In recent years, with the rapid development of satellite-based Earth observation technologies, more and more quasi-global satellite products observe the Earth. Examples are measuring terrestrial water storage with the Gravity Recovery and Climate Experiment (GRACE) Mission, the European Space Agency (ESA) Climate Change Initiative (CCI) Soil Moisture product to measure soil moisture and estimated agricultural drought for instance; or looking at cloud properties, that contributes in estimating occurrence of precipitation and its intensity with MODIS. During the last decade, the use of multi-satellites precipitation products with high spatial and temporal resolutions raised interest in hydrological and climate communities.

Knowing the locations and extents of rain and snowfall is vital to understanding how weather and climate impact our environment and Earth’s water and energy cycles, including effects on agriculture, freshwater availability, and response to natural disasters. In areas where sufficiently dense in-situ observations and radar system are not available, satellite-based precipitation estimate (SPE) products can potentially provide precipitation estimates needed for hydro-meteorological applications such as drought and flood monitoring systems. Rapid growth in computer technology and the remote sensing area help observations processed from satellites, individually, and merge them with other data sources to provide a better visualization of the spatial distribution of precipitation. However, SPEs, like all datasets are subject to errors and uncertainties. Due to inherent biases embedded in satellite precipitation estimates, the accuracy of the new SPE products should be assessed and bias-corrected before it can be employed in decision-making activities. Moreover, precipitation might occur within a region at finer scales compared with the pixel size of satellites, therefore depending on the interested application, satellite data should be sufficiently downscaled before being used as inputs to the hydro-meteorological and water management models.

According to recent studies the Global Precipitation Measurement (GPM) constellation satellites, that produce IMERG product, is one of the most reliable satellite-based precipitation products that has been developed by the American National Aeronautics and Space Administration (NASA) and Japanese Aerospace Exploration Agency (JAXA) with 30-minutes and 0.1° × 0.1° temporal and spatial resolution, respectively (Precipitation Processing System at NASA GSFC, 2017). See the animation at the bottom of the article, which shows the orbit paths of the GPM constellation satellites (Credits: NASA/Goddard/Scientific Visualization Studio).

Sharifi et al., (2016) evaluated IMERG-V03D, 3B42-V7 and ERA-Interim in daily, monthly and seasonal scales. These precipitation products evaluated against in-situ observations in four different topography and climate conditions in Iran from mid-March 2014 to February 2015. The results indicated that in overall, in daily scale (at days with observed precipitation), all three products lead to underestimation but IMERG underestimates precipitation slightly in all four selected regions. Another study examined the performance of the IMERG-V03 Final-Run and IMERG-V03 Real-Time against meteorological stations data on a multi thresholds and multi timescale such as daily and hourly time resolution over northeast Austria from mid-March 2015 till end of January 2016 (Sharifi et al., 2018). They found that in general based on the entire range of precipitation thresholds (Pr ≥ 0.1 mm), the hourly, 3- and 6-hourly IMERG-FR did not show a clear improvement of the bias over IMERG-RT, while for 12-hourly and daily precipitation estimates, the bias in IMERG-FR improved compare to IMERG-RT. In addition, despite the general low probability of detection (POD) and high false alarm ratio (FAR) skill scores within specified precipitation thresholds, both products showed relatively good values of the POD and FAR for precipitation without classification (Pr ≥ 0.1 mm) the classified precipitation thresholds. This means these two products are relatively well able to detect precipitation without classification, but they have poor results to detect precipitation in their exact precipitation categories over northeast Austria.

The rapid changes in elevation can create the microclimate which cause the obstruct of the air mass movement. Therefore, Sharifi et al., (2019) examined six new SPE products to characterize the spatial distribution of daily precipitation over Austria with respect to their performance over extremes, and different elevation categories as well. According to the correlation coefficient, Multi-Source Weighted-Ensemble Precipitation (MSWEP) followed by IMERG-V05B and IMERG-V06A performed well over the whole region as well as the alpine area, in compare to other products.

In an investigation (Sharifi et al., 2019b) developed a bias-correction method to reduce the random errors of SPE, IMERG products. It was found that IMERG precipitation estimates improved after an error adjustment and decreased the uncertainties amplitude (Figure 1).

