Here, we evaluate the TRMM ARI product in two
steps. The first step is to compare the GEV statistics
from 3B42RT with those from independent high-quality
long-term observations with similar spatial and
temporal resolutions, and the second step is to evaluate
the derived ARI for the actual events. Long-term highresolution
precipitation datasets have only become
available in recent years even for gauge-based observations.
We used two such datasets that meet these criteria:
one is the NOAA’s Climate Prediction Center
(CPC) daily U.S. unified precipitation datasets (Higgins
et al. 2000). The CPC daily unified precipitation dataset
(1948–2012) is a gauge-based, gridded, and qualitycontrolled
product derived from daily and hourly precipitation
measurements from over 13 000 (8000 before
1992) stations over CONUS with the same 0.258 spatial
resolution as the 3B42 and 3B42RT data. The other
dataset comes from the Asian Precipitation–Highly-
Resolved Observational Data Integration Toward
Evaluation of Water Resources (APHRODITE) project
(Yatagai et al. 2012). APHRODITE has produced
high-resolution quality-controlled daily gridded precipitation
datasets based on data collected at 5000–
12 000 stations covering the Asian monsoon region over
the Himalayas, Southeast Asia, and mountainous
regions of the Middle East for the years 1957–2007. This
study used their 0.258 products for the Asian monsoon
and Middle East regions.