*3.1. General Comparison of Precipitation Datasets*

Monthly, seasonal, and annual average precipitation from daily IMERG-Early, -Late, -Final, and -Monthly products together with the corresponding ground observations are shown in Figure 2, Figure 3, and Figure 4. We defined the seasons as winter (Dec-Jan-Feb), spring (Mar-Apr-May), summer (Jun-Jul-Aug), and fall (Sep-Oct-Nov). Figure 2 shows the monthly averages of daily products, including IMERG-Early, -Late, -Final, and -Monthly, compared to rain gauge-measured precipitation during the study period. It can be seen that the difference between IMERG-Early, -Late products, and observations is maximum in April and May (Figure 2). The rain gauge corrected IMERG-Final and -Monthly products showed good agreement with observations during these two months. The correlation between IMERG-Early and -Late products and ground observations in Figure 2 is 0.93 and 0.99 for IMERG-Final and -Monthly products, respectively. In other words, the IMERG-Early and -Late daily products are generally less accurate in estimating the monthly average rainfall amounts as compared to the IMERG-Final. Further, the comparison indicates that IMERG-Final and -Monthly are highly correlated during all months and consistent with the rain gauge measurements.

According to Figure 3, the IMERG-Early and -Late slightly underestimated the measured precipitation in winter (by −7% and −12%, respectively) and overestimated the measured precipitation in the spring (by 33% and 25%, respectively). In contrary, the IMERG-Final and -Monthly overestimated the rain gauge measurements in winter (by 6% and 12%, respectively) but adequately matched measured precipitation in the spring (overestimation by ~5%). Considering summer and fall, all daily products showed high performance for the entire country while the IMERG-Monthly product performed well for spring, summer, and fall with overestimation in the winter. Overall, IMERG-Final and -Monthly products showed high performance in estimating seasonal precipitation relative to measurements.

Average annual precipitation (Figure 4) showed no major discrepancy between the IMERG-Late and -Final (overestimation by only 2% and 1%, respectively) products relative to rain gauge measurements, but the IMERG-Early and -Monthly products overestimated by 7% and 5%, respectively, relative to the measurements.

**Figure 2.** Monthly averages of IMERG-Early, -Late, and -Final based on the daily time step and -Monthly products compared to ground-based precipitation measurements (Observation).

**Figure 3.** Seasonal averages of monthly and daily precipitation products (IMERG-Early, -Late, and -Final), compared to ground-based precipitation measurements (Observation).

**Figure 4.** Annual averages of monthly and daily precipitation products (IMERG-Early, -Late, and -Final), and ground-based precipitation measurements (Observation).
