Characteristics of Monsoon Rainfall around the Himalayas Revealed by TRMM Precipitation Radar

B. C. Bhatt Hydrospheric Atmospheric Research Center, Nagoya University, Nagoya, Japan

Search for other papers by B. C. Bhatt in
Current site
Google Scholar
PubMed
Close
and
K. Nakamura Hydrospheric Atmospheric Research Center, Nagoya University, Nagoya, Japan

Search for other papers by K. Nakamura in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

The climatological features of the diurnal cycle and its spatial and temporal variability are investigated around the Himalayas using hourly, 0.05° × 0.05° grid, near-surface rainfall data from the Precipitation Radar (PR) aboard the Tropical Rainfall Measuring Mission (TRMM) satellite during June–July–August (JJA) of 1998–2002. Though sampling errors inherent to TRMM PR measurements around the Himalayas could influence results, PR-observed precipitation features show agreement with previous studies in this region.

The analysis of precipitation characteristics presented here is based on two rain-rate thresholds: (a) light rain rate (≤5 mm h−1), and (b) moderate to heavy rain rate (>5 mm h−1). The results suggest that afternoon to evening precipitation is noticed as embedded convection within a large region of light rain over the south-facing slopes of the Himalayas. The moderate to heavy conditional rain rate exhibits a relatively stronger diurnal cycle of precipitation in this region. However, this may be biased because of sampling. Almost all the Tibetan Plateau shows light rain activity.

The Tibetan Plateau and northern Indian subcontinent regions are characterized by daytime maximum precipitation. From the analysis of near-surface rainfall over the finescale topography, it is observed that daytime (1200–1800 LT) precipitation is concentrated over the ridges and strong ridge–valley gradients with rain appearing over the south-facing slopes of the Himalayas. During midnight–early morning, intense rainfall concentrates over the ridges as well as in river valleys. Precipitation broadening and movement are noticed during this time period.

Corresponding author address: Bhuwan Chandra Bhatt, Hydrospheric Atmospheric Research Center, Nagoya University, Furo-cho, Chikusa-ku, Nagoya 464-8601, Japan. Email: bhatt@ihas.nagoya-u.ac.jp

Abstract

The climatological features of the diurnal cycle and its spatial and temporal variability are investigated around the Himalayas using hourly, 0.05° × 0.05° grid, near-surface rainfall data from the Precipitation Radar (PR) aboard the Tropical Rainfall Measuring Mission (TRMM) satellite during June–July–August (JJA) of 1998–2002. Though sampling errors inherent to TRMM PR measurements around the Himalayas could influence results, PR-observed precipitation features show agreement with previous studies in this region.

The analysis of precipitation characteristics presented here is based on two rain-rate thresholds: (a) light rain rate (≤5 mm h−1), and (b) moderate to heavy rain rate (>5 mm h−1). The results suggest that afternoon to evening precipitation is noticed as embedded convection within a large region of light rain over the south-facing slopes of the Himalayas. The moderate to heavy conditional rain rate exhibits a relatively stronger diurnal cycle of precipitation in this region. However, this may be biased because of sampling. Almost all the Tibetan Plateau shows light rain activity.

The Tibetan Plateau and northern Indian subcontinent regions are characterized by daytime maximum precipitation. From the analysis of near-surface rainfall over the finescale topography, it is observed that daytime (1200–1800 LT) precipitation is concentrated over the ridges and strong ridge–valley gradients with rain appearing over the south-facing slopes of the Himalayas. During midnight–early morning, intense rainfall concentrates over the ridges as well as in river valleys. Precipitation broadening and movement are noticed during this time period.

Corresponding author address: Bhuwan Chandra Bhatt, Hydrospheric Atmospheric Research Center, Nagoya University, Furo-cho, Chikusa-ku, Nagoya 464-8601, Japan. Email: bhatt@ihas.nagoya-u.ac.jp

1. Introduction

The Himalayan range is one of the major highly elevated land areas in the world. The unique uplifting of the Himalayas and the Tibetan Plateau modulate the regional monsoon climate. In the recent years, attention has been drawn over this landscape regarding climate change. Some elements of atmospheric circulation, for example, monsoon depressions, westerly disturbances, heating over the Tibetan Plateau, etc., are important to the climate of this region. The climate of south Asia is dominated by the summer monsoon circulation system. The Himalayas are prone to abundant rainfall and flooding episodes during the summer monsoon, particularly over south-facing slopes and in the Gangetic Plains of India. Landslides are one of the most devastating hazards of monsoon rainfall. The agricultural importance of this rainfall is also substantial, as the region’s population depends greatly on monsoon rains.

Several studies have focussed on large-scale modes of variability in the Asian summer monsoon (e.g., Ye 1981; Chen et al. 1985; Li and Yanai 1996; Krishnamurti and Kishtawal 2000). Most of these studies show that convection in the Tibetan Plateau and its surroundings affects the circulation and sustains the monsoon through the release of latent heat. From a small-scale perspective, there are few ground- as well as satellite-based investigations of the monsoon region that can be found in the literature. Previous studies have not focused on the south-facing slopes of the Himalayas where vigorous interactions among climate, geomorphology, and hydrology take place (Burbank et al. 2003). Few meteorological observations have been carried out at specific locations over the south-facing slopes of the Himalayas over the past few years. In recent years, a few studies (e.g., Shrestha 2000; Barros et al. 2000) on rainfall at some locations over the south-facing slopes of the Himalayas have noted large rain totals on the order of 300–400 cm yr−1. Necessarily, such large rain totals contribute to the latent heating of the regional atmosphere. However, these studies were from a set of fixed, sparsely distributed rain gauges. The complexity of the orography, as well as financial constraints, were behind this poor arrangement of the rainfall monitoring network. For the cloud field, one of the most recent studies on the small-scale aspect over the south-facing slopes of the Himalayas was carried out by Barros and Lang (2003). However, cloud field studies can only be considered as a proxy measurement of rainfall. The prevalent area-averaged techniques for estimating precipitation over this landscape, which consists of a curved barrier comprising many small-scale ridges and valleys, are erroneous. This is because most of the rain gauges are placed in river valleys, which lack sufficient information on ridge precipitation. For the numerical model studies, this topography is too steep and variable to be fully represented in a numerical grid system. Hence, there exists a large estimation error for precipitation in this region. To address these concerns, precipitation data with high spatial and temporal resolution is in great demand over this remote and rugged orography.

