Partitioning Solid and Liquid Precipitation over the Tibetan Plateau Based on Satellite Radar Observations

Ping Song aShandong Meteorological Service Center, Jinan, Shandong, China
bDepartment of Earth, Ocean and Atmospheric Science, Florida State University, Tallahassee, Florida

Search for other papers by Ping Song in
Current site
Google Scholar
PubMed
Close
and
Guosheng Liu bDepartment of Earth, Ocean and Atmospheric Science, Florida State University, Tallahassee, Florida

Search for other papers by Guosheng Liu in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

Whether precipitation falls in the form of rain or snow is of great importance to glacier accumulation and ablation. Assessments of phase-aware precipitation have been lacking over the vast area of the Tibetan Plateau (TP) due to the scarcity of surface measurements and the low quality of satellite estimates in this region. In this study, we attempt a satellite radar-based method for this precipitation partition, in which the CloudSat radar is used for snowfall while the Global Precipitation Measurement Mission radar is used for rainfall estimation. Assuming that 11-yr snowfall and 5-yr rainfall estimates represent the mean states of precipitation at each phase, the phase partition characteristics, including its annual mean, spatial pattern, seasonal dependence, and variation with elevations, are then discussed. Averaged over the highland area (over 1 km above mean sea level) in TP, the annual total precipitation is estimated to be around 400 mm, of which about 40% falls as snow. The snowfall mass fraction is about 45% in the northern and 30% in the southern part of TP, and about 80% in the cold and 30% in the warm half of the year. Surface elevation is found to be a high-impact factor on total precipitation and its phase partition, generally with total precipitation decreasing but snowfall fraction increasing with the increase of elevation. While there are some shortcomings, the current approach in combining snowfall and rainfall estimates from two satellite radars presents a useful pathway to assessing phase-aware precipitation over the TP region.

Significance Statement

Whether precipitation falls in the form of rain or snow is of great importance to glacier accumulation and ablation over the Tibetan Plateau region, and yet a plateau-wide assessment of the phase-resolved precipitation has been lacking due to the scarcity of surface measurements in this region. In this study, we attempt a satellite radar-based method to separately estimate liquid and solid phase precipitation. The phase-resolved precipitation characteristics, including its annual mean, spatial pattern, seasonal dependence, and variation with elevation, are then analyzed and discussed. To the authors’ knowledge, this is the first satellite observation-based study to evaluate a plateau-wide, phase-resolved annual mean precipitation, and its dependence on season, location, and elevation.

© 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: Guosheng Liu, gliu@fsu.edu

Abstract

Whether precipitation falls in the form of rain or snow is of great importance to glacier accumulation and ablation. Assessments of phase-aware precipitation have been lacking over the vast area of the Tibetan Plateau (TP) due to the scarcity of surface measurements and the low quality of satellite estimates in this region. In this study, we attempt a satellite radar-based method for this precipitation partition, in which the CloudSat radar is used for snowfall while the Global Precipitation Measurement Mission radar is used for rainfall estimation. Assuming that 11-yr snowfall and 5-yr rainfall estimates represent the mean states of precipitation at each phase, the phase partition characteristics, including its annual mean, spatial pattern, seasonal dependence, and variation with elevations, are then discussed. Averaged over the highland area (over 1 km above mean sea level) in TP, the annual total precipitation is estimated to be around 400 mm, of which about 40% falls as snow. The snowfall mass fraction is about 45% in the northern and 30% in the southern part of TP, and about 80% in the cold and 30% in the warm half of the year. Surface elevation is found to be a high-impact factor on total precipitation and its phase partition, generally with total precipitation decreasing but snowfall fraction increasing with the increase of elevation. While there are some shortcomings, the current approach in combining snowfall and rainfall estimates from two satellite radars presents a useful pathway to assessing phase-aware precipitation over the TP region.

Significance Statement

Whether precipitation falls in the form of rain or snow is of great importance to glacier accumulation and ablation over the Tibetan Plateau region, and yet a plateau-wide assessment of the phase-resolved precipitation has been lacking due to the scarcity of surface measurements in this region. In this study, we attempt a satellite radar-based method to separately estimate liquid and solid phase precipitation. The phase-resolved precipitation characteristics, including its annual mean, spatial pattern, seasonal dependence, and variation with elevation, are then analyzed and discussed. To the authors’ knowledge, this is the first satellite observation-based study to evaluate a plateau-wide, phase-resolved annual mean precipitation, and its dependence on season, location, and elevation.

© 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: Guosheng Liu, gliu@fsu.edu

1. Introduction

The Tibetan Plateau (TP) covers a vast area of over 2.5 million km2 with a mean elevation of 4500 m above mean sea level (MSL) (Zhang et al. 2002). A regional map of the Tibetan Plateau and its surrounding region (hereafter, we refer to this entire region as the TP region for convenience) is shown in Fig. 1. As often called the “Third Pole” of the world, it is estimated that the TP region contains about 36 800 glaciers (Yao et al. 2007). Glaciers are natural buffers of hydrological seasonality, in particular releasing meltwater during summer and early autumn (Bookhagen and Burbank 2010; Immerzeel et al. 2010). Studies indicated that glacial retreat over TP has intensified in the past decades, which has caused an increase in river runoffs and rising lake levels (Bolch et al. 2012; Kääb et al. 2012; Yao et al. 2012).

Fig. 1.
Fig. 1.

Map of the study area with elevations (m MSL) shown by colors. Geographical locations mentioned in the text are also indicated.

Citation: Journal of Hydrometeorology 22, 11; 10.1175/JHM-D-21-0018.1

The impact of precipitation characteristics such as its amount, type, seasonality, and variability on glaciers has been investigated in several previous studies (e.g., Bookhagen and Burbank 2010; Barros et al. 2000; Maussion et al. 2014; Fujita 2008). The main process for glacier accumulation is snowfall. Rainfall, unless it freezes, does not constitute significant accumulation (Cogley et al. 2011). Due to the high elevation, a substantial portion of the precipitation is snowfall in the TP region (Viste and Sorteberg 2015; Lang and Barros 2004). In the central Himalayas, it is estimated that snowfall contributes up to 25%–35% of annual precipitation at stations of high elevation of 3000 m MSL and above (Lang and Barros 2004). However, the snowfall contribution over the entire TP region and its dependence on season and altitude still need to be investigated.

Precipitation in the TP region is largely influenced by two season-dependent weather systems: monsoon circulation in the summer months (Webster et al. 1998; Gadgil 2003; Wang 2006) and westerly disturbances during winter months (Barros et al. 2006; Lang and Barros 2004; Schiemann et al. 2009). The phase of precipitation varies with season and elevation. In the western region, snow accumulates during winter, while the summer is the main snow-melting season. However, in the central and eastern Himalayas, summer is the main season for both snow melting and accumulation (Rees and Collins 2006; Viste and Sorteberg 2015). Whether the precipitation phase is solid (snowfall) or liquid (rainfall), is of profound importance because of their distinct impact on snowpack and glaciers. Therefore, partitioning rainfall and snowfall in the total precipitation is very important in understanding the hydrological balance and glacier accumulation processes in this region.

Observational studies on snowfall are very limited over TP, and mostly are based on in situ data from weather stations (Lang and Barros 2004; Deng et al. 2017; Lu and Liu 2010; Immerzeel et al. 2010). From these limited studies, it is found that since the early 1980s there has been a rapid decline in snowfall over most of TP, especially in the east portion and at moderate elevations around 2000 m MSL. Above 3000 m MSL, however, most areas have shown an increase in snowfall (Deng et al. 2017). However, due to the paucity of permanent weather stations (Qin et al. 2009), these studies were not able to fully capture an accurate picture of precipitation, in particular separately as rainfall or snowfall. In addition, a serious issue in using ground-based weather station data for climatological analysis is that these stations are not uniformly distributed spatially, but rather preferably situated at locations with easy human access (Barros and Lettenmaier 1994; Qin et al. 2009), which unavoidably leaves their representativeness for the whole region in doubt.

Satellite observations, on the other hand, have the advantages of sampling data nearly uniformly over the satellite covered latitudes regardless of surface. Several global satellite-based precipitation products have been developed based on data from space-borne radars, radiometers, or both combined (e.g., Iguchi and Haddad 2020; Wood and L’Ecuyer 2018; Kummerow et al. 2015; Grecu et al. 2016; Huffman et al. 2007; Joyce and Xie 2011; Kubota et al. 2020). Among satellite-borne instruments, passive sensors (radiometers) measure the combined contribution from the atmosphere and surface to the upwelling radiation; precipitation retrievals relying on these measurements commonly suffer a great error over complex terrains such as the TP region because of the uncertainties related to surface emissivity and possible orographic precipitation enhancement (Yin et al. 2008; Ferraro et al. 2013; Behrangi et al. 2014; Ebtehaj et al. 2016; Shige et al. 2013). On the other hand, active radars collect data with individual range gates, so that radar reflectivity at one gate is mostly not affected by radar signals at other gates with the exceptions of contaminations by ground clutter near the surface and attenuation by cloud and precipitation between the target gate and satellite. Therefore, radars are generally advantageous over radiometers in measuring precipitation over complex terrains. To conduct the analysis using the best possible satellite data to date, we choose to use radar-only precipitation retrievals in this study. Specifically, we use the Global Precipitation Measurement Mission (GPM) Dual-Frequency Precipitation Radar (DPR) observations for rainfall, while using CloudSat Cloud Profiling Radar (CPR) observations for snowfall estimation.

