TRMM-Observed Summer Warm Rain over the ... - Springer Link

NO.3

371

QIN Fang and FU Yunfei

TRMM-Observed Summer Warm Rain over the Tropical and Subtropical Pacific Ocean: Characteristics and Regional Differences QIN Fang (

) and FU Yunfei∗ (

)

School of Earth and Space Sciences, University of Science and Technology of China, Hefei 230026 (Received November 20, 2015; in final form March 24, 2016)

ABSTRACT Based on the merged measurements from the TRMM Precipitation Radar and Visible and Infrared Scanner, refined characteristics (intensity, frequency, vertical structure, and diurnal variation) and regional differences of the warm rain over the tropical and subtropical Pacific Ocean (40 ◦ S–40◦ N, 120◦ E–70◦ W) in boreal summer are investigated for the period 1998–2012. The results reveal that three warm rain types (phased, pure, and mixed) exist over these regions. The phased warm rain, which occurs during the developing or declining stage of precipitation weather systems, is located over the central to western Intertropical Convergence Zone, South Pacific Convergence Zone, and Northwest Pacific. Its occurrence frequency peaks at midnight and minimizes during daytime with a 5.5-km maximum echo top. The frequency of this warm rain type is about 2.2%, and it contributes to 40% of the regional total rainfall. The pure warm rain is characterized by typical stable precipitation with an echo top lower than 4 km, and mostly occurs in Southeast Pacific. Although its frequency is less than 1.3%, this type of warm rain accounts for 95% of the regional total rainfall. Its occurrence peaks before dawn and it usually disappears in the afternoon. For the mixed warm rain, some may develop into deep convective precipitation, while most are similar to those of the pure type. The mixed warm rain is mainly located over the ocean east of Hawaii. Its frequency is 1.2%, but this type of warm rain could contribute to 80% of the regional total rainfall. The results also uncover that the mixed and pure types occur over the regions where SST ranges from 295 to 299 K, accompanied by relatively strong downdrafts at 500 hPa. Both the mixed and pure warm rains happen in a more unstable atmosphere, compared with the phased warm rain. Key words: warm rain, frequency, diurnal variation, radar reflective factor, trade wind Citation: Qin Fang and Fu Yunfei, 2016: TRMM-observed summer warm rain over the tropical and subtropical Pacific Ocean: Characteristics and regional differences. J. Meteor. Res., 30(3), 371– 385, doi: 10.1007/s13351-016-5151-x.

1. Introduction Most of the global atmospheric water falls to the ground in the form of liquid precipitation. Warm rain is one of the primary precipitation types, which refers to the rain produced within clouds with cloud top temperature above 273 K (0℃). It is mainly formed by the coalescence of water droplets with different falling terminal velocities (Beard and Ochs, 1993). Previous studies suggest that warm rain processes are widely spread throughout the tropics and subtropics; however, due to sufficient liquid water and updraft required to sustain collision-coalescence processes, warm rain not only is restricted to low- or mid-

level clouds, but also may occur in deep convective systems (Schumacher and Houze, 2003). Related investigations show that warm rain is able to provide heat and moisture for the development of low-level convection and balance the atmospheric stability (Bretherton and Wyant, 1997; Kodama et al., 2009). It also impacts upon the spatiotemporal distribution of organized convective systems in the tropics (Johnson and Lin, 1997; Johnson et al., 1999; Mapes, 2000; Del Genio and Kovan, 2002; Wu, 2003). Studying the spatiotemporal distribution of warm rain and its structural characteristics could improve our understanding of cloud and precipitation thermodynamic processes, as well as radiative forcing. Such studies also provide

Supported by the National Natural Science Foundation of China (41230419, 91337213, 40730950, and 40375018), and China Meteorological Administration Special Public Welfare Research Fund (GYHY201306077). ∗ Corresponding author: [email protected]. ©The Chinese Meteorological Society and Springer-Verlag Berlin Heidelberg 2016

372

JOURNAL OF METEOROLOGICAL RESEARCH

a reference base for synoptic model simulations. Precipitation from shallow cloud generally occurs in the form of light rain from stratocumulus or stratus (Austin et al., 1995), and the rain from welldeveloped trade wind cumuli (Baker, 1993) over the tropical and subtropical ocean. Observation of warm rain within stratocumulus/stratus and trade wind cumuli has been a major focus in many field experiments over the past 50 years. From the 1960s to 1980s, a number of projects were carried out, such as the Warm Rain Project (Lavoie, 1967), the Barbados Oceanographic and Meteorological Experiment (Holland and Rasmussen, 1973), and the First International Satellite Cloud Climatology Project (ISCCP) regional experiment (Cox et al., 1987). Takahashi (1977) used aircraft to detect the cloud microphysical features of warm rain and its atmospheric environment near Hawaii. Later, Takahashi (1981) summarized the synoptic situation and wind field during the occurrence of warm rain. Petty (1995) combined shipborne collection and the Comprehensive Ocean–Atmosphere Data Set (COADS) to study the types and intensities of precipitation, as well as warm precipitation detection (Woodruff et al., 1987). Other subsequent studies of the COADS data provided the climatological patterns of stratocumulus/stratus and trade wind cumuli regimes (Warren et al., 1988), including their seasonal cycle (Klein and Hartmann, 1993; Norris, 1998) and long-term variability (Bajuk and Leovy, 1998). The Atlantic Stratocumulus Transition Experiment field campaign (Albrecht et al., 1995; Bretherton et al., 1995) was also carried out to analyze the dynamical features during the transformation process from stratocumulus/stratus to trade wind cumuli. With the progress of detection technology, using satellite remote sensing technology to detect warm rain has become an advantage. Liu et al. (1995) combined retrieved passive microwave infrared and radiances imagers to analyze warm rain over the western Pacific Ocean. They found that 14% of rainfall was associated with cloud-top infrared brightness temperatures of greater than 273 K. Some researchers combined visible/infrared data with microwave remote sensing data from ISCCP and revealed that the pro-