One of the most distinct features of precipitation over land is the diurnal variation. There have been numerous studies of diurnal precipitation characteristics using various sensor data over South Asia. Nesbitt and Zipser (2003) examined the large-scale characteristics of the diurnal cycle using Precipitation Radar–Tropical Rainfall Measuring Mission (TRMM) Microwave Imager (PR–TMI) combined precipitation features over 10° grids for a 3-yr period. They noted afternoon and (before) midnight maximum rainfall over this region. Some of the researchers (e.g., Ueno et al. 2001) used rain gauge data of the Khumbu region in the eastern Nepal Himalayas and found an after-midnight peak in rainfall. Recent studies using rain gauge data have shown a nocturnal peak in rainfall over the Himalayan range of central Nepal during monsoon onset (Barros and Lang 2003). Such nocturnal behavior was hypothesized as a result of maximized atmospheric instability just before rainfall peak time as revealed by collocated rain gauge/radiosonde observations. Some studies of this region have relied on visible/infrared observations. For example, Murakami (1983) used Geostationary Meteorological Satellite data to examine the phase and amplitude of the diurnal convection activity. He found that during the summer monsoon months, the diurnal convection over the Tibetan Plateau is enhanced during the afternoon hours and suppressed during the early-morning hours. He noted heavy convection in the early-morning hours and suppressed convection in the afternoon hours over the eastern Himalayan foothills. Ohsawa et al. (2001) used the same sensor data and rain gauge data on the Indo–China peninsula during the monsoon season and noted late-night–early-morning maxima in rainfall all along the eastern Himalayan foothills. Kurosaki and Kimura (2002) analyzed the Geostationary Meteorological Satellite (visible) data and noted that low-level clouds were predominant over the south-facing slopes of the eastern Himalayas, both in premonsoon and monsoon periods in the daylight hours. In their research, the premonsoon and monsoon seasons were defined as 1–31 May and 27 June–31 July of 1998, respectively. Premonsoon cloud activity showed strong diurnal variation as compared to the monsoon season both over the Tibetan Plateau and the south-facing slopes of the eastern Himalayas. A recent study (Barros et al. 2004), using high-resolution infrared satellite imagery Meteosat-5 IR data over the south-facing slopes of the Himalayas, has shown high degrees of cloudiness on the ridges at night (2000–0700 LT) centered at 0000–0300 LT during the monsoon season. Although cloud field studies with Meteosat-5 IR were conducted over wide areas, a mechanism similar to that hypothesized for nocturnal rainfall over the Himalayan range of central Nepal (Barros et al. 2004) was speculated. It stated that high nocturnal rainfall on the Himalayan ridges results from the synergistic interaction of orographic gravity waves and atmospheric instability.

The nocturnal peak of rainfall over the Himalayan mountainous region is a unique behavior in the sense that most land areas around the globe are characterized by a daytime maximum of rainfall. The daytime maximum can be linked to surface heating. The mechanism of midnight–early-morning rainfall still remains difficult to explain. We hypothesize that strong meridional circulation in the lower atmosphere may be important for midnight–early-morning rainfall over the south-facing slopes of the Himalayas. When mesoscale depressions originating from the Bay of Bengal interact with the easterly vertical shear forced by the mountains, strong near-surface flow and weaker return flow in the midlevel probably establish cyclonic circulation, mostly meridional in nature. Most of the past studies in this region were dependent on passive sensor data, and their technique often averaged over a large spatial domain that may have ignored local features of the diurnal cycle. Rainfall, particularly on a daily time scale, typically displays complex spatial patterns. These are often related to topography and prevailing wind direction. Some locations over the south-facing slopes of the Himalayas are prone to large rain totals as stated in previous studies. A better understanding of precipitation characteristics over the widespread range of the Himalayas is extremely important for improving our understanding of the regional monsoon climate.

TRMM (Kummerow et al. 1998) carries a unique suite of instruments, including a PR and TMI. The PR is the first spaceborne radar capable of measuring the detailed, three-dimensional structure of precipitation. The PR’s reflectivity–rain-rate algorithm (Iguchi et al. 2000) offers more direct remote sensing estimates of near-surface rainfall than other methods (IR or microwave brightness temperature). The PR has provided a nearly homogenous dataset covering the entire region, irrespective of the leeward side, snowcapped tops, or deep gorges over this landscape. But temporal variability of precipitation by PR could be observed to only a certain extent due to its narrow swath of 215 km and low sampling frequency. Discrepancies in the PR-observed precipitation features may arise because of poor temporal samplings combined with the relatively large footprint of the PR (4.3 km) and the complex terrain in this region.

In this paper, we aim to examine monsoon precipitation features around the Himalayas by utilizing hourly 0.05° × 0.05° grid data from the TRMM PR. There are fewer possibilities to study temporal variability in detail from PR measurements especially at a small spatial scales (5 km) because of sampling. Aware of this sampling constraint, we make use of accumulation over 6 h of local time to estimate the horizontal distribution of the diurnal cycle of precipitation. The rain total and rain-conditioned rain-rate estimates are used to express spatial variability in precipitation. Time scales of variability in precipitation ranging from diurnal to interannual are addressed. A simple characterization of rainfall using rain-rate thresholds and climatological features of the diurnal cycle of rainfall during June–July–August (JJA) of 1998–2002 is developed.

2. Data and method

The data source for this study is the TRMM PR2A25 version 5.0 (Iguchi et al. 2000) during the period of 1998–2002. The PR2A25 is a data product that has vertical profiles of attenuation-corrected radar reflectivity factors and rainfall rates. The TRMM PR data volume is huge, that is, 250 Mbyte for one revolution. For convenience, a new subset was generated in order to cover from 18.0°–36.0°N, 58.0°–92.0°E. For our data processing, the “near-surface rainfall rate” contained in PR2A25 was accumulated and binned to hourly local times with 0.05° grid data. The grid size of 0.05° is nearly the size of a PR pixel. For all the grids, rain frequency, peak time of rain, total rain amount, and rain-conditioned rain rate1 were calculated for each season separately, as well as for the entire period.

The PR near-surface mean stratiform and convective daily rainfall amount histogram exhibits a mode around 130–160 mm day−1 (F. Furuzawa 2003, Nagoya University, personal communication) over the Himalayan region for 0.1° × 0.1° grid data during JJA. Below this range of 130–160 mm day−1, slight differences appear in distributions for both stratiform and convective rainfall. There is a maximum number of counts below this range of 130–160 mm day−1. In the histogram, convective shows a wider distribution than stratiform beyond this range. The threshold of ≤5 mm h−1 (120 mm day−1) was taken as light rain, and >5 mm h−1 was taken as moderate to heavy rain. Hence, both the light and moderate to heavy rain are mostly a mixture of stratiform and convective rainfall. We seek to answer the question of how well will our threshold method show rainfall characteristics including diurnal variations. In case studies from TRMM PR, Lang and Barros (2002) have shown the existence of embedded convection in a large region of light rain (<5 mm h−1) at the Marsyandi network over the central Nepal Himalayas during monsoon onset. It is interesting to know from the PR measurements whether these precipitation characteristics would hold in different regions of the Himalayas. Rain intensity is one of the more highly variable quantities, and the technique adopted here is expected to address intensity-based precipitation characteristics from TRMM PR observations. It is expected that such a technique has not been attempted so far.

Other than PR data, we used topography data and wind data. For topographic data, the GTOPO30 dataset of the U.S. Geological Survey was utilized, and wind field data were extracted from the National Centers for Environmental Prediction–National Center for Atmospheric Research (NCEP–NCAR) reanalysis (http://www.cdc.noaa.gov/cdc/data.ncep.Reanalysis).

3. Topography

The topography of the analysis domain is shown in Fig. 1. The elevation of the Tibetan Plateau is more than 4000 m MSL. One of the characteristics of the topography over the Tibetan Plateau is the large undulation in the north–south direction. The major mountain ranges and valleys are oriented in an east–west direction over the Tibetan Plateau. On the other hand, the Gangetic Plains in India is a low-elevated land in this domain. Smaller mountains also exist around the northern Indian subcontinent. The Himalayas, an approximately SE–NW trending mountain range, are located on the verge of 27.0°–32.0°N, 77.0°–92.0°E, in south Asia. East of 84.0°E, the mountains run roughly E–W, while to the west of 84.0°E the barrier is arranged ESE–WNW. The southern slope of the Himalayas is steep. Most of the river valleys are approximately oriented in the north–south direction and orography has large undulations in this region. The Hindkus Mountains lie to the west of the Himalayas and extend up north of Pakistan. The Himalayan range provides a huge barrier to the flow in this region. The summer monsoon depressions collide with them from the south-southeast direction and undergo strong orographic capture because of the favorable geometry of the Himalayas. Conversely, westerly disturbances encroach them from the west-southwest direction.

