1. Introduction
Ground-based weather radar networks, which can provide atmospheric surveillance data at a high temporal (5–6 min) and spatial (~1 km) resolution, are essential for weather-related applications such as natural hazard detection and warning systems. China started construction on the China New Generation Doppler Weather Radar (CINRAD) network in 1998, and more than 200 radars consisting of wavelengths of 10 cm (i.e., the S band) and 5 cm (i.e., the C band) have been utilized in operational observations. There has been extensive research on the application of observations from such ground-based radar nets to studies of weather processes, particularly with regard to monitoring and forecasting severe weather (e.g., Xiao and Liu 2006; Wang et al. 2012). In recent years, the Chinese Meteorological Academy has established several operational systems including the Severe Weather Automatic Nowcast System (SWAN) by the National Meteorological Center, the Radar Operational Software Engineering (ROSE) system by the Meteorological Observation Center, the Regional Three-Dimensional (3D) Mosaic Digital System by the Chinese Academy of Meteorological Sciences, and the China Integrated Meteorological Information Service System (CIMISS) by the National Meteorological Information Center, all of which are used to supply national and regional CINRAD network products used in weather monitoring and forecasting. With advanced quality-control (e.g., Liu et al. 2007), data-mosaic (e.g., Xiao et al. 2007; Wang et al. 2009; Yang et al. 2009), and rainfall-estimation techniques, these systems are capable of generating real-time (or non-real-time) quantitative precipitation estimations (QPE) at high temporal and spatial resolution. The ground-based weather radar network does have limitations (Maddox et al. 2002; Qi and Zhang 2013); for example, it makes large QPE errors in areas consisting of complex terrain where the radar beam is blocked or partially blocked. In this condition, the beam of ground-based radar, which gets in or above the higher atmospheric layers with long-range beam broadening, is usually used as the near-surface precipitation information. This approach will result in overestimates or underestimates of quantitative precipitation when the QPEs are based only on ground-based radar. Thus, corrections must be made to obtain more-accurate QPE products.
Because the vertical profiles of reflectivity (VPR) can reveal the structural characteristics of precipitation, researchers have been able to improve the accuracy of precipitation products by studying the variation of VPRs (e.g., Kitchen et al. 1994; Fabry and Zawadzki 1995; Vignal et al. 2000; Germann and Joss 2002; Bellon et al. 2005; Kirstetter et al. 2010; Zhang and Qi 2010; Cao et al. 2013; Zhuang et al. 2013; Gou et al. 2015). In mountainous regions with complex terrain, the CINRAD network coverage in the lower layers is limited (Yang et al. 2009; Wang et al. 2011; Zhuang et al. 2013), which often results in the absence of VPRs near the surface and in large QPE errors. Both beam blockage and beam broadening in the long-distance direction make it difficult to provide high spatial resolution and accurate VPRs. To solve this problem, spaceborne radars, with the advantages of higher vertical resolution (especially at far ranges, e.g., 100 km away from the radar site) and fewer impacts from intervening mountain blockages and beam broadening, can be used to modify the QPE from single ground-based radars (e.g., Gabella et al. 2011; Cao et al. 2013; Wen et al. 2013; Qi et al. 2013).
Known variously as the “roof of the world,” the region of “pumping of sensible heat” (Wu and Zhang 1998), and “the atmospheric water tower” (Xu et al. 2008), the Tibetan Plateau (TP) plays an important role in monsoon circulations and regional energy and water cycles over Asia (Wu and Zhang 1998; Xu et al. 2008, 2014; Wu et al. 2012). The TP has a great influence on the thermal and dynamical structure of its eastern downstream atmospheric boundary layer because of its unique topography, especially on the formation and development of precipitation microphysical processes, which are an important research topic in the fields of meteorology, hydrology, natural resources, environmental science, and so on. This study aimed to analyze the seasonal VPRs of different precipitation types by using the National Aeronautics and Space Administration (NASA)–Japan Aerospace Exploration Agency (JAXA) Tropical Rainfall Measuring Mission (TRMM) Precipitation Radar (PR) over an 11-yr period (January 2004–December 2014) in the ChuanYu (CY) megaregion, which is located in the eastern downstream region of the TP; this is where the ground-based CINRAD network encounters difficulties in accurately measuring the surface precipitation because of the complex terrain (as shown in Fig. 1). This region is also prone to disastrous floods and mudslides. The PR has a 13.8-GHz frequency (2.2-cm wavelength), with a field-of-view diameter of about 5.0 km (after the boost in August of 2001) at the nadir and a 0.25-km range resolution. The radar has a nominal sensitivity of approximately 18 dBZ (Simpson et al. 1996; Wen et al. 2013). Accurate calibration has been performed to ensure stable measurements (e.g., Kawanishi et al. 2000; Kozu et al. 2001; Takahashi et al. 2003). Extensive work involving the analysis of the storm structures according to PR measurements has been reported (e.g., Williams et al. 2007; Liu et al. 2010; Cao et al. 2013; Wen et al. 2013; Cao and Qi 2014). Most of these studies did not focus on the complex-terrain region in China, which is often affected by natural disasters. This paper mainly focuses on revealing the vertical structure of precipitation and the climatological characteristics of the VPRs in the mountainous region downstream of the TP, which will be helpful to applications aimed at merging the VPRs from spaceborne and ground-based radars for improving the QPE products there. The datasets used are introduced in section 2. Section 3 presents the detailed analysis results. The discussion and conclusions are presented in sections 4 and 5, respectively.
(top) Location of the analysis region in the southeastern downstream part of the TP (dashed black rectangle); (bottom) topographic image of the analysis region (units are meters).
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
(top) Location of the analysis region in the southeastern downstream part of the TP (dashed black rectangle); (bottom) topographic image of the analysis region (units are meters).
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
(top) Location of the analysis region in the southeastern downstream part of the TP (dashed black rectangle); (bottom) topographic image of the analysis region (units are meters).