The GPM Core Observatory satellite was launched in 2014, and the DPR observations cover most of the globe between 65°S and 65°N latitudes (Hou et al. 2014; Skofronick-Jackson et al. 2018). The DPR operates at Ku and Ka band with a minimum detectability of about 12 dBZ (Toyoshima et al. 2015). Although this detectability is a significant improvement over its predecessor Tropical Rainfall Measuring Mission (TRMM) Precipitation Radar (PR), it still misses a sizable portion of light precipitation, particularly snowfall (Casella et al. 2017; Adhikari et al. 2018; Liu 2020). By comparing a 3-yr GPM DPR data with 4-yr CloudSat CPR observations, Adhikari et al. (2018) concluded that the DPR underestimates about 52% of snowfall by volume. Similarly, Casella et al. (2017) showed that the occurrence of snowfall events correctly detected by DPR is quite small (around 5%–7%) in reference to CPR observations. Studies on radar snowfall retrievals based on CloudSat CPR observations have been conducted by several investigators (e.g., Liu 2008; Kulie and Milani 2018; Kulie et al. 2020). The CloudSat satellite was launched in 2006, and the CPR observations cover the latitudes between 82°S and 82°N at W band. Analysis of in-orbit data has shown that the CPR’s minimum detectable radar reflectivity factor is about −30 dBZ, sensible enough to detect snowfall. However, the CPR has difficulties of measuring moderate to heavy precipitation, heavy rainfall in particular, due to hydrometeor’s attenuation and Mie scattering effects, which limits the upper bound of the CPR reflectivity to around 20 dBZ. This point was discussed in Liu (2020) and will be further illustrated later in section 2.

As described above, although radar observations are better suited for precipitation estimation over complex terrains such as TP, the currently available space-borne radars have their own shortcomings, i.e., the GPM DPR lacks the sensitivity to snowfall while the CloudSat CPR cannot correctly measure heavy rainfall. Because the purpose of this study is to understand the partitioning of liquid and solid precipitation over TP, it is not feasible for us to use either DPR or CPR alone to accomplish this objective. Instead, we elect to estimate rainfall and snowfall separately from DPR and CPR. Due to data availability issues to be described in the following section, we could not use temporally well overlapped DPR and CPR data, but have to use data covering different years—nearly 11 years of CPR data from mid-2006 through mid-2017 and 5 years of DPR data from 2015 through 2019. An assumption here is that the 11-yr CPR data derived mean snowfall and the 5-yr DPR data derived mean rainfall roughly represent the “mean states” of the solid and liquid precipitation over this region, so that we can discuss the precipitation phase partition even though they come from different time periods.

2. Data and methods

The principal data used in this study for deriving solid and liquid precipitation over the TP region are the observations by the CPR on board CloudSat and the DPR on board the GPM satellite. GPM DPR and CloudSat CPR data are used, respectively, for rainfall and snowfall retrievals. DPR data are available since March 2014, while CPR data are available since June 2006, but the CPR radar had been operating in daylight-only mode due to battery issues since early 2011. At the time of this study, the CPR snowfall data are only available from mid-2006 through mid-2017; all of them will be used to calculate annual mean snowfall. It should be noted that the daylight-only mode of CPR operation could cause biases in snowfall estimates. A study by Milani and Wood (2021) showed that globally the mean snowfall rate derived from CPR in the daylight-only period is about 8% lower than that in the full operation period, implying an underestimation. To avoid the mean being skewed toward a particular season due to uneven data sampling during a year, the data are first averaged for every 5-day window within a year before calculating the annual mean. Consequently, for the CPR we use the 11 years of data from 2006 through 2017, and for DPR we use the 5 years of data from 2015 through 2019.

To demonstrate the ability of the CPR and the DPR in detecting snowfall and rainfall, in Fig. 2 we show a case coincidently observed by both radars over the TP region. The collocated data are archived at NASA’s GPM data center as 2B-CSATGPM (version 03B) product (Turk 2016), which combines CPR and DPR measurements taken within 15 min or less. Two major precipitation events are shown in the case, one over the southern slope of the plateau (CloudSat ray indices 0–150) and the other well over the plateau (CloudSat ray indices 300–370). In both events, the surface precipitation was identified as rainfall although the event well over the plateau is very close to snowfall condition. The freezing level is around 5 km altitude; radar echo above this level can be considered to be generated by snow particles. Over the plateau, CloudSat CPR observed substantial radar returns from precipitation between ray indices 300 and 370, while most of them are missed by the DPR. On the southern slope where the precipitation is heavier, both DPR and CPR observed the precipitation. However, the DPR misses most of the echoes above 5 km observed by CPR, and the CPR reflectivity profiles show that values at lower levels are clearly lower than those above them, indicating possible reduction of radar return due to attenuation by rain. Since most radar returns due to snowfall are weak and therefore prone to be missed by the DPR (Liu 2020), CPR observations are considered to be more suited for snowfall retrievals. Meanwhile, the DPR experiences less attenuation than the CPR and then is more suited for rainfall retrievals.

Fig. 2.
Fig. 2.

(a) Ground track of CloudSat–GPM coincident data as both satellites overpass the Himalayas around 0725 UTC 7 Jun 2014, (b) distance–height cross section of CPR radar reflectivity, and (c) distance–height cross section of DPR Ku radar reflectivity. The CloudSat ray index starts from south to north covering the distance shown by the thick line in (a). The digital elevation map (DEM) used in the CloudSat and GPM data are shown as a heavy black line in (b) and (c).

Citation: Journal of Hydrometeorology 22, 11; 10.1175/JHM-D-21-0018.1

a. Determination of rain or snow at surface

For a CPR or DPR radar observation, the first task in our analysis is to determine whether the scene corresponds to a rainfall or snowfall event at ground level. Although precipitation phase determination is performed in both CPR and DPR products, they are not done in a consistent way (Wood and L’Ecuyer 2018; Iguchi et al. 2018), and in the case of DPR product the phase information is not given at surface level, but rather at the levels not contaminated by ground clutters, which are usually 1.5 km above actual surface at nadir and even higher at off-nadir positions. To perform a uniform phase determination across both CPR and DPR observations, we choose to apply a rain–snow separation scheme developed by Sims and Liu (2015) with the input of environmental variables from the fifth generation European Centre for Medium-Range Weather Forecasts (ECMWF) reanalysis (ERA5; Hersbach et al. 2020).

The Sims and Liu (2015) scheme is a statistical method developed using multidecades of ground-based and shipboard manual observations of current weather over the globe. In developing the scheme, current weather reports from over 9700 land stations during 1997 and 2007 and shipboard current weather reports during 1950–2007 are used. They first studied what environmental parameters impact precipitation phase, and found that the key impact factors are 2-m wet-bulb temperature, low-level lapse rate, surface pressure (or elevation), and surface skin temperature. The scheme provides the conditional probability of solid precipitation using inputs of environmental variables such as 2-m temperature, relative humidity, and low-level temperature lapse rate. Since the output of the scheme is a conditional probability, a 50% probability of solid precipitation criterion is used to separate rain and snow in this study. It is noted that the Sims and Liu (2015) scheme has not been validated over the TP region, although station data from this region were included when developing this scheme. In this study, the environmental variables from ERA5 reanalysis are used as the input of this scheme. The ERA5 reanalysis data have a 1-h temporal, and a 0.25° × 0.25° (latitude/longitude), or approximately 30 km, horizontal resolution.

In the Sims and Liu (2015) scheme, 2-m temperature is the most important parameter for separating solid and liquid precipitation. For the TP region with complex terrains, the 30-km spatial resolution of ERA5 is not adequate to provide an accurate estimate of the 2-m temperature at the radar pixel scale (a few kilometers). To yield the 2-m temperature at the radar footprint, the Shuttle Radar Topography Mission dataset with 15-arc-s (~417 m) resolution, SRTM15, is used together with ERA5 for topographic correction to obtain a more suitable 2-m temperature value. Specifically, for each radar pixel, the elevation difference between the ERA5 grid averaged value and that at the position of radar pixel is first calculated, and then, using the lapse rate in the near-surface layer of the ERA5 data the 2-m temperature at the location of the radar pixel is interpolated. A similar method was used by Viste and Sorteberg (2015) over the TP region, Tao and Barros (2018) over the southern Appalachian Mountains, and Rouf et al. (2020) over the midwestern United States. Validation by Tao and Barros (2018) indicated that this method improves the root-mean-square error of 2-m temperature by 1 K (up to 5 K in winter). By comparing to in situ data observed at surface stations, Rouf et al. (2020) pointed out the benefit of using a dynamic lapse rate, i.e., varying with time and location rather than a fixed value, to improve the downscaled 2-m temperature. However, validation of this approach over the more rugged TP terrains has not been done so far. One example of so derived 2-m temperature along a CloudSat track is shown in Fig. 3, together plotted with 2-m temperature contained in the CloudSat data product, which is from a different version of ECMWF reanalysis (ECMWF-AUX; Cronk and Partain 2017). Clearly, our topography corrected 2-m temperature resolves more detailed spatial variations, which closely corresponds to the variation of surface elevation at the position of radar observations.

Fig. 3.
Fig. 3.

The 2-m temperatures corrected by using SRTM15 data (orange) as compared with those contained in CloudSat product (blue). The surface elevation at CPR pixel location is interpolated from SRTM15 data. The example is for a CloudSat pass over the TP on 21 Jan 2010.

