Effect of Land–Atmosphere Interactions on the IHOP 24–25 May 2002 Convection Case

Teddy R. Holt Marine Meteorology Division, Naval Research Laboratory, Monterey, California

Search for other papers by Teddy R. Holt in
Current site
Google Scholar
PubMed
Close
,
Dev Niyogi Departments of Agronomy and Earth and Atmospheric Sciences, Purdue University, West Lafayette, Indiana

Search for other papers by Dev Niyogi in
Current site
Google Scholar
PubMed
Close
,
Fei Chen National Center for Atmospheric Research, Boulder, Colorado

Search for other papers by Fei Chen in
Current site
Google Scholar
PubMed
Close
,
Kevin Manning National Center for Atmospheric Research, Boulder, Colorado

Search for other papers by Kevin Manning in
Current site
Google Scholar
PubMed
Close
,
Margaret A. LeMone National Center for Atmospheric Research, Boulder, Colorado

Search for other papers by Margaret A. LeMone in
Current site
Google Scholar
PubMed
Close
, and
Aneela Qureshi Department of Marine, Earth, and Atmospheric Sciences, North Carolina State University, Raleigh, North Carolina

Search for other papers by Aneela Qureshi in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

Numerical simulations are conducted using the Coupled Ocean/Atmosphere Mesoscale Prediction System (COAMPS) to investigate the impact of land–vegetation processes on the prediction of mesoscale convection observed on 24–25 May 2002 during the International H2O Project (IHOP_2002). The control COAMPS configuration uses the Weather Research and Forecasting (WRF) model version of the Noah land surface model (LSM) initialized using a high-resolution land surface data assimilation system (HRLDAS). Physically consistent surface fields are ensured by an 18-month spinup time for HRLDAS, and physically consistent mesoscale fields are ensured by a 2-day data assimilation spinup for COAMPS. Sensitivity simulations are performed to assess the impact of land–vegetative processes by 1) replacing the Noah LSM with a simple slab soil model (SLAB), 2) adding a photosynthesis, canopy resistance/transpiration scheme [the gas exchange/photosynthesis-based evapotranspiration model (GEM)] to the Noah LSM, and 3) replacing the HRLDAS soil moisture with the National Centers for Environmental Prediction (NCEP) 40-km Eta Data Assimilation (EDAS) operational soil fields.

CONTROL, EDAS, and GEM develop convection along the dryline and frontal boundaries 2–3 h after observed, with synoptic-scale forcing determining the location and timing. SLAB convection along the boundaries is further delayed, indicating that detailed surface parameterization is necessary for a realistic model forecast. EDAS soils are generally drier and warmer than HRLDAS, resulting in more extensive development of convection along the dryline than for CONTROL. The inclusion of photosynthesis-based evapotranspiration (GEM) improves predictive skill for both air temperature and moisture. Biases in soil moisture and temperature (as well as air temperature and moisture during the prefrontal period) are larger for EDAS than HRLDAS, indicating land–vegetative processes in EDAS are forced by anomalously warmer and drier conditions than observed. Of the four simulations, the errors in SLAB predictions of these quantities are generally the largest.

By adding a sophisticated transpiration model, the atmospheric model is able to better respond to the more detailed representation of soil moisture and temperature. The sensitivity of the synoptically forced convection to soil and vegetative processes including transpiration indicates that detailed representation of land surface processes should be included in weather forecasting models, particularly for severe storm forecasting where local-scale information is important.

