Experiment (GRACE) mission measurements, to assess and quantify the ongoing drought over the SFRB. We found a water loss rate of 3.30 km³/year over the time-span of April 2002 to March 2015. In addition, the TWS drought index (TWSDI) showed the extension of the recent drought that has jeopardized the SFRB since January 2012, and which reached its maximum in July 2015 (the end of TWS time series). In this sense there seems to be a linkage between the TWSDI (wetness/dryness) and the El Niño Southern Oscillation (ENSO), in terms of the wavelet coherence, at the semi-annual and biennial bands, suggesting a relationship between the two. While acknowledging that further investigation is needed, we believe that our findings should contribute to the water management policies by quantifying the impact of this drought event over the SFRB.

The measurement of total basin discharge along coastal regions is necessary for understanding the hydrological and oceanographic issues related to the water and energy cycles. However, only the observed streamflow (gauge-based observation) is used to estimate the total fluxes from the river basin to the ocean, neglecting the portion of discharge that infiltrates to underground and directly discharges into the ocean. Hence, the aim of this study is to assess the total discharge of the Yangtze River (Chang Jiang) basin. In this study, we explore the potential response of total discharge to changes in precipitation (from the Tropical Rainfall Measuring Mission—TRMM), evaporation (from four versions of the Global Land Data Assimilation—GLDAS, namely, CLM, Mosaic, Noah and VIC), and water-storage changes (from the Gravity Recovery and Climate Experiment—GRACE) by using the terrestrial water budget method. This method has been validated by comparison with the observed streamflow, and shows an agreement with a root mean square error (RMSE) of 14.30 mm/month for GRACE-based discharge and 20.98 mm/month for that derived from precipitation minus evaporation (P − E). This improvement of approximately 32% indicates that monthly terrestrial water-storage changes, as estimated by GRACE, cannot be considered negligible over Yangtze basin. The results for the proposed method are more accurate than the results previously reported in the literature.

Reconstructing missing streamflow data can be challenging when additional data are not available, and missing data imputation of real-world datasets to investigate how to ascertain the accuracy of imputation algorithms for these datasets are lacking. This study investigated the necessary complexity of missing data reconstruction schemes to obtain the relevant results for a real-world single station streamflow observation to facilitate its further use. This investigation was implemented by applying different missing data mechanisms spanning from univariate algorithms to multiple imputation methods accustomed to multivariate data taking time as an explicit variable. The performance accuracy of these schemes was assessed using the total error measurement (TEM) and a recommended localized error measurement (LEM) in this study. The results show that univariate missing value algorithms, which are specially developed to handle univariate time series, provide satisfactory results, but the ones which provide the best results are usually time and computationally intensive. Also, multiple imputation algorithms which consider the surrounding observed values and/or which can understand the characteristics of the data provide similar results to the univariate missing data algorithms and, in some cases, perform better without the added time and computational downsides when time is taken as an explicit variable. Furthermore, the LEM would be especially useful when the missing data are in specific portions of the dataset or where very large gaps of ‘missingness’ occur. Finally, proper handling of missing values of real-world hydroclimatic datasets depends on imputing and extensive study of the particular dataset to be imputed.