1. Introduction
The atmospheric boundary layer (ABL) appears as the result of surface forcings. Therefore, examination of the characteristics of the ABL can clarify the impacts of land surfaces on the atmosphere. In terrestrial areas, the ABL shows a strong diurnal change caused by land surface forcings, and a convective boundary layer (CBL) develops during daytime. Below the top of the CBL, thermal updrafts and interthermal downdrafts form vertical circulation, which transports heat, water vapor, momentum, and other quantities and creates a quasi-uniform layer. Such behavior of the CBL changes with the forcings of the underlying land surface (Stull 1988; Garratt 1992; Kaimal and Finnigan 1994).
The height of the CBL top is a key parameter in describing the CBL. Tennekes (1973) presented a conceptual model for calculating the height of the CBL top from the sensible heat provided from the land surface. This model is called a “bulk” or “slab” model because the CBL is idealized as a layer of uniform air. Kaimal et al. (1976) showed a typical time series of the sensible heat flux (SHF) and the height of the CBL top based on radiosonde observations. Continuous observation of the CBL top became possible with the development of remote sensors, such as wind profiler radar (WPR) (Ecklund et al. 1988; Angevine et al. 1994; Hashiguchi et al. 1995; Beyrich and Görsdorf 1995; Grimsdell and Angevine 1998; Cohn and Angevine 2000). Yi et al. (2001) reported the seasonal change of diurnal variation of the surface flux and the CBL top estimated by WPR. The maximum CBL top coincided with the maximum surface sensible heat flux.
Vertical circulation plays a central role in the vertical transport in the CBL. The statistical characteristics of this circulation have been reported using data collected from aircraft observations (e.g., Young 1988a, b). These thermal updrafts are driven by the buoyancy flux in the CBL. In terrestrial areas, the buoyancy flux depends largely on the heat flux (e.g., Stull 1988).
Extensive studies of the CBL over terrestrial areas have been conducted for the large, relatively dry plains of North America (e.g., Kaimal et al. 1976; Stull 1988; Young 1988a, b; Yi et al. 2001). In contrast to the wide, dry plains of North America, vast paddy fields cover large parts of eastern China, according to a land use map derived from satellite observations (cf. Matsuoka et al. 2007). Paddy fields supply large amounts of water vapor to the atmosphere (cf. Oue 2005), but little research has been conducted on the CBL over such humid terrestrial areas.
The Huaihe River basin of China has large areas of farmland on which double cropping (wheat and rice) is practiced. In early summer, the vegetation changes from mature wheat fields to paddy fields. Several studies of deep convection were conducted in this region (e.g., Shinoda and Uyeda 2002; Shusse et al. 2005; Shusse and Tsuboki 2006) during the Global Energy and Water Experiment (GEWEX) Asian Monsoon Experiment (GAME)/Huaihe River Basin Experiment (HUBEX) (Zhao and Takeda 1998). However, no study of CBL processes has been conducted in the area.
To investigate the CBL processes over the Huaihe River basin, we conducted the Lower Atmosphere and Precipitation Study (LAPS) project as part of the Core Research for Evolutional Science and Technology (CREST) program, which implemented continuous monitoring and intensive observations in the area from August 2003 to January 2006. Tanaka et al. (2007) described the seasonal change of parameters with regard to the land surface (e.g., albedo, soil moisture), physical values near the surface [e.g., sensible heat flux, latent heat flux (LHF)], and the height of the CBL top. The height of the CBL top was found to correlate with the sensible heat provided from the land surface, which was related to vegetation change; the CBL top was low when paddy fields covered the land surface. The sensible heat flux was the primary controlling factor, as reported by previous studies of relatively dry terrestrial areas (e.g., Kaimal et al. 1976; Stull 1988; Yi et al. 2001). However, although Tanaka et al. (2007) investigated the characteristics of the bulk structure of the CBL, the vertical circulation in the CBL has not been studied.
In the present study, we clarify the characteristics of the CBL observed over the Huaihe River basin in early summer, focusing on the circulation and the vertical transport of heat and moisture in the CBL. First, we describe the development of the CBL and surface heat flux based on observations. Next, we describe the results of numerical simulations to quantitatively evaluate the vertical fluxes of heat and moisture. Finally, we discuss the characteristics of the CBL over humid terrestrial areas.
2. Study site and measurements
Under the LAPS project, continuous observations were performed using WPR, a Doppler sodar, a microwave radiometer, and a flux tower from August 2003 to January 2006 at Shouxian, Anhui Province, China (Fig. 1). The study site was located on a large plain along the Huaihe River. Being nearly flat and uniform, the site was suitable for a boundary layer experiment to study the effects of the land surface. In this study, we used the WPR and surface data obtained during an intensive observation period (IOP) in early summer of 2004 (IOP-2004: 24 May–16 July). During this season, a surface transition is caused by the change in the crop in the double-cropping system. For a detailed description of the observations, see Tanaka et al. (2007).