Corresponding author address: Dr. Teddy R. Holt, Code 7533, Naval Research Laboratory, Monterey, CA 93943-5502. Email: holt@nrlmry.navy.mil

Abstract

Numerical simulations are conducted using the Coupled Ocean/Atmosphere Mesoscale Prediction System (COAMPS) to investigate the impact of land–vegetation processes on the prediction of mesoscale convection observed on 24–25 May 2002 during the International H2O Project (IHOP_2002). The control COAMPS configuration uses the Weather Research and Forecasting (WRF) model version of the Noah land surface model (LSM) initialized using a high-resolution land surface data assimilation system (HRLDAS). Physically consistent surface fields are ensured by an 18-month spinup time for HRLDAS, and physically consistent mesoscale fields are ensured by a 2-day data assimilation spinup for COAMPS. Sensitivity simulations are performed to assess the impact of land–vegetative processes by 1) replacing the Noah LSM with a simple slab soil model (SLAB), 2) adding a photosynthesis, canopy resistance/transpiration scheme [the gas exchange/photosynthesis-based evapotranspiration model (GEM)] to the Noah LSM, and 3) replacing the HRLDAS soil moisture with the National Centers for Environmental Prediction (NCEP) 40-km Eta Data Assimilation (EDAS) operational soil fields.

CONTROL, EDAS, and GEM develop convection along the dryline and frontal boundaries 2–3 h after observed, with synoptic-scale forcing determining the location and timing. SLAB convection along the boundaries is further delayed, indicating that detailed surface parameterization is necessary for a realistic model forecast. EDAS soils are generally drier and warmer than HRLDAS, resulting in more extensive development of convection along the dryline than for CONTROL. The inclusion of photosynthesis-based evapotranspiration (GEM) improves predictive skill for both air temperature and moisture. Biases in soil moisture and temperature (as well as air temperature and moisture during the prefrontal period) are larger for EDAS than HRLDAS, indicating land–vegetative processes in EDAS are forced by anomalously warmer and drier conditions than observed. Of the four simulations, the errors in SLAB predictions of these quantities are generally the largest.

By adding a sophisticated transpiration model, the atmospheric model is able to better respond to the more detailed representation of soil moisture and temperature. The sensitivity of the synoptically forced convection to soil and vegetative processes including transpiration indicates that detailed representation of land surface processes should be included in weather forecasting models, particularly for severe storm forecasting where local-scale information is important.

Corresponding author address: Dr. Teddy R. Holt, Code 7533, Naval Research Laboratory, Monterey, CA 93943-5502. Email: holt@nrlmry.navy.mil

1. Introduction

The effect of land–vegetative processes and the corresponding dynamical impact on land–atmosphere interactions is investigated for simulations of the 24–25 May mesoscale convection event that was observed during the International H2O Project (IHOP_2002) field experiment (Weckwerth et al. 2004). Land–vegetative processes, as driven by features such as surface heterogeneity (Pielke 2001) or soil moisture gradients (Zhang and Anthes 1982; Segal et al. 1989; Chang and Wetzel 1991; Doran and Zhong 1995) have been shown to be important mechanisms in the development of convection. Chang and Wetzel (1991) and Shaw et al. (1997) show that vegetation gradients can also be important in the formation of drylines, or narrow north–south regions of large horizontal gradients of atmospheric boundary layer (BL) moisture not associated with density gradients (McGuire 1962). Strong gradients in surface fluxes resulting from these inhomogeneities can drive mesoscale circulations along the dryline. Drylines have long been known to be preferential areas of convection initiation (CI) in the southern Great Plains (SGP) region (Rhea 1966; Miller 1967; Schaefer 1986).

The objective of this study is to investigate the sensitivity of land–vegetative processes in a SGP frontal and dryline region. This study deals with the impact of land–atmosphere interactions in strongly forced mesoscale convection, in contrast to previous work, which deals with weakly forced synoptic conditions (e.g., Trier et al. 2004; Clark and Arritt 1995; Segal et al. 1995; Segal and Arritt 1992; Mahfouf et al. 1987; McCumber and Pielke 1981). The synoptic forcing dictates the timing and location of the frontal boundaries on the larger scale. However, the sensitivity of frontal development and propagation to land–vegetative processes in such a scenario is not well known. Section 2 describes the experiment design, including the synoptic scenario and the numerical model simulations. The impact of data assimilation on the initial conditions and forecast characteristics of the front and dryline is discussed in section 3. The results from the sensitivity simulations are given in section 4, and the land–atmosphere interactions are discussed in section 5.