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
2. Region and datasets
The study area is located in the eastern downstream region of the TP at longitudes 104°–110°E and latitudes 28°–32°N. Most of the area is surrounded by vast mountains, especially in the mideastern, southeastern, northern, and western regions (Fig. 1, bottom panel). Some research has revealed that the heat sources over the areas ranging from the eastern area of the TP via the Yangtze River basin to Bohai Bay and its surrounding areas and from the Bay of Bengal to the Indochina peninsula are key factors that determine the climate and weather over the study region. The unique climate and surface conditions in this area are very different from those in the vast plain area of the Huaihe River basin [HRB, as shown in Fig. 1 of Cao and Qi (2014)] in China and have caused frequent flooding, drought, storm, and mudslide disasters (Wu and Zhang 1998; Duan et al. 2005; Xu et al. 2008, 2014; Wei et al. 2009; Wu et al. 2012; Ceng et al. 2014). Given that the vertical structure of precipitation there has not been analyzed in detail in previous studies, new research using TRMM PR observations could enhance the understanding of precipitation characteristics in this region.
The level-2 2A25 and 2A23 products of TRMM, version 7 (V7), from January 2004 to December 2014 were used in this study. The 2A25 products provide attenuation-corrected profiles of radar reflectivity and corresponding rain-estimation parameters (Meneghini et al. 2004; Iguchi et al. 2009). Both the brightband (BB) detection and quantification and precipitation-type classification are included in the 2A23 algorithms (Awaka et al. 2009). Figures 2 and 3 respectively show the statistics for the seasonal rainfall amounts and number of PR scans with precipitation (≥0 mm h−1) measured by the PR over the analysis region. The total precipitation is actually a mean 1-h precipitation computed by averaging the precipitation rates available in each grid box and converting millimeters per hour to millimeters. The average amount of rainfall reaches the maximum value in summer, and high values are concentrated in the north-central CY region within complex mountain terrain. The amounts and spatial distributions of rainfall in spring and autumn are similar, and the precipitation systems move toward the northern mountainous area in autumn. There is much less rainfall in winter than in the other seasons, and this precipitation occurs mainly in the eastern region.
Statistics for the TRMM PR dataset used in the analysis; data show the annual average precipitation (mm) for PR pixels within each grid box. The spatial resolution is 0.1° × 0.1°.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
Statistics for the TRMM PR dataset used in the analysis; data show the annual average precipitation (mm) for PR pixels within each grid box. The spatial resolution is 0.1° × 0.1°.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
Statistics for the TRMM PR dataset used in the analysis; data show the annual average precipitation (mm) for PR pixels within each grid box. The spatial resolution is 0.1° × 0.1°.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
Statistics for the TRMM PR dataset used in the analysis; data show the total number of passes with precipitation (≥0 mm h−1) over the analysis region in the four seasons. The spatial resolution is 0.1° × 0.1°.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
Statistics for the TRMM PR dataset used in the analysis; data show the total number of passes with precipitation (≥0 mm h−1) over the analysis region in the four seasons. The spatial resolution is 0.1° × 0.1°.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
Statistics for the TRMM PR dataset used in the analysis; data show the total number of passes with precipitation (≥0 mm h−1) over the analysis region in the four seasons. The spatial resolution is 0.1° × 0.1°.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
According to Awaka et al. (2009), there are more than 30 subcategories for the classification of rain types in the V7 2A25 algorithm. For example, precipitation subcategories can be classified as “stratiform” (the flags in 2A25 are 100 and 110), “stratiform maybe” (the flags in 2A25 are 120, 130, 140, 152, 160, and 170), “convective” (the flags in 2A25 are 200, 210, 220, and 251–291), “convective maybe” (the flag in 2A25 is 240), and “other” (the flags in 2A25 are 300, 312, and 313) types (see Table 1). As is well known, both convective and stratiform precipitation represent normal, typical rain types. In this paper, only convective and convective-maybe and stratiform and stratiform-maybe precipitation are discussed. We put convective and convective maybe together as the convective-precipitation category because the number of the latter events was very small. The analysis methods used in this study are similar to those described in former work (Cao et al. 2013; Cao and Qi 2014).
Codes/flags for different rain types in 2A23 and 2A25 products.
3. Analysis results
a. Vertical profiles of reflectivity
The VPRs can reveal both the vertical structure and the microphysical processes of a storm by linking the surface precipitation to the radar observations at higher levels. The VPRs also provide useful information for investigations of microphysical processes and the quantitative retrieval of the liquid water content of precipitation (Willis and Heymsfield 1989; Fabry and Zawadzki 1995; Smith et al. 2009). Figure 4 gives the vertical profiles that were calculated in a way that is similar to that of Fig. 3 in Cao et al. (2013). The thick solid line represents the 50th-percentile curve. For the calculation of the occurrence frequency, the intervals of reflectivity and altitude were set to 0.1 dBZ and 250 m, and the total numbers of different precipitation categories for the study area are listed in Table 2. In this region, the convective type occurs at a lower probability than the stratiform type, and nearly 60% of the latter have unobvious BB features. The maximum reflectivity of the stratiform-maybe type is less than 35 dBZ. A total of 90% of the BB peak reflectivity in stratiform precipitation is less than 32 dBZ, and 40% of the maximum reflectivity of convective precipitation exceed 35 dBZ. Both the convective and stratiform types mostly occur at 2–5 km in the central region and 5–6 km in the northeastern area because of the complex terrain. As compared with percentile curves observed in the plain regions of China within the HRB (Cao and Qi 2014), the complex terrain in the eastern downstream region of the TP has data availability issues for the low levels (e.g., <1.5 km), and this situation has resulted in biased slopes in the lower portion of percentile curves. Moreover, the occurrence frequency of relatively large reflectivity is lower in the study region than in the HRB, which implies that the convective type is much shallower and weaker in the eastern downstream region of the TP.