Table 1 shows the operating specifications of the WPR. The WPR obtained data on the signal-to-noise ratio (SNR) and three-dimensional wind components. The frequency (1290 MHz) is suitable for observation of the lower atmosphere. The height and temporal resolution were 100 m and approximately 59 s, respectively. Surface fluxes were calculated from data obtained by flux sensors and radiation sensors set at a height of 32 m on the flux tower.
The SNR of the WPR is related to the structure parameter of the refractive index C2n, which depends on the turbulence intensity, and the vertical gradient of the refractive index. In the lower atmosphere, C2n mainly depends on the vertical gradient of water vapor (White et al. 1991). Therefore, the top of the CBL is detected as a large SNR because of the turbulence and the large gradient of water vapor between the humid boundary layer and dry free atmosphere (Angevine et al. 1994). In this study, we used the concept outlined above but also added a limit to the possible height range (Tanaka et al. 2007). The CBL top height was estimated within a variable range to avoid large SNRs near the surface or above the actual CBL top. The variable range was set based on the assumption that the CBL top starts to develop just after sunrise at around 0600 local standard time (0600 LST), reaches a maximum height between 1200 and 1600 LST, and does not change largely after 1600 LST. This assumption was used to determine the variable range, and thus the actual CBL top did not have to follow the assumed development.
The CBL top was determined by the following procedure using the SNR data interpolated every 30 s (cf. Tanaka et al. 2007): first, the tentative maximum CBL top height on the day ht,max was determined. The height of the maximum SNR h30s(t) was selected for each time step. The tentative CBL top height at every hour ht,1h(t) was calculated as the median value of h30s(t) using a 4-h centered difference; ht,max was defined as the maximum ht,1h(t) between 1200 and 1600 LST. Second, based on ht,max, the central value of the variable range at every hour hc,1h (t) was defined as follows: between 1200 and 1600 LST, hc,1h (t) was given as the median of h30s (t) for 2 h between 0.5 ht,max and 1.2 ht,max. Similarly, between 0900 and 1100 LST, hc,1h (t) was the median of h30s(t) for 2 h between 400 m and 1.2 hc,1h (t + 1 h); for example, 1.2 hc,1h (1200 LST) was used to determine hc,1h (1100 LST). The initial hc,1h (0600 LST) was 100 m. The hc,1h (0700 LST) and hc,1h (0800 LST) were determined assuming linear development between 0600 and 0900 LST; hc,1h (1700 LST) and hc,1h (1800 LST) were equal to hc,1h (1600 LST), and hc,1h (t) was linearly interpolated for each time step using a 30-s interval. Finally, the variable range was determined as (1 ± A) hc,1h (t), where the constant A = 0.00–0.25 between 0600 and 0900 LST, 0.25 between 0900 and 1600 LST, and 0.25–0.30 between 1600 and 1800 LST. Within the height range, the height of the maximum SNR h30s(t) was selected again as the CBL top height zi.
3. Numerical model and experimental setup
Table 2 illustrates the configuration of the large-eddy simulation (LES) conducted by the Cloud Resolving Storm Simulator (CReSS: Tsuboki and Sakakibara 2002). The CReSS model is a nonhydrostatic model that uses a 1.5-order turbulence kinetic energy (TKE) closure scheme and is applicable to the simulation of boundary layer processes (e.g., Liu et al. 2006). The simulation spanned a domain of 20 km × 20 km × 5888 m with a mesh of 200 × 200 × 120 points. The horizontal grid size was 100 m, which was sufficiently smaller than the horizontal scale of the thermal updraft estimated by the WPR observations. The vertical grid size was 30 m below 3000 m in height. The periodic boundary condition was applied to the lateral boundary, and the rigid boundary condition was applied to the top boundary with a spongy layer. The surface process was a simple bulk method with prediction of soil temperature by thermal diffusion (Louis et al. 1981). Land surface parameters were set as uniform (Table 3). These parameters were determined from the flux tower and manual observations to reproduce the observed surface fluxes from bare fields and paddy fields. The initial conditions were based on radiosonde data observed at 0000 UTC (0800 LST) at Fuyang (Fig. 1), which was the nearest sounding point. The simulations were executed for an integrated time of 12 h from 0800 LST.
4. Results
a. Observation
During the IOP-2004, land use changed from mature wheat fields to bare fields and then to paddy fields. Figure 2 shows the time series of the precipitation rate at the land surface, net radiation, and sensible and latent heat fluxes at 32 m above ground level (AGL). The small net radiation on rainy days was caused by the interruption by clouds. There was an obvious difference between the first period, with mature wheat or bare fields, and the second period, with paddy fields. Although net radiation was a little larger in the second period than in the first period, the sensible heat flux was much smaller. In the second period, much energy was partitioned to the latent heat flux because a large amount of water existed on the surface of the paddy fields. As a result, while the sensible heat flux was nearly equal to the latent heat flux in the first period, the latent heat flux was predominant in the second period.