4. TRMM PR sampling

TRMM rotates the earth about 16 times a day in a low-inclination orbit. TRMM visits a region at a given hour of the day and its next observation at the same hour takes 23 days at the equator and 46 days at the highest latitude in its coverage. Because of its orbital characteristics, accumulation of data over a number of days (depending on the latitude of the area) is required to cover the entire diurnal cycle. It should be noted that at small spatial scales (4.3 km), TRMM PR has more intermittent samplings. Even when we utilize hourly PR data around the Himalayas over 5 yr (1998–2002), there exist sampling errors. Aware of this sampling constraint, we generally make use of accumulations over 6 h of local time in this analysis. Figure 2 shows the result from JJA of the 5-yr period (1998–2002) of PR sampling accumulated for periods from 3 to 12 h at 0.05° resolution. It is evident that 3- and 6-h accumulations provide a spatially even sampling pattern (see the area enclosed by an open ellipse in Fig. 2). The diurnal cycle of precipitation can be computed with acceptable statistical confidence.

5. Results and discussion

a. Interannual variability

To begin, we examine the spatial and temporal organization of near-surface rain for individual years. The rainfall totals in this region depend on the appearance of monsoon depressions in the Bay of Bengal and the trajectory taken by the depressions. According to the climatology, monsoon depressions from the Bay of Bengal travel in the west-northwest direction and the Himalayan range acts as a boundary barrier impeding the flow (e.g., Das 1987; Lang and Barros 2002). Influence of the western disturbances—that is, westerly upper-tropospheric synoptic-scale disturbances—cannot be neglected in this region, although they only contribute small rain totals. Generally, western disturbances cause precipitation during December–January–February (DJF) (Mani 1981). The influence of these two circulation systems (summer and winter monsoon circulation) is not evenly distributed over the Himalayas, with summer rainfall greater in the southeast and winter rainfall greatest in the northwest (Mani 1981).

Figure 3 shows yearly horizontal distributions of total rainfall (millimeters per day) from 1998 to 2002 for 0100–2400 LT. Refer to Fig. 1 for topographical as well as surface hydrological features. From the figures, it is clear that the southeastern part of the Tibetan Plateau shows more rain activity than the northwestern part. The Thar Desert area also exhibits less rain activity. Around the location of 35.0°N, 66.0°E, there appears smaller rain totals. High mountains with elevations ranging from 3500 to 4500 m MSL are present around this location. It should be mentioned that the spatial variation of precipitation is similar among individual years, though relatively less rain activity appeared during 2000. This is particularly true over the south-facing slopes of the Himalayas. This interannual variation is consistent with results shown by Lang and Barros (2002) in which they noticed less rain activity with depressed rainfall during the monsoon period of 2000. According to Lang and Barros (2002), the storm track was situated more southerly of the Himalayas during the monsoon period of 2000.

Precipitation features at smaller temporal scales for individual years are analyzed by examining the horizontal distributions of mean conditional rain rates (millimeters per hour) for four time periods a day (see Fig. 4 for 1200–1800 and 0000–0600 LT). The spatial variability in the observed diurnal cycle can be considered robust if only rain-conditioned overpasses for any single year are utilized. Over the Hindkus Mountains, intense rainfall is apparent between 1200 and 1800 LT in individual years. In 2000, less rain activity appears between 1200 and 1800 LT over the Tibetan Plateau. During the same year, weak conditional rain is apparent over the south-facing slopes of the Himalayas and over the Himalayan foothills, along the eastern Himalayan foothills in particular. The Narayani River watershed located in the central Nepal Himalayan foothills has similar rain patterns in individual years, that is, intense rainfall especially between 0000 and 0600 LT. Compared with the fine-resolution topography map, our yearly analysis shows more widespread distribution of rainfall on the ridges of the south-facing slopes of the Himalayas between 1200 and 1800 LT. Generally, evening rainfall (1800–2400 LT) appears localized near the layout of high mountains over the south-facing slopes of the Himalayas (not shown). The early-morning (0000–0600 LT) intense rainfall is more common over low lands or river basins over the south-facing slopes of the Himalayas. During late morning (0600–1200 LT), weak rain activity appears over the south-facing slopes of the Himalayas (not shown). As a whole, the relative spatial variations are noticed to be similar among individual years.

b. Monsoon rainfall

Next, we analyzed seasonal variations of the diurnal cycle of precipitation. We focused on the summer monsoon season (JJA). The rain-conditioned overpasses for a 1-h local time bin do not exceed 6. In this region, the increase in rain samples of PR during JJA is due to more frequent rain. The horizontal distributions of total rainfall (millimeters per day) summarized for 0100–2400 LT during JJA of 1998–2002 shows that the Bay of Bengal region receives larger rain totals (not shown) compared to other regions (e.g., Tibetan Plateau, northern Indian subcontinent) in the domain. The enhanced rainfall also appears over the south-facing slopes and adjacent foothill region of the Himalayas. The orography of this region is complex, and mountains do not form a uniform obstacle to flow. Hence, strong spatial gradients in rainfall are expected to occur around the Himalayas. A detailed investigation on spatial variability in precipitation in this region remains incomplete due to the lack of uniformity in terrain compared to the PR footprint (4.3 km). However, we investigated spatial variability in rain totals from PR observations over the south-facing slopes of the Himalayas. This region was divided into two subregions: the middle Himalayas and high Himalayas. The middle Himalayas span from approximately 500 to 2000 m MSL and comprise successions of many small-scale ridges and valleys. On the other hand, the high Himalayas rise to 8848 m MSL succeeding the middle Himalayas with steep slopes. These selected subregions [77°–90°E; run nearly SE–NW with approximately a 60-km width (north–south)] belong to the climatic regions (Shrestha et al. 2000) and roughly correspond with the geometry of the Himalayas. We performed meridional averaging in mean daily total rainfall during JJA over these two subregions. The result is shown in Fig. 5. Meridionally averaged topographic data are also shown, and approximate positions of river basins are shown by arrows. There appears some evidence of rainfall modulation caused by small but complex terrains (in the middle Himalayas). The averaged rain amount is always higher throughout the range in the middle Himalayas. This feature can be seen in Fig. 5 (bottom). The two series (rain-total series and mountain series) exhibit association between landform and rain totals. For example, there are larger rain totals in the spatial location of lowlands as shown in Fig. 5 (top and middle), while generally, the smaller rain totals coincide with the spatial location of high mountains. The eastern Himalayas exhibit a greater difference in rain totals between the two subregions in comparison with the western Himalayas. The spatial arrangement of the topography is very complex, and altitudinal gradients are stronger in the eastern Himalayan region.