Vertical profiles (in kilometers above mean sea level) of the occurrence frequency of radar reflectivity for the whole PR dataset: (a) convective, (b) stratiform, and (c) stratiform-maybe precipitation. The color scale indicates the number of occurrences N for 0.1-dBZ reflectivity and 250-m height intervals. Nine solid lines indicate the 10th–90th percentiles, with an interval of 10%. The thick solid line represents the 50th-percentile curve.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
Vertical profiles (in kilometers above mean sea level) of the occurrence frequency of radar reflectivity for the whole PR dataset: (a) convective, (b) stratiform, and (c) stratiform-maybe precipitation. The color scale indicates the number of occurrences N for 0.1-dBZ reflectivity and 250-m height intervals. Nine solid lines indicate the 10th–90th percentiles, with an interval of 10%. The thick solid line represents the 50th-percentile curve.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
Vertical profiles (in kilometers above mean sea level) of the occurrence frequency of radar reflectivity for the whole PR dataset: (a) convective, (b) stratiform, and (c) stratiform-maybe precipitation. The color scale indicates the number of occurrences N for 0.1-dBZ reflectivity and 250-m height intervals. Nine solid lines indicate the 10th–90th percentiles, with an interval of 10%. The thick solid line represents the 50th-percentile curve.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
Numbers of different precipitation categories by season for the study area from 2004 to 2014.
Figures 5–7 illustrate the variation of VPRs for the three types of precipitation in spring (March, April, and May), summer (June, July, and August), autumn (September, October, and November), and winter (December, January, and February). The solid lines represent the mean VPRs, which are categorized from left to right by the surface rainfall rates (mm h−1) in a way that is similar to that of Fig. 5 in Cao et al. (2013). Most of the VPRs displayed notable BB features when the surface rainfall intensity was less than 3 mm h−1 in autumn and summer and less than 10 mm h−1 in winter, especially when the rainfall intensity was larger. It is clear that the ratio of the BB peak reflectivity to its value in the lower rain region of the VPR profile was larger when the surface rainfall was weak than when the surface rainfall was heavy. The mean height of the BB bottom was about 1.5–2 km higher in summer than in the other seasons. The reflectivity decreased quickly with increasing height in spring, summer, and autumn, thus indicating that the ice–snow aggregation was very fast in the stratiform precipitation. The change of reflectivity amounted to 7–10 dBZ km−1 within the region 2–3 km above the BB, which was much larger than the value that was reported by Cao et al. (2013) for a mountainous region in the United States. The mean reflectivity was less than 18 dBZ from 3 km above the BB to the storm top with unobvious variations according to the increasing height; these data suggest that slow aggregation rates in this level contribute little to the surface rainfall intensity. Most of the conclusions from this study are similar to those of Cao et al. (2013) except that winter was not mentioned in their research. In winter, the precipitation system was shallow, and its top heights were below 8 km with a much lower BB. The variation rate of reflectivity between the BB and cloud top was much slower than that of the other seasons, which indicates that low temperatures and dry circumstances often result in low aggregation rates of hydrometeors in clouds. From Figs. 5–7, we can speculate that when the surface rainfall intensity is less than 5 mm h−1 the aggregation of hydrometeors throughout all of the levels should contribute to it but that when the surface rainfall intensity is more than 5 mm h−1 the aggregation of hydrometeors at 3 km above the BB represents the main contribution.
(top), (top middle) Seasonal variation of VPR for the type of stratiform precipitation. Solid lines represent the mean curves of VPRs, which are categorized from left to right by the surface rainfall rates (mm h−1): 0.4, 0.6, 1, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 12.0, 15.0, 20.0, 25.0, 30.0, 40.0, 60.0, and 80.0. The variation of the surface rainfall rate was set to 20% for the calculation of the mean VPR. (bottom middle),(bottom) The mean value of standard deviations for each mean VPR.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
(top), (top middle) Seasonal variation of VPR for the type of stratiform precipitation. Solid lines represent the mean curves of VPRs, which are categorized from left to right by the surface rainfall rates (mm h−1): 0.4, 0.6, 1, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 12.0, 15.0, 20.0, 25.0, 30.0, 40.0, 60.0, and 80.0. The variation of the surface rainfall rate was set to 20% for the calculation of the mean VPR. (bottom middle),(bottom) The mean value of standard deviations for each mean VPR.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
(top), (top middle) Seasonal variation of VPR for the type of stratiform precipitation. Solid lines represent the mean curves of VPRs, which are categorized from left to right by the surface rainfall rates (mm h−1): 0.4, 0.6, 1, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 12.0, 15.0, 20.0, 25.0, 30.0, 40.0, 60.0, and 80.0. The variation of the surface rainfall rate was set to 20% for the calculation of the mean VPR. (bottom middle),(bottom) The mean value of standard deviations for each mean VPR.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
As in Fig. 5, but for convective precipitation.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
As in Fig. 5, but for convective precipitation.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
As in Fig. 5, but for convective precipitation.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
As in Fig. 5, but for stratiform-maybe precipitation.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
As in Fig. 5, but for stratiform-maybe precipitation.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
As in Fig. 5, but for stratiform-maybe precipitation.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
Figure 6 gives the seasonal mean VPRs and mean standard deviation (STD) curves of convective precipitation. Relative to the stratiform type (see Fig. 5), the height of the storm top was much higher. One distinct feature of convective VPRs was that the reflectivity increased from the bottom of the freezing level to the surface. The increasing rates of their ice–snow aggregation were smaller than those of stratiform VPRs. Most of the VPRs had larger variations (5–6 dBZ) above the freezing level and smaller variations (1–2 dBZ) below the freezing level, which indicates that convective patterns with similar surface rainfall intensity have a similar distribution of raindrop size below the 2-km height level, but large variations exist above the 4-km height level, which can influence the convective depth and its precipitation intensity in spring, summer, and autumn. The STD curve in winter was very different from those of the other three seasons, with monotonic increases from the bottom to the top of clouds, as well as larger variation at the low level. The curve shows that the distribution of drop size and the microphysical processes of convective events are complicated and unstable in winter with the increase of heights. The VPRs were similar in spring and autumn. For the same surface rainfall intensity, there was no notable difference between the seasons for the mean VPRs at the heights close to the freezing level.
Figure 7 gives the VPRs of the stratiform-maybe precipitation, whose structure is similar to Fig. 5 except for the unobvious BB feature. The maximum reflectivity of most VPRs was less than 28 dBZ, and this was associated with a near-surface rainfall intensity of less than 2.5 mm h−1. Thus, the VPRs were mainly weak stratiform echoes, and their BB feature was not as apparent.
b. Height of the storm top
Some research has indicated that the higher the storm top is, the heavier the near-surface rainfall intensity is for a given precipitation type (Fu et al. 2006). Figure 8 shows the spatial distribution of the average storm top above mean sea level in the analysis region. These data are different from the results obtained by Cao et al. (2013) for the U.S. mountainous region. The height of the storm top varied along with the terrain height only in spring, and the terrain effect caused a 1.5–2.0-km difference between the highest and lowest regions. Storms in autumn often had higher top heights than those in spring, possibly because of the updrafts. The occurrence of storm systems moved toward the east in winter, with much lower heights (below 5 km) than those in the other seasons. A similar result to Cao et al. (2013) was that the storm tops were the highest in summer.
