Hydrological Evaluation of High-Resolution Precipitation Estimates from the WRF Model in the Third Pole River Basins

He Sun aKey Laboratory of Tibetan Environment Changes and Land Surface Processes, Institute of Tibetan Plateau Research, Chinese Academy of Sciences, Beijing, China
bUniversity of Chinese Academy of Sciences, Beijing, China

Search for other papers by He Sun in
Current site
Google Scholar
PubMed
Close
,
Fengge Su aKey Laboratory of Tibetan Environment Changes and Land Surface Processes, Institute of Tibetan Plateau Research, Chinese Academy of Sciences, Beijing, China
bUniversity of Chinese Academy of Sciences, Beijing, China
cCAS Center for Excellence in Tibetan Plateau Earth Sciences, Beijing, China

Search for other papers by Fengge Su in
Current site
Google Scholar
PubMed
Close
,
Zhihua He dCentre for Hydrology, University of Saskatchewan, Saskatoon, Saskatchewan, Canada

Search for other papers by Zhihua He in
Current site
Google Scholar
PubMed
Close
,
Tinghai Ou eRegional Climate Group, Department of Earth Sciences, University of Gothenburg, Gothenburg, Sweden

Search for other papers by Tinghai Ou in
Current site
Google Scholar
PubMed
Close
,
Deliang Chen eRegional Climate Group, Department of Earth Sciences, University of Gothenburg, Gothenburg, Sweden

Search for other papers by Deliang Chen in
Current site
Google Scholar
PubMed
Close
,
Zhenhua Li dCentre for Hydrology, University of Saskatchewan, Saskatoon, Saskatchewan, Canada

Search for other papers by Zhenhua Li in
Current site
Google Scholar
PubMed
Close
, and
Yanping Li dCentre for Hydrology, University of Saskatchewan, Saskatoon, Saskatchewan, Canada

Search for other papers by Yanping Li in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

In this study, two sets of precipitation estimates that are based on the regional Weather Research and Forecasting (WRF) Model—the high Asia refined analysis (HAR) and outputs with a 9-km resolution from WRF (WRF-9km)—are evaluated at both basin and point scales, and their potential hydrological utilities are investigated by driving the Variable Infiltration Capacity (VIC) large-scale land surface hydrological model in seven Third Pole (TP) basins. The regional climate model (RCM) tends to overestimate the gauge-based estimates by 20%–95% in annual means among the selected basins. Relative to the gauge observations, the RCM precipitation estimates can accurately detect daily precipitation events of varying intensities (with absolute bias < 3 mm). The WRF-9km exhibits a high potential for hydrological application in the monsoon-dominated basins in the southeastern TP (with NSE of 0.7–0.9 and bias from −11% to 3%), whereas the HAR performs well in the upper Indus and upper Brahmaputra basins (with NSE of 0.6 and bias from −15% to −9%). Both of the RCM precipitation estimates can accurately capture the magnitudes of low and moderate daily streamflow but show limited capabilities in flood prediction in most of the TP basins. This study provides a comprehensive evaluation of the strength and limitation of RCMs precipitation in hydrological modeling in the TP with complex terrains and sparse gauge observations.

© 2021 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Corresponding author: Fengge Su, fgsu@itpcas.ac.cn

Abstract

In this study, two sets of precipitation estimates that are based on the regional Weather Research and Forecasting (WRF) Model—the high Asia refined analysis (HAR) and outputs with a 9-km resolution from WRF (WRF-9km)—are evaluated at both basin and point scales, and their potential hydrological utilities are investigated by driving the Variable Infiltration Capacity (VIC) large-scale land surface hydrological model in seven Third Pole (TP) basins. The regional climate model (RCM) tends to overestimate the gauge-based estimates by 20%–95% in annual means among the selected basins. Relative to the gauge observations, the RCM precipitation estimates can accurately detect daily precipitation events of varying intensities (with absolute bias < 3 mm). The WRF-9km exhibits a high potential for hydrological application in the monsoon-dominated basins in the southeastern TP (with NSE of 0.7–0.9 and bias from −11% to 3%), whereas the HAR performs well in the upper Indus and upper Brahmaputra basins (with NSE of 0.6 and bias from −15% to −9%). Both of the RCM precipitation estimates can accurately capture the magnitudes of low and moderate daily streamflow but show limited capabilities in flood prediction in most of the TP basins. This study provides a comprehensive evaluation of the strength and limitation of RCMs precipitation in hydrological modeling in the TP with complex terrains and sparse gauge observations.

© 2021 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Corresponding author: Fengge Su, fgsu@itpcas.ac.cn

1. Introduction

Precipitation is the key driver of the terrestrial hydrological system and the most important atmospheric input in land surface hydrological models (Beven and Hornberger 1982; Su et al. 2008; Tong et al. 2014b). However, direct meteorological observations are either sparse or nonexistent in many remote high mountainous areas because of their high elevation, complex terrain, and inaccessibility. This is especially true for the Third Pole (TP) (Qiu 2008), which is the high-elevation area in Asia centered on the Tibetan Plateau and the origin of major Asian rivers (Fig. 1).

Fig. 1.
Fig. 1.

Topography and boundaries of nine upper river basins in the TP. UYE, UYA, UM, US, UB, UI, UAMD, USRD, and UYK denote the upper regions of the Yellow, Yangtze, Mekong, Salween, Brahmaputra, Indus, Amu Darya, Syr Darya, and Yarkant river basins, respectively. The green dots denote the national meteorological stations used in this study.

Citation: Journal of Hydrometeorology 22, 8; 10.1175/JHM-D-20-0272.1

Gridded datasets obtained by interpolating limited observations at low altitudes tend to largely underestimate precipitation in high mountain regions due to strong orographic precipitation gradients and wind-induced undercatch of solid precipitation (Yang et al. 2005; Ma et al. 2015), leading to substantial uncertainties in streamflow simulations with land surface hydrological models (Tong et al. 2014a,b; Immerzeel et al. 2015; Dahri et al. 2016; Kan et al. 2018; Sun and Su 2020). Many efforts have been made to explore the potential utility of climate inputs from different data sources (i.e., satellite remote sensing, atmospheric reanalysis, and high-resolution climate model outputs) other than sparse ground observations in hydrological modeling studies for high mountainous regions (Tong et al. 2014a; Ma et al. 2016; He et al. 2017; Qi et al. 2018). Evaluation studies suggest that purely satellite-based estimates without corrections tend to overestimate the gauge observations in the southeastern TP (Gao and Liu 2013; Tong et al. 2014a), due to errors resulting from the gaps in revisiting times, sampling uncertainties, and even the retrieval algorithms (Gao and Liu 2013; Ma et al. 2016). Atmospheric reanalysis datasets tend to largely overestimate the gauge observed mean annual precipitation (by 40%–160%) in the TP (Wang and Zeng 2012; Tong et al. 2014b), although they are generally capable of capturing the annual cycle and broad spatial–temporal distributions of observed precipitation (Wang and Zeng 2012; Tong et al. 2014b).

Climate models, including global climate models (GCMs) and regional climate models (RCMs), provide valuable meteorological information to study climate changes in the TP (Gao et al. 2015; Li et al. 2019). However, precipitation in the CMIP5 GCMs generally has consistent wet biases by overestimating the observed climatological annual means in the TP by 100%–183% in 16 out of the 24 GCMs in 1961–2005 (Su et al. 2013), which can introduce significant errors into hydrological simulations and projections of the region.

RCMs are used to conduct dynamic downscaling using lateral boundary conditions provided by GCM outputs or reanalysis to simulate detailed climatological information over the region of interest, which is critical but poorly represented in the GCMs (Gao et al. 2015; Norris et al. 2015; Y. Wang et al. 2020). For example, the Weather Research and Forecasting (WRF) Model, which is a widely used dynamic downscaling model, is both a mesoscale numerical weather prediction system and a regional climate model (Skamarock et al. 2005). The WRF provides many different physical parameterization options suitable for applications at scales of meters to thousands of kilometers for both operational forecasting and atmospheric research (Skamarock et al. 2005). The WRF Model has been used in climate and hydrological studies to provide high-resolution meteorological variables to compensate for the scarcity of in situ meteorological measurements in high mountain regions (Maussion et al. 2014; Gao et al. 2015; Li et al. 2017; Gao et al. 2020a). However, long-term precipitation datasets with fine spatial–temporal resolutions for the entire TP have been relatively rare until recently.

The High Asia Refined analysis (HAR) data (Maussion et al. 2011, 2014) is a pioneering effort. It offers precipitation estimate based on the WRF dynamic downscaling of coarser global analysis using 30- and 10-km nested domains and covers the entire High Asia with 30 km and most of the TP with 10 km for the period of 2000–14. The HAR shows great potential in describing high-altitude water fluxes (Curio et al. 2015; Curio and Scherer 2016), glacier variability (Mölg et al. 2012, 2013), and large-scale spatial–temporal patterns of precipitation in the TP (Maussion et al. 2014; Curio et al. 2015). Recently, a new version of HAR (HAR v2) for 2004–18 has been released, which produces slightly higher precipitation amounts than its previous version (X. Wang et al. 2020). Another precipitation estimate was developed by the Regional Climate Group at the University of Gothenburg using the WRF Model, and it provides detailed, process-based precipitation fields with a 9-km resolution (the WRF-9km hereinafter) for the entire TP for 2001–08 (Ou et al. 2020). This WRF-9km product can accurately reproduce the spatial patterns and the diurnal cycles of summer precipitation in the TP when compared with both satellite and gauge observations (Ou et al. 2020). However, the potential capabilities of HAR and WRF-9km precipitation estimates in hydrological simulations have not been well evaluated in TP river basins.