Citation: Journal of Hydrometeorology 22, 11; 10.1175/JHM-D-21-0018.1

b. CPR snowfall estimation

The CPR on the CloudSat satellite is a nadir-looking W-band radar that observes a curtain of the atmosphere with each satellite pass. Although the spatial coverage is limited due to its nonscanning mechanism, the CPR radar possesses an excellent sensitivity to light precipitation.It has a minimum detectable reflectivity factor of about −30 dBZ at the range of Earth surface, which makes it capable to detect even very light snow. The standard snowfall product, 2C-SNOW-PROFILE (version R05), was created from CPR radar reflectivity profiles based on an algorithm described by Wood and L’Ecuyer (2018). The radar reflectivity is sampled to 150 bins in the vertical with a bin size of about 240 m. The footprint size of radar reflectivity profiles is 1.4 km cross track × 2.5 km along track. For those bins deemed as solid precipitation, radar reflectivity is converted to snowfall rate with an optimization algorithm. In this study, we decide to perform precipitation phase identification ourselves instead of relying on what is given in the CloudSat product in order to use the same method to determine precipitation phase for both CloudSat CPR and GPM DPR observations. Due to contamination by surface clutters, it is impossible to obtain useful snowfall retrievals for several bins near the surface. Our decision is to use the retrieval at the lowest uncontaminated bin to represent surface snowfall rate after it is determined by the method described in section 2a that the surface precipitation is in solid phase. The quality flag contained in the 2C-SNOW-PROFILE product is used to remove data possibly contaminated by ground clutter.

To assess the difference made by applying our phase identification method, in Fig. 4 we show the scatterplot of annual mean surface snowfall rates (mm day−1) determined in this study versus those in the original 2C-SNOW-PROFILE product; each point in this figure is an average of 11-yr observations in a 1° × 2° grid. While there are some scatters, the two estimates are quite consistent with the correlation coefficient, bias, and root-mean-square error (RMSE) being 0.96, −0.01 mm day−1 (2C-SNOW-PROFILE slightly higher) and 0.1 mm day−1, respectively. Therefore, we conclude that the phase identification used in CloudSat 2C-SNOW-PROFILE product is similar to the one described in this study, while applying our identification allows us to use the same method for both CPR and DPR observations.

Fig. 4.
Fig. 4.

Scatterplot of annual mean surface snowfall rate (mm day−1) calculated from 2C-SNOW-PROFILE surface snowfall rate and by our method. Each point is an average of 11-yr data in a 1° (latitude) × 2° (longitude) grid box. The values of correlation coefficient (r), bias, and RMSE are also given.

Citation: Journal of Hydrometeorology 22, 11; 10.1175/JHM-D-21-0018.1

c. DPR rainfall estimation

The DPR on the GPM Core Observatory satellite is a cross-track scanning radar that operates at Ku and Ka bands. The Ku-band radar swath covers 245 km with a horizontal resolution of approximately 5 km and a vertical resolution of 250 m in the normal scan mode. The Ka-band radar has two scan modes with a slightly different scan patterns before and after 21 May 2018 (Iguchi et al. 2018). The minimum detectable reflectivity was designed to be 18 and 12 dBZ, respectively, for the Ku- and Ka-band radar. In orbit, however, observational data analysis indicates that the postlaunch minimum detectability of about 12 dBZ for the Ku band has far exceeded prelaunch estimates and is comparable to that of the Ka band (Hamada and Takayabu 2016). In this study, the normal scan mode data with 49 beams from the DPR level 2A precipitation retrievals at version 6 (Iguchi et al. 2018) are used. The retrieval algorithm consists of steps of precipitation type characterization, correction for attenuation by atmosphere and hydrometeors, estimation of particle size distribution, and retrieval of precipitation profiles. While precipitation phase is given in the product, this information is only valid for the levels uncontaminated by ground clutters. The lowest level without ground clutter contamination commonly locates 1.5 km or higher above surface. Therefore, determination of precipitation phase at surface is needed for our data analysis. Similar to the case of CloudSat snowfall, we perform the precipitation phase identification using procedures described in section 2a, i.e., applying the Sims and Liu (2015) method with the input of environmental variables from ERA5 and correcting 2-m temperature at pixel’s surface elevation based on SRTM15 data. Once the surface precipitation is determined to be rain, the DPR precipitation retrieval at the lowest uncontaminated bin is used as surface rainfall.

It should be mentioned that the approach of using only CPR observations to quantify snowfall and only DPR observations to quantify rainfall causes errors in the precipitation estimation. Attenuation of CPR reflectivity by heavy snowfall may cause underestimation. To quantify this underestimation, we studied coincident CPR and DPR snowfall retrievals using the coincident dataset of Turk (2016). However, the results did not show a systematic underestimation by CPR against DPR retrievals as snowfall becomes heavier. On the contrary, DPR snowfall retrievals seem to be systematically lower than those of CPR over all the snowfall rate range. Without a better reference to correct the possible CPR attenuation problem, the CPR snowfall retrieval was not corrected for possible attenuation. Considering the high elevation of the plateau (over 4000 m above sea level on average), we think that snowfall precipitation in this region should be less heavy than those over lowland. Therefore, attenuation problem should be less severe over the plateau. In addition, error in rainfall estimates occurs because of the inability for DPR to detect light rain due to its low detectability (~12 dBZ), which corresponds to about 0.3 mm h−1 in rainfall rate. Rainfall below this intensity will be missed in our rainfall estimates. This error of underestimation will stay in the results of our analysis.

d. Other datasets

For comparison purposes, we also used two other precipitation datasets in the analysis. One is the Asian Precipitation–Highly Resolved Observational Data Integration Towards Evaluation of Extreme Events (APHRODITE-2) developed based on surface gauges (Yatagai et al. 2012), and the other is precipitation from the ERA5 reanalysis data.

The APHRODITE-2 project aims to create daily gridded precipitation data over the Asian land areas with a high spatial and temporal resolution based on rain gauge observations (Yatagai et al. 2012). Extensive quality checks and robust interpolation are conducted to improve the quality and applicability of the datasets. The latest APHRODITE-2 product is available for a period of 18 years from 1998 to 2015. The data used in this study are 0.25° × 0.25° gridded data for 4 years from 2007 to 2010. No precipitation phase information is available in the APHRODITE-2 product.

Surface precipitation (rain and snow) data from ERA5 (Hersbach et al. 2020) from 2006 to 2017 are also used for comparison. The reanalysis data are at a 1-hourly time step and with 0.25° × 0.25° spatial resolution.

e. Data sample statistics

The focused area of this study is 26°–39°N, 71°–104°E, which covers the entire Tibetan Plateau region. The total number of CloudSat CPR radar reflectivity profiles in the region during the 11-yr study period is 84 590 463, of which 3 751 692, or 4%, contain surface snowfall. Meanwhile, there are 93 420 500 GPM DPR radar reflectivity profiles in the region during the 5-yr study period, of which 2 053 589, or 2%, are deemed as having surface rainfall. The numbers of CPR and DPR observations in each 1° latitude × 2° longitude box are shown in Fig. 5. It is seen that the density of samples is greater than 2 × 105 per 1° × 2° area for CPR and 3.8 × 105 per 1° × 2° area for DPR observations.

Fig. 5.
Fig. 5.

Number of (a) CPR and (b) DPR observations in each 1° (latitude) × 2° (longitude) grid box during the study period.

Citation: Journal of Hydrometeorology 22, 11; 10.1175/JHM-D-21-0018.1

3. Results

In this section, we show the partition of solid and liquid precipitation over the TP region derived from satellite radar observations. As mentioned earlier, since the CloudSat CPR derived snowfall data and GPM DPR derived rainfall data do not well overlap in time, an assumption made in this study is that the 11-yr snowfall and the 5-yr rainfall averages represent the mean states of snowfall and rainfall, respectively, so that they can be used to calculate total precipitation and snowfall (or rainfall) fractions.

a. Mean distribution and phase partition of precipitation

The distributions of mean precipitation rate averaged over a 1° × 2° grid from the satellite radars (sum of the snowfall rate from CPR and the rainfall rate from DPR), APHRODITE-2, and ERA5, as well as comparisons of the satellite radar retrieval to the other two estimates, are shown in Fig. 6. All three datasets show more precipitation in the southern than in the northern region, with the 33°N latitude as a rough boundary. Hereafter, we will refer the northern and the southern region within our study area by the 33°N latitude as a divider for convenience. Several previous studies revealed the same contrast between regions north and south of 33°N latitude in spatial precipitation patterns (Tian et al. 2001; Lu and Liu 2010). This contrast is often explained by the different precipitation mechanisms between the two regions. That is, in the dry climate of the northern TP, strong local evaporation and convective precipitation dominate the hydrological processes, whereas precipitation in the southern TP is largely controlled by the Indian monsoon and ocean originated moisture (e.g., Webster et al. 1998; Tian et al. 2001; Barros et al. 2006). In addition, the steep topography can cause a strong enhancement of monsoonal rainfall on the southern slope of the Himalayas (Fu et al. 2018). The pattern of differences is also believed to be related to the North Atlantic Oscillation (NAO). The upstream westerly winds derived from the NAO generate dynamic bifurcation flows reaching 33°N, producing a precipitation contrast between the northern and southern TP (Liu and Yin 2001).

Fig. 6.
Fig. 6.