Seasonal and spatial variation of the storm-top height (meters above mean sea level). The spatial resolution is 0.1° × 0.1°.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
Seasonal and spatial variation of the storm-top height (meters above mean sea level). The spatial resolution is 0.1° × 0.1°.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
Seasonal and spatial variation of the storm-top height (meters above mean sea level). The spatial resolution is 0.1° × 0.1°.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
c. Height of the BB
The melting layer is one of the major factors that can increase the uncertainty of ground-based radar QPE products (Kirstetter et al. 2010; Zhang and Qi 2010; Cao et al. 2013; Qi and Zhang 2013; Zhuang et al. 2013). It is easy to interpret whether the radar beam intercepts or overshoots a melting layer for a given elevation angle and within a specific range away from the radar site when the BB height is known. Figure 9 gives the seasonal and spatial distribution of the BB-peak height above mean sea level. The heights of BB peaks arrived at 4.5–5 km in most of the area during summer. An obvious feature was that the topography could affect the BB height in spring whereas no such influence occurred in autumn. In the study region, the difference between the highest and the lowest terrain was about 1–1.5 km, and the difference between the BB-peak heights in such varied terrain was similar (i.e., 1–1.5 km). In winter, the mean height of BB peaks was much lower than that of the other seasons because of the low temperature. The reason for the influence of the topography on the mean heights of the BB in summer was not obvious in that the bottom of the melting layer was normally several kilometers above the surface. Figure 10 illustrates the seasonal mean heights of the BB peaks for different rainfall intensities. This figure shows that the average BB-peak height reaches the maximum and minimum values in summer and winter, respectively. The surface precipitation intensity increased in association with increasing BB heights in winter, and it increased with decreasing BB heights in summer and autumn when the rainfall intensity was larger than 10 mm h−1. No obvious relationship was found between the precipitation intensity and BB-peak heights in spring.
As in Fig. 8, but for BB-peak height.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
As in Fig. 8, but for BB-peak height.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
As in Fig. 8, but for BB-peak height.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
Mean heights of BB peaks for stratiform precipitation in the study region. The x axis gives the different rainfall intensities (mm h−1), and the y axis gives the heights above mean sea level (km). The four colored lines are for spring (blue), summer (red), autumn (green), and winter (purple).
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
Mean heights of BB peaks for stratiform precipitation in the study region. The x axis gives the different rainfall intensities (mm h−1), and the y axis gives the heights above mean sea level (km). The four colored lines are for spring (blue), summer (red), autumn (green), and winter (purple).
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
Mean heights of BB peaks for stratiform precipitation in the study region. The x axis gives the different rainfall intensities (mm h−1), and the y axis gives the heights above mean sea level (km). The four colored lines are for spring (blue), summer (red), autumn (green), and winter (purple).
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
d. Slope in the rain region
Knowledge of rain intensification/weakening below the melting layer would be helpful to improve the QPE products not only for ground radars but also for spaceborne radars (Cao et al. 2013; Qi and Zhang 2013; Cao and Qi 2014; Qi et al. 2013). The slope of the rainy region, in a way that was similar to the method used for Fig. 10 in Cao et al.(2013), was calculated and analyzed. Negative and positive slope values mean that the reflectivity increases and decreases, respectively, with decreases in altitude. As is well known, the intensification of reflectivity below the BB usually results from the accretion or collision of raindrops, and the weakening of reflectivity below the BB is often caused by the evaporation and breakup of raindrops. Figure 11 shows the spatial variation of VPR slopes over the analysis domain for stratiform, stratiform-maybe, and convective precipitation during summer and autumn. A notable feature, which agrees with former results (Cao et al. 2013), was that most of the slopes were positive for stratiform precipitation, and this result will be helpful to infer the raindrop sizes in the low layers ranging from the BB to the rainy region, which may become smaller because of the process of evaporation. Meanwhile, the slope trends were mostly negative for convective precipitation in the high terrain, which reveals the significant influence of topography on changes in the vertical updraft structures. This mechanism is different from that of stratiform precipitation because there are usually strong updrafts that occur in association with convective precipitation, which can accelerate raindrop collision or coalescence in the low layers. The VPR slope for convective precipitation was negative, especially in the northeastern part of the region with high terrain (−4 dB km−1). In autumn, some positive values appeared at the central and western regions where the terrain was very low. The reason may have been that relative humidity was low in the lower layers, which could have enhanced the evaporation in the area affected by the TP. This finding is different from that of Cao et al. (2013, their Fig. 10) whereby the terrain effect on the VPR slope seemed weak in the mountainous region in the United States, but it was very evident in the eastern downstream region of the TP.
Seasonal and spatial variation of the VPR slope in the rainy region: convective, stratiform, and stratiform-maybe precipitation for (left) summer and (right) autumn. The spatial resolution is 0.5° × 0.5°.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
Seasonal and spatial variation of the VPR slope in the rainy region: convective, stratiform, and stratiform-maybe precipitation for (left) summer and (right) autumn. The spatial resolution is 0.5° × 0.5°.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
Seasonal and spatial variation of the VPR slope in the rainy region: convective, stratiform, and stratiform-maybe precipitation for (left) summer and (right) autumn. The spatial resolution is 0.5° × 0.5°.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
Figure 12 illustrates the effect of the terrain’s altitude on the slope of the rainy region according to different rainfall rates. The effect of the terrain was not notable except for a slight increase with positive slopes when the rainfall rates were less than 2.0 mm h−1 in areas of higher terrain for stratiform and stratiform-maybe precipitation. Strong updrafts will enhance the raindrop growth at the low level. Hence, the terrain height had a marked influence on the convective-precipitation VPR slope. Another important factor that affected the VPR slope was the relative humidity at the low level; for example, when the surface rainfall intensity was weak (e.g., <2.5 mm h−1), the VPR slope of stratiform precipitation tended to be positive because the dry conditions would have enhanced the evaporation. Overall, the VPR slope tended to be negative during strong updrafts, conditions with much more moisture, weak evaporation, and collision at low levels for heavy convective precipitation. The VPR results also show the convective vertical structural feature when the rainfall intensity became large. It is easy to infer that weak convective precipitation often occurs in this region; in addition, it is evident that the TRMM PR cloud-type algorithms still encounter some challenges in the mountainous region around the TP.