In this study, the HAR and WRF-9km precipitation estimates are compared with gauge-based estimates in nine selected basins over the TP in terms of annual, seasonal, and spatial patterns of precipitation during 2001–08. The ability of the RCMs in detecting precipitation events at different thresholds are evaluated using daily records from 37 gauges for 2001–08. The potential hydrological utilities of the RCM precipitation estimates are investigated with help from a large-scale land surface hydrological model. The aim of this study is to assess the strengths and limitations of the RCM precipitation estimates for hydrological modeling in the TP, which is characterized by complex terrain and sparse gauge observations.

2. Study area

The study area focuses on the upstream of nine rivers in the TP, including the upper Yangtze (UYA), Yellow (UYE), Mekong (UM), Salween (US), Brahmaputra (UB), Indus (UI), Yarkant (UYK), Amu Darya (UAMD), and Syr Darya (USRD) basins (Fig. 1). The basin areas range from 46 704 to 284 800 km2 (Table 1), with a total area of about 127.7 × 104 km2, accounting for about 25.6% of the entire TP. The mean basin elevations range from 1450 to 4881 m, with a mean value of 3750 m (Table 1).

Table 1.

Characteristic of major upstream river basins in the Third Pole and precipitation data used in this study. UYA, UYE, UM, US, UB, UYK, UI, UAMD, and USRD denote the upper regions of the Yangtze, Yellow, Mekong, Salween, Brahmaputra, Yarkant, Indus, Amu Darya, and Syr Darya river basins, respectively.

Table 1.

Five of the nine basins (the UYE, UYA, UM, US, and UB) are located in the southeastern part of the TP, which is monsoon dominated with more than 70% of annual precipitation occurring in June–September (Ma et al. 2018). The remaining four basins (the UI, UYK, UAMD, and USRD) in the western TP are mostly affected by the westerlies system (Ma et al. 2018). The UI, UAMD, and USRD have similar precipitation regimes with more than 70% of annual precipitation occurring in November–April, and more than 55% of annual precipitation occurs in May–August over the UYK basin (Mölg et al. 2013; Kan et al. 2018).

3. Data and method

a. Precipitation data

1) RCM precipitation estimates

The HAR (Maussion et al. 2014) and WRF-9km (Ou et al. 2020) precipitation datasets are both generated by dynamical downscaling using the WRF Model. There are two versions of HAR datasets with spatial resolutions of 10 km (HAR-10km) and 30 km (HAR-30km) for October 2000–September 2011. The HAR-10km covers most of the TP, while the HAR-30km covers all of Central Asia. The daily HAR-10km precipitation is used for the UYA, UM, US, UB, UI, and UYK basins, while the daily HAR-30km data are used for the UYE, UAMD, and USRD basins, which are outside of the coverage of HAR-10km. Both the HAR-10km and HAR-30km datasets are accessed from a public data server (https://www.klima.tu-berlin.de/), and they are hereinafter referred to as HAR. The daily WRF-9km precipitation data are available from the National Tibetan Plateau Data Center (https://data.tpdc.ac.cn). Because HAR and WRF-9km have different simulation periods, the two datasets during the overlapping period of January 2000–December 2008 are evaluated in this study.

The model configuration and physical parameterization schemes adopted in the generation of the HAR and WRF-9km are listed in Table 2. The major differences between the HAR and WRF-9km lie in the cumulus parameterization and forcing strategy (Table 2). The cumulus parameterization of the new Grell–Devenyi 3 scheme and the forcing strategy of the daily reinitialization driven by global tropospheric analyses are used in HAR to avoid the growth of errors in the planetary to synoptic scale features within the simulation domain (Hong and Kanamitsu 2014). The WRF-9km is simulated using a continuous integration forcing strategy driven by global reanalysis, with spectral nudging to prevent the simulation from drifting away from the large-scale driving fields (Hong and Kanamitsu 2014). Cumulus parameterization is not used in WRF-9km since cumulus parameterizations tend to overestimate precipitation at a 9-km resolution over the TP (Ou et al. 2020).

Table 2.

The HAR and WRF-9km model strategies.

Table 2.

2) Gauge observations

Daily precipitation records from a total of 37 meteorological stations run by the China Meteorological Administration (CMA, http://data.cma.cn/) for 2001–08 (green dots in Fig. 1) are used to evaluate the RCM estimates at point scales. A total of 35 stations are located in the five monsoon-dominated basins, and two stations are in the westerlies-dominated UI and UYK basins. These gauge data are directly used without further bias adjustment since they have undergone quality control procedures to eliminate erroneous and homogenous assessment (http://data.cma.cn/).

3) Gauge-based gridded data

Four different gauge-based gridded precipitation datasets for 2000–08 are used to compare with the RCM estimates at basin scales. Given the relatively dense station coverage in the southeastern regions of TP, daily gridded precipitation at 1/12° × 1/12° (10 km × 10 km) grids interpolated from the CMA meteorological stations is used in the four monsoon basins of UYA, UYE, UM, and US, which has been previously proved to have adequate accuracy in hydrological simulations (Zhang et al. 2013; Su et al. 2016; Meng et al. 2019).

In the UB, a newly reconstructed daily gridded precipitation dataset is used (Sun and Su 2020), which includes precipitation gradient and linear corrections based on 262 gauges in the basin. The daily gridded data in the westerlies-dominated UYK are adopted from Kan et al. (2018), where the data are generated based on precipitation gradient observations and CMA stations in the UYK. The rationalities of these two precipitation datasets have been extensively evaluated and validated through hydrological modeling using discharge, and glacier and snow observations in the basins (Sun and Su 2020; Kan et al. 2018).

Data from the Asian Precipitation-Highly Resolved Observational Data Integration Toward the Evaluation of Water Resources (APHRODITE’s Water Resources) Project (APHRODITE, http://www.chikyu.ac.jp/precip/) are used in the westerlies-dominated UI, UAMD, and USRD basins. The APHRODITE dataset is the only long-term continental-scale daily product that contains a dense network of 5000–12 000 daily rain gauge data for Asia (Yatagai et al. 2012). However, the APHRODITE dataset has been reported to largely underestimate the real precipitation in the UI when compared with observed runoff (Immerzeel et al. 2015; Dahri et al. 2016; Shafeeque et al. 2019), which has been a common problem in gauge-based gridded data. Therefore, the APHRODITE data are basically used as a reference instead of a “truth” for comparison with the RCM precipitation estimates in the UI, UAMD, and USRD. However, the gauged-based gridded data used in the other basins could be considered to be reliable.

5) Other gridded datasets

Monthly precipitation estimates from the Integrated Multisatellite Retrievals for Global Precipitation Measurement (GPM IMERG) V06 at 0.1° × 0.1° spatial resolution and from ERA5 reanalysis data developed by the European Centre for Medium-Range Weather Forecasts (ECMWF) at 0.25° × 0.25° grids are also used to compare with the RCM estimates for 2000–08 at basin scales.

The algorithm for GPM is intended to intercalibrate, merge, and interpolate “all” satellite microwave precipitation estimates, together with microwave-calibrated infrared (IR) satellite estimates, precipitation gauge analyses, and potentially other precipitation estimators at fine time and space scales for GPM eras over the entire globe (https://gpm.nasa.gov/data/directory). Current studies suggest that the GPM product outperforms TRMM at all spatial scales and elevation ranges in detecting daily rainfall accumulation (Ma et al. 2016; Xu et al. 2017; He et al. 2017). ERA5 is the fifth generation of ECMWF atmospheric reanalyses of the global climate, which will be generated to replace ERA-Interim. ERA5 has important changes relative to the former ERA-Interim atmospheric reanalysis including higher spatial and temporal resolutions as well as a more recent model and data assimilation system (https://www.ecmwf.int/en/forecasts/datasets/reanalysis-datasets/era5).

b. Hydrological model

Hydrological models offer a useful tool for inversely evaluating the potential utility of precipitation datasets (Duethmann et al. 2013; Immerzeel et al. 2015; Wortmann et al. 2018; Sun and Su 2020). The Variable Infiltration Capacity (VIC) hydrological model (Liang et al. 1994, 1996) is used to evaluate the hydrological utility of the RCM precipitation estimates in the TP basins. The VIC model is a grid-based land surface scheme that parameterizes the dominant hydrometeorological processes taking place at the land surface–atmosphere interface. The model solves both the surface water and energy balances within the grid cell. The VIC model uses a mosaic representation of the land cover and a parameterization for infiltration that accounts for subgrid-scale heterogeneities in the land surface hydrologic processes. It has been previously used in flow simulations in major river basins in the TP (Zhang et al. 2013; Tong et al. 2014a; Su et al. 2016; Kan et al. 2018; Meng et al. 2019; Zhao et al. 2019). The required input data for the VIC model include daily meteorological forcing data (precipitation, maximum and minimum temperatures, and wind speeds), soil texture, and vegetation types. The gauge-based gridded forcing data for 2001–08 that are described above are used to drive the VIC model constituting reference model simulations to compare with those driven with the RCMs precipitation. The modeling frameworks and parameters for the UYA, UYE, US, UM, UI, and UB basins are adopted from Zhang et al. (2013) and Sun and Su (2020), while the model setup of the westerlies-dominated UYK basin is adopted from Kan et al. (2018) without further calibration. In addition, to adjust the model internal stores of energy and water from the initial condition to an equilibrium state, the VIC model is first run for the years of 2001–04 for warming up, then is rerun for the entire period of 2001–08.