c. Diurnal variation

We next present the horizontal distribution of the diurnal variation of precipitation. Examples of the JJA total rainfall (millimeters per day) from 5 yr (1998–2002) of PR observations for 6-h periods are shown in Fig. 6. Over the south-facing slopes of the Himalayas, widespread rain activity appears between 1200 and 1800 LT. The early-morning (0000–0600 LT) rain totals appear more common over river basins. However, the ridge locations also exhibit similar behavior during this time period. The late-morning rain (0600–1200 LT) appears dominant over the foothill region of the Himalayas. There appear to be disagreement with daytime (0900–1500 LT) cloud field studies by Kurosaki and Kimura (2002) in terms of spatial distribution of rainfall and cloudiness, particularly above 3000 m MSL over the south-facing slopes of the eastern Himalayas. Our results show that large rain totals appear below 3000 m MSL over the south-facing slopes of the Himalayas during the monsoon season. According to Kurosaki and Kimura (2002), greater cloud cover frequency (>75%) of low clouds was observed above 3000 m MSL over the south-facing slopes of the eastern Himalayas during the premonsoon and monsoon seasons. One reason for the disagreement may have been the existence of nonprecipitating-type clouds along the tallest ridges in the eastern Himalayas. This is because their analysis did not discriminate precipitating clouds from nonprecipitating clouds. Despite this, the spatial variability between the reported cloud field and the PR-observed rainfall distribution are in agreement for the south-facing slopes and the Tibetan Plateau region. The conditional rain frequency (overpasses with conditional rain) over the eastern Himalayan region is higher than over the western Himalayan region (not shown). The PR observations show that strong/weak conditional rain frequency exists mornings/afternoons along the eastern Himalayan foothill region. This inference is consistent with results shown by Kurosaki and Kimura (2002). Overall, the daytime (1200–1800 LT) widespread distribution of precipitation over the south-facing slopes of the Himalayas conforms well with the reported cloud field (e.g., high percentage in cloud coverage) by Kurosaki and Kimura (2002). It is noticed that larger rain totals appear over the Narayani River watershed in central Nepal, along the eastern Himalayan foothills and over the Karakoram Valley region of the Hindkus Mountains during early morning (0000–0600 LT). Ohsawa et al. (2001) also noticed convective activity along the eastern Himalayan foothills in the morning during monsoon period.

Widespread distribution of precipitation is apparent along the major mountain ranges over the Tibetan Plateau during evening hours (1800–2400 LT). There appears to be agreement with results by Kuwagata et al. (2001), who discussed horizontal transport of moisture by thermally induced circulations over regional-scale topography by a simple two-dimensional numerical model. Kuwagata et al. (2001) reported that water vapor is transported from valley areas to the mountainous areas during daytime and clouds often appear during evening along the major mountain ranges over the Tibetan Plateau. Krishnamurti and Kishtawal (2000) examined the diurnal mode of the Asian summer monsoon using data from Meteosat-5 and TMI. The Tibetan high and tropical easterly jet were noticed as prerequisites for the diurnal oscillation of monsoon in their studies. Our results show the concentration of large rain totals around the northern Indian subcontinent between 1800 and 2400 LT and over the north of the Bay of the Bengal between 0600 and 1200 LT. However, widespread distribution of precipitation appears between 1200 and 1800 LT around the northern Indian subcontinent. This is in agreement with TMI-derived results shown by Krishnamurti and Kishtawal (2000). The difference in timing of maximum precipitation over these two regions (e.g., Bay of Bengal and northern Indian subcontinent) appears to be related to continental-scale oscillations as noted by Krishnamurti and Kishtawal (2000). The PR observations have shown early-morning (0000–0600 LT) appearance of rainfall over the western Himalayan region. This inference is also consistent with results shown by Krishnamurti and Kishtawal (2000) for 0000 LT.

High-resolution plots of mean conditional rain rate over the south-facing slopes of the eastern and western Himalayas with the topographic map are shown in Figs. 7 and 8. In comparison with the topographic map, precipitation patterns depicted by PR show that mean conditional rainfall is confined to the ridges over the slope area in the daytime (1200–1800 LT) and spatial variability in rain patterns are consistent with orography. There appears to be concentrations of intense rainfall on the ridges with 20–30-km spatial scale. There seems to be some evidence of ridge–valley gradients in rainfall over the south-facing slopes of the Himalayas during daytime. The rain totals also exhibit similar behavior (not shown). This is true, for example, along the eastern Himalayan slope region at 27.1°N, 91.0°E; at 27.2°N, 88.2°E; in the Koshi River basin area; and at 28.0°N, 85.5°E in the Narayani River watershed. The conditional rain rate on the ridge locations is stronger than adjacent lowlands. Likewise, the western Himalayan slope area also exhibits this trend in ridge–valley gradients. Relatively weak conditional rain rate is noticed at 28.0°N, 82.5°E, near the Mahakali River; near the Ganga River at 30.0°N, 79.0°E; in between the Ganga River and the Yamuna River; and to the west of the Yamuna River during daytime. The adjacent ridge locations show stronger rainfall, which can be seen in Figs. 7 and 8. A feature worthy of noting is evidence of a correlation in mean conditional rain rate and elevation. It is reasonable to mention that daytime thermally induced circulations over complex terrain transport water vapor from valley regions to the high ridges. The ridge areas, with elevations from nearly 2000–3500 m MSL, exert control on daytime precipitation over the south-facing slopes of the Himalayas. Early-morning (0000–0600 LT) rainfall concentrates on ridges as well as over the lowlands over the south-facing slopes of the Himalayas. But the Koshi River watershed and surroundings in the eastern Himalayan slope area exhibit weak conditional rain rate during early morning. The Mahakali River, Karnali River, and the Yamuna River in the slope region of the south-facing slopes of the western Himalayas especially show a high mean conditional rain rate during this time period. Overall, the slope region of the western Himalayas shows more rain activity than the eastern Himalayan slope region during early morning. Strong meridional circulation in the lower atmosphere may be important for midnight–early-morning rainfall over the south-facing slopes of the Himalayas. We discuss this hypothesis later.

d. Diurnal variations of light and moderate to heavy rainfall

One principal feature in the horizontal distribution of precipitation is the difference between the south-facing slopes and adjacent foothill region of the Himalayas; that is, precipitation begins over the south-facing slopes during midnight–early morning while it appears in the adjacent foothill region during late morning. We investigated precipitation features at a smaller temporal scale (e.g., 3-h periods). However, 3-h accumulations provide reduced sampling in comparison to 6-h accumulations in this region. We seek to answer a question of whether the observed light rain diurnal variability was typical.

Figure 9 represents JJA light mean conditional rain-rate images from observations over 5 yr (1998–2002) during 3-h intervals. The most remarkable feature of these images is the gradual development of convection in the Karakoram Valley, especially from 2100–2400 to 0600–0900 LT, with the spatial scale nearly that of the valley width. The valley region is highlighted in Fig. 9 (see the area enclosed by circle with an arrow on it). The early-morning maximum in this valley is probably produced by the convergence of downslope winds from nearby higher terrain, which agrees well with the results obtained by Kousky (1980) in the Rio Sao Francisco Valley in Brazil. The large-scale precipitation features associated with layout of the mountain peaks over the south-facing slopes of the Himalayas do not appear clearly between 0600 and 1200 LT. The broadening of the rainband can be seen especially in after-midnight hours. For example, between 0000 and 0300 LT a narrower rainband (the main area of precipitation) appears along the south slopes (at high elevations) of the Himalayas. That means light rain initiates around 2500–3500 m MSL over the south-facing slopes of the Himalayas. More interestingly, it broadens along the slope area and moves with time in the west-southwest direction. It can be seen clearly in the western Himalayan region. Between 0600 and 0900 LT, the foothills of the western Himalayas show more rain activity. But the slope area of the western Himalayas show less rain activity at this time period. Enhanced rainfall on the western bank of the Yamuna River is probably due to the presence of the Aravalli Mountain range. Topography and prevailing winds exert a significant influence on precipitation patterns. The effect of the rivers, the Ganga, Yamuna, Narayani, and Ghagara, in triggering convection in the early morning seems apparent, although rainfall enhancement does not exactly converge at the rivers as shown in Fig. 9. Barros et al. (2004) also noticed dominant cloudiness over these river basins during the monsoon season by pattern covariance analysis of the cloud field. The possible mechanism of river basins on rainfall production is not yet known. Between 0900 and 1200 LT, the rainband moves farther southwestward, crossing the Aravalli range, and becomes weaker. In the slope region, a fuzzy distribution of rainfall is apparent at this time period as compared to preceding hours. Turning to daytime (1200–1500 LT), a narrower rainband appears nearly at the same place. The eastern Himalayan region exhibits more intense rainfall in the slope area. More widespread distribution of rainfall appears between 1500 and 1800 LT along the entire slope region of the Himalayas. Between 1800 and 2100 LT, the rainband seems narrower or perhaps becomes the narrowest among clear rainbands existing during local hours. Less rain activity in the slope region is apparent during this time period.