Since distinct characteristics of the surface fluxes appeared in the first and second periods, two clear days were selected for detailed analysis. For these clear days, downward shortwave radiation showed a smooth curve with time, and the lack of clouds was confirmed by hemispherical photographs taken at the observation site. Two days, 31 May and 22 June 2004, were selected to represent the first and second periods and are referred to as the dry case and wet case, respectively.
The diurnal change of the surface fluxes and the CBL structure for the dry case is shown in Fig. 3. In this case, the wheat had already been harvested, and the land surface was bare. Since the surface soil was easily heated because of its small heat capacity, the sensible heat flux became positive with the positive net radiation around 0630 LST (Fig. 3c). The sensible heat flux remained between 190 and 390 W m−2 from 0900 to 1600 LST, and the latent heat flux stayed around 200 W m−2. With the sensible heat flux, a deep CBL developed rapidly from early morning (Fig. 3a). The height of the CBL top began to increase at 0730 LST, reaching 1600 m at 1000 LST, 1900 m at 1200 LST, and a maximum height of 2250 m at about 1400 LST. Updraft and downdraft occurred in the entire CBL until about 1630 LST, thus forming circulation (Fig. 3b). The maximum value of the thermal updraft was about 3 m s−1. The horizontal scale of the thermal updrafts was about 720–3600 m, assuming the drift with a 5 m s−1 time-averaged horizontal wind speed. The horizontal scale of the circulation (distance between updrafts) was at least 3 km based on the same assumption. After around 1630 LST, the average thermal top lowered gradually, although the thermals reaching high altitude were still present.
For the wet case, a shallow CBL developed slowly from late morning (Fig. 4). Although the net radiation became positive at 0600 LST, the sensible heat flux remained nearly zero until 1000 LST (Fig. 4c). Between 1200 and 1300 LST, the net radiation reached its maximum value of 700 W m−2, while the sensible heat flux was only about 100 W m−2. In contrast to the sensible heat flux, the latent heat flux maintained large values between 320 and 560 W m−2 from 1000 to 1600 LST. There were large amounts of water on the surface of the paddy fields, and a large amount of energy was consumed in evapotranspiration. The CBL started to develop at 1000 LST, 2.5 h later than in the dry case, and gradually increased its top height (Fig. 4a). The height of the CBL top was 700 m at 1200 LST, 1100 m at 1400 LST, and stopped developing at about 1740 LST, when the sensible heat flux neared zero. The maximum height of the CBL top was about 1400 m. Repetition of weak vertical motion was seen in the CBL (Fig. 4b). The maximum value of the thermal updraft was about 1.5 m s−1, the horizontal scale of the thermal updrafts was approximately 720–3600 m, and the horizontal scale of the circulation was at least 1.3 km. The height of thermal top tended to be low after around 1630 LST.
Table 4 summarizes the results of the observations. The characteristics of the surface fluxes, the development of the CBL, and the structure of the circulation in the CBL in each case are presented.
b. Simulation
We conducted a numerical LES to clarify the three-dimensional structure of the CBL, including the temperature and humidity, and to evaluate the vertical fluxes of heat, moisture, and buoyancy.
Figures 5 and 6 show the initial conditions for the dry case and the wet case, respectively. Potential temperatures were stably stratified for dry convection. Humidity clearly differed between the two cases, with more humidity in the lower troposphere in the wet case than in the dry case. The mixing ratio of water vapor had a maximum near the land surface and was not mixed yet because the CBL was not active at 0800 LST.
Figure 7 presents time–height sections of potential temperature (Fig. 7a) and vertical velocity (Fig. 7b) with the mixing ratio of water vapor (contours in Figs. 7a and 7b) at the center of the model domain and time series of the surface fluxes (Fig. 7c) for the dry case. The large vertical gradient of water vapor (dense contours in Figs. 7a,b) corresponds to the top of the CBL. The sensible heat flux was slightly larger than the latent heat flux (Fig. 7c), and a deep CBL developed rapidly (Figs. 7a,b). The maximum height of the CBL top was about 2400 m, which is only 150 m higher than that of the observed CBL top shown in Fig. 3. The potential temperature (Fig. 7a) did not change with time above the CBL top and, therefore, the free atmosphere remained stably stratified. In the CBL, the potential temperature was uniform with height because of the mixing by updrafts and downdrafts (Fig. 7b), and gradually increased with time.
For the wet case (Fig. 8), a shallow CBL developed slowly (Figs. 8a,b) as a result of the small sensible heat flux and large latent heat flux from the land surface (Fig. 8c). The maximum height of the CBL top was about 1300 m, which is only 100 m lower than that of the observed CBL top shown in Fig. 4. The potential temperature (Fig. 8a) was stratified above the CBL top and was also uniform in the CBL, as in the dry case. The large water vapor in the CBL (Figs. 8a,b) resulted not only from the initial condition, but also from the supply from paddy fields. The gradient of water vapor at the CBL top was larger than that for the dry case.