To exclude the impact of glacier runoff on the precipitation evaluation, the offline glacier scheme, which is included in previous applications of the VIC-Glacier model, is not used in this study. Therefore, the differences in simulated streamflow obtained for the same set of VIC model parameters are entirely attributed to the differences in the precipitation inputs (Su et al. 2008; Tong et al. 2014a; He et al. 2019). The RCM precipitation-driven VIC simulations are compared with the observed streamflow data at seven outlets (Fig. 1; Table 1) of the monsoon-dominated UYE, UYA, UM, US, and UB and the westerlies-dominated UI and UYK for 2001–08, while the UAMD and USRD are excluded from this comparison because of a lack of observed streamflow data for the selected time periods.

c. Statistical indexes

The statistical indexes (Table 3) of correlation coefficient (CC), relative bias (bias; %), and Nash–Sutcliffe efficiency (NSE) are used to quantify the agreement and the systemic deviation between the RCM and gauge-based precipitation estimates at basin scales, as well as between the simulated and observed streamflow at the seven outlets of selected basins (Fig. 1, Table 1).

Table 3.

Statistical evaluation metrics. For the equations in this table, n is the total number of dates, i is the ith date, Oi is the observations, Si is the simulated streamflow, O¯i is the average of the observations, Pi is the precipitation estimates of the RCMs, P¯ is the average of the precipitation estimates of the RCMs, A is the number of precipitation days on which both the RCM precipitation estimates and the observations detected precipitation, B is the number of precipitation days on which the RCM precipitation estimates detected precipitation but the observations did not, and C is the number of precipitation days on which the observations detected precipitation but the RCM precipitation estimates did not.

Table 3.

To quantify the ability of RCMs in detecting daily precipitation events, the probability of detection (POD), false alarm ratio (FAR), and critical success index (CSI) are computed (Table 3). The POD represents the ratio of the number of precipitation events that are correctly predicted by the RCM to the total number of events measured at the gauge station, and it indicates how well the RCM data detects the occurrence of precipitation events. The FAR is the ratio of the number of times the RCM detects a precipitation event that is not measured at the gauge station to the total number of events detected. The CSI is the fraction of the RCM precipitation events that are correctly predicted during the entire period. Typically, a threshold value of 0.1 mm day−1 is adopted to identify the rain/no-rain events (Tian et al. 2009; Tang et al. 2020).

The probability distribution functions (PDF) of precipitation are computed to evaluate the abilities of the RCM precipitation estimates to detect the occurrence frequency in comparison with the gauge observations. The PDF is the percentage of the number of detected events under a specific rain-rate rank out of all of the rain events (Li et al. 2013). The discrepancies between the amounts in the specific rain-rate ranks detected by the RCM and the gauge observations are quantified by bias values. To evaluate the effectiveness of the RCM precipitation estimates in detecting the frequency and bias at different precipitation thresholds, they are classified into five ranks based on intensity thresholds: two light precipitation ranks with intensities of 0.1–5 and 5–10 mm day−1, a moderate precipitation rank with an intensity of 10–25 mm day−1, a heavy precipitation rank with intensities of 25–50 mm day−1, and an extreme heavy precipitation rank with an intensity of greater than 50 mm day−1 (Tan et al. 2015; He et al. 2017; Xu et al. 2017).

The flow duration curve (FDC) is used to evaluate the potential usefulness of the RCM precipitation in the probability of daily streamflow simulation (Cigizoglu and Bayazit 2000; Lutz et al. 2016). The FDC typically represents the relationship between the magnitude and frequency of the daily streamflow in a particular river basin, providing an estimate of the percentage of time during the entire study period. The 95% probability of exceedance is defined as the threshold of flood events simulated by gauge-based estimates in this study, which is used as a reference to compare the flow simulations with the RCM precipitation.

4. Comparison and evaluation of the RCM precipitation estimates

a. Comparison at basin scales

Figure 2 shows the time series of annual precipitation from the RCMs and gauge-based estimates in the nine selected basins for 2001–08. Overall, the RCMs are in good agreement with the gauge-based estimates in annual variations, with CCs of 0.43–0.86 for seven of the nine basins for the HAR, and CCs of 0.52–0.90 for six of the basins for the WRF-9km. However, the HAR tends to overestimate the gauge-based estimates in annual means in both monsoon and westerly basins (with biases of 39%–95%) except in the UB where the HAR has an underestimate of 14% (Fig. 2). The WRF-9km also exhibits an overall overestimate in all basins but tends to have a smaller positive bias in the monsoon basins (bias of 1%–41% with the smallest in the US and UM) than in the westerly basins (46%–129%) (Fig. 2). The two RCMs agree well with each other in terms of the annual pattern in the westerly basins (CC > 0.8 and bias within ±23%; Figs. 2f–i), but they show little agreement in the monsoon basins (Figs. 2a–e), except for the UYE (CC = 0.69 and bias = 14%; Fig. 2b).

Fig. 2.
Fig. 2.

Time series of annual precipitation from the RCMs and gauge-based precipitation estimates for the nine selected basins in the TP for 2001–08 (G: gauge-based precipitation estimates; H: HAR; W: WRF-9km). An asterisk indicates 95% significance confidence level.

Citation: Journal of Hydrometeorology 22, 8; 10.1175/JHM-D-20-0272.1

At seasonal scales (Fig. 3), both the RCMs exhibit consistent precipitation seasonality with the gauge-based estimates in the monsoon basins, where more than 70% of annual precipitation occurs in June–September (Figs. 3a–e). However, the HAR tends to overestimate the gauge-based estimates in all seasons (26%–76% in warm seasons and 96%–333% in cold seasons) of the monsoon basins, except for the UB (Fig. 3e) where all seasons are underestimated, except for winter. The WRF-9km produced overestimations in the UYA and UYE (20%–49% in warm seasons and 38%–264% in cold seasons), and underestimations in the UM and US in summer (bias of 12%–21%, Figs. 3c,d). The WRF-9km produces overestimations in all of the seasons in the UB (bias of 30%–367%), except for summer (Fig. 3e).

Fig. 3.
Fig. 3.

Seasonal cycles of the RCM and gauge-based precipitation estimates for the nine basins in the TP for 2001–08.

Citation: Journal of Hydrometeorology 22, 8; 10.1175/JHM-D-20-0272.1

Both RCMs exhibit a winter–spring precipitation maximum in all the westerly basins (Figs. 3f–i), while the seasonal pattern is only present in the UAMD and USRD (Figs. 3g,h) in the gauge-based estimates, which show a bimodal intra-annual hyetograph in spring and summer in the UI (Fig. 3f), and a summer maximum in the UYK (Fig. 3i), due to the orographic barrier created by the Pamir-Tian Shan mountains, which restricts the moisture from the westerlies (Khromova et al. 2006; Baldwin and Vecchi 2016; Chen et al. 2020). The HAR tends to largely overestimate the gauge-based estimates in all the months (by 44%–213%), especially the cold seasons (bias of 78%–213%), while the WRF-9km tends to underestimate the gauge-based estimates in summer (bias < 30%) (Figs. 3f–i).

Figure 4 shows the spatial patterns of the mean annual and seasonal precipitation estimates from the RCM and gauge-based estimates for the nine basins for 2001–08. The mean annual precipitation from the gauge-based estimates increases gradually from the northeast (300–500 mm yr−1) to southwest (600–1400 mm yr−1) in the monsoon basins and increases gradually from the southeast (<200 mm yr−1) to northwest (300–800 mm yr−1) in the westerly basins (Fig. 4a). The spatial patterns of mean annual precipitation obtained from the RCM estimates are similar to those of the gauge-based estimates, while the RCM estimates present more details on the spatial variability of the precipitation in comparison with the gauge-based data (Figs. 4d,g). In June–September (Figs. 4b,e,h), all the precipitation estimates can detect the monsoon signals well, which is consistent with the results in Fig. 3. While in the nonmonsoon seasons (i.e., October–May), the RCM precipitation estimates tend to have much stronger westerlies signals than the gauge-based estimates, which have much less winter–spring precipitation than the RCMs in the western basins (Figs. 4c,f,i).

Fig. 4.
Fig. 4.

Spatial patterns of the mean annual and seasonal gauge-based, HAR, and WRF-9km precipitation estimates for the nine basins in the TP for 2001–08.