The appearance of the narrowest rainband between 1800 and 2100 LT suggests consistency with the results of Barros and Lang (2003). Likewise, less rain activity during late morning corresponds with their results. Barros and Lang (2003) examined the diurnal cycle of rainfall by utilizing June 2001 rain gauge data over the central Himalayan region at the Marsyandi network and noted a lull in rain rate during the time period 0600–1200 LT both at high-altitude (>2000 m MSL) and low-altitude (<2000 m MSL) stations. The rainfall patterns depicted by PR over the south-facing slopes of the Himalayas during after-midnight hours are consistent with Meteosat-5 IR cloudiness patterns studied by Barros et al. (2004). They noted high cloudiness at night centered between 0000 and 0300 LT over the south-facing slopes of the entire Himalayas. However, the slope region of the western Himalayas shows more rain activity than the eastern Himalayan slope region during early morning, as mentioned earlier. Insufficient sampling is one of the major issues with PR observations. We also examined precipitation characteristics at 3-h time resolutions for JJA over the 2-yr period of 1998–99 and the 3-yr period of 1998–2000 (see Fig. 10). There appears to be some similarity for the spatial variability in the observed diurnal cycle patterns.

For moderate to heavy rain, hourly rain-conditioned overpasses do not exceed nearly 3. So, we could utilize only a few available samples from moderate to heavy rain. The moderate to heavy rain horizontal distributions for JJA of 1998–2002 for 6 h are shown in Fig. 11. From these figures, we notice that the Tibetan Plateau shows nearly no moderate to heavy rain activity between 0000 and 0600 LT. On the other hand, the Karakoram Valley shows nearly no rain activity between 1200 and 1800 LT in the domain. It is noteworthy that the foothill area of the Himalayan range shows maximum rain activity and intense rainfall between 0600 and 1200 LT. The Aravalli Mountain range and nearby river basin areas in the foothills exhibit this behavior. The south-facing slopes show nearly no rain except a few rain-occupied grids in the central Nepal foothill region at this time period. This is not similar to light conditional rain distributions in terms of occurrence. Compared to light rain, horizontal distributions of moderate to heavy rain rate exhibit a stronger diurnal variation over the south-facing slopes of the Himalayas.

So far, we have discussed horizontal distributions of light and moderate to heavy rain rate. We now focus on investigating rainfall distribution, particularly in the layout of high mountains over the south-facing slopes of the Himalayas. For this analysis, we make use of mean conditional rain rate in the cross section. We selected an altitudinal cross section (2500–3500 m MSL; 77°–90°E) over the south-facing slopes of the Himalayas. Here, we utilized all conditional rain rates since moderate to heavy rain occurrences were rare for 3-h accumulations. Figure 12 shows a line graph representing meriodionally averaged conditional rain rates over the cross section for JJA over the period 1998–2002. The suppressed rain with less normalized rain count between 0600–0900 and 0900–1200 LT can be seen from the figures. They suggest that large-scale precipitation patterns always exist in this altitude range during the monsoon season. The western Himalayan region exhibits intense rainfall in comparison to the eastern Himalayan region. Generally, late-afternoon–evening rainfall appears stronger in the cross section. Probably, the moist upslope flow blocked in the layout of the tallest mountains enhance convection due to an influx of low-level moisture and lifting of conditionally unstable air. However, this process may be retarded during late morning because of weak boundary layer activity.

An alternate way of presenting the diurnal behavior of rainfall is shown by creating a rain occurrence contour. We demonstrated the space–time variability of precipitation in three subregions (refer to section 5b for description on subregions) of the Himalayas. The lower Himalayas subregion was selected in a similar fashion just beneath the middle Himalayas. We utilized total rain occurrence since moderate to heavy rain occurrence was rare for 3-h accumulations. The contour plot thus created represents the diurnal cycle in each of the regions. Figure 13 shows the normalized percentage of rainy grids for 3-h time resolutions at each subregion during JJA 1998–2002. The remarkable feature is the northward progression of precipitation during afternoon hours. Over the middle Himalayas, maximum precipitation frequency appears between 0300 and 0600 LT. A midnight–early-morning southward shift is also evident. The second peak in precipitation frequency appears between 1500 and 1800 LT. Figure 14 shows a schematic illustration of the PR-observed precipitation features for 3- and 6-h accumulations over the south-facing slopes of the Himalayas. The spatial preference (gray boxes for morning hours and white boxes for the other periods) and movement (arrows) of precipitation are shown. The number of rainy grids with moderate to heavy rain shows a stronger diurnal cycle of precipitation over the south-facing slopes of the Himalayas. Analysis of TRMM PR data indicates that the south-facing slopes of the Himalayas can be characterized as a weak convection region where persistent light rainfall is noticed. So far, the question as to which mechanism dominates the physics behind midnight–early-morning convection over the south-facing slopes of the Himalayas remains an open one.

e. Wind distribution

Wind is one of the most important factors in the diurnal cycle of rainfall, because it governs the transport and distribution of moisture. The seasonal wind climatology of the Himalayas suggests low-level southerly–southeasterly flow during the summer and westerly–southwesterly flow during the winter monsoon season. We utilized JJA 1998–99 combined wind field data from NCEP–NCAR reanalysis. Figure 15 shows the averaged wind profiles (north–south component) for 0000 UTC (0600 LT) and 1200 UTC (1800 LT) for the Himalayan region grids. The NCEP–NCAR reanalysis data is on a 2.5° × 2.5° grid and uses coarse topography in the model. It cannot possibly resolve mesoscale features close to the mountains, but it can give some evidence. Near the surface, northerly flow is dominant, although it is weak for the mountain region grids (e.g., H, I notation in Figs. 1 and 15) at 0600 LT. For both these times, there seems to be weak but clear diurnal variation of wind just above the surface in the tropospheric layer. Generally, wind is stable, which means day and night monsoon flow is roughly constant. During the daytime, interaction between upvalley flow and large-scale flow restrict convergence at low elevations. This is because of the weakening in spatial gradients of wind. But upslope flow during daytime aids the formation of convection at higher elevations. At night, strong near-surface winds and relatively weaker winds aloft probably establish meridional circulation in the lower atmosphere within areas bounded to the south by the Ganga River valley area and the north by the south-facing slopes of the Himalayas. The meridional circulation thus created probably aids in favoring convection. However, the influence of mountain breezes in triggering convection cannot be neglected in this region. The daytime upvalley and nocturnal downvally tendency of wind has been reported in some locations over the Himalayas (Ueno et al. 2001; Barros and Lang 2003). The north–south component of wind supports this speculation of meridional circulation only to a certain extent.

Once the rainfall activity has begun over the higher elevations, these areas begin to cool due to precipitation. Cool outflow from rainstorms increases the downvalley tendency of wind at low levels according to Barros and Lang (2003). Similar characteristics were reported by Steiner et al. (2003) in the Alps. The midnight–early-morning rain patterns suggest an analogy to that of low-level convergence zone over windward slopes, especially in the western Himalayan region during the monsoon season. The midnight–early-morning convergence could be aided by the interaction between mountain-forced gravity waves and atmospheric instability as speculated by Barros and Lang (2003). Contrary to the north–south wind component, the east–west wind component shows weak diurnal variation, both at 0600 and 1800 LT and throughout the atmospheric layer (not shown).