2. Experiment design

a. COAMPS configuration

The atmospheric component of the Naval Research Laboratory's (NRL) Coupled Ocean/Atmosphere Mesoscale Prediction System (COAMPS;1 Hodur 1997; http://www.nrlmry.navy.mil/coamps-web/web/home/) with nonhydrostatic dynamics is used for the numerical model simulations. For this study COAMPS is configured with two one-way interactive nests of 12 km (201 × 181 grid points) over the central United States and 4 km (244 × 247 grid points) centered over the IHOP_2002 observation region (Fig. 1). The emphasis is on the higher-resolution 4-km nest, so all subsequent figures and discussion, with the exception of the synoptic discussion, will pertain to nest 2. The model has 40 vertical sigma-z levels from 10 to 25 790 m, with increased vertical resolution in the lower levels. There are 10 levels below 900 m, with the lowest four levels at 10, 30, 55, and 90 m above ground level (AGL).

Four sets of numerical simulations are conducted in which three key components—the land surface model, the canopy resistance/transpiration formulation, and the soil assimilation system—are varied. Table 1 summarizes the simulations. The control simulation (referred to hereafter as CONTROL) includes the Weather Research and Forecasting (WRF) Noah land surface model (LSM), the WRF canopy resistance formulation (Noilhan and Planton 1989; Jacquemin and Noilhan 1990), and a soil data assimilation system high-resolution land data assimilation system (HRLDAS). The WRF Noah land surface/hydrology model (Pan and Mahrt 1987; Chen et al. 1996; Chen and Dudhia 2001; Ek et al. 2003) is based on the coupling of the diurnally dependent Penman potential evaporation approach of Mahrt and Ek (1984), the multilayer soil model of Mahrt and Pan (1984), and the one-layer canopy model of Pan and Mahrt (1987). The canopy resistance formulation has been extended by Chen et al. (1996) to include the modestly complex Jarvis-type canopy resistance parameterization (Jarvis 1976; Niyogi and Raman 1997).

The HRLDAS (Chen et al. 2004; Trier et al. 2004) uses observation-based analyses to drive the WRF Noah LSM in a decoupled mode on the same grids as in the coupled atmosphere/LSM model configuration (Fig. 1), preventing a mismatch of terrain height, land use, soil texture, LSM climatology, or LSM physics between HRLDAS and the coupled forecast system. For this study the HRLDAS is initialized with data from 0000 UTC 1 January 2001, and run uncoupled with four soil layers (thickness of each layer from the ground surface to the bottom of 0.1, 0.3, 0.6, and 1.0 m, respectively) with a 1-h time step for 18 months, to 24 May 2002 to reach its equilibrium state. The mesoscale variability of vegetation and soil characteristics in the region is illustrated in Fig. 2 showing the COAMPS 4-km vegetation categories from the United States Geological Survey (USGS) 24-category 30-s dataset and the soil texture derived from the U.S. Department of Agriculture 16-category State Soil Geographic Database (STATSGO).

The simulation SLAB examines the sensitivity to the land surface model. It uses a bare ground, slab soil model with a force–restore surface energy budget with predictive equations for surface skin temperature and ground wetness as described in Hodur (1997), and thus has no canopy resistance formulation or soil assimilation. The initial ground moisture availability is estimated from the HRLDAS 10-cm soil moisture to provide similar initial soil conditions as CONTROL.

The simulation EDAS examines the sensitivity to the soil assimilation system. It is the same as CONTROL but replaces the HRLDAS with the coarser 40-km horizontal resolution (but same vertical resolution) National Centers for Environmental Prediction (NCEP) Eta Data Assimilation (EDAS) operational soil temperature and moisture fields. In contrast to several prior synthetic studies on the sensitivity of drylines to soil moisture by uniformly varying the amount to values less than 100% (Grasso 2000; Shaw et al. 1997; Ziegler et al. 1995), this resolution degradation allows a realistic assessment of the impact of high-resolution soil moisture.