Citation: Journal of Hydrometeorology 22, 8; 10.1175/JHM-D-20-0272.1

b. Evaluation at stations

Figure 5 shows comparisons of the precipitation estimates between the 37 gauges and the corresponding RCMs grids for 2001–08 at daily, monthly, and annual scales. The HAR and WRF-9km estimates exhibit reasonable performances at the monthly and annual scales, with CCs of 0.5–0.8 (p < 0.05) and biases within ±3.3% (Figs. 5c–f). At the daily scale, both the HAR and WRF-9km exhibit poor correlations with the gauge records, with CCs of <0.4 (Figs. 5a,b). However, the RCMs exhibit a good ability in detecting daily precipitation occurrence, with high PODs of 0.7–0.8, CSIs of 0.5–0.6, and a small FAR of 0.4 (Figs. 5a,b).

Fig. 5.
Fig. 5.

Precipitation estimates from the 37 gauges compared with the corresponding (left) HAR and (right) WRF-9km grids for 2001–08 at (a),(b) daily; (c),(d) monthly; and (e),(f) annual scales. An asterisk indicates 95% significance confidence level.

Citation: Journal of Hydrometeorology 22, 8; 10.1175/JHM-D-20-0272.1

The PDF of the HAR, WRF-9km, and gauge estimates at five precipitation intensities for 2001–08 and the corresponding absolute bias relative to the gauge estimates are shown in Fig. 6. Both the RCM estimates accurately capture the distribution of the PDF indicated by the gauge observations (Fig. 6a), with more than 90% of precipitation events falling within the light precipitation category (<5 mm day−1), and only about 3% of precipitation events falling within heavy precipitation category (>25 mm day−1). Both the HAR and WRF-9km precipitation estimates tend to overestimate the heavy precipitation events (15–20 mm), but they yield smaller absolute bias values (<3 mm) for the precipitation accumulations of the other precipitation intensity categories (Fig. 6b). The above suggests that the HAR and WRF-9km can effectively reproduce precipitation events of various intensities, except for heavy precipitation.

Fig. 6.
Fig. 6.

(a) Probability distribution functions (PDF; %) for the HAR, WRF-9km, and gauge precipitation estimates at five precipitation intensities for 2001–08, and (b) the corresponding absolute biases (mm) relative to the gauge observations.

Citation: Journal of Hydrometeorology 22, 8; 10.1175/JHM-D-20-0272.1

c. Comparison of the RCM with other precipitation estimates

Owing to the lack of direct meteorological observations, large spreads are commonly reported in precipitation estimates from different data sources in the TP regions (Wang and Zeng 2012; Gao and Liu, 2013; Tong et al. 2014b; You et al. 2015; Bai et al. 2016). Table 4 compares the mean annual precipitation estimates of the RCMs used in this study with those of gauge-based, satellite-based GPM IMERG V6, and ERA5 reanalysis data, which have been widely used in previous studies in the TP. Mean annual precipitation estimates from satellite-based GPM precipitation estimates tend to be close to the gauge-based estimates with smallest biases of −68% to 20% in the nine TP basins, followed by the RCM estimates (HAR and WRF-9km) with biases of 39%–95%, and the ERA5 reanalysis data tend to largely overestimate the gauge-based estimates by 64%–310% in the basins (Table 4).

Table 4.

Mean annual precipitation estimates from the RCMs and other data sources in the TP basins.

Table 4.

Note that the GPM is corrected using the gauge-based global precipitation climatology project (GPCP) precipitation on a monthly basis, which explains the smallest bias relative to the gauge observations in the monsoon basins (bias from −13% to 10%). However, in the westerly basins where the gauges are sparse or nonexistent, the GPM shows the largest bias from −68% to 20% relative to the gauge-based estimates. Tong et al. (2014a) evaluate the performance of the purely satellite-derived TRMM real-time version (3B42RT) with gauged data in the TP, and suggest a large positive bias of 100%–500% in the UYA and UYE, which is much higher than in the RCMs estimates when compared with the gauge-based precipitation (bias of 40%–95%; Fig. 2). The above suggest that mean annual the HAR and WRF-9km precipitation estimates are closer to the gauge-based estimates than the ERA5 reanalysis and purely real-time satellite-based estimates, showing encouraging potential for use in climate and hydrological studies in high mountainous basins.

5. Hydrological evaluation of the RCM precipitation estimates

In this section, the VIC model simulated streamflow simulations driven by the RCMs and gauge-based precipitation estimates are compared with each other and with flow observations for 2001–08.

a. Monthly simulation

Figure 7 shows mean monthly simulated streamflow driven by the RCM precipitation in seven TP basins where observed streamflow data are available for 2001–08. All the model results successfully reproduce the seasonal pattern in the observed streamflow, with peak flow occurring in warm seasons. However, large discrepancies exist between simulated and observed streamflow.

Fig. 7.
Fig. 7.

Simulated mean monthly streamflow driven by the HAR and WRF-9km precipitation in seven TP basins for 2001–08.

Citation: Journal of Hydrometeorology 22, 8; 10.1175/JHM-D-20-0272.1

The HAR-driven simulations tend to overestimate the observations by 23%–147% among the monsoon basins of UYA, UYE, UM, and US (Figs. 7a–d), where the WRF-9km-driven performs differently with overestimates (by 14%–60%) in the UYA and UYE (Figs. 7a,b) and underestimates (from −11% to −7%) in the UM and US (Figs. 7c,d). In the UB (Fig. 7e), both the HAR and WRF-9km exhibit a plausible performance in the flow simulations in terms of NSE (0.6–0.8) and bias (from −15% to 3%), especially in the summer.

In the westerly basins of UI and UYK (Figs. 7f,g), the WRF-9km-driven simulations tend to underestimate the observations by 4%–15%, while the HAR-driven simulations have an underestimate of 33% in the UI and overestimate of 14% in the UYK basin. The glacier contribution to the total runoff is estimated to be about 26%–48% in the UI basin (Zhang et al. 2013; Lutz et al. 2014) and about 52% in the UYK basin (Kan et al. 2018). If we take these numbers as the glacier runoff contribution, the HAR precipitation shows a good potential for hydrological simulations in the UI, while large overestimates may still exist in the WRF-9km precipitation for the UI and in both RCMs for the UYK (Figs. 7f,g).

b. Daily simulation

Figure 8 compares the daily streamflow simulated using the HAR and WRF-9km precipitation in the seven TP basins for 2001–08. Because there is a lack of daily observed streamflow data, the model results driven by gauge-based precipitation are used as a reference. Both simulations with HAR and WRF-9km precipitation show a good correspondence with the reference in daily flow variations in all the selected basins (with CCs of 0.64–0.87). However, they tend to overestimate the summer flows by 11%–185%. This is mostly attributed to the large overestimates of the RCM precipitation inputs in these basins, except for the UB.

Fig. 8.
Fig. 8.

Simulated daily streamflow driven by the HAR, WRF-9km, and gauge-based precipitation in the seven TP basins for 2001–08.

Citation: Journal of Hydrometeorology 22, 8; 10.1175/JHM-D-20-0272.1

Figure 9 shows the FDC of simulated daily streamflow with HAR, WRF-9km, and gauge-based precipitation in the seven selected basins for 2001–08. The HAR- and WRF-9km-driven simulations generally capture well the magnitude and frequency of low and moderate daily flow (exceedance probability of <50%) in the gauge-based simulations in all the basins, except for the UI (Fig. 9f). While both RCMs tend to overestimate the magnitude of flood events (exceedance probability of >95%), with about 16%–36% in HAR- and WRF-9km-driven simulations and 5% in the gauge-based estimates. When compared with the simulated FDCs forced with gauge-based, the simulations with WRF-9km precipitation match better than HAR-driven simulations in most of the basins, especially in the UM and US basins (Figs. 9c,d).

Fig. 9.
Fig. 9.

Flow duration curves for the simulated daily streamflow driven by the HAR, WRF-9km, and gauge-based precipitation in the seven TP basins for 2001–08.

Citation: Journal of Hydrometeorology 22, 8; 10.1175/JHM-D-20-0272.1

This suggests that the HAR and WRF-9km can successfully capture the magnitudes of low and moderate daily streamflow (exceedance probability of <50%), but they are unable to successfully catch flood events because of large precipitation overestimations in most the basins, except for the good performance of the HAR and WRF-9km in the UB and that of the WRF-9km in the UM and US basins.

6. Discussion

a. Uncertainties in the generation of RCMs

RCMs are sensitive to initial and time-dependent meteorological lateral boundary conditions, horizontal resolution, cumulus, planetary boundary layer (PBL), and microphysics schemes, as well as model versions (Huang and Gao 2018; Laprise 2008; Ou et al. 2020; Xu et al. 2019; Yu 2013). The HAR and WRF-9km differ from each other in terms of the abovementioned model settings.