The effect of the terrain’s altitude on the low-level VPR slope with different rainfall rates for convective, stratiform, and stratiform-maybe precipitation for (left) summer and (right) autumn. The x axis shows the 24 classes of rainfall rate denoted as 0.4, 0.6, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 12.0, 15.0, 20.0, 25.0, 30.0, 40.0, 60.0, and 80.0 mm h−1.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
The effect of the terrain’s altitude on the low-level VPR slope with different rainfall rates for convective, stratiform, and stratiform-maybe precipitation for (left) summer and (right) autumn. The x axis shows the 24 classes of rainfall rate denoted as 0.4, 0.6, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 12.0, 15.0, 20.0, 25.0, 30.0, 40.0, 60.0, and 80.0 mm h−1.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
The effect of the terrain’s altitude on the low-level VPR slope with different rainfall rates for convective, stratiform, and stratiform-maybe precipitation for (left) summer and (right) autumn. The x axis shows the 24 classes of rainfall rate denoted as 0.4, 0.6, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 12.0, 15.0, 20.0, 25.0, 30.0, 40.0, 60.0, and 80.0 mm h−1.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
4. Discussion
The statistical analysis of VPR characteristics above helped us to understand the vertical structural features and microphysical and dynamical processes of precipitation in the eastern downstream region of the TP. In the study region, the detection capability and QPE accuracy of the ground-based CINRAD network are limited, especially for low levels, because of the beam blockage associated with the complex terrain and the beam broadening far away from radar sites. The VPRs from spaceborne measurements are less affected by those factors and usually reveal the storm structures with much higher vertical resolution than that from ground-based weather radars. This space-based VPR analysis thus provides useful information for VPR correction methods that could better estimate the near-surface rainfall by combining spaceborne radar and CINRAD network data if the spatially and temporally representative VPRs are known for the different precipitation types. Certain research work aimed at improving ground-based radar QPEs with real-time/climatological PR data in the U.S. mountainous region (defined as the VPR-IE/CVPR-IE method) has been reported in recent years (e.g., Cao et al. 2013; Wen et al. 2013, 2016).
The ground radar reflectivity observed at a height above/within the BB can be related to the surface reflectivity and used to estimate the surface rainfall rates after applying the conversion method from Ku band to S band (Kirstetter et al. 2010; Cao et al. 2013). The establishment of the representative climatological VPRs (CPRs) of different seasons and regions on the basis of spaceborne radar observations can be used to merge data with the VPR profiles from ground weather radar in complex-terrain conditions. Figure 13 shows the CPRs derived from TRMM PR data for stratiform precipitation in the study region. It gives the correction ratios (dB) for different rainfall intensities in spring, summer, autumn, and winter that were derived through sorting and averaging the VPRs with different near-surface reflectivity. The point located at 2 km below the freezing level was assumed to have near-surface rainfall and was chosen as the reference point. Five curves for every season gave the correction numbers, which were added to the ground-based weather radar observations. For example, the correction ratio was about −7 dB at 2 km for the 30-dBZ curve in spring (Fig. 13). If the ground-based weather radar measures 20 dBZ at 2 km above the freezing level, the near-surface reflectivity can be considered as 27 dBZ by adding 7 dB to the measured 20 dBZ. Thus, the ground-based weather radar measurements in the ice layer can be corrected to the rain region. Moreover, we can see in Figs. 4–7 that the space-based VPR profiles were often not available at the lower layers close to the surface because of ground clutter. So to avoid the effect of the ground clutter on the near-surface VPR, values below 1 km from the reference layer are better set as a constant. This illustrates why such data could be helpful to improve single ground-based weather radar QPE products. Because of the variable vertical structure and much more comprehensive characteristics of convective precipitation, however, it is hard to establish representative CPRs for convective precipitation. The method is currently only suitable for stratiform precipitation systems.
The S-band climatological VPRs (i.e., correction ratio in decibels) for the type of stratiform precipitation in spring, summer, autumn, and winter. The x axis denotes the difference (dB) between radar reflectivity at the given height and the reference height. The y axis indicates the height relative to the freezing level (0°C isotherm). The solid lines are calculated from the VPRs, which have been sorted by near-surface reflectivity of 20, 25, 30, 35, and 40 dBZ.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
The S-band climatological VPRs (i.e., correction ratio in decibels) for the type of stratiform precipitation in spring, summer, autumn, and winter. The x axis denotes the difference (dB) between radar reflectivity at the given height and the reference height. The y axis indicates the height relative to the freezing level (0°C isotherm). The solid lines are calculated from the VPRs, which have been sorted by near-surface reflectivity of 20, 25, 30, 35, and 40 dBZ.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
The S-band climatological VPRs (i.e., correction ratio in decibels) for the type of stratiform precipitation in spring, summer, autumn, and winter. The x axis denotes the difference (dB) between radar reflectivity at the given height and the reference height. The y axis indicates the height relative to the freezing level (0°C isotherm). The solid lines are calculated from the VPRs, which have been sorted by near-surface reflectivity of 20, 25, 30, 35, and 40 dBZ.