The simulation GEM examines the sensitivity to the canopy resistance–transpiration formulation. It is the same as CONTROL but replaces the WRF Noah formulation with a photosynthesis model, the gas exchange/photosynthesis-based evapotranspiration model (GEM) (Niyogi 2000; Niyogi et al. 2004, manuscript submitted to J. Appl. Meteor., hereafter NAR). Canopy resistance is a measure of difficulty for soil moisture to be released to the atmosphere via transpiration, which is one of the most efficient means of water loss from the vegetated land surface.

The canopy resistance of the WRF Noah scheme is a function of minimal stomatal resistance (vegetation-type based), leaf area index (calculated after Walko et al. 2000), and effects of solar radiation, water stress, vapor pressure deficit, and air temperature as defined in Noilhan and Planton (1989). In GEM the vegetation model is based on the Ball–Woodrow–Berry leaf model (Ball et al. 1987; Niyogi and Raman 1997) and the Collatz et al. (1991, 1992) photosynthesis scheme. The GEM canopy resistance is calculated as a function of the net carbon assimilation (photosynthesis) rate, relative humidity, and CO2 concentration at the leaf surface. Physiological variables at the leaf surface in GEM are estimated using transpiration/photosynthesis relationships at the leaf scale, and then scaled up using simple sun-shade and scaling parameterizations as discussed in Campbell and Norman (1998). A photosynthesis-based stomatal resistance scheme such as GEM is expected to be more responsive to atmospheric changes than the WRF Noah scheme, and in turn provide quicker thermodynamic changes in the surface layer (Sellers et al. 1996; Niyogi and Raman 1997; Niyogi et al. 1998; Calvet et al. 1998). This GEM simulation is one of the first tests of the sensitivity of mesoscale convection to a photosynthesis land surface scheme.

b. Data assimilation

An initial 2-day spinup is performed for each of the four simulations. A series of 12-h simulations every 12 h from 0000 UTC 22 May to 1200 UTC 23 May 2002 is performed using intermittent data assimilation in which routinely available observations are blended with model first-guess fields using a multivariate optimum interpolation (MVOI; Barker 1992) scheme after quality control checks (Baker 1992). For the first simulation only (0000 UTC 22 May), initial conditions (i.e., model first-guess fields) are obtained by interpolating the 1° Navy Operational Global Atmospheric Prediction System (NOGAPS) data to the COAMPS domain. Subsequent first-guess fields for all other simulations use the previous COAMPS 12-h forecast. After this spinup period, a 36-h COAMPS simulation for the period of interest from 0000 UTC 24 May 2002 is then performed. Boundary conditions for all simulations are derived from 6-hourly NOGAPS forecasts.

Figure 1b shows the mesonet surface stations used for model validation of low-level temperature, mixing ratio, winds, and solar radiation (see the IHOP data management siteat http://www.joss.ucar.edu/ihop/dm/ for mesonet details). These stations are the 115 Oklahoma Mesonet stations (http://www.mesonet.ou.edu/) (of which 100 stations have soil moisture and temperature data), and the 28 west Texas Mesonet stations (http://www.mesonet.ttu.edu/). These mesonet data are not used in the data assimilation and are thus available for independent verification of the model forecasts.

c. Synoptic scenario

The weather for 24–25 May 2002 over the SGP region is dominated by a slow-moving cold frontal system and upper-level short-wave trough (Fig. 3). The surface front extends from the Kansas–Oklahoma border to the Texas panhandle at 0000 UTC 24 May, with maxima in surface moisture convergence along and just south of the wind shift as shown in the COAMPS analysis (Fig. 3a). The front slowly moves southeastward, reaching the southeast corner of Oklahoma by 1200 UTC. A low pressure center lies over western Texas (1003 hPa at 0000 UTC 24 May and 1007 hPa at 0000 UTC 25 May) (Fig. 3c). As is typical of dryline environments in this region, there is substantial confluence of moist southeasterly flow from the Gulf of Mexico region with southwesterly flow from the drier plateaus of southern New Mexico at levels typically below 700 hPa. Between 0000 UTC 24 May and 0000 UTC 25 May, the 500-hPa short-wave trough moves eastward and amplifies (Figs. 3b,d), tightening the height gradients and increasing the deep layer shear over the Oklahoma–Texas panhandle. The 250-hPa flow (not shown) indicates cyclonic vorticity advection in the same region at 0000 UTC 25 May.