The HAR is generated by WRF-ARW, version 3.3.1, using the National Centers for Environmental Prediction (NCEP) Global Tropospheric Analyses, which consist of consecutive reinitialized model runs, while the WRF-9km is generated by WRF-ARW, version 3.7.1, using ERA5 reanalysis data, which is continued with nudging. Different schemes as well as their combinations have a great impact on precipitation simulations (X. Wang et al. 2020; Y. Wang et al. 2020), especially in summer season, because small-scale processes play a larger role in triggering convective precipitation under weak synoptic-scale forcing in summer than under strong synoptic-scale forcing such as in winter (Awan et al. 2011). Different PBL schemes have shown an important impact on precipitation pattern in summer over the TP (Wu et al. 2015). The Yonsei University (YSU) PBL scheme used in the WRF-9km shows better abilities than the Mellor–Yamada–Janjić (MYJ) PBL scheme used in the HAR in simulating precipitation pattern in summer in the east and middle of the TP (Wu et al. 2015). In addition, the WRF double moment 6-class (WDM6) microphysics scheme, used in the WRF-9km, better simulates the vertical structure of warm-type heavy rain than that of the Thompson used by HAR (Song and Sohn 2018). This may lead to better captured precipitation in the WRF-9km over the relatively warmer eastern TP than that of the HAR. This may partly explain why the WRF-9km exhibits a high potential for hydrological application in the monsoon-dominated basins in the southeastern TP. In addition, lake configuration may affect precipitation (Gao et al. 2020a,b). HAR uses the inland water module, which was found to be a cause for the great overestimation in the TP. However, the nearest SST method, which is used in the WRF-9km simulation, may greatly overestimate precipitation around big lakes.

The coarse resolution of the RCMs may be the cause of wet biases in precipitation estimates in the TP (Lin et al. 2018; Y. Wang et al. 2020; Gao et al. 2020b). RCMs with horizontal grid spacings of less than 4 km, also called convection permitting modeling (CPM), can resolve the convective processes and avoid error-prone convection parameterization by explicitly resolving the cumulus plumes (Sun et al. 2016; Li et al. 2019). CPM offers the advantage of improving the representation of fine-scale variations in orography and in land surface characteristics, which is especially beneficial in mountainous regions and in areas with heterogeneous land surfaces (Gao et al. 2020b). Thus, an alternative way to improve the quality of the RCM simulated precipitation over the TP is to conduct CPMs, i.e., models with horizontal resolutions of less than 4 km, to explicitly resolve convective precipitation. In fact, the capacity of CPMs to simulate precipitation over the TP is being explored through the newly endorsed Coordinated Regional Downscaling Experiment Flagship Pilot Studies (CORDEX-FPS) with a focus on the TP (http://rcg.gvc.gu.se/cordex_fps_cptp).

b. Uncertainties in the evaluation of RCMs

When comparing the precipitation products with the gauge observations at sites, the spatial density of the rain gauges is an important source of uncertainty in the evaluation of the precipitation estimates (Tang et al. 2018; Tian et al. 2018; Zhang et al. 2020). The CC and Bias exhibit consistent upward and downward trends, respectively, as the number of rain gauges increases (Tang et al. 2018; Tian et al. 2018; Zhang et al. 2020). For example, Sun and Su (2020) reconstruct the precipitation estimate for the entire UB based on 262 rain gauges, showing that the basin-averaged mean annual precipitation increased from 465 mm in the original interpolated data by 16 CMA stations to 709 mm in the reconstructed data for 1961–2016, and streamflow simulations with reconstructed estimates are significantly improved when compared with those forced by the original data.

Fortunately, numerous field research campaigns have recently set up rain gauges throughout wide regions and at different altitudes. Furthermore, most of the existing rain gauges operated by different Institutes have recently been compiled by the Second Tibetan Plateau Scientific Expedition and Research (STEP) Project led by the Institute of Tibetan Plateau Research, Chinese Academy of Sciences (http://data.tpdc.ac.cn), which will provide a unique observation basis for improving the evaluation of the performance of RCM precipitation estimates and for correcting the bias of the RCM datasets. In addition, the newly released HAR version2 has been extended to cover the entire TP (Fig. 1) available from 2004 to the present, and will be updated from 1979 to the present. A revised version of WRF-9km will also be updated from 1979 to 2016 in the near future. Both of the two products are driven by ERA5. Therefore, the long-term HAR v2 and WRF-9km precipitation dataset will be further evaluated based on the high-density gauge network and currently used hydrological modeling evaluation framework in the remote and data-sparse TP region, improving our understanding of climate-related and hydrological processes in high mountainous regions.

We hope the work reported here will improve communications between the hydrology community and regional precipitation product developers and eventually lead to improvements in the accuracy of regional simulated precipitation estimates for hydrologic applications in the TP.

7. Conclusions

In this study, precipitation estimates from two RCMs (HAR and WRF-9km) are evaluated at both basin and point scales in the TP for 2001–08. The potential hydrological utility of RCMs is investigated by driving the VIC large-scale land surface hydrological model in seven TP basins. The main results of this study are summarized below.

  1. Both RCM precipitation estimates exhibit precipitation seasonality consistent with the gauge-based estimates in the monsoon-dominated basins in the southeastern TP, while they reflect much stronger westerlies signals relative to the gauge-based data in the western TP basins. The RCMs tend to overestimate mean annual gauge-based estimates by 20%–95% in most of the TP basins, with the WRF-9km tending to have a smaller positive bias in the monsoon basins (<41%) than in the westerlies-dominated basins (46%–129%).

  2. Both the RCM precipitation estimates exhibit good abilities in detecting daily precipitation occurrences of varying intensities, and the magnitudes of specific rain-rate ranks indicated in the gauge observations, except for heavy precipitation (25–50 mm day−1).

  3. The RCMs precipitation estimates display an inconsistent hydrological simulation performance in the TP basins. The WRF-9km exhibits a high potential for streamflow simulations in the monsoon-dominated basins in the southeastern TP (with NSE of 0.7–0.9 and bias from −11% to 3%), while the HAR performs well in the UI and UB basins (NSE of 0.6 and bias from −15% to −9%).

  4. The RCM precipitation estimates can capture accurately the magnitudes of low and moderate daily streamflow (exceedance probability of <50%) but show limited capabilities in flood prediction (exceedance probability of >95%) in most of the TP basins.

Acknowledgments

This work is financially supported by the National Natural Science Foundation of China (91747201 and 41871057), the Second Tibetan Plateau Scientific Expedition and Research (STEP) program (2019QZKK0201), and the State Scholarship Fund of the China Scholarship Council (201904910329). The computations of the WRF-9km are enabled by resources provided by the Swedish National Infrastructure for Computing (SNIC) at the National Supercomputer Centre in Sweden (NSC) partially funded by the Swedish Research Council through Grant Agreement 2018-05973, and the Swedish Foundation for International Cooperation in Research and Higher Education (CH2019-8377). Authorship credits are as follows—He Sun: conceptualization, formal analysis, investigation, method, resources, visualization, and writing draft; Fengge Su: conceptualization, formal analysis, investigation, method, resources, visualization, funding acquisition, and writing (review and editing); Zhihua He: Formal analysis and writing (review and editing); Tinghai Ou and Deliang Chen: writing (review and editing) and providing WRF-9km data; Zhenhua Li and Yanping Li: writing (review and editing). The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