Figure 9 shows the vertical cross section of vertical velocity (shading) and potential temperature (contours) at 1400 LST, when the vertical circulation was active, for the dry case and the wet case. The CBL top appears as the large vertical gradient of the potential temperature around 2100 m for the dry case and around 900 m for the wet case. In the CBL, the gradient of potential temperature was small, and thermal updrafts and interthermal downdrafts were active. The CBL top was high over the updraft region and low over the downdraft region. The CBL top height fluctuated in a range of about 300 m for both cases. Weak vertical motion of the gravity wave extended above the CBL top (cf. Young et al. 2002), which is also represented in the time-height sections (Figs. 7, 8). This circulation was similar to the WPR results when the circulation was assumed to drift with the mean horizontal wind.
Figure 10 shows the horizontal distribution of vertical velocity, the fluctuations of potential temperature, and the mixing ratio of water vapor at the middle level of the CBL (z/zi ∼ 0.5, where zi is the height of the CBL top) at 1400 LST for the dry case. The fluctuations of potential temperature and water vapor had a similar pattern to that of the vertical velocity (Figs. 10a–c). According to Figs. 10d–f, which show close-up views of the black squares in Figs. 10a–c, the potential temperature and mixing ratio of water vapor had large (small) fluctuations in areas of updrafts (downdrafts). Vertical velocity correlated positively with the fluctuation of potential temperature and water vapor. This means that the circulation transported heat and water vapor. For the wet case (Fig. 11), the potential temperature and mixing ratio of water vapor had smaller horizontal-scale structures. Vertical velocity also had small horizontal scale for the updraft and circulation. In areas of updraft, the fluctuations of water vapor showed positive values similar to those of the dry case; however, the fluctuations of potential temperature tended to be negative. The correlation of vertical velocity and potential temperature was weaker than that of vertical velocity and water vapor and was sometimes negative.
The vertical fluxes of heat, moisture, and buoyancy were calculated from the fluctuations of temperature and water vapor. For compatibility with buoyancy flux, we show the heat flux 
The initial profiles of potential temperature and water vapor were modified by the CBL process. Figure 14 shows the vertical profiles of the simulation and the sounding data for Fuyang at 0800 LST (0000 UTC; initial condition) and 2000 LST (1200 UTC; end of the simulation) in the dry case. The initial conditions of the simulations (broken lines) were given by the sounding at 0800 LST at Fuyang (solid squares); thus, the two profiles are the same. After integration for 12 h, the simulated profiles at 2000 LST (solid lines) had been modified by the vertical flux including the surface flux and the entrainment flux. However, the simulated profiles did not completely correspond to the soundings (open circles). The simulation shows well-mixed profiles while the soundings had a small vertical gradient. For the wet case (Fig. 15), the water vapor in the CBL increased largely as a result of the large water vapor supply from the land surface. The sounding profile showed a larger vertical gradient.
5. Discussion
a. Reproducibility of the simulation
The simulated surface flux, the development of the CBL, and the circulation in the CBL agreed well with the observations, although there were some differences. The starting time of the simulated development of the CBL was 1.5 h later than the observed time for the dry case (Figs. 3, 7). The delayed development of the CBL was due to the use of 0800 LST as the initial time. Sounding data for the initial condition were only available at 0800 LST, but the observed sensible heat flux showed a positive value after 0630 LST, and the observed CBL had already started to develop at 0730 LST (Fig. 3), much earlier than the initial time of 0800 LST used for the simulation.
For the wet case, the CBL started to develop 1 h earlier than the observed time (Figs. 4, 8). The earlier development of the CBL appears to have been caused by the complex land surface process of paddy fields. The simulation did not reproduce the observed late rise of the sensible heat flux. The layer of shallow water that covers the surface of flooded paddy fields has a large heat capacity. Because the surface temperature of such water bodies increases slowly, the rise of the sensible heat flux was delayed. The slow development of the observed sensible heat flux (Fig. 4c) was not reproduced by the simulation (Fig. 8c) using the simple bulk model (Louis et al. 1981). For further improvement of the simulation of the CBL over paddy fields, a land surface scheme that considers the water heat storage of paddy fields (e.g., Kim et al. 2001; Tanaka 2004) is required.
The simulated profiles also did not completely correspond to the soundings. For the dry case (Fig. 14), the simulated profiles showed a well-mixed layer and sharp inversion, while the sounding profiles showed a small gradient. The difference was larger for the wet case (Fig. 15) than for the dry case. The difference can be explained by active entrainment. The simulated entrainment was apparently smaller than the entrainment suggested by the sounding. The vertical gradients of the observed profiles in the wet case are possibly caused by large entrainment flux induced by shallow cumulus clouds. The sounding data imply the appearance of shallow cumulus clouds around Fuyang; however, the cumulus were not found at Shouxian. On the other hand, the simulated CBL was clear (i.e., no cloud) and, however, had abundant water vapor. In this humid condition, a small difference of surface flux exerts an influence on the generation of the cumulus. Another possibility is influence of the descent of the thermal top height (mixing height) in late afternoon. The descent of the mixing height was clearer in observed cases (Figs. 3, 4) than in the simulations (Figs. 7, 8). A fine-gridded simulation dealing with large-scale subsidence is necessary for precise evaluation of the influence.