Citation: Journal of Applied Meteorology and Climatology 56, 8; 10.1175/JAMC-D-16-0382.1
On the other hand, any kind of observation will not be perfect. Some reports have showed the limitations of TRMM PR in improving QPE products when the terrain is complex (Barros et al. 2000; Lang and Barros 2002; Barros and Tao 2008; Prat and Barros 2010; Rasmussen et al. 2013; Duan et al. 2015). All of this research focused more on the retrieved rainfall products of TRMM PR than on the echo reflectivity. Different from the echo reflectivity, retrieved rainfall products always introduce more retrieval errors (errors in the assumed drop size distribution, incorrect physical assumptions like freezing-level height and hydrometeor temperatures, possible contamination by surface backscatter, poor reliability in the physical basis of the stratiform–convective classification, etc.) in addition to the inevitable nonuniform beam-filling effects and attenuation. Moreover, the low precipitation amounts and the low number of pixels associated with precipitation in the winter season (shown in Figs. 2 and 3) may be related not only to the environmental conditions but also to the inability of the TRMM PR to detect light rain and snowfall because of its low sensitivity (18 dBZ). All of these errors will make it hard to improve the ground-based weather radar QPE in complex terrain by applying spaceborne radar observations.
5. Summary and conclusions
This study investigated the variability of the vertical structures of precipitation at CY in the eastern downstream region of the TP, which is often impacted by disastrous floods and mudslides because of its special climate and complex terrain. The major conclusions are briefly summarized below.
Precipitation type is an important influencing factor on the VPRs. In particular, large differences from the ice region to the rainy region can be observed for convective and stratiform precipitation because of their different microphysical and dynamical processes. In the study area, convective precipitation occurs with less probability than stratiform precipitation, and nearly 60% of the latter has an unobvious BB feature. A total of 90% of the BB peak reflectivity of the stratiform precipitation was less than 32 dBZ, and 40% of the maximum reflectivity of convective precipitation exceeded 35 dBZ. The intensity of surface rainfall rates also depended on the shapes of the VPRs. For the stratiform precipitation, the reductions in reflectivity above the melting layer were much weaker in the light surface rainfall rates than were those in moderate and heavy rainfall rates because of their different ice/snow aggregation rates. Hydrometeors aggregated much more slowly in winter because of the low temperatures and moisture. When the rainfall intensity was less than 5 mm h−1, hydrometeor conversion of the whole vertical layers contributed to the near-surface rainfall, whereas if the intensity reached more than 5 mm h−1, the aggregation from the BB peak to 3 km above the melting layer mainly determined the surface rainfall.
The structures of VPRs were similar between spring and autumn. Both the storm-top and BB-peak heights increased during the summer months. When the surface rainfall intensity was less than 4 mm h−1 in the other seasons as well as in winter, higher BB peaks were associated with heavier surface rainfall. Both the heights of storm tops and BB peaks varied with the underlying terrain because the heating of the surface and topography-induced updrafts affected the VPR shapes. Most stratiform precipitation in the CY was weaker than 2 mm h−1. The convective type was weaker and shallower than that in both the plain region in China (Cao and Qi 2014) and the mountainous region in the United States reported in Cao et al. (2013). The VPRs showed convective vertical structural features when the rainfall intensity was large. It is easy to infer that weak convective systems often occur in the study region. In addition, it is apparent that the TRMM PR cloud-type algorithms still encounter some challenges in the mountainous region around the TP.
All of the statistical analyses of VPR characteristics suggest that a typical representative VPR model can be created on the basis of the integration of normalized VPR shape for a given area and the precipitation type. Similar to the research in a mountainous area in the United States (Cao et al. 2013; Wen et al. 2013) and a plains area in China (Cao and Qi 2014), there is an urgent need to apply the spaceborne representative CPRs to improve the single ground-based radar QPE products in the mountainous region around the TP.
We mention also that the current CPRs and adherent correction method are based on the TRMM satellite, for which the coverage is between 36°N and 36°S. The Global Precipitation Measurement (GPM) satellite is also known as the post-TRMM satellite, and it carries a Ku/Ka-band dual-frequency radar with the advantage of wide coverage (68°N–68°S); this satellite observes the detailed structures of clouds, precipitation, and hydrometeor particle phases (Schwaller and Morris 2011; Tang et al. 2015). Thus, it will better capture the microphysical processes and will provide more insightful information on storm vertical structures, which potentially will improve the representative CPRs, especially for solid precipitation on and around the TP. The detailed methods of applying real-time or CPRs from TRMM PR and GPM dual-frequency precipitation radar observations to improve the ground-based CINRAD QPE products in the mountainous region around the TP will be addressed in a follow-on paper.
Acknowledgments
This study was sponsored by the National Natural Science Fund (91437214) and the Third Tibetan Plateau Atmospheric Scientific Experiment (GYHY201406001). The authors are thankful for the support from Jiafeng Zheng, Yabin Gou, and Qin Zhou. We are also grateful to the NASA TRMM and appreciate the NASA scientists and engineers who have made the TRMM PR data available.
REFERENCES
Awaka, J., T. Iguchi, and K. Okamoto, 2009: TRMM PR standard algorithm 2A23 and its performance on bright band detection. J. Meteor. Soc. Japan, 87A, 31–57, doi:10.2151/jmsj.87A.31.
Barros, A. P., and K. Tao, 2008: A space-filling algorithm to extrapolate narrow-swath instantaneous TRMM microwave rain-rate estimates using thermal IR imagery. J. Atmos. Oceanic Technol., 25, 1901–1920, doi:10.1175/2008JTECHA1019.1.
Barros, A. P., M. Joshi, J. Putkonen, and D. W. Burbank, 2000: A study of the 1999 monsoon rainfall in a mountainous region in central Nepal using TRMM products and rain gauge observations. Geophys. Res. Lett., 27, 3683–3686, doi:10.1029/2000GL011827.
Bellon, A., G.-W. Lee, and I. Zawadzki, 2005: Error statistics of VPR corrections in stratiform precipitation. J. Appl. Meteor., 44, 998–1015, doi:10.1175/JAM2253.1.
Cao, Q., and Y. Qi, 2014: The variability of vertical structure of precipitation in Huaihe River basin of China: Implications from long-term spaceborne observations with TRMM Precipitation Radar. Water Resour. Res., 50, 3690–3705, doi:10.1002/2013WR014555.
Cao, Q., Y. Hong, J. J. Gourley, Y. Qi, J. Zhang, Y. Wen, and P. E. Kirstetter, 2013: Statistical and physical analysis of the vertical structure of precipitation in the mountainous west region of the United States using 11+ years of spaceborne observations from TRMM Precipitation Radar. J. Appl. Meteor. Climatol., 52, 408–424, doi:10.1175/JAMC-D-12-095.1.