REFERENCES

  • Awan, N. K., H. Truhetz, and A. Gobiet, 2011: Parameterization-induced error characteristics of MM5 and WRF operated in climate mode over the alpine region: An ensemble-based analysis. J. Climate, 24, 31073123, https://doi.org/10.1175/2011JCLI3674.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bai, P., and Coauthors, 2016: Evaluation of streamflow simulation results of land surface models in GLDAS on the Tibetan plateau. J. Geophys. Res. Atmos., 121, 12 18012 197, https://doi.org/10.1002/2016JD025501.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Baldwin, J., and G. Vecchi, 2016: Influence of the Tian Shan on arid extratropical Asia. J. Climate, 29, 57415762, https://doi.org/10.1175/JCLI-D-15-0490.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Beven, J., and M. Hornberger, 1982: Assessing the effect of spatial pattern of precipitation in modeling stream flow hydrographs. J. Amer. Water Resour. Assoc., 18, 823829, https://doi.org/10.1111/j.1752-1688.1982.tb00078.x.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chen, F., and Coauthors, 2020: The role of the westerlies and orography in Asian hydroclimate since the late Oligocene. Geology, 48, 728732, https://doi.org/10.1130/G47400.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cigizoglu, H. K., and M. Bayazit, 2000: A generalized seasonal model for flow duration curve. Hydrol. Processes, 14, 10531067, https://doi.org/10.1002/(SICI)1099-1085(20000430)14:6<1053::AID-HYP996>3.0.CO;2-B.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Curio, J., and D. Scherer, 2016: Seasonality and spatial variability of dynamic precipitation controls on the Tibetan Plateau. Earth Syst. Dyn., 7, 767782, https://doi.org/10.5194/esd-7-767-2016.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Curio, J., F. Maussion, and D. Scherer, 2015: A 12-year high-resolution climatology of atmospheric water transport over the Tibetan Plateau. Earth Syst. Dyn., 6, 109124, https://doi.org/10.5194/esd-6-109-2015.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dahri, Z. H., and Coauthors, 2016: An appraisal of precipitation distribution in the high-altitude catchments of the Indus basin. Sci. Total Environ., 548–549, 289306, https://doi.org/10.1016/j.scitotenv.2016.01.001.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Duethmann, D., and Coauthors, 2013: Evaluation of areal precipitation estimates based on downscaled reanalysis and station data by hydrological modelling. Hydrol. Earth Syst. Sci., 17, 24152434, https://doi.org/10.5194/hess-17-2415-2013.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gao, Y., J. Xu, and D. Chen, 2015: Evaluation of WRF mesoscale climate simulations over the Tibetan Plateau during 1979–2011. J. Climate, 28, 28232841, https://doi.org/10.1175/JCLI-D-14-00300.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gao, Y., and Coauthors, 2020a: Understanding precipitation recycling over the Tibetan Plateau using tracer analysis with WRF. Climate Dyn., 55, 29212937, https://doi.org/10.1007/s00382-020-05426-9.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gao, Y., F. Chen, and Y. Jiang, 2020b: Evaluation of a convection-permitting modeling of precipitation over the Tibetan Plateau and its influences on the simulation of snow-cover fraction. J. Hydrometeor. 21, 15311548, https://doi.org/10.1175/JHM-D-19-0277.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gao, Y. C., and M. F. Liu, 2013: Evaluation of high-resolution satellite precipitation products using rain gauge observations over the Tibetan Plateau. Hydrol. Earth Syst. Sci. 17, 837849, https://doi.org/10.5194/hess-17-837-2013.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • He, Z., and Coauthors, 2017: Intercomparisons of rainfall estimates from TRMM and GPM multisatellite products over the upper Mekong river basin. J. Hydrometeor., 18, 413430, https://doi.org/10.1175/JHM-D-16-0198.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • He, Z., and Coauthors, 2019: Constraining hydrological model parameters using water isotopic compositions in a glacierized basin, Central Asia. J. Hydrol., 571, 332348, https://doi.org/10.1016/j.jhydrol.2019.01.048.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hong, S.-Y., and M. Kanamitsu, 2014: Dynamical downscaling: Fundamental issues from an NWP point of view and recommendations. Asia-Pac. J. Atmos. Sci., 50, 83104, https://doi.org/10.1007/s13143-014-0029-2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Huang, D., and S. Gao, 2018: Impact of different reanalysis data on WRF dynamical downscaling over China. Atmos. Res., 200, 2535, https://doi.org/10.1016/j.atmosres.2017.09.017.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Immerzeel, W. W., and Coauthors, 2015: Reconciling high-altitude precipitation in the upper Indus basin with glacier mass balances and runoff. Hydrol. Earth Syst. Sci., 19, 46734687, https://doi.org/10.5194/hess-19-4673-2015.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kan, B., and Coauthors, 2018: Generation of high mountain precipitation and temperature data for a quantitative assessment of flow regime in the upper Yarkant Basin in the Karakoram. J. Geophys. Res. Atmos., 123, 84628486, https://doi.org/10.1029/2017JD028055.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Khromova, T. E., and Coauthors, 2006: Changes in glacier extent in the Eastern Pamir, Central Asia, determined from historical data and ASTER imagery. Remote Sens. Environ., 102, 2432, https://doi.org/10.1016/j.rse.2006.01.019.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Laprise, R., 2008: Regional climate modelling. J. Comput. Phys., 227, 36413666, https://doi.org/10.1016/j.jcp.2006.10.024.

  • Li, L., and Coauthors, 2017: Evaluating the present annual water budget of a Himalayan headwater river basin using a high-resolution atmosphere-hydrology model. J. Geophys. Res. Atmos., 122, 47864807, https://doi.org/10.1002/2016JD026279.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Li, Y., and Coauthors, 2019: High-resolution regional climate modeling and projection over Western Canada using a weather research forecasting model with a pseudo-global warming approach. Hydrol. Earth Syst. Sci., 23, 46354659, https://doi.org/10.5194/hess-23-4635-2019.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Li, Z., D. Yang, and Y. Hong, 2013: Multi-scale evaluation of high-resolution multi-sensor blended global precipitation products over the Yangtze River. J. Hydrol., 500, 157169, https://doi.org/10.1016/j.jhydrol.2013.07.023.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Liang, X., and Coauthors, 1994: A simple hydrologically based model of land-surface water and energy fluxes. J. Geophys. Res., 99, 14 41514 428, https://doi.org/10.1029/94JD00483.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Liang, X., D. P. Lettenmaier, and E. F. Wood, 1996: One-dimensional statistical dynamic representation of subgrid spatial variability of precipitation in the two-layer variable infiltration capacity model. J. Geophys. Res., 101, 21 40321 422, https://doi.org/10.1029/96JD01448.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lin, C., and Coauthors, 2018: Impact of model resolution on simulating the water vapor transport through the central Himalayas: Implication for models’ wet bias over the Tibetan Plateau. Climate Dyn., 51, 31953207, https://doi.org/10.1007/s00382-018-4074-x.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lutz, A. F., and Coauthors, 2014: Consistent increase in High Asia’s runoff due to increasing glacier melt and precipitation. Nat. Climate Change, 4, 587592, https://doi.org/10.1038/nclimate2237.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lutz, A. F., and Coauthors, 2016: Climate change impacts on the upper Indus hydrology: Sources, shifts and extremes. PLOS ONE, 11, e0165630, https://doi.org/10.1371/journal.pone.0165630.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ma, Y., and Coauthors, 2015: Precipitation bias variability versus various gauges under different climatic conditions over the Third Pole Environment (TPE) region. Int. J. Climatol., 35, 12011211, https://doi.org/10.1002/joc.4045.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ma, Y., and Coauthors, 2016: Similarity and error intercomparison of the GPM and its predecessor-TRMM multisatellite precipitation analysis using the best available hourly gauge network over the Tibetan Plateau. Remote Sens., 8, 569, https://doi.org/10.3390/rs8070569.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ma, Y., and Coauthors, 2018: Atmospheric moisture transport versus precipitation across the Tibetan Plateau: A mini-review and current challenges. Atmos. Res., 209, 5058, https://doi.org/10.1016/j.atmosres.2018.03.015.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Maussion, F., and Coauthors, 2011: WRF simulation of a precipitation event over the Tibetan Plateau, China - An assessment using remote sensing and ground observations. Hydrol. Earth Syst. Sci., 15, 17951817, https://doi.org/10.5194/hess-15-1795-2011.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Maussion, F., and Coauthors, 2014: Precipitation seasonality and variability over the Tibetan Plateau as resolved by the High Asia Reanalysis. J. Climate, 27, 19101927, https://doi.org/10.1175/JCLI-D-13-00282.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Meng, F., and Coauthors, 2019: Changes in terrestrial water storage during 2003–2014 and possible causes in Tibetan Plateau. J. Geophys. Res. Atmos., 124, 29092931, https://doi.org/10.1029/2018JD029552.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mölg, T., and Coauthors, 2012: The footprint of Asian monsoon dynamics in the mass and energy balance of a Tibetan glacier. Cryosphere, 6, 14451461, https://doi.org/10.5194/tc-6-1445-2012.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mölg, T., F. Maussion, and D. Scherer, 2013: Mid-latitude westerlies as a driver of glacier variability in monsoonal High Asia. Nat. Climate Change, 4, 6873, https://doi.org/10.1038/nclimate2055.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Norris, J., and Coauthors, 2015: WRF simulations of two extreme snowfall events associated with contrasting extratropical cyclones over the western and central Himalaya. J. Geophys. Res. Atmos., 120, 31143138, https://doi.org/10.1002/2014JD022592.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ou, T., and Coauthors, 2020: Simulation of summer precipitation diurnal cycles over the Tibetan Plateau at the gray-zone grid spacing for cumulus parameterization. Climate Dyn., 54, 35253539, https://doi.org/10.1007/s00382-020-05181-x.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Qi, W., J. Liu, and D. Chen, 2018: Evaluations and improvements of GLDAS2.0 and GLDAS2.1 forcing data’s applicability for basin scale hydrological simulations in the Tibetan Plateau. J. Geophys. Res. Atmos., 123, 13 12813 148, https://doi.org/10.1029/2018JD029116.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Qiu, J., 2008: China: The third pole. Nature, 454, 393396, https://doi.org/10.1038/454393a.

  • Shafeeque, M., and Coauthors, 2019: Revealing vertical distribution of precipitation in the glacierized upper Indus Basin based on multiple datasets. J. Hydrometeor., 20, 22912314, https://doi.org/10.1175/JHM-D-19-0081.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Skamarock, W. C., J. B. Klemp, J. Dudhia, D. O. Gill, D. M. Barker, W. Wang, and J. G. Powers, 2005: A description of the Advanced Research WRF version 2. NCAR Tech. Note NCAR/TN-468+STR, 88 pp., https://doi.org/10.5065/D6DZ069T.