b. Characteristics of the CBL over a humid terrestrial area
The characteristics of the surface fluxes, the development of the CBL, and the structure of the circulation clearly differed between the wet case and the dry case (Table 4). The sensible heat flux was smaller for the wet case than for the dry case because a large amount of energy was partitioned to latent heat flux. The height of the CBL top mainly depends on surface sensible heat flux (cf. Tennekes 1973). Therefore, the CBL top for the wet case became lower than that for the dry case. The slight sensible and huge latent heat fluxes and the slow development of the low CBL top, due to the presence of surface water, should be a feature of the CBL over humid terrestrial areas. Tanaka et al. (2007) also confirmed the presence of a low CBL top over paddy fields; that study showed seasonal variation in surface conditions and the CBL top height for a period including the IOP-2004.
The amount of surface flux also affects the circulation in the CBL. A forced air mass that is positively buoyant ascends and forms thermal updrafts. The thermal updraft is weaker in the wet case than in the dry case, so the thermal updraft has less buoyancy in the wet case than in the dry case.
The source of the buoyancy was examined using the simulation results. For the dry case, the sensible heat flux from bare fields was slightly larger than the latent heat flux (Fig. 7), and the buoyancy flux largely depended on the heat flux in the CBL (Fig. 12). The relationship between the buoyancy flux and the surface heat flux was similar to that reported for relatively dry terrestrial areas (Stull 1988). Under these conditions, the sensible heat flux from the land surface was much more important than the latent heat flux in the development of the CBL (Tennekes 1973; Kaimal et al. 1976; Stull 1988; Yi et al. 2001).
On the other hand, for the wet case, paddy fields provided small sensible and large latent heat fluxes to the CBL (Fig. 8), and the contribution of moisture to the buoyancy flux was as large as that of heat due to the large supply of water vapor from the paddy fields (Fig. 13). This condition is particular to humid terrestrial areas and differs from that in the dry case in the present study and the relatively dry terrestrial sites of previous studies. In humid terrestrial areas, where the latent heat flux is much larger than the sensible heat flux, the large amount of water vapor is as important as heat in the development of the CBL.
In addition, the contribution of heat decreases with time. Figure 16 shows the time variation of the buoyancy flux and the contributions of heat and moisture in the wet case. Near the surface, the contribution of heat was larger than that of moisture at 1200 LST, nearly equal to it at 1400 LST, and below zero, except near the surface at 1600 LST. The negative value of the heat flux was caused by the entrainment of air from the free atmosphere having high potential temperature. The positive value of the buoyancy flux was yielded only by the contribution of moisture at 1600 LST.
The negative value of the heat flux and the contribution of moisture to most of the buoyancy flux have also been found in CBL processes over the tropical and subtropical oceans (e.g., Nicholls and LeMone 1980). However, the vertical fluxes in the oceanic CBL were one order smaller than those of the wet case, and the oceanic CBL was in a quasi-steady state; there was no diurnal variation such as that found in the wet case in the present study. Therefore, CBL processes over humid terrestrial areas are similar to those over tropical oceanic areas with respect to the large contribution of moisture to buoyancy, but also have diurnal variation caused by the larger surface flux. These characteristics should be applicable to dry (no condensation) CBL over large paddy fields in the eastern part of China, which receive enormous latent heat flux and small sensible heat flux in early summer.
The profiles of the potential temperature and mixing ratio of water vapor were modified by the heat and moisture fluxes as a result of CBL processes (Figs. 14, 15). This humidification by CBL processes was one of the effective factors for the development of the deep convection over the Huaihe River basin (Shinoda and Uyeda 2002). The vertical moisture transport by dry (thermals) and moist (shallow cumulus) convection is important for the maintenance of the moisture field composed of a deep moist layer in this region, and the deep moist layer is an essential source of water vapor for the mei-yu front (Shinoda et al. 2005). Within this regional water circulation, the present study quantitatively describes the dry CBL processes, including the vertical flux and the redistribution of heat and water vapor. The moist CBL processes associated with latent heat release by shallow cumulus convection remain to be discussed. Moist convection is driven by latent heat release, transports water vapor higher, and maintains the deep moist layer. Therefore, as an expansion of the present study, further research on CBL processes should include examination of shallow cumulus convection.
6. Summary and conclusions
Although many past studies have examined the CBL over relatively dry terrestrial areas, few have examined CBL processes in detail over humid terrestrial areas. The aim of this study was to clarify the characteristics of vertical circulation in the CBL, including the vertical fluxes of heat and moisture caused by that circulation, over a humid terrestrial area: the Huaihe River basin of China. Data for early summer 2004 were obtained from flux tower and WPR observations during the IOP-2004 of the LAPS project, and numerical LES using the CReSS model were performed.
