Obtaining accurate and reliable rainfall information is very crucial for regional and global water resource management, and agricultural water use (Yong et al., 2010). Rainfall variability in its rate, amount, and distribution substantially determine the hydrological cycle and earth’s ecosystem (Stillman et al., 2014). Thus, accurate measurement of rainfall is vital to analyze the spatial and temporal patterns of rainfall at various scales and advance our understanding of the effect of rainfall on agriculture, hydrology, and climatology (Awulachew et al., 2007; Ayalew et al., 2012). Conventionally, the primary source of rainfall data is rain gauge, which has been the most accurate and reliable approach for rainfall measurement (Ayehu et al., 2018). However, in developing countries, weather stations are characterized by sparse, uneven distribution, poor data quality, temporally inconsistent and unavailability of updated data (Katsanos et al., 2016; Fenta et al., 2018). These problems are more common in inaccessible and rugged areas (Rivera et al., 2018) like the highlands of Ethiopia, where rainfall is extremely variable over short distances (Haile et al., 2009; Rientjes et al., 2012). Because of this sparse distribution of gauging stations, estimating the spatial distribution of rainfall over remote parts remains difficult, and the dependability of rain gauge data to estimate areal rainfall and spatial distribution of rainfall over large areas of Ethiopia is considerably reduced (Ayehu et al., 2018; Belay et al., 2019). Therefore, evaluation of spatio-temporal variability of rainfall solely based on station data is a challenging task in most developing countries such as Ethiopia (Hirpa et al., 2010; Igbal et al., 2018). Furthermore, analysis using point-based rain gauge observations is limited to the given particular location, which provides a poor estimation of spatial distribution of rainfall for areas such as Ethiopia, where rainfall is extremely variable over short distances. To overcome these limitations, the recently developed long-term and spatially distributed satellite-based rainfall estimates have become important source to analyze the spatial and temporal variability of rainfall, especially for data-sparse regions (Ayehu et al., 2018; Fenta et al., 2018; Alemu and Bawoke, 2019).
Satellite rainfall products (SRPs) provide synoptic data at finer temporal and spatial resolutions (Park et al., 2017). It is well known that due to the indirect nature of estimates the satellite rainfall values are just estimates that are subject to a variety of certain bias and error sources attributed with gaps in revisit times, the poor direct relationship between remotely sensed signals and rainfall rate, atmospheric effects that modify the radiation field and retrieval algorithms, sampling frequency and satellite instrument sensors (Fenta et al., 2018; Alemu and Bawoke, 2019). Therefore, performance evaluation of SRPs in different regions is essential to enable users, the scientific communities and algorithm developers to better understand and quantify such errors & uncertainties and to identify the best SRPs for a site-specific application (Belay et al., 2019). Moreover, validation of SRPs at varying spatial and temporal resolutions is essential before integration into operational applications such as rainfall pattern and variability study (Dinku et al., 2007; Jiang et al., 2012; Ayehu et al., 2018).
Recently, many studies on validation of different satellite rainfall products have been performed to assure the quality in using the satellite products for different applications. Each validation found different performance at different geographical regions, topographic conditions and seasons (Katsanos et al., 2016; Hobouchian et al., 2017; Lakew et al., 2017; Zambrano-Bigiarini et al., 2017; Iqbal and Athar, 2018; Lekula et al., 2018; Rivera et al., 2018; Alemu and Bawoke, 2019). The results show that the Climate Hazards Group InfraRed Precipitation with Stations (CHIRPS) product (a new multi-source precipitation climatology database, which has relatively high spatial resolution and long-term time series) performs better than most long-term satellite rainfall products. Nevertheless, site-specific local validation of SRPs at diverse physiographic settings has been recommended before using it for any operational applications. Thiemig et al. (2012) compared six SRPs against rain gauge data over four African river basins. They found that SRPs showed higher performance over the tropical wet and dry zone compared to semiarid mountainous regions, low accuracy in detecting heavy rainfall events over semiarid areas, general underestimation of heavy rainfall events, and overestimation of the number of rainy days in the tropics. Hessel (2015) compared ten satellite rainfall products over the Nile basin and CHIRPS products were best performed and recommended for the basin. The likely reason for its better performance could be a fine spatial resolution. Besides, the performance of SRPs have been evaluated in different climatic zones including Northern Tanzania (Dinku et al., 2011; Mashingia et al., 2014), Nzoia Basin along Lake Victoria (Li et al., 2009), Ethiopian highlands (Hirpa et al., 2010; Gebrechorkos et al., 2017), and western region of Uganda (Asadullah et al., 2008; Diem et al., 2014).
With regard to the Upper Tekeze-Atbara River Basin (UTARB), very few studies were conducted from small watershed to some parts of the basin (Seleshi and Zanke, 2004; Gebremichael et al., 2017)), Moreover, the previous studies overlook comprehensive validation of spatio-temporal rainfall variability from satellite rainfall estimates with observed rainfall value at basin scale. For example, Seleshi and Zanke (2004) attempted to investigate the pattern of rainfall over the upper part of Tekeze River basin by considering only one climatic station. Their result demonstrated that the amount of rainfall remained constant for the past 40 years (1962–2002). Gebremichael et al. (2017) evaluated the performance of eight SRPs (TRMM, CHIRPS, RFEv2, ARC2, PERSIANN, GPCP, CMAP and CMORPH) in UTARB. They evaluated the performance of SRPs at temporal (daily, monthly, seasonal) and spatial (point, sub-basin, basin) scales over the period 2002–2015. They compared SRPs which have different spatial resolution from finer (0.05° to 0.25°) to coarser (1° to 2.5°), and validated for only wet season and decadal timescale on 45,694 km2 area of Tekeze-Atbara basin.
The spatial and temporal resolutions and measurement accuracy of these products are continuously improving because of advancement in sensor technologies and estimation techniques. A number of higher resolutions SRPs are now available at a quasi-global scale (Behrangi et al., 2011; Jiang et al., 2012). Among the many available SRPs, the three widely used precipitation datasets are Climate Hazards Group Infrared Rainfall with Stations (CHIRPS) (Funk et al., 2015), Precipitation Estimation from Remotely Sensed Information Using Artificial Neural Networks-Climate Data Record (PERSIANN_CDR) (Ashouri et al., 2015), and The Tropical Rainfall Measurement Mission (TRMM) Multi-satellite Precipitation Analysis (TMPA) (Huffman et al., 2007). These SRPs have fine spatial resolutions ((0.25°).
Therefore, the main purpose of the study is to evaluate the performance of these SRPs through statistical indices over Upper Tekeze-Atbara River Basin using gauged rainfall observations. In particular, this study can address the following research questions: (1) What is the overall performance of each SRPs? and (2) Which SRPs performs best for different spatial and temporal timescales? The accuracy of these products was evaluated by using 45 rain gauges at daily, monthly, and seasonal scale from 2007 to 2017. This study presents a more comprehensive evaluation across Upper-Tekeze-Atbara River Basin by considering a lot of gauge stations, the same spatial and temporal resolution of satellite products, and different aspects of evaluation. Moreover, this study presents the first attempts in the spatio-temporal validation and evaluation of the rain-detection ability of each of the SRPs over Upper-Tekeze-Atbara River Basin.