VOL.30

portion of warm rain in total precipitation was about 10%–20% (Schiffer and Rossow, 1983; Rossow and Schiffer, 1991; Lin and Rossow, 1997). Petty (1999) collocated surface synoptic reports of precipitation with satellite infrared observations from Japan’s Geostationary Meteorological Satellite, and found that 20%–40% of the precipitation was associated with warmer infrared temperature (> 273 K) over most of the ocean east of Australia. Tokay et al. (1999) suggested that 7% of total precipitation was from warm cloud in equatorial west Pacific regions. Short and Nakamura (2000) used the relationship between rainfall intensity and storm height in combination with the mixed lognormal distribution to estimate that shallow precipitation comprised approximately 20% of the total precipitation over tropical oceans. Lau and Wu (2003) found that warm rain accounted for 31% of the total rain amount and 72% of the total rain area in the tropics. Liu and Zipser (2009) indicated that raining pixels with a cloud-top temperature above 0℃ contributed 20% of the rainfall over the tropical ocean, based on their TRMM precipitation features database. In summary, there are still inconsistent results regarding the warm rain contribution, especially over the tropics and subtropics, and the regional differences of warm rain over the oceans are not clear either. Variation in the methods used for rainfall identification and the lack of precipitation information are the primary reasons for warm rain estimation discrepancy. Cloud-top radiant brightness temperature detected by infrared remote sensing has been a key parameter to identify warm rain. As suggested by Fu et al. (2011), a lower cloud-top brightness temperature value could also reflect a higher cloud top. Generally, warm rain is defined as the rain with a cloud-top brightness temperature greater than 273 K (Liou, 2004). As we know, the precipitation radar (PR) of TRMM can acquire the horizontal distribution and vertical structure of precipitation. However, it would be better to acquire warm rain information by combining the data detected by PR with the Visible and Infrared Scanner (VIRS). On this basis, regional differences of warm rain in boreal summer over the tropical and subtropical Pacific Ocean are investigated in the present study by using

NO.3


merged measurements of PR and VIRS from 1998 to 2012. The frequency, intensity, vertical structure, and diurnal variation of warm rain will be analyzed in detail, and then the reasons for warm rain differences in the context of the large-scale environmental background are discussed. The results should help provide refined information on warm rain structure, distribution, and daily variation over the tropical and subtropical Pacific Ocean. 2. Data and methods The data used in this study are the TRMM version 7 dataset released by the Goddard Space Flight Center. The TRMM satellite was launched in November 1997. It operates at an altitude of 400 km (after a boost in 2001) and samples between 38◦ S and 38◦ N with 16 swaths each day. PR and VIRS are both carried onboard the satellite of TRMM. The bands of VIRS include 0.63, 1.6, 3.7, 10.8, and 12.0 µm. One of the TRMM standard products, 1B01, derived from the VIRS, offers upward radiation signals in the visible and thermal infrared bands, with a horizontal resolution of about 2.1 km at nadir (Kummerow et al., 1998). Another standard dataset 2A25, the PR level 2 product, includes near-surface rain rates and vertical profiles of precipitation. The swath of PR is narrower than that of VIRS, with a horizontal and vertical resolution of 4.3 km and 250 m, respectively (Awaka et al., 1998; Iguchi et al., 2000). PR’s minimum detectable echo intensity is about 17 dBZ (Iguchi et al., 2000). This implies a rain rate of about 0.3 mm h−1 when it is assumed as uniform beam filling (Fisher, 2004). The similar cross-track scanning mode of PR and VIRS generates little time lag in detecting the same target, which confirms the availability of data matching at the PR spatial resolution. The weight-averaged method is used to establish the merged dataset. The resolution difference between these two detection instruments is taken into account by using VIRS pixelby-pixel weighting in each PR detecting pixel, which makes each channel’s resolution of VIRS (approximately 2 km) reduce to the 2A25 resolution (approxi-

373

mately 4.3 km). The results preserve the information of PR and VIRS at the native resolution of PR. The detail regarding the merger method can be found in Fu et al. (2011). Liu et al. (2010) indicated that this merger method does not cause distortion to the original observation of VIRS. Based on the merged data, firstly, minimum infrared brightness temperatures at 10.8 µm channel of VIRS warmer than 273 K are used to define warm rain (Petty, 1999; Short and Nakamura, 2000; Chen et al., 2011). Then, we combine the PR radar echo intensity greater than 17 dBZ with the restrictive near-surface rain rate (> 0.3 mm h−1 ) to detect warm rain. In order to analyze the climate background where warm rain occurs, NCEP monthly averaged reanalysis data on a 2.5◦ horizontal resolution are used, including potential temperature, geopotential height, horizontal wind, and vertical wind at 17 standard isobaric surfaces and SST (Kalnay et al., 1996). TRMM is the non-geostationary satellite, meaning that satellite dishes have to track across the sky, rather than stay in the same spot. In some cases, the detected precipitation is associated with a rainfall system in its preliminary development phase, and the rain is likely to develop into deep convective system precipitation later, so it may be termed as “phased” warm rain. In other cases, the detected precipitation is just in a steady state, being “pure” warm rainfall during the whole rainy period. Furthermore, part of the “pure” warm rain may develop into an unstable state and then the warm rain in the region may be termed as “mixed”, with coexistence of both “pure” and “phased” warm rains. The warm rain studied in this paper covers all these three types. Figure 1 shows two precipitation cases identified by using the merged data. The VIRS brightness temperature in the 10.8 µm channel (BT4 ) in case one (Fig. 1a) varies from 240 to 294 K. The highest rain rate is about 6.5 mm h−1 , with lower BT4 . As shown in Fig. 1e, warm pixel points (BT4 > 273 K) are mainly distributed on one side of the precipitation cell, and the other side is mixed-phase (233 K 6 BT4 < 273 K) or cold-phase precipitation (BT4 < 233 K). Figure 1b shows that the BT4 of case two ranges from 284

