6. Conclusions
The overall objective of this study was to evaluate the hydrological
application potential of TRMM, IMERG, CMADS, and CFSR in the YRSR. The
major findings of this study are summarized as follows.
(1) At the basin-scale, the TRMM, IMERG, CMADS, and CFSR have higher
detection accuracy in the warm season, and the PBIAS and CC values of
each precipitation product are characterized by small warm season and
large cold season values. Among the four precipitation products, IMERG
had the smallest deviation (average RMSE = 13.71 mm), while CFSR had the
best correlation (average CC = 0.73)
(2) At the grid scale, among the four precipitation products, CMADS has
the best performance for precipitation observation, with PBIAS of
-27.22%2.48 %, CC of 0.430.58, and RMSE of 2.684.96
(mm/d), followed by IMERG, CFSR, and TRMM. CFSR has the best performance
for precipitation events, with POD of 0.900.98, FAR of 0.290.51,
and CSI of 0.480.69, followed by CMADS, IMERG and TRMM.
(3) Taken together, IMERG has the best performance, followed by CMADS,
CFSR, and TRMM. TRMM severely overestimated high rainfall of
> 10 mm/day. CFSR obviously overestimated moderate
precipitation events of 110 mm/d, while CMADS underestimated the
precipitation events of 120 mm/d.
(4) Models using the GO as input resulted in satisfactory performance
during 20082013, and precipitation products have poor simulation
results. The results of simulation using CMADS significantly
underestimated the runoff during the dry season, but the performance in
the validation periods (R 2 = 0.78, NSE = 0.53
at TNH; R 2 = 0.64, NSE = 0.53 at JM) was best
among those scenarios analyzed. The runoff simulated using TRMM and CFSR
is significantly overestimated, especially when using TRMM. Although the
model using IMERG as input yielded unsatisfactory performance during
20142016, it did not affect the use of IMERG as a potential data
source for YRSR.
(5) After bias correction, the quality of CFSR improves significantly
with increases to R 2 and NSE of 0.25 and 0.31
at TNH, respectively. SWAT model driven by the combination of GO and
CMADS precipitation was the best across all scenarios. The simulation
results at TNH yielded satisfactory performance
(R 2 = 0.77, NSE = 0.72). Although the
simulation results at JM yielded an unsatisfactory performance, they
were close to being deemed satisfactory (R 2 =
0.53, NSE = 0.48).
In
summary, although the satellite and reanalysis precipitation products
represented by TRMM and CFSR have been widely used in hydrological
modeling, the quality of these products could be significantly improved
when applied to alpine basins. In contrast, IMERG has a better
performance in observing solid precipitation due to the more advanced
GPM microwave imager sensor and the dual-frequency precipitation radar
mounted on the GPM satellites (Yang et al ., 2020). The findings
of this assessment provide valuable reference and feedback for satellite
and reanalysis precipitation product development for use in alpine
basins. In addition, snowfall is the main form of precipitation in the
YRSR from September to May, however, such an assessment was not
fulfilled due to the lack of snowfall observation site, a task that
warrants investigation and inclusion in future research.