Evaluation of GPM Day-1 IMERG and TMPA Version-7 legacy products over Mainland China at multiple spatiotemporal scales

Highlights

• The study is among the earliest evaluation and comparison of IMERG and 3B42V7.

• Evaluations were conducted at multiple spatiotemporal scales.

• IMERG outperforms 3B42V7 over Mainland China in most cases.

• Results reveal error characteristics of IMERG and 3B42V7 products.

Summary

The post-real time product of Day-1 Integrated Multi-satellitE Retrievals for Global Precipitation Measurement (IMERG) is evaluated over Mainland China from April to December 2014 at the hourly timescale, against data from hourly ground-based observations. In addition, the IMERG product is compared with its predecessor-the Version-7 post-real-time 3B42 (3B42V7) product of Tropical Rainfall Measuring Mission (TRMM) Multisatellite Precipitation Analysis (TMPA) at its original 3-hourly and then daily timescales for the same period. All the products are cross-evaluated at gridded, regional, and national scales. Results show that: (1) the Day-1 IMERG shows appreciably better performance than 3B42V7 at both sub-daily and daily timescales, and all the three spatial scales. The gap between the two products is more significant at the sub-daily resolution; (2) Out of the six sub-regions of China, IMERG especially performs better than 3B42V7 at the mid- and high-latitudes, as well as relatively dry climate regions; (3) IMERG can better reproduce the probability density function (PDF) in terms of precipitation intensity, particularly in the low ranges; and (4) although IMERG better captures the precipitation diurnal variability, both products have room to further improve their capability, particularly in the dry climate and high-latitude regions. This study is among the earliest evaluation and comparison of IMERG and 3B42V7 products, which could be valuable in providing reference for the development of IMERG algorithms, associated global products, and various applications as well.

Introduction

Precipitation is one of the most important components of global water and energy cycles, playing an important role in the interactions between hydrosphere, atmosphere, and biosphere (Kidd and Huffman, 2011). Accurate precipitation measurement or estimation is vital to water resource management, weather prediction, disaster monitoring, controlling, and so on (Hou et al., 2014). In addition, precipitation input has great impacts on the performance of a range of hydrological, climatic, and atmospheric models (Shen and Xiong, 2015). However, obtaining accurate precipitation has always been challenging for scientists. Currently, there are three mainstream methods to measure precipitation, i.e., gauge, weather radar, and satellite-based sensors (Li et al., 2013). Gauges provide the most straightforward and accurate precipitation observations so far (Ma et al., 2015). However, gauge data are provided at specific sites, and various interpolation methods would result in potential errors when obtaining continuous spatial precipitation estimates. In addition, networks of rain gauges are sparse over most of continents, and few gauges are located over the ocean (Kidd and Huffman, 2011). In regard to the weather radar, it can provide the internal structure of storms, and real-time high-resolution monitoring over large areas (Doviak and Zrnic, 2006, Germann et al., 2006). However, the radar also suffers from various sources of errors, including mean-field systematic error, range-dependent systematic error, and random error (Dinku et al., 2002). The radar network is often not dense enough over most parts of the world. The only practical way to achieve comprehensive estimation of precipitation on a global basis relies on earth observation satellites (Hong et al., 2012, Hou et al., 2014, Villarini and Krajewski, 2008).

In recent years, a large number of quasi-global satellite precipitation products with various temporal and spatial resolutions have been developed and released to the public, such as TMPA (Huffman et al., 2007), Climate Prediction Center (CPC) MORPHing technique (CMORPH) (Joyce et al., 2004), Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks (PERSIANN) (Sorooshian et al., 2000), and Global Satellite Mapping of Precipitation (GSMap) (Kubota et al., 2007). Those free and open access products have been widely studied and applied globally (Hong et al., 2007a, Hong et al., 2007b, Hong et al., 2007c, Long et al., 2015a, Yong et al., 2015), and regionally (Bitew et al., 2012, Kirstetter et al., 2013, Long et al., 2014, Long et al., 2015b, Xue et al., 2013), and could potentially bring substantial scientific and societal benefits (e.g., disaster forecast and monitoring, and water resource management).