3. Initial conditions and forecasts of the front and dryline

The 2-day data assimilation spinup period prior to the 36-h forecast for each of the four simulations provides initial model conditions that more closely resemble observations than simulations initialized from just an interpolation of larger-scale data. For example, the positive impact of this spinup on the 24 May initial conditions is evident in the CONTROL 0000 UTC surface analysis shown in Fig. 4. The surface boundaries evident in the observations (surface and satellite) include a weakly defined north–south dryline in west Texas and the quasi-stationary east–west front extending through the Oklahoma panhandle northeastward into southern Kansas (Figs. 4a,b). The front is correctly replicated in the CONTROL analysis except for a slight northward displacement in eastern Kansas (Fig. 4c). Likewise, the CONTROL dryline in west Texas is correctly positioned considering the 9 g kg−1 surface mixing ratio contour (Schaefer 1986), with southwesterly flow to the west of the dryline and southerly flow to the east. The modeled northeasterly–southwesterly banded cloud structure and convective cell near Amarillo, Texas, also agrees well with observations.

The southeastward movement of the cold front across Oklahoma from 1800 UTC 24 May to 0600 UTC 25 May is depicted by the solid lines in Fig. 4a estimated from surface observations. The dryline remains in approximately the same location as given in Fig. 4a until 0000 UTC 25 May when the cold front moves far enough south to merge with the dryline. A comparison of model low-level temperatures and moisture to mesonet observations indicates that all the simulations are slow in developing and propagating the front. Figure 5 shows the observed and modeled radar reflectivity (dBZ) valid at 0000 UTC 25 May. The box shows the estimated orientation of the observed cold-frontal precipitation band over Oklahoma and north-central Texas. The simulations with the LSM (CONTROL, GEM, and EDAS) each show a precipitation band oriented similar to observations, but ∼100–200 km to the west and lagging by approximately 2–3 h. The SLAB simulation has not developed convection, indicating that land surface processes reinforce synoptic processes to initiate and propagate the frontal precipitation. The synoptics dictate on the larger scale the timing and location of the front; however, a proper characterization of land surface processes can be crucial in developing the associated convection.

Figure 6 shows the bias and root-mean-square error (rmse) of 2-m air temperature and mixing ratio for the four simulations computed using only Oklahoma Mesonet surface data located ahead of the observed surface frontal position (see Fig. 4a). This region and time period (0900 to 2000 LT) is chosen to isolate the model sensitivity to prefrontal land–atmospheric processes. Both model and mesonet observations are averaged over a 1-h time window centered on the hour. For 2-m temperature statistics (Fig. 6a), all simulations show a warm bias from 15 to 24 h. GEM typically shows the smallest bias and rmse (∼1 and ∼1.5–2.0 K, respectively) and EDAS and SLAB the largest. A positive bias of surface shortwave radiation for each of the simulations (maximum of approximately 200 W m−2 at 2000 UTC) (figure not shown), indicating a general underprediction of clouds, would account for a large portion of the warm bias during the daytime. EDAS and SLAB both show much larger bias and rmse, particularly from ∼19 to 24 h (∼2–3 and ∼2.5–3.2 K, respectively) when convection was prominent along the front. The 2-m mixing ratio dry bias prior to approximately 22 to 24 h (Fig. 6b) corresponds with the low-level warm bias. For each simulation the bias approaches zero and the rmse decreases significantly after ∼26 h when the front and associated convection begins to weaken. There is no clear indication of one simulation consistently performing best; however, for example, the differences in the mean bias and rmse values between the EDAS, SLAB, and GEM and CONTROL are found to be significant at the 95% level at 18, 21, and 24 h using a Wilcoxon signed-rank test (Wilks 1995) on the hourly mean values from each of the mesonet stations. This indicates that each of the sensitivity experiments have different distributions from CONTROL. Generally SLAB is the driest, showing the largest rmse (indicating more large errors), particularly during the afternoon, and GEM is the wettest.