Distributions of the mean precipitation rates calculated from (a) satellite radars (CPR and DPR), (b) APHRODITE-2, and (c) ERA5. Comparisons between satellite radar and APHRODITE-2 and ERA5 estimates are also shown by a log-difference of (d) log(Ps/Pg) and (e) log(Ps/Pe). Here, Ps, Pg, and Pe are, respectively, the satellite, the APHRODITE-2, and the ERA5 total precipitation estimates.

Citation: Journal of Hydrometeorology 22, 11; 10.1175/JHM-D-21-0018.1

The ERA5 shows much larger total precipitation rates than the satellite radar and the APHRODITE-2 gauge estimates (Note that the different color scales are used for the plots). For a better visual comparison, we plotted the distribution of log-difference, log(Ps/Pg), in Fig. 6d and log(Ps/Pe) in Fig. 6e, where Ps, Pg, and Pe are, respectively, the satellite radar, the APHRODITE-2 gauge, and the ERA5 reanalysis total precipitation estimates. In Fig. 6d, it is shown that the satellite retrieval overestimates precipitation against the gauge estimates in the northern region along the Karakoram Range and Kunlun Mountains where the averaged elevation is higher than 5500 m MSL and precipitation gauge stations are sparse. Note that the total precipitation in this region is low. The large log-difference here is caused by the small values of Pg and Pe. The precipitation values in the southern region are comparable between the two estimates with a mix of positive and negative log-difference values scattered over the area. While the patterns of spatial distributions are similar, the magnitude of the ERA5 precipitation is much larger than the satellite radar counterpart in most areas as shown in Fig. 6e. As will be indicated later, this “excess” precipitation in the ERA5 reanalysis is mainly due to the greater rainfall amount.

The satellite-derived mean snowfall rate, rainfall rate, snowfall frequency, and snowfall mass fraction are shown in Fig. 7. Snowfall frequency is defined as the snowfall pixel number divided by the total observation pixel number within a 1° × 2° grid box. Snowfall mass fraction is defined by the mean snowfall rate divided by total precipitation rate. It is seen that the precipitation over the Karakoram Range and Kunlun Mountains is mostly from snowfall, where snowfall mass fraction reaches more than 80% (Fig. 7d), and snowfall frequency is also relatively large (Fig. 7c).In addition, large snowfall rates and snowfall mass fractions (over 50%) are also observed in the Himalayan and Nyainqentanglha Mountains, consistent with a previous study by Maussion et al. (2014) based on their High Asia Reanalysis data. Moreover, most of the large mean liquid precipitation rates are observed at low elevation areas, such as in the southwest edge and the eastern part of TP.

Fig. 7.
Fig. 7.

(a) Mean snowfall rate calculated from CPR, (b) mean rainfall rate calculated from DPR, (c) snowfall frequency, which is snowfall sample number divided by total observation sample number, and (d) snowfall mass fraction, which is the mean snowfall rate divided by sum of mean snowfall rate and rainfall rate.

Citation: Journal of Hydrometeorology 22, 11; 10.1175/JHM-D-21-0018.1

A similar plot is also shown for the ERA5 data (Fig. 8). The mean snowfall rate from ERA5 shows a similar distribution, but with a somewhat larger magnitude, to that from CPR estimates (Fig. 8a versus Fig. 7a). Since ERA5 rainfall values (Fig. 8b) are even larger than those from DPR estimates (Fig. 7b), the snowfall mass fraction derived from satellite observations (Fig. 7d) turns to be somewhat larger than that derived from ERA5 (Fig. 8c).

Fig. 8.
Fig. 8.

(a) Mean snowfall rate, (b) mean rainfall rate, and (c) snowfall mass fraction based on ERA5 data, and (d) log-difference of mean snowfall rates calculated from CPR and ERA5, log(Ss/Se), where Ss and Se are mean snowfall rates from satellite and ERA5, respectively.

Citation: Journal of Hydrometeorology 22, 11; 10.1175/JHM-D-21-0018.1

b. Seasonal characteristics of rainfall and snowfall partition

Recent studies based on surface station measurements over TP indicated that mean daily snow depth and the number of snow-covered days have been decreasing in recent decades, and the trend is particularly significant during summer and fall when air temperature is the controlling variable to ground snow cover. In contrast, during winter the change of ground snow cover is largely determined by precipitation (Xu et al. 2017). Deng et al. (2017) found that winter snowfall has significantly increased, so has it in spring but to a less extent. Summer and fall, on the other hand, show a significant decrease in snowfall except in the coldest area. Clearly, the seasonal difference in precipitation amount and phase has important implications to ground snow accumulations.

To avoid spreading the data samples too thin, in this study, we decide to divide a year into only two seasons, i.e., the cold season refers to from 15 October to the next 15 April, and warm season refers to 16 April–14 October. The mean rainfall and snowfall distributions in the two seasons are shown in Fig. 9. In the cold season, snowfall mainly appears along the Karakoram Range and Himalayas, and over the Nyainqentanglha Mountains. There is no significant amount of rainfall at elevations higher than 2000 m. In the warm season, high snowfall values are not only observed over the high mountain areas mentioned above, but also near the Kunlun Mountains in the northern part of TP. Interestingly, the majority of snowfall in the Kunlun Mountains occurs in the warm season. Compared to the cold season, the warm season rainfall covers a broad area; rainfall rates greater than 200 mm yr−1 occur over the eastern and central TP and Tarim basin, in addition to the low-elevation areas at the foothills and south of TP. Moreover, the pattern of mean rainfall rate in the warm season is similar to that of the annual mean rainfall rate shown in Fig. 7, especially in high-elevation areas.

Fig. 9.
Fig. 9.

Mean snowfall rate from CPR in (a) the cold season and (b) the warm season, and mean rainfall rate from DPR in (c) in the cold season and (d) the warm season.

Citation: Journal of Hydrometeorology 22, 11; 10.1175/JHM-D-21-0018.1

The seasonal changes of the snowfall and rainfall mass fractions are also studied, and the results are shown in Fig. 10. It is clearly seen that almost all precipitation over the highland areas of TP falls as snow during the cold season. Snowfall mass fraction is high in most of high mountain areas even during the warm season, with over 60% precipitation falling as snowfall over Karakoram, Kunlun, and a large portion of the Himalayas. This illustrates the important role of snowfall in the glacier accumulation in high mountain areas.

Fig. 10.
Fig. 10.

Snowfall mass fractions in (a) cold and (b) warm seasons, and rainfall mass fractions in (c) cold and (d) warm seasons.

Citation: Journal of Hydrometeorology 22, 11; 10.1175/JHM-D-21-0018.1

c. Snowfall fraction versus elevation

Several studies have indicated that in a warming climate the rate of warming is amplified with elevation, that is, surface air temperature increases more rapidly at higher elevations (Mountain Research Initiative EDW Working Group 2015). Wang et al. (2016) studied that dependence of the trend of precipitation on elevations over TP based on meteorological station measurements and found a clear decreasing trend of snowfall above 3000 m MSL while there is a relative flat snowfall variation below this elevation. While the limited duration of our satellite data does not allow us to study the trend, it can provide the mean state on how the precipitation phase partition depends on elevation.

In Fig. 11 we show the scatterplots of the elevation dependence of total precipitation rate, snowfall rate, snowfall frequency, and snowfall mass fraction. Each point represents a mean value over a 1° × 2° grid. While the scatter of data points is tremendous in any of the four plots, there exists a general trend, that is, while total precipitation decreases with elevation, snowfall frequency, mean snowfall rate, and snowfall mass fraction all increase as elevation increases. Below 1 km, precipitation is mostly rainfall; both snowfall frequency and snowfall rate are close to 0. These are the data from the foothill of the plateau and lowland areas in the south. If we take the median value at each elevation, snowfall mass fraction appears to increase more than 10% with every increase of 1 km in elevation from near 0% below 1 km to about 60% at 5-km elevation (Fig. 11d). Interestingly, the elevation of 3000 m MSL seems to be a level marking the transition of precipitation phase, above which there is a sudden jump of the number of grid boxes with high values of snowfall frequency, snowfall rate, and snowfall mass fraction. This sudden transition at 3000-km elevation is similar to the snowfall trend transition level found by earlier investigators (e.g., Wang et al. 2016; Deng et al. 2017).

Fig. 11.
Fig. 11.

Elevation dependence of (a) precipitation rate, (b) snowfall frequency, (c) snowfall rate, and (d) snowfall mass fraction. Each point is an average for a 1° (latitude) × 2° (longitude) grid. Points are color coded by snowfall rate with values shown in (c) in mm yr−1. Median values of snowfall mass fraction are also shown by connected “x” marks in (d). The five elevation bands are defined as <1.5, 1.5–2.5, 2.5–3.5, 3.5–4.5, and >4.5 km.

Citation: Journal of Hydrometeorology 22, 11; 10.1175/JHM-D-21-0018.1

d. Summary of phase partition of total precipitation

As a summary of the precipitation phase partition characteristics over the study area (26°–39°N, 71°–104°E), we computed the satellite derived annual rainfall, snowfall, and snowfall mass fraction values averaged for those 1° × 2° grids with mean elevation greater than 1 km MSL. The 1-km elevation threshold is used to exclude data from foothills and lowland areas in the south (ref. Fig. 1), which by definition are not part of the plateau. The results are listed in Table 1, along with the same attributes derived from the APHRODITE-2 and ERA5 data for comparison. Annually, it is estimated that the mean precipitation total is 359 mm over the plateau, of which 41% are in solid phase. The satellite estimated precipitation total is quite similar to the APHRODITE-2 estimate (within 10%), although the satellite retrievals give ~13% more in the northern part but ~28% less precipitation in the southern part of the plateau. Snowfall is the dominant form of precipitation in the cold season (half year) with a mass fraction of 83%, while snowfall mass fraction is about 28% in the warm season. In the northern part (north of 33°N) of the plateau, the annual precipitation is almost equally partitioned to liquid and solid phases, while in the south snowfall contributes about one-third (32%) of the precipitation total. For comparison purposes, the precipitation statistics derived from the ERA5 reanalysis are also included in Table 1. It is seen that the total precipitation from the ERA5 in the high elevation areas is about twice of satellite radar estimate, and a large portion of the overestimation results from rainfall in the warm season.