374


VOL.30

Fig. 1. Two typical cases of warm rain identified by using the merged measurements of VIRS and PR in (a, c, e) Northwest Pacific Ocean on 1 August 1998 and (b, d, f) Southeast Pacific Ocean on 6 June 1998: (a, b) brightness temperature detected by VIRS channel 4; (c, d) near-surface rain rates measured by PR; (e, f) locations of the brightness temperature pixel for BT4 > 273 K (red), 233 K 6 BT4 < 273 K (green), and BT4 < 233 K (blue).

to 294 K, which represents the situation with warmer cloud or clear sky. The rain rate is lower than 3.5 mm h−1 (Fig. 1d), with a scattered and isolated distribution (Fig. 1f). Although lacking other effective observation data, we presume that the warm rain within these two cases shows differences. The warm rain of case one belongs to the phased type, while the other case is more likely the pure type of warm rain. These may represent the typical precipitation types in this particular region. Following the case analysis, the spatial distribution of the detected warm rain is analyzed statistically, as well as the averaged SST (Fig. 2). The results in-

dicate that the number of warm rain events is more

Fig. 2. The distribution of warm rain pixels in each grid (color-filled) and the SST (contours; ℃) during summer 1998–2012.

NO.3


than 102 in each 0.5◦ grid, except the cold water area between the two sides of the eastern Intertropical Convergence Zone (ITCZ). Only sample sizes greater than 50 are used to ensure statistical significance. As shown in Fig. 2, there are more than 1500 warm rain samples in the central ITCZ, as well as in the mid–west ITCZ flanks. Thus, region A (15◦ –25◦ N, 150◦ E–160◦ W), region B (5◦ –7◦ N, 180◦ –130◦ W), and region C (7◦ –15◦ S, 180◦ –140◦ W) are selected as the regions upon which we further focus. Regions A and C are in prevailing westerlies, accompanied by higher SST. Region B is located in the ITCZ, which is the high-frequency zone for precipitation. Schumacher and Houze (2003) also suggested that isolated and shallow precipitation often occurs in the middle–east ITCZ flanks, which represents the transition from the warm water area to the cold water area. In order to reveal the regional differences of warm rain, two other regions are also selected on both sides of the ITCZ, accompanied by the northeasterly and southeasterly trade winds (Fig. 5a): region D (18◦ –30◦ N, 150◦ –140◦ W) and region E (5◦ – 20◦ S, 130◦ –100◦ W). In this paper, a number of rainfall characteristic parameters will be analyzed in these five typical regions; namely, rain echo-top height, occurrence frequency, rain rate, ratio of warm rain amount to total precipitation, vertical structure of echo intensity, and diurnal variation. As suggested by Liu et al. (2013), precipitation echo-top height is defined as the maximum height of echo intensity greater than 17 dBZ. Due to the distribution of TRMM detection being zonally non-uniform, the occurrence frequency of warm rain is defined as the ratio of the number of identified warm rain pixels to the total number of pixels within each 0.5◦ grid. Whereas, the ratio of warm rain is the ratio of the number of warm rain pixels to the total number of detected precipitation pixels. The relationships among the averaged rain rate, averaged daily rain rate, and the contribution to total rainfall are also taken into account for warm rain in each 0.5◦ grid. The averaged rain rate (mm h−1 ) is the arithmetic average of precipitation intensity during 15 summers. Then, we define the averaged daily rain rate (mm day−1 ) as the warm rain frequency multiplied by the averaged rain

375

rate, and then convert to daily data (Fu et al., 2008). The contribution to total rainfall is represented by the ratio of the warm rain rate to the total precipitation rain rate. 3. Results 3.1 Geographical distribution To reveal the geographical distribution of warm rain in the tropical and subtropical Pacific Ocean, the echo-top height and BT4 of all types of precipitation (Figs. 3a and 3c) and warm rain (Figs. 3b and 3d) are firstly shown. The echo-top height of precipitation (Fig. 3a) in the central ITCZ is lower by about 0.5 km than that in the east and west, with about 5 km in summer (region B). In the flanks of the central ITCZ to the west (regions A and C), the echo-top height increases from 4.5 to 5 km due to the impact of the easterly trade wind. This distribution, however, is different in the eastern Pacific. The northeasterly trade wind and southeasterly trade wind blow from the northeastern and southeastern Pacific to the central Pacific, with an echo top height from 2.5 to 4 km in regions D and E. Along with the variation in echo-top height, BT4 in these regions also demonstrates obvious regional differences. In the western and eastern ITCZ, BT4 is lower, at 245 K, compared with the central area. The decrease in BT4 in regions A and C is from 280 K in the east to 245 K in the west. The BT4 in regions D and E is almost higher than 273 K. The precipitation with higher BT4 and lower echo-top height in these regions is likely to be warm rain. Compared with all types of precipitation, warm rain has a lower echo-top height of about 3 km. The region south of 20◦ S is in austral winter, and the echotop height of warm rain is lower than 3 km, with a 276-K BT4 . Besides, the BT4 in most regions is higher than 278 K. Specifically, warm rain in the flanks of the middle–east ITCZ has a much lower echo-top height, with higher BT4 than in other regions of the ITCZ. Fu et al. (2006) analyzed a case of deep convection in eastern China during summer 1998, and found that the precipitation echo-top height was much closer to the cloud-top height with an increase in the near-sur-

376


VOL.30

Fig. 3. Distributions of (a, b) mean storm-top height and (c, d) mean cloud-top brightness temperature for (a, c) precipitating clouds and (b, d) warm rain.