The statistics for the west Texas dryline region given in Fig. 7 for the time period from 1500 UTC 24 May until it was impacted by the front (2100 UTC) show much larger temperature and moisture bias and rmse for SLAB than the other simulations. SLAB has a large cold and moist bias (∼−4°–6°C and ∼3–5 g kg−1, respectively) and much larger rmse. The markedly different statistical characteristics for SLAB indicate the importance of land–vegetative processes in the LSM, even for synoptically driven systems. The impact of land–vegetative processes on model simulations is discussed in section 4.

4. Sensitivity to land–vegetative processes

a. SLAB simulation

The 21-h forecast of low-level moisture and winds shown in Fig. 8 illustrates some differences between SLAB and CONTROL. CONTROL shows a classic contraction of the mixing ratio field with moisture convergence concentrated along the dryline (not shown). Boundary layer depths greater than 2 km AGL are located in the westerly flow behind the dryline and closely mirror the significantly drier land over the elevated terrain of west Texas and eastern New Mexico. For SLAB (Fig. 8b) mixing ratios indicate an area of significant humidity gradient in a similar location to CONTROL, though with a more northeast–southwest orientation, and significantly weaker. The SLAB BL depth is shallower behind the surface moisture gradient, with depths over 2 km AGL confined to northeastern New Mexico and not extending into Texas as in CONTROL. The development of a stronger nocturnal stable layer through the model data assimilation cycle in SLAB limits BL development the following day, as similarly noted in modeling studies of Findell and Eltahir (2003) and Segal et al. (1995), as well as contributes to the daytime cold bias (Fig. 7a). The resulting cooler, shallower BL leads to more convective inhibition (CIN) west of the dryline for SLAB at 2100 UTC. Similarly, convective available potential energy (CAPE) at 2100 UTC is slightly lower for SLAB (∼1900 J kg−1) compared to CONTROL (∼2600 J kg−1).

Figure 9 shows the vertical structure of virtual potential temperature (θυ), mixing ratio (q), and circulation vectors for the two simulations along the east–west cross section A–B (shown in Fig. 8). The coinciding sharp gradients of θυ and q along the dryline have been frequently described as a common characteristic of the dryline (Shaw et al. 1997; Ziegler and Hane 1993; Ogura and Chen 1977). CONTROL shows a strong updraft core (vertical velocity ∼0.5–0.8 m s−1) concentrated at the dryline (x = 260 km on the abscissa), extending as high as 3.5 km AGL. The BL depth across the dryline exhibits the classical east–west gradient, with depths suppressed to the east (x = 260 to 500 km), with values ∼1 km AGL with significant vertical gradients in both θυ and q, and depths to the west (x = 0 to 260 km) up to 2.5 to 3 km AGL. The deeper BL to the west results from more rapid heating and from the resulting thermally direct circulation. The vertical shear associated with strong upper-level (∼4 km AGL) westerlies to the west (x = 0 to 260 km) enhances entrainment and hence also contributes to deepening and drying of the BL.

An elevated region of increased moisture, or “moisture bulge” (after Ziegler et al. 1995), is located in the approximately 50-km-wide zone east of the dryline (x = 260 to 310 km) at heights of ∼1 to 2.5 km AGL for CONTROL. This feature has been previously observed and modeled (Weiss and Bluestein 2002; Atkins et al. 1998; Ziegler et al. 1995) and associated with the “inland sea breeze” circulation. This circulation is characterized by low-level, upslope, moist inflow from the southeast (x = 300–400 km), the strong updraft core that transports moisture aloft, and upper-level return westerly flow. Within this moisture bulge in CONTROL there exists a distinct vertical rotor circulation with a downdraft core approximately 10 km east of the main dryline updraft, similar to that noted by Weiss and Bluestein (2002). The downdraft at x ∼ 300 km delineates the easterly extent of the ∼50 km wide bulge.