Table 1.

Statistics of total precipitation and phase partition for grids with elevation greater than 1 km. The statistics of ERA5 are listed in parentheses. Total precipitation rates from APHRODITE-2 are also included as the second value in the parentheses.

Table 1.

4. Discussions and conclusions

Precipitation is the most important process for glacier accumulation over the TP region, and precipitation phase is particularly a key factor. Based on satellite radar observations, the goal of this study is to assess the precipitation over TP not only by its total amount but also its partition between solid and liquid phases. The advantage of using satellite radar observations is twofold, i.e., relatively uniform spatial distribution of samples (unlike measuring by rain gauges) and less contamination by complex surface characteristics (unlike using passive sensors). A shortcoming is the short time period of data availability for satellite radar observations. In particular, we have to assess snowfall and rainfall separately using radars on two different platforms, i.e., CloudSat and GPM, from which the most suitable time periods for snowfall and rainfall retrievals are not well overlapped. An assumption has to be made in this study, i.e., the 11-yr CloudSat and 5-yr GPM observations respectively represent the mean states of snowfall and rainfall, so that the precipitation total and snowfall (or rainfall) fraction can be assessed by additions and taking ratios. Although this assumption is a weakness of this study, the combination of observations from the two space-borne radars with their best sensitivity to different hydrometeors presents a creative way to assess precipitation with phase information resolved.

To examine the magnitude of error due to the time mismatch in the snowfall and rainfall retrievals, we studied the precipitation differences in ERA5 between 2007–10 (when CPR was in full operation) and 2015–19 (when DPR rainfall data are used in our analysis). The results are shown in Fig. 12, in which total precipitation and snowfall between the two time periods are compared by both difference (Figs. 12e,f) and relative difference (Figs. 12g,h). It is seen that for most part of the region the absolute value of relative difference is less than 10%, similar magnitude to the uncertainty of DPR rainfall estimates (Oki et al. 2020). In fact, if averaging data over the entire region, the differences are 20 mm yr−1 for total precipitation and 3 mm yr−1 for snowfall, which translates to 3% and 2% relative differences, respectively. In other words, the interannual variation is much smaller than the mean values for both total precipitation and snowfall. This result lends some support to our assumption of using snowfall and rainfall derived from different time periods to represent their mean states.

Fig. 12.
Fig. 12.

Comparison of ERA5 total precipitation and snowfall between the time periods of 2007–10 and 2015–19. The difference is defined by mean value of 2006–10 minus that of 2015–19, and relative difference is the difference divided by the mean of total precipitation or snowfall rates of the two time periods.

Citation: Journal of Hydrometeorology 22, 11; 10.1175/JHM-D-21-0018.1

In this study, the total precipitation is determined by combining snowfall retrievals from CloudSat CPR observations averaged over the time period of 2006–17 and rainfall retrievals from GPM DPR observations averaged over the time period of 2015–19. Determination for whether a radar observation corresponds to a scene of snowfall or rainfall at surface is consistently done for both satellite sensors using a scheme primarily relying on 2-m temperature, which in this study is interpolated based on data of ERA5 reanalysis and digital elevation at the location of radar pixel. As a sanity check, the satellite radar derived total precipitation is compared with gauge-based APHRODITE-2 and ERA5 reanalysis estimates. When averaged over the areas with elevations above 1 km MSL within 26°–39°N and 71°–104°E, the total precipitation from satellite retrievals is very close to that of APHRODITE-2, although there is a difference between the two estimates over the northern mountainous areas where snowfall dominates the total precipitation according to the satellite estimates (no phase information in APHRODITE-2 product). The similarity between the satellite and gauge estimates provides some confidence to the validity of the satellite retrievals. However, the ERA5 total precipitation is about twice of the satellite estimates, and the largest discrepancy between the two occurs for rainfall in the southern part of the region. Although more investigations are still needed, the comparable magnitude seen between the satellite radar and APHRODITE-2 gauge-based estimates makes us to conclude that precipitation in the ERA5 reanalysis may be too high for the TP region.

Analyzing the satellite radar retrievals, we found the following precipitation characteristics over the TP region. Averaged over the highland area (1 km MSL and above) within 26°–39°N and 71°–104°E, the annual total precipitation is around 400 mm, of which about 40% fall as snowfall. If we use 33°N latitude to divide the region into two, the snowfall mass fraction is about 45% in the northern and 30% in the southern portion. If we divide the whole year into cold (from 15 October to the next 15 April) and warm (rest of the year) seasons, the snowfall mass fraction is about 80% in the cold and 30% in the warm season. The spatial distribution of the satellite derived total precipitation generally agrees with earlier studies, as well as with those derived in this study from gauge and reanalysis data, indicating much more precipitation in the south of 33°N than in the north. This spatial contrast was earlier explained by the difference in the dominant hydrological process in each region. There are several snowfall dominant high mountain regions where snowfall mass fraction exceeds 50% or even 80%, i.e., the Karakoram, Kunlun, Himalayan, and Nyainqentanglha Mountains. Interestingly, of these regions, the snowfall in the Kunlun Mountains mainly occurs in the warm season. Except for over the southern slopes and foothills of the Karakoram and Himalayas, rainfall is chiefly observed during the warm season with the largest amount (over 500 mm yr−1 when averaged over 1° × 2° grid) appearing over the southern slopes of the plateau. Right over the plateau in the Nyainqentanglha Mountains, the 1° × 2° grid averaged rainfall rate during warm season can also reach over 400 mm yr−1. Surface elevation is clearly found to be a high-impact factor to total precipitation and its phase partition, generally with total precipitation decreasing but snowfall fraction increasing with the increase of elevation. The 3-km elevation seems to be a critical level, above which both snowfall rate and snowfall mass fraction jump higher, a result also consistent with previous gauge-based studies (e.g., Wang et al. 2016).

In this study, we combined the best available satellite estimates to show a possible passage of partitioning the solid and liquid precipitation over TP and neighboring regions, which is useful for better estimating glacier changes and hydrological processes over mountainous highland regions. While presently there are still unavoidable shortcomings of this approach, particularly, due to the difference in time periods covered by the two satellite radars, the consistency of some current results with earlier gauge-based studies provides some support to this approach. With additional space-borne radars planned in the future, such as Earth Clouds, Aerosol and Radiation Explorer (EarthCARE; Illingworth et al. 2015) and Aerosol, Cloud, Convection and Precipitation (ACCP; NASA 2020), this approach can be further tested and improved to derive more accurate precipitation characteristics over high mountain regions.

Acknowledgments

This work has been carried out during the first author’s (PS) visit to Florida State University with the support from The China Scholarship Council. GL’s participation is supported by NASA PMM Grant 80NSSC19K0718. The ERA5 data are obtained from the Copernicus Climate Change Service (C3S) Climate Data Store (CDS) cloud server (https://cds.climate.copernicus.eu/). CloudSat CPR data are obtained from CloudSat Data Center at Colorado State University (http://www.cloudsat.cira.colostate.edu). GPM DPR data are obtained from NASA PMM data archive center (https://gpm.nasa.gov/data).

REFERENCES

  • Adhikari, A., C. Liu, and M. S. Kulie, 2018: Global distribution of snow precipitation features and their properties from 3 years of GPM observations. J. Climate, 31, 37313754, https://doi.org/10.1175/JCLI-D-17-0012.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Barros, A. P., and D. P. Lettenmaier, 1994: Dynamic modeling of orographically induced precipitation. Rev. Geophys., 32, 265284, https://doi.org/10.1029/94RG00625.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Barros, A. P., M. Joshi, J. Putonen, and D. W. Burbank, 2000: A study of the 1999 monsoon rainfall in a mountainous region in the central Nepal using TRMM products and rain gauge observations. Geophys. Res. Lett., 27, 36833686, https://doi.org/10.1029/2000GL011827.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Barros, A. P., and Coauthors, 2006: From weather to climate – Seasonal and interannual variability of storms and implications for erosion processes in the Himalaya. Tectonics, Climate, and Landscape Evolution, S. Willett et al., Eds., GSA Special Papers, Vol. 398, Geological Society of America, 17–38, https://doi.org/10.1130/S2006.2398(02).

    • Crossref
    • Export Citation
  • Behrangi, A., K. Andreadis, J. B. Fisher, F. J. Turk, S. Granger, T. Painter, and N. Das, 2014: Satellite-based precipitation estimation and its application for streamflow prediction over mountainous western U.S. basins. J. Appl. Meteor. Climatol., 53, 28232842, https://doi.org/10.1175/JAMC-D-14-0056.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bolch, T., and Coauthors, 2012: The state and fate of Himalayan glaciers. Science, 336, 310314, https://doi.org/10.1126/science.1215828.