face rain rate. However, for shallow precipitation, this difference would change considerably. Figures 3b and 3d show that there is no one-to-one correspondence between the distributions of BT4 and echo-top height. This indicates that the echo-top height detected by PR differs to the BT4 of VIRS, and the thermodynamic and microphysical processes may be more complex within the warm cloud. Besides, the regional difference between the echo-top height and BT4 could also be obvious under the influence of SST. For instance, in terms of the precipitation in the western ITCZ and its flanks, known as the warm pool, the warmer SST may lead to a higher cloud-top and echo-top height. However, the lower SST along the west coast of America is influenced by ocean currents and generates less precipitation. When the region extends to the eastern ITCZ flanks, the SST gradually rises. However, the temperature is not high enough for precipitation to obtain the energy necessary to develop into a deep convective system. Precipitation in these regions is often accompanied by a lower echo-top height and cloud top. The distributions of echo-top height and BT4 from Fig. 3 provide a preliminary picture of the re-

gional differences in the precipitation characteristics in summer over the tropical and subtropical Pacific. In order to obtain more specific information on the warm rain regional characteristics, we also analyze the diurnal cycle of echo-top height and BT4 of precipitation in the five regions. As shown in Fig. 4, the echo-top height in regions A, B, and C has an obvious diurnal cycle: the peak value appears at 0700 LT (local time) (region A), 1100 LT (region B), and 1000 LT (region C), with its amplitude between 4.5 and 5.5 km. Corresponding to the BT4 diurnal cycle of these three regions, the appearance time of the minimum value is in accordance with the peak times of echo-top height. This diurnal variation suggests that the warm rain in these regions is more likely the phased warm rain. In contrast, precipitation in regions D and E has no obvious diurnal cycle. The echo-top height changes between 2.8 and 4.0 km, with a 1–10-K BT4 variation in a day. In region D, both echo-top height and BT4 have a 6-h oscillation. This indicates that the rain in region D is most likely the mixed warm rain. Precipitation in region E shows a stable characteristic of warm rain, with an echo-top height of about 3 km and a 280-K BT4 . This indicates that pure warm rain pro-

NO.3


377

Fig. 4. The diurnal cycle of (a) mean storm-top height and (b) BT4 of the cloud top, for precipitating clouds in the five selected regions.

bably exists in this region. It can be seen that the precipitation in regions A, B, and C differs clearly to that in regions D and E. We thus infer that the warm rain in the five regions exhibits distinct regional differences. Furthermore, we analyze how much warm rain occurs over the tropical and subtropical Pacific Ocean. Short and Nakamura (2000) revealed the different modes of shallow precipitation in the tropics, and Schumacher and Houze (2003) also found the regional characteristics of warm rain. Figure 5 indicates that the distributions of the occurrence frequency of warm rain and its ratio overlap with atmospheric flow at 1000 hPa and geopotential height at 500 hPa. The distributions illustrate a frequency of warm rain up to 2.4%. This is higher than the frequency of deep convective precipitation (1.5%) and lower than that of stratiform precipitation (8%) in the same regions (Liu and Fu, 2001). The distribution of warm rain frequency indicates that there are three high-frequency zones in summer over the tropical and subtropical Pa-

cific; namely, the central ITCZ and its flanks with easterly trade wind, which also refers to regions A, B, and C, respectively. This is consistent with previous studies (Warren et al., 1988; Baker, 1993). Compared with Figs. 3b and 3d, the warm rain frequency in Fig. 5a does not collocate well with the echo-top height and BT4 , meaning that the frequency of warm rain has differences with its cloud-top information, based on its statistical properties. However, the region belonging to the high-frequency zones mostly has a ratio of less than 50%, and regional differences exist. As in region B, the ratio of warm rain is less than 30%, which means that other types of precipitation (such as convective precipitation, stratiform precipitation, and so on) are the main rainfall generators in this region. Warm rain in the western regions of A and C has a ratio of less than 50%, while it is more than 90% on their eastern sides. This ratio is close to that of the warm rain in regions D and E, indicating that the ratio of warm rain is affected by the atmospheric wind field. Figure 5 also shows that the

Fig. 5. Distributions of (a) warm rain occurrence frequency overlapped with the atmospheric wind field (m s −1 ) at 1000 hPa and (b) the warm rain ratio overlapped with geopotential height (gpm) at 500 hPa.

378


ratio in region E is nearly 100%, although the frequency is low. It is clear that almost all the precipitation is pure warm rain in this region. These results are markedly different to those of Liu and Zipser (2009), who reported a ratio of only 20% of warm rain to total precipitation over the tropical ocean. Based on the PR detection results, the mean rain rate, daily mean amount of warm rain, and its contribution to total precipitation are shown in Fig. 6. The mean rain rate is mainly distributed in the South Pacific Convergence Zone (in which region C lies), whose value is about 2.3–2.7 mm h−1 . Moreover, the rain rate in region A decreases from 2.4 mm h−1 in the east to 2.0 mm h−1 in the west, and is lower than 2.4 mm h−1 in the central ITCZ (region B). But in the middle–east ITCZ flanks, the warm rain rate is lower than 2.0 mm h−1 . Comparing with the distribution of SST in Fig. 2, we also find that the colder the SST,

VOL.30

the lower the rain rate. The daily mean precipitation amount is shown in Fig. 6b, and can be compared to the rain rate (Fig. 6a) and frequency of warm rain (Fig. 5a). The daily amount displays three high-value zones, which are consistent with the frequency distribution (Fig. 5a). This characteristic indicates that the daily mean precipitation amount is determined by the frequency of warm rain instead of its mean rain rate. For instance, the warm rain in region A has a lower rain rate but a higher frequency than in region C, leading to the similar contribution to total precipitation in these two regions. Meanwhile, the warm rain in regions D and E has a lower frequency and rain rate, which makes the daily mean precipitation amount very low (< 0.6 mm day−1 ). There are different types of precipitation in summer over the tropical and subtropical Pacific, such as convective or stratiform precipitation (Liu et al., 2013). Figure 6c shows the warm rain contribution to total precipitation. It is seen that over the middle–east ITCZ flanks (including regions D and E) warm rain is the main contributor, accounting for more than 95% of the total regional rainfall. However, the warm rain contribution in region B is lower than 20%. These results indicate the uniqueness of warm rain in the middle–east ITCZ flanks of the trade wind region, especially region E. 3.2 Vertical structure

Fig. 6. Distributions of the warm rain: (a) mean rain rate, (b) daily mean amount, and (c) its contribution to total precipitation.