The IOP-2004 was divided into two periods according to the condition of the surface vegetation. In the first period, mature wheat fields or bare fields provided almost the same amounts of sensible and latent heat fluxes. In the second period, paddy fields provided small sensible heat flux and enormous latent heat flux. Two clear days (31 May and 22 June 2004) were selected as typical cases for each period and identified as the dry case and the wet case, respectively. For the dry case, the sensible heat flux (390 W m−2 at maximum) was larger than the latent heat flux (300 W m−2 at maximum), although the difference between both was small. The CBL started to develop rapidly from 0730 LST, and the top attained a maximum height of 2250 m. For the wet case, the sensible heat flux was slight (150 W m−2 at maximum), and the latent heat flux was very large (560 W m−2 at maximum). The development of the CBL started slowly at 1000 LST, and the maximum height was 1400 m. The updrafts in the CBL were strong for the dry case (3 m s−1) and weak for the wet case (1.5 m s−1).
The LES was conducted to investigate the three-dimensional structure of the CBL, including the temperature and humidity, and to evaluate the vertical fluxes caused by the circulation in the CBL. The surface heat fluxes and the development of the CBL for both the dry case and the wet case were adequately reproduced. The vertical velocity showed a correlation with potential temperature and the mixing ratio of water vapor, indicating vertical transport of heat and moisture. The vertical heat and moisture fluxes in the CBL were also estimated. For the dry case, heat contributed nearly all of the buoyancy flux, as in previous studies of relatively dry terrestrial areas. For the wet case, on the other hand, the contribution of moisture was equal to that of heat. In particular in late afternoon, the moisture contribution yielded most of the positive value of the buoyancy flux. The results suggest that the large contribution of moisture to the buoyancy is a characteristic of the CBL in humid terrestrial areas. Such a structure of the CBL should be common in the Asian monsoon region.
Acknowledgments
This work was supported by the Core Research for Evolutional Science and Technology (CREST) program of the Japan Science and Technology Agency (JST). The authors are grateful to Sumitomo Electric Industries, Ltd., and Climatec, Inc., for the installation and maintenance of the observation equipment and to Shouxian Meteorological Observatory, Anhui Meteorological Bureau, and the China Meteorological Administration (CMA) for their support in performing the observations. The sounding data were obtained from the University of Wyoming. The authors express their thanks to Mr. A. Sakakibara of Chuden CTI, Ltd., for his help in using the CReSS model and to the Information Technology Center of the University of Tokyo for performing the simulation. We acknowledge Prof. A. Higuchi of Chiba University, Dr. H. Fujinami and Mr. K. Yamamoto of Nagoya University, and Mr. S. Ikeda of Okayama University for conducting the observations, and the members of the LAPS project, Laboratory of Meteorology, and Laboratory of Eco-Hydrometeorology, Hydrospheric Atmospheric Research Center, Nagoya University, for their helpful support and valuable comments. This work was partly supported by the 21st Century COE Program “Dynamics of the Sun-Earth-Life Interactive System” (SELIS) of Nagoya University. Generic Mapping Tools (GMT) software was used to draw the figures.
REFERENCES
Angevine, W. M., A. B. White, and S. K. Avery, 1994: Boundary-layer depth and entrainment zone characterization with a boundary-layer profiler. Bound.-Layer Meteor., 68 , 375–385.
Beyrich, F., and U. Görsdorf, 1995: Composing the diurnal cycle of mixing height from simultaneous sodar and wind profiler measurements. Bound.-Layer Meteor., 76 , 387–394.
Cohn, S. A., and W. M. Angevine, 2000: Boundary layer height and entrainment zone thickness measured by lidars and wind-profiling radars. J. Appl. Meteor., 39 , 1233–1247.
Ecklund, W. L., D. A. Carter, and B. B. Balsley, 1988: A UHF wind profiler for the boundary layer: Brief description and initial results. J. Atmos. Oceanic Technol., 5 , 432–441.
Garratt, J. R., 1992: The Atmospheric Boundary Layer. Cambridge University Press, 316 pp.
Grimsdell, A. W., and W. M. Angevine, 1998: Convective boundary layer height measurement with wind profilers and comparison to cloud base. J. Atmos. Oceanic Technol., 15 , 1331–1338.
Hashiguchi, H., M. D. Yamanaka, T. Tsuda, M. Yamamoto, T. Nakamura, T. Adachi, and S. Fukao, 1995: Diurnal variations of the planetary boundary layer observed with an L-band clear-air Doppler radar. Bound.-Layer Meteor., 74 , 419–424.
Kaimal, J. C., and J. J. Finnigan, 1994: Atmospheric Boundary Layer Flows: Their Structure and Measurement. Oxford University Press, 289 pp.
Kaimal, J. C., J. C. Wyngaard, D. A. Haugen, O. R. Cotè, and Y. Izumi, 1976: Turbulence structure in the convective boundary layer. J. Atmos. Sci., 33 , 2152–2169.