  • Bookhagen, B., and D. W. Burbank, 2010: Toward a complete Himalaya hydrological budget: Spatiotemporal distribution of snowmelt and rainfall and their impact on river discharge. J. Geophys. Res., 115, F03019, https://doi.org/10.1029/2009JF001426.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Casella, D., G. Panegrossi, P. Sanò, A. C. Marra, S. Dietrich, B. T. Johnson, and M. S. Kulie, 2017: Evaluation of the GPM DPR snowfall detection capability: Comparison with CloudSat-CPR. Atmos. Res., 197, 6475, https://doi.org/10.1016/j.atmosres.2017.06.018.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cogley, J. G., and Coauthors, 2011: Glossary of glacier mass balance and related terms. IHP-VII Tech. Doc. in Hydrology 86, IACS Contribution 2, UNESCO-IHP, 114 pp., https://wgms.ch/downloads/Cogley_etal_2011.pdf.

  • Cronk, H., and P. Partain, 2017: CloudSat ECMWF-AUX auxiliary data product process description and interface control document, product version: P_R05. NASA Doc., 15 pp., http://www.cloudsat.cira.colostate.edu/sites/default/files/products/files/ECMWF-AUX_PDICD.P_R05.rev0_.pdf.

  • Deng, H., N. C. Pepin, and Y. Chen, 2017: Changes of snowfall under warming in the Tibetan Plateau. J. Geophys. Res. Atmos., 122, 73237341, https://doi.org/10.1002/2017JD026524.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ebtehaj, A. M., R. L. Bras, and E. Foufoula-Georgiou, 2016: Evaluation of ShARP passive rainfall retrievals over snow-covered land surfaces and coastal zones. J. Hydrometeor., 17, 10131029, https://doi.org/10.1175/JHM-D-15-0164.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ferraro, R. R., and Coauthors, 2013: An evaluation of microwave land surface emissivities over the continental United States to benefit GPM-era precipitation algorithms. IEEE Trans. Geosci. Remote Sens., 51, 378398, https://doi.org/10.1109/TGRS.2012.2199121.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Fu, Y., and Coauthors, 2018: Precipitation characteristics over the steep slope of the Himalayas in rainy season observed by TRMM PR and VIRS. Climate Dyn., 51, 19711989, https://doi.org/10.1007/s00382-017-3992-3.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Fujita, K., 2008: Effect of precipitation seasonality on climatic sensitivity of glacier mass balance. Earth Planet. Sci. Lett., 276, 1419, https://doi.org/10.1016/j.epsl.2008.08.028.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gadgil, S., 2003: The Indian monsoon and its variability. Annu. Rev. Earth Planet. Sci., 31, 429467, https://doi.org/10.1146/annurev.earth.31.100901.141251.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Grecu, M., W. S. Olson, S. J. Munchak, S. Ringerud, L. Liao, Z. S. Haddad, B. L. Kelley, and S. F. McLaughlin, 2016: The GPM combined algorithm. J. Atmos. Oceanic Technol., 33, 22252245, https://doi.org/10.1175/JTECH-D-16-0019.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hamada, A., and Y. N. Takayabu, 2016: Improvements in detection of light precipitation with the Global Precipitation Measurement dual-frequency precipitation radar (GPM DPR). J. Atmos. Oceanic Technol., 33, 653667, https://doi.org/10.1175/JTECH-D-15-0097.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hersbach, H., and Coauthors, 2020: The ERA5 global reanalysis. Quart. J. Roy. Meteor. Soc., 146, 19992049, https://doi.org/10.1002/qj.3803.

  • Hou, A. Y., and Coauthors, 2014: The Global Precipitation Measurement mission. Bull. Amer. Meteor. Soc., 95, 701722, https://doi.org/10.1175/BAMS-D-13-00164.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Huffman, G. J., and Coauthors, 2007: The TRMM Multi-Satellite Precipitation Analysis: Quasi-global, multi-year, combined-sensor precipitation estimates at fine scale. J. Hydrometeor., 8, 3855, https://doi.org/10.1175/JHM560.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Iguchi, T., and Z. Haddad, 2020: Introduction to radar rain retrieval methods. Satellite Precipitation Measurement, Vol. 1, V. Levizzani et al., Eds., Advances in Global Change Research, Vol. 67, Springer, 169–182.

    • Crossref
    • Export Citation
  • Iguchi, T., and Coauthors, 2018: GPM/DPR level-2. Algorithm Theoretical Basis Doc., 127 pp., https://gpm.nasa.gov/sites/default/files/2019-05/ATBD_DPR_201811_with_Appendix3b.pdf.

  • Illingworth, A. J., and Coauthors, 2015: The EarthCARE satellite: The next step forward in global measurements of clouds, aerosols, precipitation, and radiation. Bull. Amer. Meteor. Soc., 96, 13111332, https://doi.org/10.1175/BAMS-D-12-00227.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Immerzeel, W. W., L. P. H. van Beek, and M. F. P. Bierkens, 2010: Climate change will affect the Asian water towers. Science, 328, 13821385, https://doi.org/10.1126/science.1183188.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Joyce, R. J., and P. Xie, 2011: Kalman filter-based CMORPH. J. Hydrometeor., 12, 15471563, https://doi.org/10.1175/JHM-D-11-022.1.

  • Kääb, A., E. Berthier, C. Nuth, J. Gardelle, and Y. Arnaud, 2012: Contrasting patterns of early twenty-first-century glacier mass change in the Himalayas. Nature, 488, 495498, https://doi.org/10.1038/nature11324.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kubota, T., and Coauthors, 2020: Global Satellite Mapping of Precipitation (GSMaP) products in the GPM era. Satellite Precipitation Measurement, Vol. 1, V. Levizzani et al., Eds., Advances in Global Change Research, Vol. 67, Springer, 355–373.

    • Crossref
    • Export Citation
  • Kulie, M. S., and L. Milani, 2018: Seasonal variability of shallow cumuliform snowfall: A CloudSat perspective. Quart. J. Roy. Meteor. Soc., 144, 329343, https://doi.org/10.1002/qj.3222.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kulie, M. S., L. Milani, N. R. Wood, and T. S. L’Ecuyer, 2020: Global snowfall detection and measurement. Satellite Precipitation Measurement, Vol. 2, V. Levizzani et al., Eds., Advances in Global Change Research, Vol. 69, Springer, 699–716.

    • Crossref
    • Export Citation
  • Kummerow, C. D., D. L. Randel, M. Kulie, N. Wang, R. Ferraro, S. J. Munchak, and V. Petkovic, 2015: The evolution of the Goddard profiling algorithm to a fully parametric scheme. J. Atmos. Oceanic Technol., 32, 22652280, https://doi.org/10.1175/JTECH-D-15-0039.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lang, T. J., and A. P. Barros, 2004: Winter storm in the central Himalayas. J. Meteor. Soc. Japan, 82, 829844, https://doi.org/10.2151/jmsj.2004.829.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Liu, G., 2008: Deriving snow cloud characteristics from CloudSat observations. J. Geophys. Res., 113, D00A09, https://doi.org/10.1029/2007JD009766.

  • Liu, G., 2020: Radar snowfall measurement. Satellite Precipitation Measurement, Vol. 1, V. Levizzani et al., Eds., Advances in Global Change Research, Vol. 67, Springer, 277–295.

    • Crossref
    • Export Citation
  • Liu, X., and Z.-Y. Yin, 2001: Spatial and temporal variation of summer precipitation over the eastern Tibetan Plateau and the North Atlantic Oscillation. J. Climate, 14, 28962909, https://doi.org/10.1175/1520-0442(2001)014<2896:SATVOS>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lu, H., and G. Liu, 2010: Trends in temperature and precipitation on the Tibetan Plateau, 1961–2005. Climate Res., 43, 179190, https://doi.org/10.3354/cr00909.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Maussion, F., D. Scherer, T. Mölg, E. Collier, J. Curio, and R. Finkelnburg, 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
  • Milani, L., and N. B. Wood, 2021: Biases in CloudSat falling snow estimates resulting from daylight-only operations. Remote Sens., 13, 2041, https://doi.org/10.3390/rs13112041.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mountain Research Initiative EDW Working Group, 2015: Elevation-dependent warming in mountain regions of the world. Nat. Climate Change, 5, 424430, https://doi.org/10.1038/nclimate2563.

    • Search Google Scholar
    • Export Citation
  • NASA, 2020: Aerosol, Cloud, Convection and Precipitation. Accessed 30 December 2020, https://vac.gsfc.nasa.gov/accp/home.htm.

  • Oki, R., T. Iguchi, and K. Nakamura, 2020: The GPM DPR validation program. Satellite Precipitation Measurement, Vol. 2, V. Levizzani et al., Eds., Advances in Global Change Research, Vol. 69, Springer, 503–514.