The Global Precipitation Measurement (GPM) mission is an international constellation of satellites, including one Core Observatory satellite and approximately ten partner satellites. As the successor to the TRMM satellite, which was launched by the National Aeronautics and Space Administration (NASA) and National Space Development Agency (NASDA) on November 27, 1997, the GPM Core Observatory was deployed on February 28, 2014 by a joint effort of NASA and the Japan Aerospace Exploration Agency (JAXA), marking a transition from the TRMM era to the GPM era. The GPM Core Observatory carries a dual-frequency precipitation radar (DPR; the Ku-band at 13.6 GHz and Ka-band at 35.5 GHz) and a conical-scanning multichannel GPM Microwave Imager (GMI; frequencies range between 10 and 183 GHz). GPM extends the sensor package compared to TRMM instruments, which had a single-frequency precipitation radar (PR; the Ku-band at 13.8 GHz) and a multichannel TRMM Microwave Imager (TMI; frequencies range between 10 and 85.5 GHz). Therefore, the GPM sensors can detect light and solid precipitation more accurately than TRMM sensors (Hou et al., 2014). This study will focus on the Level-3 product, provided by the Day-1 Integrated Multi-satellitE Retrievals for GPM (IMERG) algorithm, which is intended to intercalibrate, merge, and interpolate all microwave (MW) estimates of the GPM constellation, infrared (IR) estimates, gauge observations, and other data from potential sensors at 0.1° × 0.1° and half-hour temporal resolutions (Huffman et al., 2014). IMERG provides three kinds of products, including the near real time “Early” and “Late” run products, and the post real time “Final” run product. The “Final” run is the research level product used in this study.

As GPM’s precursor, the TRMM satellite re-entered the Earth’s atmosphere on June 15, 2015 after over 17 years of productive data gathering (http://pmm.nasa.gov/gpm-news/trmm-spacecraft-re-enters-over-tropics). The TMPA is intended to provide the “best” satellite precipitation estimate (Huffman et al., 2007), which has been studied and applied widely over the past years. Once the TRMM satellite has decommissioned, the TMPA products are intended to be generated using other calibrators instead of TRMM satellite instruments, until the GPM products can totally substitute TMPA products (Huffman et al., 2015). It is necessary and meaningful to evaluate the Day-1 IMERG product and compare it with TMPA products, which will shed light on research and application during the transition from TMPA to IMERG, and future improvement and retrospective construction of IMERG. There have been a multitude of statistical and hydrological studies comparing and evaluating various satellite precipitation products (Bitew et al., 2012, Li et al., 2015, Scheel et al., 2011, Su et al., 2008). However, the majority of these studies were conducted at relatively coarse temporal scales (e.g., daily or monthly scales), which may be far less than satisfactory from revealing the characteristics of those products comprehensively at 3 hourly or even finer temporal resolutions. The IMERG products are characterized by high temporal and spatial resolutions (half-hour and 0.1° × 0.1°), and its capability to detect light and solid precipitation (Hou et al., 2014, Huffman et al., 2015). Due to the lack of ground snowfall observations, this study only focuses on the evaluation and comparison of total precipitation (including liquid, solid, and mixed precipitation). Further studies are needed to evaluate the quality of IMERG solid precipitation products.

Therefore, the objectives of this study are twofold: (1) evaluating the quality of Day-1 IMERG “Final” run precipitation products over Mainland China at hourly, 0.1° × 0.1° resolutions against hourly ground-based observations from the China Meteorological Administration (CMA) over Mainland China, and (2) comparing IMERG and 3B42V7 products synchronously at 3-hourly and daily timescales, to explore the continuity and differences between the two products in both GPM and TRMM eras. This study will reveal characteristics of hourly, 3-hourly and daily errors of IMERG and TMPA products at gridded, regional, and national scales, and provide insight into subsequent studies and applications of IMERG precipitation products at high temporal and spatial resolutions. The remaining parts of the paper are organized as follows: Section 2 introduces the study area, precipitation datasets, and metrics. Section 3 evaluates the quality of IMERG products at 0.1° × 0.1° and hourly resolutions. Section 4 compares IMERG and TMPA 3B42 products at multiple spatiotemporal scales. Finally, discussion and summary are involved in Section 5.