6. Summary and conclusions

This paper characterizes rainfall around the Himalayas during the monsoon season using TRMM PR data. Climatological features of the diurnal cycle, intensity-based precipitation characteristics, and spatiotemporal variability were investigated using hourly data. However, the limitation of only daily TRMM PR measurement around the Himalayas precludes the ability to use these data to investigate the diurnal cycle in detail. Discrepancies in the PR-observed precipitation features may arise due to poor temporal sampling combined with the relatively large footprint of the PR (4.3 km) and the complex terrain in this region. Though sampling errors were believed to possibly influence our results, PR-observed precipitation features have shown agreement with previous studies.

The diurnal cycle of rainfall over the Tibetan Plateau exhibits maximum activity during afternoon to evening hours, and a midnight–early-morning maximum is noticed over the Karakoram Valley and over the south-facing slopes of the Himalayas. The northern Indian subcontinent shows daytime activity of rainfall, while north of the Bay of Bengal is characterized by late-morning maximum in rainfall.

A comparative analysis between the finescale topography and overlying surface rainfall suggests that a ridge scale of nearly 20–30 km controls daytime precipitation over the south-facing slopes of the Himalayas. Strong ridge–valley gradients in rain intensity are also noticed in deep valleys. During midnight–early-morning hours, intense rainfall concentrates over the ridges and lowlands, while during late-morning hours minimum activity over the south-facing slopes of the Himalayas is exhibited.

A low-level convergence zone is speculated over the windward slopes of the western Himalayas during midnight–early morning. Precipitation broadening and movement are noticed over the south-facing slopes of the western Himalayas during early morning. For example, between 0300 and 0600 LT, a rainband appears over the slope region and moves to the foothills between 0600 and 0900 LT. Strong meridional circulation in the lower atmosphere probably aids in favoring convection over the south-facing slopes of the Himalayas during midnight–early-morning hours. However, the influence of the mountain breezes in triggering convection cannot be neglected in this region. The midnight–early-morning convergence could be aided by the interaction between mountain-forced gravity waves and atmospheric instability as speculated by Barros and Lang (2003).

Compared to light rain, moderate to heavy rain has more diurnal variation. It may be biased because of the small number of samples of moderate to heavy rain rate available from PR observations. Persistent light rain (≤5 mm h−1) appears over the south-facing slopes of the Himalayas. More specifically, late-morning (0600–1200 LT) rainfall strictly adheres to a threshold of ≤5 mm h−1. But daytime (1200–1800 LT) rainfall suggests few embedded convective cores within a large region of light rain.

The PR fine-resolution data have shown what previous studies lack, that is, real precipitation patterns and their characteristics in the entire region of the Himalayas. Study on the temporal distribution of precipitation, especially from TRMM PR measurements alone, is still incomplete. Analysis on precipitation characteristics would benefit from analysis of PR data with data from other sensors. The seasonal analysis would foster our understanding of the precipitation characteristics around the Himalayas.

Acknowledgments

The authors would like to thank Dr. A. Higuchi, Mr. M. Hirose, and Mr. M. Yamamoto for their generous cooperation in maintaining the computer system. Prof. T. T. Wilheit and Dr. T. N. Rao kindly checked the manuscript. The constructive criticisms of anonymous reviewers greatly improved this manuscript. The data used in this paper were provided by the National Space Development Agency of Japan, which is now renamed as the Japan Space Exploration Agency.

REFERENCES

  • Barros, A. P., and T. Lang, 2003: Monitoring the monsoon in the Himalayas: Observations in central Nepal, June 2001. Mon. Wea. Rev., 131 , 14081427.

    • Search Google Scholar
    • Export Citation
  • 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 , 36833686.

    • Search Google Scholar
    • Export Citation
  • Barros, A. P., G. Kim, E. Williams, and S. W. Nesbitt, 2004: Probing orographic controls in the Himalayas during the monsoon using satellite imagery. Nat. Hazards Earth Syst. Sci., 4 , 2951.

    • Search Google Scholar
    • Export Citation
  • Burbank, D. W., A. E. Blythe, J. Putkonen, B. Pratt-Sitaula, E. Gabet, M. Oskin, A. Barros, and T. P. Ojha, 2003: Decoupling of erosion and precipitation in the Himalayas. Nature, 426 , 652655.

    • Search Google Scholar
    • Export Citation
  • Chen, L., E. R. Reiter, and Z. Feng, 1985: The atmospheric heat source over the Tibetan Plateau: May–August 1979. Mon. Wea. Rev., 113 , 17711790.

    • Search Google Scholar
    • Export Citation
  • Das, P. K., 1987: Short- and long-range monsoon prediction in India. Monsoons, J. S. Fein and P. L. Stephens, Eds., John Wiley and Sons, 549–578.

    • Search Google Scholar
    • Export Citation
  • Iguchi, T., T. Kozu, R. Meneghini, J. Awaka, and K. Okamoto, 2000: Rain-profiling algorithm for the TRMM precipitation radar. J. Appl. Meteor., 39 , 20382052.

    • Search Google Scholar
    • Export Citation
  • Kousky, V., 1980: Diurnal rainfall variation in northwest Brazil. Mon. Wea. Rev., 108 , 488498.

  • Krishnamurti, T. N., and C. M. Kishtawal, 2000: A pronounced continental-scale diurnal mode of the Asian summer monsoon. Mon. Wea. Rev., 128 , 462473.

    • Search Google Scholar
    • Export Citation
  • Kummerow, C., W. Barnes, T. Kozu, J. Shuie, and J. Simpson, 1998: The Tropical Rainfall Measuring Mission (TRMM) sensor package. J. Atmos. Oceanic Technol., 15 , 809816.

    • Search Google Scholar
    • Export Citation
  • Kurosaki, Y., and F. Kimura, 2002: Relationship between topography and daytime cloud activity around Tibetan Plateau. J. Meteor. Soc. Japan, 80 , 339355.

    • Search Google Scholar
    • Export Citation
  • Kuwagata, T., A. Numaguti, and N. Endo, 2001: Diurnal variation of water vapor over the central Tibetan Plateau during summer. J. Meteor. Soc. Japan, 79, B , 401418.

    • Search Google Scholar
    • Export Citation
  • 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 , 12991316.

    • Search Google Scholar
    • Export Citation
  • Li, C., and M. Yanai, 1996: The onset and interannual variability of the Asian summer monsoon in relation to land–sea thermal contrast. J. Climate, 9 , 358375.

    • Search Google Scholar
    • Export Citation
  • Mani, A., 1981: The climate of Himalaya. The Himalaya: Aspects of Changes, J. S. Lall and A. D. Moddie, Eds., Oxford University Press, 3–15.

    • Search Google Scholar
    • Export Citation
  • Murakami, M., 1983: Analysis of the deep convective activity over the western Pacific and Southeast Asia. Part I: Diurnal variation. J. Meteor. Soc. Japan, 61 , 6077.

    • Search Google Scholar
    • Export Citation
  • Nesbitt, S. W., and E. J. Zipser, 2003: The diurnal cycle of rainfall and convective intensity according to three years of TRMM measurements. J. Climate, 16 , 14561475.

    • Search Google Scholar
    • Export Citation
  • Ohsawa, T., H. Ueda, T. Hayashi, A. Watanabe, and J. Matsumoto, 2001: Diurnal variations of convective activity and rainfall in tropical Asia. J. Meteor. Soc. Japan, 79 , 333352.

    • Search Google Scholar
    • Export Citation
  • Shrestha, A. B., C. P. Wake, J. E. Dibb, and P. A. Mayewski, 2000: Precipitation fluctuations in the Nepal Himalaya and its vicinity and relationship with some large scale climatological parameters. Int. J. Climatol., 20 , 317327.