Ceng, S., Y. Gong, and X. Lai, 2014: Impact of heat source over Qinghai-Xizang Plateau and its surrounding areas on rainfall in Sichuan-Chongqing basin in summer. Plateau Meteor., 33, 1182–1189, doi:10.7522/j.issn.1000-0534.2013.00122.
Duan, A. M., Y. M. Liu, and G. X. Wu, 2005: Heating status of the Tibetan Plateau from April to June and rainfall and atmospheric circulation anomaly over east Asia in midsummer. Sci. China, 48D, 250–257, doi:10.1360/02yd0510.
Duan, Y., A. M. Wilson, and A. P. Barros, 2015: Scoping a field experiment: Error diagnostics of TRMM Precipitation Radar estimates in complex terrain as a basis for IPHEx2014. Hydrol. Earth Syst. Sci., 19, 1501–1520, doi:10.5194/hess-19-1501-2015.
Fabry, F., and I. Zawadzki, 1995: Long-term radar observations of the melting layer of precipitation and their interpretation. J. Atmos. Sci., 52, 838–851, doi:10.1175/1520-0469(1995)052<0838:LTROOT>2.0.CO;2.
Fu, Y., J. Feng, H. Zhu, R. Li, and D. Liu, 2006: Precipitation structures of a thermal convective system happened in the central western subtropical Pacific anticyclone (in Chinese). Acta Meteor. Sin., 20, 232–243.
Gabella, M., E. Morin, and R. Notarpietro, 2011: Using TRMM spaceborne radar as a reference for compensating ground-based radar range degradation: Methodology verification based on rain gauges in Israel. J. Geophys. Res., 116, D02114, doi:10.1029/2010JD014496.
Germann, U., and J. Joss, 2002: Mesobeta profile to extrapolate radar precipitation measurements above the Alps to the ground level. J. Appl. Meteor., 41, 542–557, doi:10.1175/1520-0450(2002)041<0542:MPTERP>2.0.CO;2.
Gou, Y. B., L. P. Liu, D. Wang, L. Z. Zhong, and C. Chen, 2015: Evaluation and analysis of the Z–R storm-grouping relationships fitting scheme based on storm identification (in Chinese). Torrential Rain Disasters, 34, 1–8.
Iguchi, T., T. Kozu, J. Kwiatkowski, R. Meneghini, J. Awaka, and K. Okamoto, 2009: Uncertainties in the rain profiling algorithm for the TRMM Precipitation Radar. J. Meteor. Soc. Japan, 87A, 1–30, doi:10.2151/jmsj.87A.1.
Kawanishi, T., and Coauthors, 2000: TRMM Precipitation Radar. Adv. Space Res., 25, 969–972, doi:10.1016/S0273-1177(99)00932-1.
Kirstetter, P.-E., H. Andrieu, G. Delrieu, and B. Boudevillain, 2010: Identification of vertical profiles of reflectivity for correction of volumetric radar data using rainfall classification. J. Appl. Meteor. Climatol., 49, 2167–2180, doi:10.1175/2010JAMC2369.1.
Kitchen, M., R. Brown, and A. G. Davies, 1994: Real-time correction of weather radar data for the effects of bright band, range and orographic growth in widespread precipitation. Quart. J. Roy. Meteor. Soc., 120, 1231–1254, doi:10.1002/qj.49712051906.
Kozu, T., and Coauthors, 2001: Development of Precipitation Radar on board the Tropical Rainfall Measuring Mission (TRMM) satellite. IEEE Trans. Geosci. Remote Sens., 39, 102–116, doi:10.1109/36.898669.
Lang, T. J., and A. P. Barros, 2002: An investigation of the onsets of the 1999 and 2000 monsoons in central Nepal. Mon. Wea. Rev., 130, 1299–1316, doi:10.1175/1520-0493(2002)130<1299:AIOTOO>2.0.CO;2.
Liu, L. P., L. L. Wu, and Y. M. Yang, 2007: Development of fuzzy-logical two-step ground clutter detection algorithm (in Chinese). Acta Meteor. Sin., 65, 252–260.
Liu, P., Y. F. Fu, S. Feng, A. Q. Cao, Y. J. Yang, and T. Y. Li, 2010: A comparison of the precipitation from rain gauge observations with from TRMM PR measurements in the southern China (in Chinese). Acta Meteor. Sin., 68, 822–835.
Maddox, R. A., J. Zhang, J. J. Gourley, and K. W. Howard, 2002: Weather radar coverage over the contiguous United States. Wea. Forecasting, 17, 927–934, doi:10.1175/1520-0434(2002)017<0927:WRCOTC>2.0.CO;2.
Meneghini, R., J. A. Jones, T. Iguchi, K. Okamoto, and J. Kwiatkowski, 2004: A hybrid surface reference technique and its application to the TRMM Precipitation Radar. J. Atmos. Oceanic Technol., 21, 1645–1658, doi:10.1175/JTECH1664.1.
Prat, O. P., and A. P. Barros, 2010: Assessing satellite-based precipitation estimates in the southern Appalachian Mountains using rain gauges and TRMM PR. Adv. Geosci., 25, 143–153, doi:10.5194/adgeo-25-143-2010.
Qi, Y., and J. Zhang, 2013: Correction of radar QPE errors associated with partial brightband observations near the ground. J. Hydrometeor., 14, 1933–1943, doi:10.1175/JHM-D-13-040.1.
Qi, Y., J. Zhang, Q. Cao, Y. Hong, and X. Hu, 2013: Correction of radar QPE errors for nonuniform VPRs in mesoscale convective systems using TRMM observations. J. Hydrometeor., 14, 1672–1682, doi:10.1175/JHM-D-12-0165.1.
Rasmussen, K. L., S. L. Choi, M. D. Zuluaga, and R. A. Houze, 2013: TRMM precipitation bias in extreme storms in South America. Geophys. Res. Lett., 40, 3457–3461, doi:10.1002/grl.50651.
Schwaller, M. R., and K. R. Morris, 2011: A ground validation network for the Global Precipitation Measurement mission. J. Atmos. Oceanic Technol., 28, 301–318, doi:10.1175/2010JTECHA1403.1.