Kim, W., T. Arai, S. Kanae, T. Oki, and K. Musiake, 2001: Application of the Simple Biosphere Model (SiB2) to a paddy field for a period of growing season in GAME-Tropics. J. Meteor. Soc. Japan, 79 , 387–400.
Liu, A. Q., G. W. K. Moore, K. Tsuboki, and I. A. Renfrew, 2006: The effect of the sea-ice zone on the development of boundary-layer roll clouds during cold air outbreaks. Bound.-Layer Meteor., 118 , 557–581.
Louis, J. F., M. Tiedtke, and J. F. Geleyn, 1981: A short history of the operational PBL parameterization at ECMWF. Proc. ECMWF Workshop on Planetary Boundary Layer Parameterization, Reading, United Kingdom, ECMWF, 59–79.
Matsuoka, M., T. Hayasaka, Y. Fukushima, and Y. Honda, 2007: Land cover in East Asia classified using Terra MODIS and DMSP OLS products. Int. J. Remote Sens., 28 , 221–248.
Nicholls, S., and M. A. LeMone, 1980: The fair weather boundary layer in GATE: The relationship of subcloud fluxes and structure to the distribution and enhancement of cumulus clouds. J. Atmos. Sci., 37 , 2051–2067.
Oue, H., 2005: Influence of meteorological and vegetational factors on the partitioning of the energy of a rice paddy field. Hydrol. Processes, 19 , 1567–1583.
Shinoda, T., and H. Uyeda, 2002: Effective factors in the development of deep convective clouds over the wet region of eastern China during the summer monsoon season. J. Meteor. Soc. Japan, 80 , 1395–1414.
Shinoda, T., H. Uyeda, and K. Yoshimura, 2005: Structure of moist layer and sources of water over the southern region far from the Meiyu/Baiu front. J. Meteor. Soc. Japan, 83 , 137–152.
Shusse, Y., and K. Tsuboki, 2006: Dimension characteristics and precipitation efficiency of cumulonimbus clouds in the region far south from the mei-yu front over the eastern Asian continent. Mon. Wea. Rev., 134 , 1942–1953.
Shusse, Y., K. Tsuboki, B. Geng, H. Minda, and T. Takeda, 2005: Structure and evolution of deeply developed convective cells in a long-lived cumulonimbus cloud under a weak vertical wind-shear condition. J. Meteor. Soc. Japan, 83 , 351–371.
Stull, R. B., 1988: An Introduction to Boundary Layer Meteorology. Kluwer Academic, 666 pp.
Tanaka, H., and Coauthors, 2007: Surface flux and atmospheric boundary layer observations from the LAPS project over the middle stream of the Huaihe River Basin in China. Hydrol. Processes, 21 , 1997–2008. doi:10.1002/hyp.6706.
Tanaka, K., 2004: Development of the new land surface scheme SiBUC commonly applicable to basin water management and numerical weather prediction model. Ph.D. dissertation, Kyoto University, 292 pp.
Tennekes, H., 1973: A model for the dynamics of the inversion above a convective boundary layer. J. Atmos. Sci., 30 , 558–567.
Tsuboki, K., and A. Sakakibara, 2002: Large-scale parallel computing of Cloud Resolving Storm Simulator. High Performance Computing, H. P. Zima et al., Eds., Springer, 243–259.
White, A. B., C. W. Fairall, and D. W. Thomson, 1991: Radar observation of humidity variability in and above the marine atmospheric boundary layer. J. Atmos. Oceanic Technol., 8 , 639–658.
Yi, C., K. J. Davis, and B. W. Berger, 2001: Long-term observations of the continental planetary boundary layer. J. Atmos. Sci., 58 , 1288–1299.
Young, G. S., 1988a: Turbulence structure of the convective boundary layer. Part I: Variability of normalized turbulence statistics. J. Atmos. Sci., 45 , 719–726.
Young, G. S., 1988b: Turbulence structure of the convective boundary layer. Part II: PHOENIX 78 aircraft observations of thermals and their environment. J. Atmos. Sci., 45 , 727–735.
Young, G. S., D. A. R. Kristovich, M. R. Hjelmfelt, and R. C. Foster, 2002: Rolls, streets, waves and more: A review of quasi-two-dimensional structures in the atmospheric boundary layer. Bull. Amer. Meteor. Soc., 83 , 997–1001.
Zhao, B., and T. Takeda, 1998: Huaihe River Basin Experiment (GAME/HUBEX). Beijing University Press, 24 pp.
Topography around the observation point, Shouxian (shown by the diamond).
Citation: Journal of Applied Meteorology and Climatology 47, 11; 10.1175/2008JAMC1769.1
Time series of the rain rate at the surface, net radiation, and SHF (black line) and LHF (gray line) observed at 32 m AGL during the IOP from 24 May [day of year (DOY) = 145] to 16 Jul (DOY = 199) 2004. “Wheat,” “Bare,” and “Paddy” indicate the periods when vegetation around the observation site was a mature wheat field, bare field, and paddy field, respectively.