    • Crossref
    • Export Citation
  • Song, H.-J., and B.-J. Sohn, 2018: An evaluation of WRF microphysics schemes for simulating the warm-type heavy rain over the Korean Peninsula. Asia-Pac. J. Atmos. Sci., 54, 225236, https://doi.org/10.1007/s13143-018-0006-2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Su, F., Y. Hong, and D. P. Lettenmaier, 2008: Evaluation of TRMM Multisatellite Precipitation Analysis (TMPA) and its utility in hydrologic prediction in the La Plata Basin. J. Hydrometeor., 9, 622640, https://doi.org/10.1175/2007JHM944.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Su, F., and Coauthors, 2013: Evaluation of the global climate models in the CMIP5 over the Tibetan Plateau. J. Climate, 26, 31873208, https://doi.org/10.1175/JCLI-D-12-00321.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Su, F., and Coauthors, 2016: Hydrological response to future climate changes for the major upstream river basins in the Tibetan Plateau. Global Planet. Change, 136, 8295, https://doi.org/10.1016/j.gloplacha.2015.10.012.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sun, H., and F. Su, 2020: Precipitation correction and reconstruction for streamflow simulation based on 262 rain gauges in the upper Brahmaputra of southern Tibetan Plateau. J. Hydrol., 590, 125484, https://doi.org/10.1016/j.jhydrol.2020.125484.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sun, X., and Coauthors, 2016: An evaluation of dynamical downscaling of Central Plains summer precipitation using a WRF-based regional climate model at a convection-permitting 4 km resolution. J. Geophys. Res. Atmos., 121, 13 80113 825, https://doi.org/10.1002/2016JD024796.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Tan, M., and Coauthors, 2015: Evaluation of six high-resolution satellite and ground-based precipitation products over Malaysia. Remote Sens., 7, 15041528, https://doi.org/10.3390/rs70201504.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Tang, G., and Coauthors, 2018: Accounting for spatiotemporal errors of gauges: A critical step to evaluate gridded precipitation products. J. Hydrol., 559, 294306, https://doi.org/10.1016/j.jhydrol.2018.02.057.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Tang, G., and Coauthors, 2020: Have satellite precipitation products improved over last two decades? A comprehensive comparison of GPM IMERG with nine satellite and reanalysis datasets. Remote Sens. Environ., 240, 111697, https://doi.org/10.1016/j.rse.2020.111697.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Tian, F., and Coauthors, 2018: How does the evaluation of the GPM IMERG rainfall product depend on gauge density and rainfall intensity? J. Hydrometeor., 19, 339349, https://doi.org/10.1175/JHM-D-17-0161.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Tian, Y., and Coauthors, 2009: Component analysis of errors in satellite-based precipitation estimates. J. Geophys. Res., 114, D24101, https://doi.org/10.1029/2009JD011949.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Tong, K., and Coauthors, 2014a: Evaluation of satellite precipitation retrievals and their potential utilities in hydrologic modeling over the Tibetan Plateau. J. Hydrol., 519, 423437, https://doi.org/10.1016/j.jhydrol.2014.07.044.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Tong, K., and Coauthors, 2014b: Tibetan Plateau precipitation as depicted by gauge observations, reanalyses and satellite retrievals. Int. J. Climatol., 34, 265285, https://doi.org/10.1002/joc.3682.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wang, A., and X. Zeng, 2012: Evaluation of multireanalysis products with in situ observations over the Tibetan Plateau. J. Geophys. Res., 117, D05102, https://doi.org/10.1029/2011JD016553.

    • Search Google Scholar
    • Export Citation
  • Wang, X., and Coauthors, 2020: WRF-based dynamical downscaling of ERA5 reanalysis data for High Mountain Asia: Towards a new version of the High Asia refined analysis. Int. J. Climatol., 41, 743762, https://doi.org/10.1002/joc.6686.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wang, Y., and Coauthors, 2020: Synergy of orographic drag parameterization and high resolution greatly reduces biases of WRF-simulated precipitation in central Himalaya. Climate Dyn., 54, 17291740, https://doi.org/10.1007/s00382-019-05080-w.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wortmann, M., and Coauthors, 2018: Comparison and correction of high-mountain precipitation data based on glacio-hydrological modeling in the Tarim River headwaters (High Asia). J. Hydrometeor., 19, 777801, https://doi.org/10.1175/JHM-D-17-0106.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wu, Y., Y. Li, and X. Jiang, 2015: Influence of two planetary boundary layer parameterization schemes on summer rain in 2013 on Tibetan Plateau by WRF model. Plateau Mountain Meteor. Res., 33, 716.

    • Search Google Scholar
    • Export Citation
  • Xu, L., and Coauthors, 2019: The assessment of the planetary boundary layer schemes in WRF over the central Tibetan Plateau. Atmos. Res., 230, 104644, https://doi.org/10.1016/j.atmosres.2019.104644.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Xu, R., and Coauthors, 2017: Ground validation of GPM IMERG and TRMM 3B42V7 rainfall products over southern Tibetan Plateau based on a high-density rain gauge network. J. Geophys. Res. Atmos., 122, 910924, https://doi.org/10.1002/2016JD025418.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yang, D., and Coauthors, 2005: Bias corrections of long-term (1973–2004) daily precipitation data over the northern regions. Geophys. Res. Lett., 32, L19501, https://doi.org/10.1029/2005GL024057.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yatagai, K., K. Kamiguchi, O. Arakawa, A. Hamada, N. Yasutomi, and A. Kitoh, 2012: APHRODITE: Constructing a long-term daily gridded precipitation dataset for Asia based on a dense network of rain gauges. Bull. Amer. Meteor. Soc., 93, 14011415, https://doi.org/10.1175/BAMS-D-11-00122.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • You, Q., and Coauthors, 2015: Comparison of multiple datasets with gridded precipitation observations over the Tibetan Plateau. Climate Dyn., 45, 791806, https://doi.org/10.1007/s00382-014-2310-6.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yu, E., 2013: High-resolution seasonal snowfall simulation over Northeast China. Chin. Sci. Bull., 58, 14121419, https://doi.org/10.1007/s11434-012-5561-9.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, L., and Coauthors, 2013: Discharge regime and simulation for the upstream of major rivers over Tibetan Plateau. J. Geophys. Res. Atmos., 118, 85008518, https://doi.org/10.1002/jgrd.50665.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, L., and Coauthors, 2020: Interpolated or satellite-based precipitation? Implications for hydrological modeling in a meso-scale mountainous watershed on the Qinghai-Tibet Plateau. J. Hydrol., 583, 124629, https://doi.org/10.1016/j.jhydrol.2020.124629.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhao, Q., and Coauthors, 2019: Projecting climate change impacts on hydrological processes on the Tibetan Plateau with model calibration against the Glacier Inventory Data and observed streamflow. J. Hydrol., 573, 6081, https://doi.org/10.1016/j.jhydrol.2019.03.043.

    • Crossref
    • Search Google Scholar
    • Export Citation
Save
  • Awan, N. K., H. Truhetz, and A. Gobiet, 2011: Parameterization-induced error characteristics of MM5 and WRF operated in climate mode over the alpine region: An ensemble-based analysis. J. Climate, 24, 31073123, https://doi.org/10.1175/2011JCLI3674.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bai, P., and Coauthors, 2016: Evaluation of streamflow simulation results of land surface models in GLDAS on the Tibetan plateau. J. Geophys. Res. Atmos., 121, 12 18012 197, https://doi.org/10.1002/2016JD025501.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Baldwin, J., and G. Vecchi, 2016: Influence of the Tian Shan on arid extratropical Asia. J. Climate, 29, 57415762, https://doi.org/10.1175/JCLI-D-15-0490.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Beven, J., and M. Hornberger, 1982: Assessing the effect of spatial pattern of precipitation in modeling stream flow hydrographs. J. Amer. Water Resour. Assoc., 18, 823829, https://doi.org/10.1111/j.1752-1688.1982.tb00078.x.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chen, F., and Coauthors, 2020: The role of the westerlies and orography in Asian hydroclimate since the late Oligocene. Geology, 48, 728732, https://doi.org/10.1130/G47400.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cigizoglu, H. K., and M. Bayazit, 2000: A generalized seasonal model for flow duration curve. Hydrol. Processes, 14, 10531067, https://doi.org/10.1002/(SICI)1099-1085(20000430)14:6<1053::AID-HYP996>3.0.CO;2-B.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Curio, J., and D. Scherer, 2016: Seasonality and spatial variability of dynamic precipitation controls on the Tibetan Plateau. Earth Syst. Dyn., 7, 767782, https://doi.org/10.5194/esd-7-767-2016.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Curio, J., F. Maussion, and D. Scherer, 2015: A 12-year high-resolution climatology of atmospheric water transport over the Tibetan Plateau. Earth Syst. Dyn., 6, 109124, https://doi.org/10.5194/esd-6-109-2015.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dahri, Z. H., and Coauthors, 2016: An appraisal of precipitation distribution in the high-altitude catchments of the Indus basin. Sci. Total Environ., 548–549, 289306, https://doi.org/10.1016/j.scitotenv.2016.01.001.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Duethmann, D., and Coauthors, 2013: Evaluation of areal precipitation estimates based on downscaled reanalysis and station data by hydrological modelling. Hydrol. Earth Syst. Sci., 17, 24152434, https://doi.org/10.5194/hess-17-2415-2013.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gao, Y., J. Xu, and D. Chen, 2015: Evaluation of WRF mesoscale climate simulations over the Tibetan Plateau during 1979–2011. J. Climate, 28, 28232841, https://doi.org/10.1175/JCLI-D-14-00300.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gao, Y., and Coauthors, 2020a: Understanding precipitation recycling over the Tibetan Plateau using tracer analysis with WRF. Climate Dyn., 55, 29212937, https://doi.org/10.1007/s00382-020-05426-9.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gao, Y., F. Chen, and Y. Jiang, 2020b: Evaluation of a convection-permitting modeling of precipitation over the Tibetan Plateau and its influences on the simulation of snow-cover fraction. J. Hydrometeor. 21, 15311548, https://doi.org/10.1175/JHM-D-19-0277.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gao, Y. C., and M. F. Liu, 2013: Evaluation of high-resolution satellite precipitation products using rain gauge observations over the Tibetan Plateau. Hydrol. Earth Syst. Sci. 17, 837849, https://doi.org/10.5194/hess-17-837-2013.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • He, Z., and Coauthors, 2017: Intercomparisons of rainfall estimates from TRMM and GPM multisatellite products over the upper Mekong river basin. J. Hydrometeor., 18, 413430, https://doi.org/10.1175/JHM-D-16-0198.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • He, Z., and Coauthors, 2019: Constraining hydrological model parameters using water isotopic compositions in a glacierized basin, Central Asia. J. Hydrol., 571, 332348, https://doi.org/10.1016/j.jhydrol.2019.01.048.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hong, S.-Y., and M. Kanamitsu, 2014: Dynamical downscaling: Fundamental issues from an NWP point of view and recommendations. Asia-Pac. J. Atmos. Sci., 50, 83104, https://doi.org/10.1007/s13143-014-0029-2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Huang, D., and S. Gao, 2018: Impact of different reanalysis data on WRF dynamical downscaling over China. Atmos. Res., 200, 2535, https://doi.org/10.1016/j.atmosres.2017.09.017.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Immerzeel, W. W., and Coauthors, 2015: Reconciling high-altitude precipitation in the upper Indus basin with glacier mass balances and runoff. Hydrol. Earth Syst. Sci., 19, 46734687, https://doi.org/10.5194/hess-19-4673-2015.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kan, B., and Coauthors, 2018: Generation of high mountain precipitation and temperature data for a quantitative assessment of flow regime in the upper Yarkant Basin in the Karakoram. J. Geophys. Res. Atmos., 123, 84628486, https://doi.org/10.1029/2017JD028055.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Khromova, T. E., and Coauthors, 2006: Changes in glacier extent in the Eastern Pamir, Central Asia, determined from historical data and ASTER imagery. Remote Sens. Environ., 102, 2432, https://doi.org/10.1016/j.rse.2006.01.019.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Laprise, R., 2008: Regional climate modelling. J. Comput. Phys., 227, 36413666, https://doi.org/10.1016/j.jcp.2006.10.024.