Simpson, J., C. Kummerow, W. K. Tao, and R. F. Adler, 1996: On the Tropical Rainfall Measuring Mission (TRMM). Meteor. Atmos. Phys., 60, 19–36, doi:10.1007/BF01029783.
Smith, A. M., G. M. McFarquhar, R. M. Rauber, J. A. Grim, M. S. Timlin, B. F. Jewett, and D. P. Jorgensen, 2009: Microphysical and thermodynamic structure and evolution of the trailing stratiform regions of mesoscale convective systems during BAMEX. Part I: Observations. Mon. Wea. Rev., 137, 1165–1185, doi:10.1175/2008MWR2504.1.
Takahashi, N., H. Kuroiwa, and T. Kawanishi, 2003: Four year result of external calibration for Precipitation Radar (PR) of the Tropical Rainfall Measuring Mission (TRMM) satellite. IEEE Trans. Geosci. Remote Sens., 41, 2398–2403, doi:10.1109/TGRS.2003.817180.
Tang, G. Q., W. Wan, Z. Y. Zeng, X. L. Guo, N. Li, D. Long, and Y. Hong, 2015: An overview of the Global Precipitation Measurement (GPM) mission and its latest development. Remote Sens. Technol. Appl., 30, 607–615.
Vignal, B., G. Galli, J. Joss, and U. Germann, 2000: Three methods to determine profiles of reflectivity from volumetric radar data to correct precipitation estimates. J. Appl. Meteor., 39, 1715–1726, doi:10.1175/1520-0450-39.10.1715.
Wang, G. L., L. P. Liu, and Y. Y. Ding, 2012: Improvement of radar quantitative precipitation estimation based on real-time adjustments to Z–R relationships and inverse distance weighting correction schemes (in Chinese). Adv. Atmos. Sci., 29, 575–584, doi:10.1007/s00376-011-1139-8.
Wang, H. Y., L. P. Liu, G. L. Wang, W. Zhuang, Z. Q. Zhang, and X. L. Chen, 2009: Development and application of the Doppler weather radar 3-D digital mosaic system (in Chinese). J. Appl. Meteor. Sci., 20, 214–224.
Wang, S. D., C. Pei, Z. M. Guo, and N. Shao, 2011: Evaluation on Chinese next generation radars coverage and terrain blockage based on SRTM data (in Chinese). Climatic Environ. Res., 16, 459–468.
Wei, N., Y. F. Gong, and J. H. He, 2009: Structural variation of an atmospheric heat source over Qinghai Xizang Plateau and its influence on precipitation in northwest China. Adv. Atmos. Sci., 26, 1027–1041, doi:10.1007/s00376-009-7207-7.
Wen, Y., Q. Cao, P.-E. Kirstetter, Y. Hong, J. J. Gourley, J. Zhang, G. Zhang, and B. Yong, 2013: Incorporating NASA spaceborne radar data into NOAA National Mosaic QPE system for improved precipitation measurement: A physically based VPR identification and method. J. Hydrometeor., 14, 1293–1307, doi:10.1175/JHM-D-12-0106.1.
Wen, Y., P. Kirstetter, Y. Hong, J. J. Gourley, Q. Cao, J. Zhang, Z. Flamig, and X. Xue, 2016: Evaluation of a method to enhance real-time, ground radar–based rainfall estimates using climatological profiles of reflectivity from space. J. Hydrometeor., 17, 761–775, doi:10.1175/JHM-D-15-0062.1.
Williams, C. R., A. B. White, K. S. Gage, and F. M. Ralph, 2007: Vertical structure of precipitation and related microphysics observed by NOAA profilers and TRMM during NAME 2004. J. Climate, 20, 1693–1712, doi:10.1175/JCLI4102.1.
Willis, P. T., and A. J. Heymsfield, 1989: Structure of the melting layer in mesoscale convective system stratiform precipitation. J. Atmos. Sci., 46, 2008–2025, doi:10.1175/1520-0469(1989)046<2008:SOTMLI>2.0.CO;2.
Wu, G., and Y. Zhang, 1998: Tibetan Plateau forcing and the timing of the monsoon onset over South Asia and the South China Sea. Mon. Wea. Rev., 126, 913–927, doi:10.1175/1520-0493(1998)126<0913:TPFATT>2.0.CO;2.
Wu, G., Y. Liu, B. He, Q. Bao, A. Duan, and F.-F. Jin, 2012: Thermal controls on the Asian summer monsoon. Sci. Rep., 2, 404, doi:10.1038/srep00404.
Xiao, Y. J., and L. P. Liu, 2006: Study of methods for interpolating data from weather radar network to 3-D grid and mosaics (in Chinese). Acta Meteor. Sin., 64, 647–657.
Xiao, Y. J., L. P. Liu, and H. P. Yang, 2007: A contrast analysis of synchronous observation from regional radar network (in Chinese). Acta Meteor. Sin., 65, 919–927.
Xu, X. D., C. Lu, X. Shi, and S. T. Gao, 2008: World water tower: An atmospheric perspective. Geophys. Res. Lett., 35, 525–530, doi:10.1029/2008GL035867.
Xu, X. D., T. L. Zhao, C. G. Lu, Y. D. Guo, B. Chen, R. X. Liu, Y. Q. Li, and X. H. Shi, 2014: An important mechanism sustaining the atmospheric “water tower” over the Tibetan Plateau. Atmos. Chem. Phys., 14, 11 287–11 295, doi:10.5194/acp-14-11287-2014.
Yang, H. P., P. Y. Zhang, M. H. Cheng, B. Li, Y. Xiong, Y. C. Gao, and D. R. Chen, 2009: The valid mosaic data region of the CINRAD network (in Chinese). J. Appl. Meteor. Sci., 20, 47–55.
Zhang, J., and Y. Qi, 2010: A real-time algorithm for the correction of brightband effects in radar-derived QPE. J. Hydrometeor., 11, 1157–1171, doi:10.1175/2010JHM1201.1.
Zhuang, W., L. P. Liu, and Z. Q. Hu, 2013: Application of bright band identification and correction in radar-based QPE over Tibetan Plateau (in Chinese). Meteor. Mon., 39, 1004–1013.