Citation: Journal of Applied Meteorology and Climatology 47, 11; 10.1175/2008JAMC1769.1
(a) Time–height cross section of the SNR on 31 May 2004, referred to as the dry case. The black line indicates the CBL top estimated from the SNR profile at the time. (b) As in (a), but for vertical velocity. (c) Time series of net radiation (black line), SHF (red line), and LHF (blue line).
Citation: Journal of Applied Meteorology and Climatology 47, 11; 10.1175/2008JAMC1769.1
As in Fig. 3, but for 22 Jun 2004, referred to as the wet case.
Citation: Journal of Applied Meteorology and Climatology 47, 11; 10.1175/2008JAMC1769.1
Sounding profiles at Fuyang at 0000 UTC (0800 LST) on 31 May 2004, used as the initial condition for the numerical simulation of the dry case. (a) Profiles of potential temperature (θ, solid line) and virtual potential temperature (θυ, broken line). (b) Profiles of the mixing ratio of water vapor (qυ, solid line) and relative humidity (rh, broken line). (c) Profiles of zonal wind speed (u, solid line) and meridional wind speed (υ, broken line).
Citation: Journal of Applied Meteorology and Climatology 47, 11; 10.1175/2008JAMC1769.1
As in Fig. 5, but for 22 Jun 2004, used as the initial condition for the numerical simulation of the wet case.
Citation: Journal of Applied Meteorology and Climatology 47, 11; 10.1175/2008JAMC1769.1
(a) Time–height cross section of simulated potential temperature (shaded) with the mixing ratio of water vapor (contours, every 1 g kg−1) at the center of the model domain in the dry case (31 May 2004). (b) As in (a), but for vertical velocity (shaded). (c) Time series of net radiation (solid line), sensible heat flux (broken line), and latent heat flux (dotted line) averaged over the entire model surface.
Citation: Journal of Applied Meteorology and Climatology 47, 11; 10.1175/2008JAMC1769.1
As in Fig. 7, but for the wet case (22 Jun 2004).
Citation: Journal of Applied Meteorology and Climatology 47, 11; 10.1175/2008JAMC1769.1
Vertical cross section of simulated vertical velocity (shading) and potential temperature (contours, every 0.5 K) in a plane orthogonal to the mean horizontal wind at 14 LST for the (a) dry case and (b) wet case.
Citation: Journal of Applied Meteorology and Climatology 47, 11; 10.1175/2008JAMC1769.1
Horizontal distribution of (a), (d) simulated vertical velocity, (b), (e) fluctuations of potential temperature, and (c), (f) the mixing ratio of water vapor at a height of 1065 m (z/zi ∼ 0.5, where zi is the height of the CBL top) at 1400 LST in the dry case. Panels (a)–(c) show the whole model domain. Panels (d)–(f) show close-up distributions in the areas outlined by black squares in (a)–(c). The contours in (e) and (f) indicate updraft over 0.5 m s−1 every 1.5 m s−1.
Citation: Journal of Applied Meteorology and Climatology 47, 11; 10.1175/2008JAMC1769.1
As in Fig. 10, but for a height of 465 m (z/zi ∼ 0.5) for the wet case.
Citation: Journal of Applied Meteorology and Climatology 47, 11; 10.1175/2008JAMC1769.1
(a) Profiles of resolved heat flux (solid line) and subgrid-scale (SGS) heat flux (dotted line) at 1400 LST in the dry case. (b) As in (a), but for moisture flux. (c) Profiles of resolved buoyancy flux (solid line) and the contributions of heat (dashed line) and of moisture (dotted line) to buoyancy flux.
Citation: Journal of Applied Meteorology and Climatology 47, 11; 10.1175/2008JAMC1769.1
As in Fig. 12, but for the wet case.
Citation: Journal of Applied Meteorology and Climatology 47, 11; 10.1175/2008JAMC1769.1
(a) Profile of potential temperature obtained by sounding observations at Fuyang at 0800 LST (solid squares) and 2000 LST (open circles) on 31 May 2004 (the dry case) and those by the simulation at 0800 LST (initial condition; broken line) and 2000 LST (solid line). The range between the maximal and minimal values is shaded. (b) As in (a), but for the mixing ratio of water vapor.
Citation: Journal of Applied Meteorology and Climatology 47, 11; 10.1175/2008JAMC1769.1
As in Fig. 14, but for 22 Jun 2004 (the wet case).
Citation: Journal of Applied Meteorology and Climatology 47, 11; 10.1175/2008JAMC1769.1
Profiles of buoyancy flux (solid line) and the contributions of heat (dashed line) and of moisture (dotted line) to buoyancy flux at (a) 1200, (b) 1400, and (c) 1600 LST in the wet case.
Citation: Journal of Applied Meteorology and Climatology 47, 11; 10.1175/2008JAMC1769.1
Operating specifications of the WPR.
Configuration of the numerical simulations.
Land surface parameters for the numerical simulations.
Characteristics of the CBL for the dry case and wet case.