    • Crossref
    • Export Citation
  • Qin, J., K. Yang, S. Liang, and X. Guo, 2009: The altitudinal dependence of recent rapid warming over the Tibetan Plateau. Climatic Change, 97, 321327, https://doi.org/10.1007/s10584-009-9733-9.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Rees, H. G., and D. N. Collins, 2006: Regional differences in response of flow in glacier-fed Himalayan rivers to climatic warming. Hydrol. Processes, 20, 21572169, https://doi.org/10.1002/hyp.6209.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Rouf, T., Y. Mei, V. Maggioni, P. Houser, and M. Noonan, 2020: A physically based atmospheric variables downscaling technique. J. Hydrometeor., 21, 93108, https://doi.org/10.1175/JHM-D-19-0109.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Schiemann, R., D. Luethi, and C. Schaer, 2009: Seasonality and interannual variability of the westerly jet in the Tibetan Plateau region. J. Climate, 22, 29402957, https://doi.org/10.1175/2008JCLI2625.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Shige, S., S. Kida, H. Ashiwake, T. Kubota, and K. Aonashi, 2013: Improvement of TMI rain retrievals in mountainous areas. J. Appl. Meteor. Climatol., 52, 242254, https://doi.org/10.1175/JAMC-D-12-074.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sims, E. M., and G. Liu, 2015: A parameterization of the probability of snow–rain transition. J. Hydrometeor., 16, 14661477, https://doi.org/10.1175/JHM-D-14-0211.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Skofronick-Jackson, G., D. Kirschbaum, W. Petersen, G. Huffman, C. Kidd, E. Stocker, and R. Kakar, 2018: The Global Precipitation Measurement (GPM) mission’s scientific achievements and societal contributions: Reviewing four years of advanced rain and snow observations. Quart. J. Roy. Meteor. Soc., 144, 2748, https://doi.org/10.1002/qj.3313.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Tao, J., and A. P. Barros, 2018: Multi-year atmospheric forcing datasets for hydrologic modeling in regions of complex terrain – Methodology and evaluation over the Integrated Precipitation and Hydrology Experiment 2014 domain. J. Hydrol., 567, 824842, https://doi.org/10.1016/j.jhydrol.2016.12.058.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Tian, L., T. Yao, W. Sun, M. Stievenard, and J. Jouzel, 2001: Relationship between δD and δ18O in precipitation on north and south of the Tibetan Plateau and moisture recycling. Sci. China, 44D, 789796, https://doi.org/10.1007/BF02907091.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Toyoshima, K., H. Masunaga, and F. A. Furuzawa, 2015: Early evaluation of Ku- and Ka-band sensitivities for the Global Precipitation Measurement (GPM) Dual-frequency Precipitation Radar (DPR). Sci. Online Lett. Atmos., 11, 1417, https://doi.org/10.2151/sola.2015-004.

    • Search Google Scholar
    • Export Citation
  • Turk, J., 2016: CloudSat-GPM coincidence dataset, version 1C. 18 pp., https://gpm.nasa.gov/sites/default/files/document_files/CSATGPM_COIN_ATBD.pdf.

  • Viste, E., and A. Sorteberg, 2015: Snowfall in the Himalayas: An uncertain future from a little-known past. Cryosphere, 9, 11471167, https://doi.org/10.5194/tc-9-1147-2015.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wang, B., 2006: The Asian Monsoon. Springer, 787 pp.

  • Wang, J., M. Zhang, S. Wang, Z. Ren, Y. Che, F. Qiang, and D. Qu, 2016: Decrease in snowfall/rainfall ratio in the Tibetan Plateau from 1961 to 2013. J. Geogr. Sci., 26, 12771288, https://doi.org/10.1007/s11442-016-1326-8.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Webster, P. J., V. O. Magana, T. N. Palmer, J. Shukla, R. A. Tomas, M. Yanai, and T. Yasunari, 1998: Monsoons: Processes, predictability, and the prospects for prediction. J. Geophys. Res., 103, 14 45114 510, https://doi.org/10.1029/97JC02719.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wood, N. B., and T. S. L’Ecuyer, 2018: Level 2C snow profile process description and interface control document, product version P1_R05. NASA Doc., 26 pp., http://www.cloudsat.cira.colostate.edu/sites/default/files/products/files/2C-SNOW-PROFILE_PDICD.P1_R05.rev0_.pdf.

  • Xu, W., L. Ma, M. Ma, H. Zhang, and W. Yuan, 2017: Spatial-temporal variability of snowcover and depth in the Qinghai-Tibetan Plateau. J. Climate, 30, 15211533, https://doi.org/10.1175/JCLI-D-15-0732.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yao, T., J. Pu, A. Lu, Y. Wang, and W. Yu, 2007: Recent glacial retreat and its impact on hydrological processes on the Tibetan Plateau, China, and surrounding regions. Arct. Antarct. Alp. Res., 39, 642650, https://doi.org/10.1657/1523-0430(07-510)[YAO]2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yao, T., and Coauthors, 2012: Different glacier status with atmospheric circulations in Tibetan Plateau and surroundings. Nat. Climate Change, 2, 663667, https://doi.org/10.1038/nclimate1580.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yatagai, A., 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
  • Yin, Z.-Y., X. Zhang, X. Liu, M. Colella, and X. Chen, 2008: An assessment of the biases of satellite rainfall estimates over the Tibetan Plateau and correction methods based on topographic analysis. J. Hydrometeor., 9, 301326, https://doi.org/10.1175/2007JHM903.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, Y., B. Li, and D. Zheng, 2002: A discussion on the boundary and area of the Tibetan Plateau in China. Geogr. Res., 21, 18, https://doi.org/10.11821/yj2002010001.

    • Search Google Scholar
    • Export Citation
Save
  • Adhikari, A., C. Liu, and M. S. Kulie, 2018: Global distribution of snow precipitation features and their properties from 3 years of GPM observations. J. Climate, 31, 37313754, https://doi.org/10.1175/JCLI-D-17-0012.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Barros, A. P., and D. P. Lettenmaier, 1994: Dynamic modeling of orographically induced precipitation. Rev. Geophys., 32, 265284, https://doi.org/10.1029/94RG00625.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Barros, A. P., M. Joshi, J. Putonen, and D. W. Burbank, 2000: A study of the 1999 monsoon rainfall in a mountainous region in the central Nepal using TRMM products and rain gauge observations. Geophys. Res. Lett., 27, 36833686, https://doi.org/10.1029/2000GL011827.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Barros, A. P., and Coauthors, 2006: From weather to climate – Seasonal and interannual variability of storms and implications for erosion processes in the Himalaya. Tectonics, Climate, and Landscape Evolution, S. Willett et al., Eds., GSA Special Papers, Vol. 398, Geological Society of America, 17–38, https://doi.org/10.1130/S2006.2398(02).

    • Crossref
    • Export Citation
  • Behrangi, A., K. Andreadis, J. B. Fisher, F. J. Turk, S. Granger, T. Painter, and N. Das, 2014: Satellite-based precipitation estimation and its application for streamflow prediction over mountainous western U.S. basins. J. Appl. Meteor. Climatol., 53, 28232842, https://doi.org/10.1175/JAMC-D-14-0056.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bolch, T., and Coauthors, 2012: The state and fate of Himalayan glaciers. Science, 336, 310314, https://doi.org/10.1126/science.1215828.

  • Bookhagen, B., and D. W. Burbank, 2010: Toward a complete Himalaya hydrological budget: Spatiotemporal distribution of snowmelt and rainfall and their impact on river discharge. J. Geophys. Res., 115, F03019, https://doi.org/10.1029/2009JF001426.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Casella, D., G. Panegrossi, P. Sanò, A. C. Marra, S. Dietrich, B. T. Johnson, and M. S. Kulie, 2017: Evaluation of the GPM DPR snowfall detection capability: Comparison with CloudSat-CPR. Atmos. Res., 197, 6475, https://doi.org/10.1016/j.atmosres.2017.06.018.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cogley, J. G., and Coauthors, 2011: Glossary of glacier mass balance and related terms. IHP-VII Tech. Doc. in Hydrology 86, IACS Contribution 2, UNESCO-IHP, 114 pp., https://wgms.ch/downloads/Cogley_etal_2011.pdf.

  • Cronk, H., and P. Partain, 2017: CloudSat ECMWF-AUX auxiliary data product process description and interface control document, product version: P_R05. NASA Doc., 15 pp., http://www.cloudsat.cira.colostate.edu/sites/default/files/products/files/ECMWF-AUX_PDICD.P_R05.rev0_.pdf.

  • Deng, H., N. C. Pepin, and Y. Chen, 2017: Changes of snowfall under warming in the Tibetan Plateau. J. Geophys. Res. Atmos., 122, 73237341, https://doi.org/10.1002/2017JD026524.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ebtehaj, A. M., R. L. Bras, and E. Foufoula-Georgiou, 2016: Evaluation of ShARP passive rainfall retrievals over snow-covered land surfaces and coastal zones. J. Hydrometeor., 17, 10131029, https://doi.org/10.1175/JHM-D-15-0164.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ferraro, R. R., and Coauthors, 2013: An evaluation of microwave land surface emissivities over the continental United States to benefit GPM-era precipitation algorithms. IEEE Trans. Geosci. Remote Sens., 51, 378398, https://doi.org/10.1109/TGRS.2012.2199121.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Fu, Y., and Coauthors, 2018: Precipitation characteristics over the steep slope of the Himalayas in rainy season observed by TRMM PR and VIRS. Climate Dyn., 51, 19711989, https://doi.org/10.1007/s00382-017-3992-3.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Fujita, K., 2008: Effect of precipitation seasonality on climatic sensitivity of glacier mass balance. Earth Planet. Sci. Lett., 276, 1419, https://doi.org/10.1016/j.epsl.2008.08.028.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gadgil, S., 2003: The Indian monsoon and its variability. Annu. Rev. Earth Planet. Sci., 31, 429467, https://doi.org/10.1146/annurev.earth.31.100901.141251.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Grecu, M., W. S. Olson, S. J. Munchak, S. Ringerud, L. Liao, Z. S. Haddad, B. L. Kelley, and S. F. McLaughlin, 2016: The GPM combined algorithm. J. Atmos. Oceanic Technol., 33, 22252245, https://doi.org/10.1175/JTECH-D-16-0019.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hamada, A., and Y. N. Takayabu, 2016: Improvements in detection of light precipitation with the Global Precipitation Measurement dual-frequency precipitation radar (GPM DPR). J. Atmos. Oceanic Technol., 33, 653667, https://doi.org/10.1175/JTECH-D-15-0097.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hersbach, H., and Coauthors, 2020: The ERA5 global reanalysis. Quart. J. Roy. Meteor. Soc., 146, 19992049, https://doi.org/10.1002/qj.3803.