In order to understand the vertical structure of warm rain over the tropical and subtropical Pacific, we investigate the mean reflectivity factor for precipitation along 150◦ E to 100◦ W, averaged within the three latitude zones (18◦ –22◦ N, 4◦ –8◦ N, and 7◦ –15◦ S) respectively, in Fig. 7. The echo reflectivity is divided into two layers by the 0℃-brightness temperature (Fig. 7b). In the upper layer, the echo reflectivity quickly decreases from 26 to 22 dBZ and slows down to 18 dBZ at 15 km. This weak echo reflectivity is possibly caused by hydrometeors brought by convective updrafts. The reflectivity increases from 26 to 28 dBZ (140◦ W to the west and 120◦ W to the east) and 27 dBZ (140◦ –120◦ W) in the 1-km thickness layer below the 0℃ interface. The highest echo reflectivity

NO.3


Fig. 7. Cross-sections of the mean reflectivity factor for precipitation along 150◦ E to 100◦ W averaged within the latitudinal zones of (a) 18◦ –22◦ N, (b) 4◦ –8◦ N, and (c) 7◦ – 15◦ S.

is produced by the melting layer. The second largest increase in echo reflectivity appears in the near-surface layer, a phenomenon that could be caused by the collision-coalescence process when precipitation particles fall. Compared with the region in the ITCZ, the cross-section of mean reflectivity factor in the easterly trade wind region is non-uniform. In regions A and C, the echo height becomes lower, from 15 to 5 (region E) and 9 km (region D) across the dateline to the west. This indicates a discrepancy between the precipitation types in regions D and E. The warm rain in region E can be regarded as pure warm rain, while the warm rain in region D is of mixed types compared with region E. The Contoured Frequency by Altitude Diagram is an effective method, as suggested by Yuter and Houze (1995), to analyze the characteristics of warm rain vertical structure. The distribution can reflect the warm rain characteristics with different echo reflectivities and heights. However, as the number of samples in higher layers is low, this method would produce a flawed high-value frequency in these layers. Therefore, Contoured Rain Rate by Altitude Diagrams (CRADs) presented by Fu et al. (2003) are used in this paper.

379

The unity of denominator to echo factors is assigned to each layer and then normalized. The vertical structure of warm rain in regions A–E is extracted after the echo reflectivity normalization. As shown in Fig. 8, the warm rain in regions A, B, and C has a uniform vertical structure. They have similar distributions, with 22-dBZ maximum reflectivity corresponding to a height of 2.5 km. The distribution of reflectivity in regions D and E, meanwhile, does not always match the other three regions. In these two regions, the maximum frequency appears at approximately 2 km, with around 20 dBZ. The peak height is at 4 km, with 17 dBZ, which is lower than the height of 5.5 km in regions A, B, and C. Thus, it can be seen that the characteristics of warm rain in regions D and E are substantially different in vertical structure, compared with those in regions A, B, and C. The mean profile of radar reflectivity for warm rain is another method to examine the results regarding both microscopic and macroscopic processes, such as the growth and maintenance of cloud particles (Yin et al., 2013). Figure 9 shows the mean profile of warm rain radar reflectivity and its standard deviation in the five regions. Warm rain in the central ITCZ and its flanks has a consistent profile (Figs. 9a–c). From the near surface up to the height of 2 km, radar reflectivity remains at around 25 dBZ. This indicates that the precipitation particle size and concentration barely change at this stable level. The profile then decreases by 0.9 dBZ m−1 from 2 up to 3.5 km. Within this unstable level, the cloud particle size decreases and the concentration declines with height. Regions D and E also have similar profiles. The stable level is lower than 1.5 km, with a reflectivity of approximately 23.5 dBZ, and the rates of decline in the unstable level are approximately 1.5 dBZ m−1 in region D and 1.0 dBZ m−1 in region E. The cloud particle size and concentration are affected by complex and changeable microphysical and thermodynamic processes in the upper part, while collision-coalescence processes are the major influence in the near-surface part. Therefore, the change of radar reflectivity factor is relatively lower. 3.3 Diurnal cycle A number of studies have shown that the diur-

380


VOL.30

Fig. 8. CRADs of radar reflectivity factor for warm rain in the five regions A–E.

Fig. 9. Mean profiles and their standard deviations of radar reflectivity for warm rain in the five regions.

nal cycle also has regional characteristics, such as the midnight rainfall of Sichuan in August (Lu et al., 2008; Wang et al., 2011). In order to analyze the diurnal cycle of warm rain, the distribution of the peak and valley local time for warm rain is calculated, as shown in Fig. 10. The corresponding probability density functions (PDFs) in the five regions are also determined to further demonstrate statistically the occurrence of the peak and valley times. The diurnal cycles in Figs. 10a and 10c show that warm rain happens more frequently during nighttime (0000–0700 LT) than daytime over the tropical and subtropical Pacific. These results are consistent with some earlier studies (Bowman et al., 2005; Liu and Zipser, 2009). However, the distributions of peak time also exhibit certain regional difference. For instance, the warm precipitation of regions A, B, and C occurs mostly from 0000 to 0300 LT at night, while from 0400 to 0600 LT in regions D and E (Fig. 10c).

Warm precipitation is less frequent during 1000– 1600 LT, compared with the peak value of local time. Among the five regions, the warm rain in regions A and C occur less frequently during 1000–1100 and 1600– 1700 LT, respectively. The probability distribution of valley local time is 1200–1400 and 1400–1600 LT in regions D and E, respectively. All these variations reflect the regional differences in oceanic warm precipitation. The local peak and valley times of the rain rate for warm rain are similar to the warm rain number distribution in Fig. 10, so the related analysis is omitted here. In order to understand comprehensively the diurnal cycle in the five regions, Fig. 11 shows the ratios of the number of occurrences and intensity for warm rain. All the diurnal cycles in the five regions indicate that warm rain occurs more at nighttime than daytime, which is consistent with the distribution in Fig. 10. The amplitude variation is much higher in regions

NO.3


381

Fig. 10. Distributions of the local (a) peak time and (b) valley time for warm rain occurrence, and (c, d) their PDFs, respectively.

Fig. 11. Diurnal cycle of the relative ratios of the (a) number of occurrences and (b) intensity for warm rain in the five regions.