    • Search Google Scholar
    • Export Citation
  • Shrestha, M. L., 2000: Interannual variation of summer monsoon rainfall over Nepal and its relation to Southern Oscillation Index. Meteor. Atmos. Phys., 75 , 2128.

    • Search Google Scholar
    • Export Citation
  • Steiner, M., O. Bousquet, R. A. Houze Jr., B. F. Smull, and M. Mancini, 2003: Airflow within major alpine river valleys under heavy rainfall. Quart. J. Roy. Meteor. Soc., 129 , 411432.

    • Search Google Scholar
    • Export Citation
  • Ueno, K., and Coauthors, 2001: Meteorological observations during 1994–2000 at the Automatic Weather Station (GEN-AWS) in Khumbu region, Nepal Himalayas. Bull. Glaciol. Res., 18 , 2330.

    • Search Google Scholar
    • Export Citation
  • Ye, D., 1981: Some characteristics of the summer circulation over the Qinghai-Xizang (Tibet) Plateau and its neighborhood. Bull. Amer. Meteor. Soc., 62 , 1419.

    • Search Google Scholar
    • Export Citation

Fig. 1.
Fig. 1.

Topography of the region of interest. Surface hydrological features including all rivers and lakes are shown with black color.

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2846.1

Fig. 2.
Fig. 2.

TRMM PR sampling at 0.05° resolution for accumulation periods of 3, 6, 9, and 12 h for JJA of the 5-yr period 1998–2002. The parameter sigma/mean denotes the ratio of the std dev of the 400 by 380 points to the mean, a measure of the inhomogeneity of the sampling across the region (15°–34°N, 75°–95°E).

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2846.1

Fig. 3.
Fig. 3.

Annual mean daily rainfall amount distribution for the 5-yr period 1998–2002: (a) 1998, (b) 1999, (c) 2000, (d) 2001, and (e) 2002.

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2846.1

Fig. 4.
Fig. 4.

Annual mean 6-hourly conditional rain-rate distribution for the 5-yr period 1998–2002: 1200–1800 LT for (a) 1998, (b) 1999, (c) 2000, (d) 2001, and (e) 2002, and 0000–0600 LT for (f) 1998, (g) 1999, (h) 2000, (i) 2001, and (j) 2002.

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2846.1

Fig. 5.
Fig. 5.

Mean daily rainfall amount (mm day−1) averaged meridionally in two subregions: (bottom) the middle Himalayas and high Himalayas. Mean daily rainfall amount averaged meridionally is shown in y axis to right on logarithmic scale for (top) the middle Himalayas and (middle) the high Himalayas. The averaged (meridional) topographic series (dotted lines) is also shown.

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2846.1

Fig. 6.
Fig. 6.

Mean 6-hourly rainfall amount distribution for JJA of the 5-yr period 1998–2002: (a) 0000–0600, (b) 0600–1200, (c) 1200–1800, and (d) 1800–2400 LT.

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2846.1

Fig. 7.
Fig. 7.

High-resolution plots on conditional rain-rate distribution for (top) 0000–0600 and (middle) 1200–1800 LT. (bottom) The topographic map.

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2846.1

Fig. 8.
Fig. 8.

As in Fig. 7, but for different location.

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2846.1

Fig. 9.
Fig. 9.

Mean 3-hourly light conditional rain-rate distribution for JJA of the 5-yr period 1998–2002: (a) 0000–0300, (b) 0300–0600, (c) 0600–0900, (d) 0900–1200, (e) 1200–1500, (f) 1500–1800, (g) 1800–2100, and (h) 2100–2400 LT.

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2846.1

Fig. 10.
Fig. 10.

Mean 3-hourly light conditional rain-rate distribution: (a) 0600–0900 and (b) 1500–1800 LT for JJA of the 2-yr period 1998–99, and (c) 0600–0900 and (d) 1500–1800 LT for JJA of the 3-yr period 1998–2000.

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2846.1

Fig. 11.
Fig. 11.

Mean 6-hourly moderate to heavy conditional rain-rate distribution for JJA of the 5-yr period 1998–2002: (a) 0000–0600, (b) 0600–1200, (c) 1200–1800, and (d) 1800–2400 LT.

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2846.1

Fig. 12.
Fig. 12.

Mean 3-hourly conditional rain rate (dashed lines) averaged meriodionally in the altitudinal cross section of 2500–3500 m MSL over the south-facing slopes of the Himalayas for JJA of the 5-yr period 1998–2002: (a) 0000–0300, (b) 0300–0600, (c) 0600–0900, (d) 0900–1200, (e) 1200–1500, (f) 1500–1800, (g) 1800–2100, and (h) 2100–2400 LT. The dark solid lines and light solid lines represent smoothed occurrences and topographic series, respectively.

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2846.1

Fig. 13.
Fig. 13.

Contour of rainfall occurrence over the south-facing slopes of the Himalayas for eight time periods of a day for JJA of the 5-yr period 1998–2002. For clarity, the diurnal cycle is repeated.

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2846.1

Fig. 14.
Fig. 14.

A schematic illustration of the PR-observed diurnal precipitation features over the south-facing slopes of the Himalayas for JJA of the 5-yr period 1998–2002. More explanation is provided in the text.

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2846.1

Fig. 15.
Fig. 15.

Averaged vertical profiles of wind (north–south component) during JJA of 1998–99 combined for the Himalayan region grids (A to M in Fig. 1) from NCEP–NCAR reanalysis. The dark solid line represents 1800 LT, while the light solid line represents 0600 LT in the plots.

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2846.1

1

Rain-conditioned rain rate (conditional rain rate) is defined as the rate only when it is raining. For the threshold-conditional rain-rate discussions in this paper, it should be understood as raining above the two thresholds cited.

Save
  • Barros, A. P., and T. Lang, 2003: Monitoring the monsoon in the Himalayas: Observations in central Nepal, June 2001. Mon. Wea. Rev., 131 , 14081427.

    • Search Google Scholar
    • Export Citation
  • 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 , 36833686.

    • Search Google Scholar
    • Export Citation
  • Barros, A. P., G. Kim, E. Williams, and S. W. Nesbitt, 2004: Probing orographic controls in the Himalayas during the monsoon using satellite imagery. Nat. Hazards Earth Syst. Sci., 4 , 2951.

    • Search Google Scholar
    • Export Citation
  • Burbank, D. W., A. E. Blythe, J. Putkonen, B. Pratt-Sitaula, E. Gabet, M. Oskin, A. Barros, and T. P. Ojha, 2003: Decoupling of erosion and precipitation in the Himalayas. Nature, 426 , 652655.

    • Search Google Scholar
    • Export Citation
  • Chen, L., E. R. Reiter, and Z. Feng, 1985: The atmospheric heat source over the Tibetan Plateau: May–August 1979. Mon. Wea. Rev., 113 , 17711790.

    • Search Google Scholar
    • Export Citation
  • Das, P. K., 1987: Short- and long-range monsoon prediction in India. Monsoons, J. S. Fein and P. L. Stephens, Eds., John Wiley and Sons, 549–578.

    • Search Google Scholar
    • Export Citation
  • Iguchi, T., T. Kozu, R. Meneghini, J. Awaka, and K. Okamoto, 2000: Rain-profiling algorithm for the TRMM precipitation radar. J. Appl. Meteor., 39 , 20382052.

    • Search Google Scholar
    • Export Citation
  • Kousky, V., 1980: Diurnal rainfall variation in northwest Brazil. Mon. Wea. Rev., 108 , 488498.