The SLAB vertical circulation (Fig. 9b) is markedly different from CONTROL. Though the moisture and temperature gradients are in approximately the same location as CONTROL (x = 260 km), and the updraft strength is comparable, the vertical circulation at the gradient and the moisture bulge east of the dryline are absent. The SLAB BL depth is similar to CONTROL east of the gradient zone (x = 300 to 500 km), but much shallower west of the gradient (x = 0 to 260 km). The BL is moister to the west as compared to CONTROL because of larger surface latent heat fluxes and less entrainment of dry air.

The structure and evolution of the BL responds to changes in fluxes over a period of time, so the 4-h-averaged (17–21 h) latent and sensible heat fluxes along A–B are also shown in Fig. 9. West of the dryline, the largest difference between SLAB and CONTROL is the consistently larger SLAB latent heat fluxes, with values greater than 300 W m−2 as compared to ∼200 W m−2 for CONTROL (from x = 0 to 200 km). Chen et al. (1996) in a comparison of four land evaporation schemes against First International Satellite Land Surface Climatology Project (ISLSCP) Field Experiment (FIFE) data noted that the simple slab model overestimates evaporation during wet periods because of the lack of a canopy resistance to reduce the evaporation to less than the potential rate. This may explain the large moist bias for SLAB shown in Fig. 7b. The BL remains shallow for SLAB to the west of the dryline (Fig. 9b), in spite of larger virtual temperature flux, because of the stronger nocturnal inversion from the previous night and weaker entrainment. These results emphasize the need for physical processes governing moisture in a land surface model via detailed vegetation representation.

b. EDAS simulation

The primary difference between EDAS and CONTROL is in the soil moisture, illustrated in Fig. 10. Allowing for differences in horizontal resolution, EDAS and CONTROL soil moistures are similar in the moister, eastern portion of the domain over eastern Kansas and Oklahoma, but EDAS is much drier than CONTROL over a large region of southwestern Oklahoma, central Texas, and the Texas panhandle. For example, in the Texas panhandle, CONTROL values are ∼0.2 to 0.3, versus 0.1 to 0.2 for EDAS. Subsequently, 10-cm soil temperatures show a corresponding pattern, with EDAS typically 1 to 1.5 K warmer than CONTROL. It should be noted that these differences are also generally true for the initial (0000 UTC 24 May) CONTROL versus EDAS soil moistures and temperatures. The 4-h-averaged (17–21 h) EDAS sensible heat flux is typically larger than CONTROL over the drier soil regions (Fig. 11). The larger sensible heat flux regions correlate spatially with larger BL depth (BL depth 1200-m contour shown in Fig. 11). The region of largest sensible heat flux and deepest BL for EDAS occurs along the region extending south-southwest from the southwestern corner of Oklahoma. This region coincides with the region of enhanced convective development in EDAS as compared to CONTROL (see Fig. 5). The time–height cross section shown in Fig. 12 of θυ and q at location C in Fig. 11b illustrates the differences in the evolution of the BL between CONTROL and EDAS in this region of large fluxes. The BL deepens to over 1300 m for EDAS by 2200 UTC, as compared to only 900 m for CONTROL. This deepening is a direct response to larger sensible heating for EDAS (sensible heat flux ∼650 W m−2 versus 400 W m−2 for CONTROL). The effect of warmer, drier EDAS soil conditions is greater efficacy in developing convection due to more radiation partitioning into sensible versus latent heat flux. The preference for dry soils to enhance convection in a dryline environment with such synoptic conditions as occurred for this case study agrees with results from Findell and Eltahir (2003).