D and E than region B. For the temporal evolution of the diurnal cycle, the ratios of the number of occurrences and intensity in regions D and E begin to increase from 1300 to 1500 LT, and after the peak, the ratios decrease to approximately 4.5%. However, the variation is relatively stable during 0900– 1700 LT in regions A and C. Warm rain in the Northern Hemisphere easterly trade wind region has a relatively larger diurnal variation in amplitude than in the Southern Hemisphere. 3.4 Summarized results The above results reveal that there are three types of warm rain over the Pacific Ocean. The first type is

defined as phased warm rain. It mainly occurs during the initial development or declining stage of rainfall systems, and is located in the middle to western ITCZ and its flanks. It mostly appears at midnight and disappears in the daytime, with a 5.5-km maximum echotop height. The frequency of this type is about 2.2%, and it contributes 40% to total rainfall. The echo reflectivity factor is concentrated at 22 dBZ, with a height of 2.5 km, in the vertical CRAD. Figure 12a shows the distribution of local peak and valley time for cold rain occurrence and its corresponding PDFs in regions A, B, and C. Cold rain is defined as precipitation with BT4 less than 233 K. The local peak time for cold precipitation is mainly 0400–0600 LT, mean-

382


Fig. 12. The (a) distribution of the local peak time for cold rain occurrence in each grid and (b) PDF of the local time for peak cold rain occurrence.

ing that cold rain occurs later than warm rain. According to the progressive nature of precipitation process, we infer that nighttime warm rain will develop into deep convective rainfall in the early morning in these three regions. Thus, the warm rain in regions A, B, and C belongs to the phased warm rain. Mixed warm rain is the second type of warm precipitation. Sometimes, it may develop into relatively deeper convective precipitation, while mostly it is similar to pure warm rain. The mixed type, which is mainly located over the ocean of east Hawaii, has a frequency of 1.2%, but can contribute 80% to the total rainfall amount. The third type of warm precipitation is pure warm rain. This is the typical stable precipitation type, with an echo-top height lower than 4 km, and is located over the southeast side of the Pacific. Although its frequency is less than 1.3%, this type of warm rain accounts for 95% of total rainfall. The diurnal variation shows its peak occurrence to be at midnight and its valley in the afternoon. 4. Conclusions and discussion

VOL.30

makes it feasible to obtain information on warm rain and its cloud-top radiances in each PR cloud pixel. Using such merged datasets in the present study, warm rain is detected and analyzed for the tropical and subtropical Pacific regions during summer 1998–2012. Three types of warm rain, i.e., phased, mixed, and pure, are categorized, with distinct characteristics in rainfall intensity, frequency, vertical structure, and diurnal variation. A detailed summary of results is given in Section 3.4. Why does the tropical and subtropical Pacific Ocean feature these three different types of warm rain? Due to the limited observation dataset, reanalysis data are used to analyze the environmental and atmospheric conditions during the occurrence of warm rain. The SST (Fig. 2), horizontal wind field at 1000 hPa (Fig. 5a), geopotential height at 500 hPa (Fig. 5b), and vertical velocity at 500 hPa (Fig. 13) have already been shown above. Here, we also calculate the potential temperature difference between 700 and 1000 hPa to examine the atmospheric stability in the lower troposphere (lower tropospheric stability; Klein and Hartmann, 1993), and the results indicate the following. Phased warm rain mainly occurs in the easterly trade wind area with warm SST (Fig. 2), which is south of the subtropical high and equatorial easterlies (Fig. 5). The subtropical trough and equatorial trough are in the middle atmosphere, associated with upward or weak downward air motion and unstable atmosphere (Fig. 13).

Fig. 13. The distribution of geopotential temperature difference (contours; K) between 700 and 1000 hPa and the vertical velocity at 500 hPa (color-filled; Pa s−1 ) during

Combining the observations of PR and VIRS

summer 1998–2012.

NO.3


Mixed warm rain and pure warm rain occur over the ocean areas with SSTs of 22–26℃, which is relatively moderate. Shearing of the wind direction happens in the lower atmosphere, with a high pressure zone at 500 hPa. The atmosphere is stable along with downdraft in that region. Compared with pure warm rain, mixed warm rain is disturbed by the northeasterly trade wind. The stability of atmospheric stratification is relatively low. This type of precipitation could occasionally develop into deep convection. The detection of weak precipitation by PR is limited to its 2.2-cm wavelength. Due to the sea surface disturbance for the echo reflectivity, the radar detection capability is limited for acquiring lower clouds only, which includes warm rain (Liu et al., 2015). Therefore, more detailed warm rain research is still undergoing. Acknowledgments. Thanks go to the Japan Aerospace Exploration Agency and Goddard Space Flight Center for providing the PR 2A25 and VIRS 1B01 data. The NCEP/NCAR reanalysis data used here were obtained from the NOAA-CIRES Climate Diagnosis Center, accessible at http:// www.esrl.noaa.gov. We also appreciate the suggestions and comments of the anonymous reviewers, which were helpful in improving the overall quality of the article.

REFERENCES Albrecht, B. A., C. S. Bretherton, D. Johnson, et al., 1995: The Atlantic stratocumulus transition experiment (ASTEX). Bull. Amer. Meteor. Soc., 76, 889– 904, doi: 10.1175/1520-0477(1995)0762.0.CO;2. Austin, P., Y. N. Wang, V. Kujala, et al., 1995: Precipitation in stratocumulus clouds: Observational and modeling results. J. Atmos. Sci., 52, 2329– 2352, doi: 10.1175/1520-0469(1995)0522.0.CO;2. Awaka, J., T. Iguchi, and K. Okamoto, 1998: Early results on rain type classification by the Tropical Rainfall Measuring Mission (TRMM) precipitation radar. Proc. 8th URSI commission Final Open