Figure 1. Average Bias of daily precipitation events over 48 IMERG pixels for (a) Original IMERG, and (b) Bias-Corrected
Figure 1. Average Bias of daily precipitation events over 48 IMERG pixels for (a) Original IMERG, and (b) Bias-Corrected


Furthermore, they presented two downscaling techniques such as multiple linear regression (MLR) and artificial neural networks (ANN) by using finer satellite-based cloud variables from MODIS such as cloud effective radius (CER), cloud optical thickness (COT), and cloud water path (CWP) as the auxiliary variables (Sharifi et al., 2019a). According to the results, the downscaled products were more accurate than the original IMERG data with bias: 6.06 mm, 2.38 mm, -1.36 mm; RMSE: 13.50 mm, 11.24 mm, 9.18 mm; and CC: 0.30, 0.36, 0.49, for original IMERG, ANN, and MLR bias-corrected techniques, respectively. Furthermore, both techniques captured the spatial patterns of precipitation reasonably well with more detailed information when compared with the original IMERG. Consequently, the results lead to increase correlation and reduce mean absolute error and root mean square error (Figure 2).

Figure 2. (a) original IMERG, (b) MLR-downscaled, and (c) ANN-downscaled precipitation for 3rd September 2015 event over northeast Austria.
Figure 2. (a) original IMERG, (b) MLR-downscaled, and (c) ANN-downscaled precipitation for 3rd September 2015 event over northeast Austria.


"Using the new multi-satellite precipitation estimate products, the bias-corrected and downscaled precipitation products may be applied for fulfilling the needs of applications in natural hazard, hydro-meteorology, agricultural, climate change studies etc. where the need for high quality and fine scale precipitation data is of paramount importance" said Silas Michaelides, Ph.D., a Professor at the Cyprus Institute.

Therefore, the performance of a SPE was evaluated in recent severe floods in three basins in Iran that occurred over the March–April 2019 period (Figure 3). The results revealed that the SPE performed well in the three major flood events in spring 2019 in Iran (Aminyavari et al., 2019). This confirms the capability of using SPE as an alternative for rain-gauges and use the state-of-the-art SPE products in hydro-meteorological application to mitigate the negative impacts of heavy precipitation.

Figure 3. The spatial distribution of in situ observations on the elevation map of Iran. The three basins that were affected by the severe floods in 2019, are indicated.
Figure 3. The spatial distribution of in situ observations on the elevation map of Iran. The three basins that were affected by the severe floods in 2019, are indicated.



Aminyavari, Saleh, Bahram Saghafian, and Ehsan Sharifi. 2019. “Assessment of Precipitation Estimation from the NWP Models and Satellite Products for the Spring 2019 Severe Floods in Iran.” Remote Sensing 11 (23): 2741.

Precipitation Processing System at NASA GSFC. 2017. “GPM IMERG Final Precipitation L3 Half Hourly 0.1 Degree X 0.1 Degree V05.”.

Sharifi, Ehsan, Bahram Saghafian, and Reinhold Steinacker. 2019a. “Downscaling Satellite Precipitation Estimates with Multiple Linear Regression, Artificial Neural Networks, and Spline Interpolation Techniques.” J. Geophys. Res. Atmos. 124 (2): 789–805.
Sharifi, Ehsan, Josef Eitzinger, and Wouter Dorigo. 2019. “Performance of the State-of-the-Art Gridded Precipitation Products over Mountainous Terrain: A Regional Study over Austria.” Remote Sensing 11 (17): 2018.

Sharifi, Ehsan, Bahram Saghafian, and Reinhold Steinacker. 2019b. “Copula-Based Stochastic Uncertainty Analysis of Satellite Precipitation Products.” Journal of Hydrology 570:739–54.

Sharifi, Ehsan, Reinhold Steinacker, and Bahram Saghafian. 2016. “Assessment of GPM-IMERG and Other Precipitation Products Against Gauge Data Under Different Topographic and Climatic Conditions in Iran: Preliminary Results.” Remote Sensing 8 (2): 135.

Sharifi, Ehsan, Reinhold Steinacker, and Bahram Saghafian. 2018. “Multi Time-Scale Evaluation of High-Resolution Satellite-Based Precipitation Products over Northeast of Austria.” Atmospheric Research 206:46–63.