  • Krishnamurti, T. N., and C. M. Kishtawal, 2000: A pronounced continental-scale diurnal mode of the Asian summer monsoon. Mon. Wea. Rev., 128 , 462473.

    • Search Google Scholar
    • Export Citation
  • Kummerow, C., W. Barnes, T. Kozu, J. Shuie, and J. Simpson, 1998: The Tropical Rainfall Measuring Mission (TRMM) sensor package. J. Atmos. Oceanic Technol., 15 , 809816.

    • Search Google Scholar
    • Export Citation
  • Kurosaki, Y., and F. Kimura, 2002: Relationship between topography and daytime cloud activity around Tibetan Plateau. J. Meteor. Soc. Japan, 80 , 339355.

    • Search Google Scholar
    • Export Citation
  • Kuwagata, T., A. Numaguti, and N. Endo, 2001: Diurnal variation of water vapor over the central Tibetan Plateau during summer. J. Meteor. Soc. Japan, 79, B , 401418.

    • Search Google Scholar
    • Export Citation
  • 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 , 12991316.

    • Search Google Scholar
    • Export Citation
  • Li, C., and M. Yanai, 1996: The onset and interannual variability of the Asian summer monsoon in relation to land–sea thermal contrast. J. Climate, 9 , 358375.

    • Search Google Scholar
    • Export Citation
  • Mani, A., 1981: The climate of Himalaya. The Himalaya: Aspects of Changes, J. S. Lall and A. D. Moddie, Eds., Oxford University Press, 3–15.

    • Search Google Scholar
    • Export Citation
  • Murakami, M., 1983: Analysis of the deep convective activity over the western Pacific and Southeast Asia. Part I: Diurnal variation. J. Meteor. Soc. Japan, 61 , 6077.

    • Search Google Scholar
    • Export Citation
  • Nesbitt, S. W., and E. J. Zipser, 2003: The diurnal cycle of rainfall and convective intensity according to three years of TRMM measurements. J. Climate, 16 , 14561475.

    • Search Google Scholar
    • Export Citation
  • Ohsawa, T., H. Ueda, T. Hayashi, A. Watanabe, and J. Matsumoto, 2001: Diurnal variations of convective activity and rainfall in tropical Asia. J. Meteor. Soc. Japan, 79 , 333352.

    • Search Google Scholar
    • Export Citation
  • Shrestha, A. B., C. P. Wake, J. E. Dibb, and P. A. Mayewski, 2000: Precipitation fluctuations in the Nepal Himalaya and its vicinity and relationship with some large scale climatological parameters. Int. J. Climatol., 20 , 317327.

    • Search Google Scholar
    • Export Citation
  • Shrestha, M. L., 2000: Interannual variation of summer monsoon rainfall over Nepal and its relation to Southern Oscillation Index. Meteor. Atmos. Phys., 75 , 2128.

    • Search Google Scholar
    • Export Citation
  • Steiner, M., O. Bousquet, R. A. Houze Jr., B. F. Smull, and M. Mancini, 2003: Airflow within major alpine river valleys under heavy rainfall. Quart. J. Roy. Meteor. Soc., 129 , 411432.

    • Search Google Scholar
    • Export Citation
  • Ueno, K., and Coauthors, 2001: Meteorological observations during 1994–2000 at the Automatic Weather Station (GEN-AWS) in Khumbu region, Nepal Himalayas. Bull. Glaciol. Res., 18 , 2330.

    • Search Google Scholar
    • Export Citation
  • Ye, D., 1981: Some characteristics of the summer circulation over the Qinghai-Xizang (Tibet) Plateau and its neighborhood. Bull. Amer. Meteor. Soc., 62 , 1419.

    • Search Google Scholar
    • Export Citation
  • Fig. 1.

    Topography of the region of interest. Surface hydrological features including all rivers and lakes are shown with black color.

  • Fig. 2.

    TRMM PR sampling at 0.05° resolution for accumulation periods of 3, 6, 9, and 12 h for JJA of the 5-yr period 1998–2002. The parameter sigma/mean denotes the ratio of the std dev of the 400 by 380 points to the mean, a measure of the inhomogeneity of the sampling across the region (15°–34°N, 75°–95°E).

  • Fig. 3.

    Annual mean daily rainfall amount distribution for the 5-yr period 1998–2002: (a) 1998, (b) 1999, (c) 2000, (d) 2001, and (e) 2002.

  • Fig. 4.

    Annual mean 6-hourly conditional rain-rate distribution for the 5-yr period 1998–2002: 1200–1800 LT for (a) 1998, (b) 1999, (c) 2000, (d) 2001, and (e) 2002, and 0000–0600 LT for (f) 1998, (g) 1999, (h) 2000, (i) 2001, and (j) 2002.

  • Fig. 5.

    Mean daily rainfall amount (mm day−1) averaged meridionally in two subregions: (bottom) the middle Himalayas and high Himalayas. Mean daily rainfall amount averaged meridionally is shown in y axis to right on logarithmic scale for (top) the middle Himalayas and (middle) the high Himalayas. The averaged (meridional) topographic series (dotted lines) is also shown.

  • Fig. 6.

    Mean 6-hourly rainfall amount distribution for JJA of the 5-yr period 1998–2002: (a) 0000–0600, (b) 0600–1200, (c) 1200–1800, and (d) 1800–2400 LT.

  • Fig. 7.

    High-resolution plots on conditional rain-rate distribution for (top) 0000–0600 and (middle) 1200–1800 LT. (bottom) The topographic map.

  • Fig. 8.

    As in Fig. 7, but for different location.

  • Fig. 9.

    Mean 3-hourly light conditional rain-rate distribution for JJA of the 5-yr period 1998–2002: (a) 0000–0300, (b) 0300–0600, (c) 0600–0900, (d) 0900–1200, (e) 1200–1500, (f) 1500–1800, (g) 1800–2100, and (h) 2100–2400 LT.

  • Fig. 10.

    Mean 3-hourly light conditional rain-rate distribution: (a) 0600–0900 and (b) 1500–1800 LT for JJA of the 2-yr period 1998–99, and (c) 0600–0900 and (d) 1500–1800 LT for JJA of the 3-yr period 1998–2000.

  • Fig. 11.

    Mean 6-hourly moderate to heavy conditional rain-rate distribution for JJA of the 5-yr period 1998–2002: (a) 0000–0600, (b) 0600–1200, (c) 1200–1800, and (d) 1800–2400 LT.

  • Fig. 12.

    Mean 3-hourly conditional rain rate (dashed lines) averaged meriodionally in the altitudinal cross section of 2500–3500 m MSL over the south-facing slopes of the Himalayas for JJA of the 5-yr period 1998–2002: (a) 0000–0300, (b) 0300–0600, (c) 0600–0900, (d) 0900–1200, (e) 1200–1500, (f) 1500–1800, (g) 1800–2100, and (h) 2100–2400 LT. The dark solid lines and light solid lines represent smoothed occurrences and topographic series, respectively.

  • Fig. 13.

    Contour of rainfall occurrence over the south-facing slopes of the Himalayas for eight time periods of a day for JJA of the 5-yr period 1998–2002. For clarity, the diurnal cycle is repeated.

  • Fig. 14.

    A schematic illustration of the PR-observed diurnal precipitation features over the south-facing slopes of the Himalayas for JJA of the 5-yr period 1998–2002. More explanation is provided in the text.

  • Fig. 15.

    Averaged vertical profiles of wind (north–south component) during JJA of 1998–99 combined for the Himalayan region grids (A to M in Fig. 1) from NCEP–NCAR reanalysis. The dark solid line represents 1800 LT, while the light solid line represents 0600 LT in the plots.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 4242 1233 78
PDF Downloads 2375 358 56