383

Symp., Aveiro, Portugal, 143–146. Bajuk, L. J., and C. B. Leovy, 1998: Are there real interdecadal variations in marine low clouds? J. Climate, 11, 2910–2921, doi: 10.1175/15200442(1998)0112.0.CO;2. Baker, M., 1993: Trade cumulus observations. The Representation of Cumulus Convection in Numerical Models. Amer. Meteor. Soc., 29–37, doi: 10.1007/ 978-1-935704-13-3− 3. Beard, K. V., and H. T. Ochs, 1993: Warm-rain initiation: An overview of microphysical mechanisms. J. Appl. Meteor., 32, 608–625, doi: 10.1175/15200450(1993)0322.0.CO;2. Bowman, K. P., J. C. Collier, G. R. North, et al., 2005: Diurnal cycle of tropical precipitation in Tropical Rainfall Measuring Mission (TRMM) satellite and ocean buoy rain gauge data. J. Geophys. Res., 110, doi: 10.1029/2005JD005763. Bretherton, C. S., P. Austin, and S. T. Siems, 1995: Cloudiness and marine boundary layer dynamics in the ASTEX Lagrangian experiments. Part II: Cloudiness, drizzle, surface fluxes, and entrainment. J. Atmos. Sci., 52, 2724–2735, doi: 10.1175/1520– 0469(1995)0522.0.CO;2. Bretherton, C. S., and M. C. Wyant, 1997: Moisture transport, lower-tropospheric stability, and decoupling of cloud-topped boundary layers. J. Atmos. Sci., 54, 148–167, doi: 10.1175/1520– 0469(1997)0542.0.CO;2. Chen, R. Y., Z. Q. Li, R. J. Kuligowski, et al., 2011: A study of warm rain detection using A-Train satellite data. Geophys. Res. Lett., 38, L04804, doi: 10.1029/2010GL046217. Cox, S. K., D. S. McDougal, D. A. Randall, et al., 1987: FIRE—the first ISCCP regional experiment. Bull. Amer. Meteor. Soc., 68, 114–118, doi: 10. 1175/1520-0477(1987)0682.0.CO;2. Del Genio, A. D., and W. Kovan, 2002: Climatic properties of tropical precipitating convection under varying environmental conditions. J. Climate, 15, 2597–2615, doi: 10.1175/1520-0442(2002)0152.0.CO;2. Fisher, B. L., 2004: Climatological validation of TRMM TMI and PR monthly rain products over Oklahoma. J. Appl. Meteor., 43, 519–535, doi: 10.1175/15200450(2004)0432.0.CO;2. Fu Yunfei, Lin Yihua, Liu Guosheng, et al., 2003: Seasonal characteristics of precipitation in 1998 over

384


East Asia as derived from TRMM PR. Adv. Atmos. Sci., 20, 511–529, doi: 10.1007/BF02915495. Fu Yunfei, Feng Jingyi, Zhu Hongfang, et al., 2006: Precipitation structures of a thermal convective system in the central western subtropical Pacific anticyclone. Acta Meteor. Sinica, 20, 232–243. Fu Yunfei, Zhang Aimin, Liu Yong, et al., 2008: Characteristics of seasonal scale convective and stratiform precipitation in Asia based on measurements by TRMM Precipitation Radar. Acta Meteor. Sincia, 66, 730–746, doi: 10.11676/qxxb2008.067. (in Chinese) Fu Yunfei, Liu Peng, Liu Qi, et al., 2011: Climatological characteristics of VIRS channels for precipitating cloud in summer over the tropics and subtropics. J. Atmos. Environ. Opt., 6, 129–140, doi: 10.3969/j.issn.1673-6141.2011.02.009. (in Chinese) Holland, J. Z., and E. M. Rasmussen, 1973: Measurements of the atmospheric mass, energy, and momentum budgets over a 500-km square of tropical ocean. Mon. Wea. Rev., 101, 44–55, doi: 10.1175/1520–0493(1973)1012.3. CO;2. Iguchi, T., T. Kozu, R. Meneghini, et al., 2000: Rainprofiling algorithm for the TRMM precipitation radar. J. Appl. Meteor., 39, 2038–2052, doi: 10. 1175/1520-0450(2001)0402.0.CO;2. Johnson, R. H., and X. Lin, 1997: Episodic trade wind regimes over the western Pacific warm pool. J. Atmos. Sci., 54, 2020–2034, doi: 10.1175/1520– 0469(1997)0542.0.CO;2. Johnson, R. H., T. M. Rickenbach, S. A. Rutledge, et al., 1999: Trimodal characteristics of tropical convection. J. Climate, 12, 2397–2418, doi: 10.1175/15200442(1999)0122.0.CO;2. Kalnay, E., M. Kanamitsu, R. Kistler, et al., 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77, 437–471, doi: 10.1175/ 1520–0477(1996)0772.0.CO;2. Klein, S. A., and D. L. Hartmann, 1993: The seasonal cycle of low stratiform clouds. J. Climate, 6, 1587–1606, doi: 10.1175/1520–0442(1993)0062.0.CO;2. Kodama, Y. M., M. Katsumata, S. Mori, et al., 2009: Climatology of warm rain and associated latent heating derived from TRMM PR observations. J. Climate, 22, 4908–4929, doi: 10.1175/2009JCLI2575.1.