  • Hou, A. Y., and Coauthors, 2014: The Global Precipitation Measurement mission. Bull. Amer. Meteor. Soc., 95, 701722, https://doi.org/10.1175/BAMS-D-13-00164.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Huffman, G. J., and Coauthors, 2007: The TRMM Multi-Satellite Precipitation Analysis: Quasi-global, multi-year, combined-sensor precipitation estimates at fine scale. J. Hydrometeor., 8, 3855, https://doi.org/10.1175/JHM560.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Iguchi, T., and Z. Haddad, 2020: Introduction to radar rain retrieval methods. Satellite Precipitation Measurement, Vol. 1, V. Levizzani et al., Eds., Advances in Global Change Research, Vol. 67, Springer, 169–182.

    • Crossref
    • Export Citation
  • Iguchi, T., and Coauthors, 2018: GPM/DPR level-2. Algorithm Theoretical Basis Doc., 127 pp., https://gpm.nasa.gov/sites/default/files/2019-05/ATBD_DPR_201811_with_Appendix3b.pdf.

  • Illingworth, A. J., and Coauthors, 2015: The EarthCARE satellite: The next step forward in global measurements of clouds, aerosols, precipitation, and radiation. Bull. Amer. Meteor. Soc., 96, 13111332, https://doi.org/10.1175/BAMS-D-12-00227.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Immerzeel, W. W., L. P. H. van Beek, and M. F. P. Bierkens, 2010: Climate change will affect the Asian water towers. Science, 328, 13821385, https://doi.org/10.1126/science.1183188.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Joyce, R. J., and P. Xie, 2011: Kalman filter-based CMORPH. J. Hydrometeor., 12, 15471563, https://doi.org/10.1175/JHM-D-11-022.1.

  • Kääb, A., E. Berthier, C. Nuth, J. Gardelle, and Y. Arnaud, 2012: Contrasting patterns of early twenty-first-century glacier mass change in the Himalayas. Nature, 488, 495498, https://doi.org/10.1038/nature11324.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kubota, T., and Coauthors, 2020: Global Satellite Mapping of Precipitation (GSMaP) products in the GPM era. Satellite Precipitation Measurement, Vol. 1, V. Levizzani et al., Eds., Advances in Global Change Research, Vol. 67, Springer, 355–373.

    • Crossref
    • Export Citation
  • Kulie, M. S., and L. Milani, 2018: Seasonal variability of shallow cumuliform snowfall: A CloudSat perspective. Quart. J. Roy. Meteor. Soc., 144, 329343, https://doi.org/10.1002/qj.3222.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kulie, M. S., L. Milani, N. R. Wood, and T. S. L’Ecuyer, 2020: Global snowfall detection and measurement. Satellite Precipitation Measurement, Vol. 2, V. Levizzani et al., Eds., Advances in Global Change Research, Vol. 69, Springer, 699–716.

    • Crossref
    • Export Citation
  • Kummerow, C. D., D. L. Randel, M. Kulie, N. Wang, R. Ferraro, S. J. Munchak, and V. Petkovic, 2015: The evolution of the Goddard profiling algorithm to a fully parametric scheme. J. Atmos. Oceanic Technol., 32, 22652280, https://doi.org/10.1175/JTECH-D-15-0039.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lang, T. J., and A. P. Barros, 2004: Winter storm in the central Himalayas. J. Meteor. Soc. Japan, 82, 829844, https://doi.org/10.2151/jmsj.2004.829.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Liu, G., 2008: Deriving snow cloud characteristics from CloudSat observations. J. Geophys. Res., 113, D00A09, https://doi.org/10.1029/2007JD009766.

    • Search Google Scholar
    • Export Citation
  • Liu, G., 2020: Radar snowfall measurement. Satellite Precipitation Measurement, Vol. 1, V. Levizzani et al., Eds., Advances in Global Change Research, Vol. 67, Springer, 277–295.

    • Crossref
    • Export Citation
  • Liu, X., and Z.-Y. Yin, 2001: Spatial and temporal variation of summer precipitation over the eastern Tibetan Plateau and the North Atlantic Oscillation. J. Climate, 14, 28962909, https://doi.org/10.1175/1520-0442(2001)014<2896:SATVOS>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lu, H., and G. Liu, 2010: Trends in temperature and precipitation on the Tibetan Plateau, 1961–2005. Climate Res., 43, 179190, https://doi.org/10.3354/cr00909.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Maussion, F., D. Scherer, T. Mölg, E. Collier, J. Curio, and R. Finkelnburg, 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
  • Milani, L., and N. B. Wood, 2021: Biases in CloudSat falling snow estimates resulting from daylight-only operations. Remote Sens., 13, 2041, https://doi.org/10.3390/rs13112041.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mountain Research Initiative EDW Working Group, 2015: Elevation-dependent warming in mountain regions of the world. Nat. Climate Change, 5, 424430, https://doi.org/10.1038/nclimate2563.

    • Search Google Scholar
    • Export Citation
  • NASA, 2020: Aerosol, Cloud, Convection and Precipitation. Accessed 30 December 2020, https://vac.gsfc.nasa.gov/accp/home.htm.

  • Oki, R., T. Iguchi, and K. Nakamura, 2020: The GPM DPR validation program. Satellite Precipitation Measurement, Vol. 2, V. Levizzani et al., Eds., Advances in Global Change Research, Vol. 69, Springer, 503–514.

    • Crossref
    • Export Citation
  • Qin, J., K. Yang, S. Liang, and X. Guo, 2009: The altitudinal dependence of recent rapid warming over the Tibetan Plateau. Climatic Change, 97, 321327, https://doi.org/10.1007/s10584-009-9733-9.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Rees, H. G., and D. N. Collins, 2006: Regional differences in response of flow in glacier-fed Himalayan rivers to climatic warming. Hydrol. Processes, 20, 21572169, https://doi.org/10.1002/hyp.6209.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Rouf, T., Y. Mei, V. Maggioni, P. Houser, and M. Noonan, 2020: A physically based atmospheric variables downscaling technique. J. Hydrometeor., 21, 93108, https://doi.org/10.1175/JHM-D-19-0109.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Schiemann, R., D. Luethi, and C. Schaer, 2009: Seasonality and interannual variability of the westerly jet in the Tibetan Plateau region. J. Climate, 22, 29402957, https://doi.org/10.1175/2008JCLI2625.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Shige, S., S. Kida, H. Ashiwake, T. Kubota, and K. Aonashi, 2013: Improvement of TMI rain retrievals in mountainous areas. J. Appl. Meteor. Climatol., 52, 242254, https://doi.org/10.1175/JAMC-D-12-074.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sims, E. M., and G. Liu, 2015: A parameterization of the probability of snow–rain transition. J. Hydrometeor., 16, 14661477, https://doi.org/10.1175/JHM-D-14-0211.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Skofronick-Jackson, G., D. Kirschbaum, W. Petersen, G. Huffman, C. Kidd, E. Stocker, and R. Kakar, 2018: The Global Precipitation Measurement (GPM) mission’s scientific achievements and societal contributions: Reviewing four years of advanced rain and snow observations. Quart. J. Roy. Meteor. Soc., 144, 2748, https://doi.org/10.1002/qj.3313.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Tao, J., and A. P. Barros, 2018: Multi-year atmospheric forcing datasets for hydrologic modeling in regions of complex terrain – Methodology and evaluation over the Integrated Precipitation and Hydrology Experiment 2014 domain. J. Hydrol., 567, 824842, https://doi.org/10.1016/j.jhydrol.2016.12.058.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Tian, L., T. Yao, W. Sun, M. Stievenard, and J. Jouzel, 2001: Relationship between δD and δ18O in precipitation on north and south of the Tibetan Plateau and moisture recycling. Sci. China, 44D, 789796, https://doi.org/10.1007/BF02907091.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Toyoshima, K., H. Masunaga, and F. A. Furuzawa, 2015: Early evaluation of Ku- and Ka-band sensitivities for the Global Precipitation Measurement (GPM) Dual-frequency Precipitation Radar (DPR). Sci. Online Lett. Atmos., 11, 1417, https://doi.org/10.2151/sola.2015-004.

    • Search Google Scholar
    • Export Citation
  • Turk, J., 2016: CloudSat-GPM coincidence dataset, version 1C. 18 pp., https://gpm.nasa.gov/sites/default/files/document_files/CSATGPM_COIN_ATBD.pdf.

  • Viste, E., and A. Sorteberg, 2015: Snowfall in the Himalayas: An uncertain future from a little-known past. Cryosphere, 9, 11471167, https://doi.org/10.5194/tc-9-1147-2015.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wang, B., 2006: The Asian Monsoon. Springer, 787 pp.

  • Wang, J., M. Zhang, S. Wang, Z. Ren, Y. Che, F. Qiang, and D. Qu, 2016: Decrease in snowfall/rainfall ratio in the Tibetan Plateau from 1961 to 2013. J. Geogr. Sci., 26, 12771288, https://doi.org/10.1007/s11442-016-1326-8.