  • Li, L., and Coauthors, 2017: Evaluating the present annual water budget of a Himalayan headwater river basin using a high-resolution atmosphere-hydrology model. J. Geophys. Res. Atmos., 122, 47864807, https://doi.org/10.1002/2016JD026279.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Li, Y., and Coauthors, 2019: High-resolution regional climate modeling and projection over Western Canada using a weather research forecasting model with a pseudo-global warming approach. Hydrol. Earth Syst. Sci., 23, 46354659, https://doi.org/10.5194/hess-23-4635-2019.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Li, Z., D. Yang, and Y. Hong, 2013: Multi-scale evaluation of high-resolution multi-sensor blended global precipitation products over the Yangtze River. J. Hydrol., 500, 157169, https://doi.org/10.1016/j.jhydrol.2013.07.023.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Liang, X., and Coauthors, 1994: A simple hydrologically based model of land-surface water and energy fluxes. J. Geophys. Res., 99, 14 41514 428, https://doi.org/10.1029/94JD00483.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Liang, X., D. P. Lettenmaier, and E. F. Wood, 1996: One-dimensional statistical dynamic representation of subgrid spatial variability of precipitation in the two-layer variable infiltration capacity model. J. Geophys. Res., 101, 21 40321 422, https://doi.org/10.1029/96JD01448.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lin, C., and Coauthors, 2018: Impact of model resolution on simulating the water vapor transport through the central Himalayas: Implication for models’ wet bias over the Tibetan Plateau. Climate Dyn., 51, 31953207, https://doi.org/10.1007/s00382-018-4074-x.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lutz, A. F., and Coauthors, 2014: Consistent increase in High Asia’s runoff due to increasing glacier melt and precipitation. Nat. Climate Change, 4, 587592, https://doi.org/10.1038/nclimate2237.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lutz, A. F., and Coauthors, 2016: Climate change impacts on the upper Indus hydrology: Sources, shifts and extremes. PLOS ONE, 11, e0165630, https://doi.org/10.1371/journal.pone.0165630.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ma, Y., and Coauthors, 2015: Precipitation bias variability versus various gauges under different climatic conditions over the Third Pole Environment (TPE) region. Int. J. Climatol., 35, 12011211, https://doi.org/10.1002/joc.4045.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ma, Y., and Coauthors, 2016: Similarity and error intercomparison of the GPM and its predecessor-TRMM multisatellite precipitation analysis using the best available hourly gauge network over the Tibetan Plateau. Remote Sens., 8, 569, https://doi.org/10.3390/rs8070569.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ma, Y., and Coauthors, 2018: Atmospheric moisture transport versus precipitation across the Tibetan Plateau: A mini-review and current challenges. Atmos. Res., 209, 5058, https://doi.org/10.1016/j.atmosres.2018.03.015.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Maussion, F., and Coauthors, 2011: WRF simulation of a precipitation event over the Tibetan Plateau, China - An assessment using remote sensing and ground observations. Hydrol. Earth Syst. Sci., 15, 17951817, https://doi.org/10.5194/hess-15-1795-2011.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Maussion, F., and Coauthors, 2014: Precipitation seasonality and variability over the Tibetan Plateau as resolved by the High Asia Reanalysis. J. Climate, 27, 19101927, https://doi.org/10.1175/JCLI-D-13-00282.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Meng, F., and Coauthors, 2019: Changes in terrestrial water storage during 2003–2014 and possible causes in Tibetan Plateau. J. Geophys. Res. Atmos., 124, 29092931, https://doi.org/10.1029/2018JD029552.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mölg, T., and Coauthors, 2012: The footprint of Asian monsoon dynamics in the mass and energy balance of a Tibetan glacier. Cryosphere, 6, 14451461, https://doi.org/10.5194/tc-6-1445-2012.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mölg, T., F. Maussion, and D. Scherer, 2013: Mid-latitude westerlies as a driver of glacier variability in monsoonal High Asia. Nat. Climate Change, 4, 6873, https://doi.org/10.1038/nclimate2055.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Norris, J., and Coauthors, 2015: WRF simulations of two extreme snowfall events associated with contrasting extratropical cyclones over the western and central Himalaya. J. Geophys. Res. Atmos., 120, 31143138, https://doi.org/10.1002/2014JD022592.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ou, T., and Coauthors, 2020: Simulation of summer precipitation diurnal cycles over the Tibetan Plateau at the gray-zone grid spacing for cumulus parameterization. Climate Dyn., 54, 35253539, https://doi.org/10.1007/s00382-020-05181-x.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Qi, W., J. Liu, and D. Chen, 2018: Evaluations and improvements of GLDAS2.0 and GLDAS2.1 forcing data’s applicability for basin scale hydrological simulations in the Tibetan Plateau. J. Geophys. Res. Atmos., 123, 13 12813 148, https://doi.org/10.1029/2018JD029116.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Qiu, J., 2008: China: The third pole. Nature, 454, 393396, https://doi.org/10.1038/454393a.

  • Shafeeque, M., and Coauthors, 2019: Revealing vertical distribution of precipitation in the glacierized upper Indus Basin based on multiple datasets. J. Hydrometeor., 20, 22912314, https://doi.org/10.1175/JHM-D-19-0081.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Skamarock, W. C., J. B. Klemp, J. Dudhia, D. O. Gill, D. M. Barker, W. Wang, and J. G. Powers, 2005: A description of the Advanced Research WRF version 2. NCAR Tech. Note NCAR/TN-468+STR, 88 pp., https://doi.org/10.5065/D6DZ069T.

    • Crossref
    • Export Citation
  • Song, H.-J., and B.-J. Sohn, 2018: An evaluation of WRF microphysics schemes for simulating the warm-type heavy rain over the Korean Peninsula. Asia-Pac. J. Atmos. Sci., 54, 225236, https://doi.org/10.1007/s13143-018-0006-2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Su, F., Y. Hong, and D. P. Lettenmaier, 2008: Evaluation of TRMM Multisatellite Precipitation Analysis (TMPA) and its utility in hydrologic prediction in the La Plata Basin. J. Hydrometeor., 9, 622640, https://doi.org/10.1175/2007JHM944.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Su, F., and Coauthors, 2013: Evaluation of the global climate models in the CMIP5 over the Tibetan Plateau. J. Climate, 26, 31873208, https://doi.org/10.1175/JCLI-D-12-00321.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Su, F., and Coauthors, 2016: Hydrological response to future climate changes for the major upstream river basins in the Tibetan Plateau. Global Planet. Change, 136, 8295, https://doi.org/10.1016/j.gloplacha.2015.10.012.

    • Crossref