VOL.30

Kummerow, C., W. Barnes, T. Kozu, et al., 1998: The Tropical Rainfall Measuring Mission (TRMM) sensor package. J. Atmos. Oceanic Technol., 15, 809–817, doi: 10.1175/1520–0426(1998)0152.0.CO;2. Lau, K. M., and H. T. Wu, 2003: Warm rain processes over tropical oceans and climate implications. Geophys. Res. Lett., 30, 2290, doi: 10.1029/2003GL018567. Lavoie, R. L., 1967: The warm rain project in Hawaii. Tellus, 19, 347–347, doi: 10.1111/j.21533490.1967.tb01488.x. Lin, B., and W. B. Rossow, 1997: Precipitation water path and rainfall rate estimates over oceans using special sensor microwave imager and International Satellite Cloud Climatology Project data. J. Geophys. Res., 102, 9359–9374, doi: 10.1029/96JD03987. Liou, K. N., 2004: An Introduction to Atmospheric Radiation. China Meteorological Press, Beijing, 614 pp. Liu, C. T., and E. J. Zipser, 2009: “Warm rain” in the tropics: Seasonal and regional distributions based on 9 yr of TRMM data. J. Climate, 22, 767–779, doi: 10.1175/2008JCLI2641.1. Liu Dongyang, Liu Qi, and Zhou Lingli, 2015: Underestimation of oceanic warm cloud occurrences by the Cloud Profiling Radar aboard CloudSat. J. Meteor. Res., 29, 576–593, doi: 10.1007/s13351-015-5027-5. Liu, G. S., J. A. Curry, and R. S. Sheu, 1995: Classification of clouds over the western equatorial Pacific Ocean using combined infrared and microwave satellite data. J. Geophys. Res., 100, 13811–13826, doi: 10.1029/95JD00823. Liu, G. S., and Y. F. Fu, 2001: The characteristics of tropical precipitation profiles as inferred from satellite radar measurements. J. Meteor. Soc. Japan, 79, 131–143, doi: 10.2151/jmsj.79.131. Liu Peng, Li Chongyin, Wang Yu, et al., 2013: Climatic characteristics of convective and stratiform precipitation over the tropical and subtropical areas as derived from TRMM PR. Sci. China (Ser. D), 56, 375–385, doi: 10.1007/s11430-012-4474-4. Liu Xiantong, Liu Qi, Fu Yunfei, et al., 2010: Daytime cloud detection scheme relying on five-channel measurements from TRMM VIRS. J. Atmos. Environ. Opt., 5, 128–140, doi: 10.3969/j.issn.16736141.2010.02.007. (in Chinese)

NO.3


Lu Ping, Yu Rucong, and Zhou Tianjun, 2008: Numerical simulation of the mid-night rainfall over Sichuan basin during August 2003. Acta Meteor. Sinica, 66, 371–380, doi: 10.11676/qxxb2008.035. (in Chinese) Mapes, B. E., 2000: Convective inhibition, subgridscale triggering energy, and stratiform instability in a toy tropical wave model. J. Atmos. Sci., 57, 1515–1535, doi: 10.1175/15200469(2000)0572.0.CO;2. Norris, J. R., 1998: Low cloud type over the ocean from surface observations. Part I: Relationship to surface meteorology and the vertical distribution of temperature and moisture. J. Climate, 11, 369–382, doi: 10.1175/15200442(1998)0112.0.CO;2. Petty, G. W., 1995: Frequencies and characteristics of global oceanic precipitation from shipboard present-weather reports. Bull. Amer. Meteor. Soc., 76, 1593–1616, doi: 10.1175/1520-0477(1995) 0762.0.CO;2. Petty, G. W., 1999: Prevalence of precipitation from warm-topped clouds over eastern Asia and the western Pacific. J. Climate, 12, 220–229, doi: 10.1175/ 1520-0442(1999)0122.0.CO;2. Rossow, W. B., and R. A. Schiffer, 1991: ISCCP cloud data products. Bull. Amer. Meteor. Soc., 72, 2–20, doi: 10.1175/1520-0477(1991)0722.0. CO;2. Schiffer, R. A., and W. B. Rossow, 1983: The International Satellite Cloud Climatology Project (ISCCP): The first project of the World Climate Research Programme. Bull. Amer. Meteor. Soc., 64, 779–784. Schumacher, C., and R. A. Houze Jr., 2003: The TRMM Precipitation Radar’s view of shallow, isolated rain. J. Appl. Meteor., 42, 1519–1524, doi: 10.1175/15200450(2003)0422.0.CO;2. Short, D. A., and K. Nakamura, 2000: TRMM radar observations of shallow precipitation over the tropical oceans. J. Climate, 13, 4107–4124, doi: 10.1175/ 1520-0442(2000)0132.0.CO;2. Takahashi, T., 1977: A study of Hawaiian warm rain showers based on aircraft observation. J. Atmos.

385

Sci., 34, 1773–1790, doi: 10.1175/1520-0469(1977) 0342.0.CO;2. Takahashi, T., 1981: Warm rain study in Hawaiirain initiation. J. Atmos. Sci., 38, 347–369, doi: 10. 1175/1520-0469(1981)0382.0.CO;2. Tokay, A., D. A. Short, C. R. Williams, et al., 1999: Tropical rainfall associated with convective and stratiform clouds: Intercomparison of disdrometer and profiler measurements. J. Appl. Meteor., 38, 302–320, doi: 10.1175/15200450(1999)0382.0.CO;2. Wang Fuchang, Yu Rucong, Chen Haoming, et al., 2011: The characteristics of rainfall diurnal variation over the southwestern China. Torrential Rain Disaster, 30, 117–121, doi: 10.3969/j.issn.10049045.2011.02.003. (in Chinese) Warren, S. G., C. J. Hahn, J. London, et al., 1988: Global Distribution of Total Cloud Cover and Cloud Type Amounts over the Ocean. NCAR Tech. Note NCAR/TN-317+STR, Boulder, USA, 32–33, doi: 10.5065/D6GH9FXB. Woodruff, S. D., R. J. Slutz, R. L. Jenne, et al., 1987: A comprehensive ocean–atmosphere data set. Bull. Amer. Meteor., 68, 1239–1250. doi: 10.1175/1520– 0477(1987)0682.0.CO;2. Wu, Z. H., 2003: A shallow CISK, deep equilibrium mechanism for the interaction between large-scale convection and large-scale circulations in the tropics. J. Atmos. Sci., 60, 377–392, doi: 10.1175/15200469(2003)0602.0.CO;2. Yin Jinfang, Wang Donghai, Zhai Guoqing, et al., 2013: Observational characteristics of cloud vertical profiles over the East Asian Continent from the Cloudsat data. J. Meteor. Res., 27, 26–39, doi: 10.1007/s13351-013-0104-0. Yuter, S. E., and R. A. Houze Jr., 1995: Threedimensional kinematic and microphysical evolution of Florida cumulonimbus. Part II: Frequency distributions of vertical velocity, reflectivity, and differential reflectivity. Mon. Wea. Rev., 123, 1941–1963, doi: 10.1175/1520-0493(1995)123 2.0.CO;2.