With more extensive development to the southwest, the region of convection for EDAS shows better agreement with observations (Fig. 5); however, closer examination of thermodynamic variables for the prefrontal regions indicates that the warmer and drier EDAS conditions are not as representative of the environment as CONTROL. For example, statistics computed using Oklahoma Mesonet data for regions ahead of the surface front indicate that EDAS soil is much warmer than observations throughout the period (Fig. 13a) with biases 1–2 K and with larger rmse. EDAS soil moistures are drier than observations (Fig. 13b), with a bias ∼−0.02 to −0.03 (approximately 10% of the magnitude of soil moisture), as compared to ∼0.01 to 0.017 for CONTROL from 15 to 24 h. Thus, though there is more convective activity in EDAS, this agreement with observations is fortuitously aided by the anomalously warm and dry soil. The HRLDAS assimilation of CONTROL provides a more realistic depiction of soil conditions than EDAS and hence better overall performance as indicated by the temperatures and moisture statistics for the region (see Fig. 6).

c. GEM simulation

The primary difference between CONTROL and GEM is illustrated in the 21-h forecast of canopy resistance to water vapor exchange (Fig. 14). Resistances are generally higher in the western part of the domain, which is dominated by grassland and shrubs. The corresponding transpiration rates generally coincide with resistance variations, with lower resistances resulting in higher transpiration. The GEM resistances vary with vegetation (Fig. 2a), particularly as regards to C3 (grasses and trees) versus C4 (certain grasses and crops) photosynthesis types. The CONTROL resistances show much less spatial variability and less correlation with vegetation variability. This is expected to contribute to much stronger land–vegetative interaction for the GEM resistance scheme. Photosynthesis-scheme-based resistance formulations, such as GEM, are generally more responsive to vegetation type, atmospheric conditions, and soil state.

The differences in GEM and CONTROL resistances affect the model surface heat fluxes via transpiration changes. Figure 15 shows the CONTROL and GEM 4-h-averaged (17–21 h) latent heat fluxes. The largest fluxes for both occur in central Oklahoma southward to northeastern Texas where moisture is large and low-level winds are strong from the south (∼7–8 m s−1; see Fig. 8a). Latent heat fluxes for GEM (Fig. 15b) correlate spatially with low-resistance areas (Fig. 14b), as well as regions of larger 2-m mixing ratio (indicated by the contour lines in Fig. 15b). In contrast, CONTROL fluxes are less than GEM, less spatially correlated to the canopy resistances, and show less relationship to low-level moisture. Differences computed from 2-m hourly averaged model and observed mixing ratios (used in statistics given in Fig. 6b) are also shown for five Oklahoma Mesonet stations representative of CONTROL and GEM differences (solid circles in Fig. 15). The moisture difference is less for each of the stations for GEM versus CONTROL (reflected in the reduced bias in Fig. 6b). Regions in central and western Oklahoma show the largest improvement (differences reduced ∼1 g kg−1), where fluxes are generally larger for GEM than CONTROL, with much more spatial variability as represented in the natural variability of the vegetation.

5. Discussion

To develop a better understanding of how land–vegetative processes impact the simulation of the dynamical response, model results are examined in light of frontogenetic forcing. This is accomplished using the frontogenesis function as originally proposed by Miller (1948) and used subsequently by others (Sanders 1955; Ziegler et al. 1995). The mixing ratio forcing (Fq) is considered here because of its importance in defining the dryline:
i1520-0493-134-1-113-e1
where the time tendency has been neglected following Ziegler et al. (1995) because the soil moisture gradients are on a larger scale than the dryline. The first two terms on the right-hand side of (1) represent the horizontal deformation (Fhdef), and the third is the tilting term (Ftilt).

Figure 16 shows the two forcing terms for CONTROL, SLAB, GEM, and EDAS for the cross section along A–B in Fig. 8. Forcing for LSM simulations (CONTROL, GEM, and EDAS) shows that horizontal deformation (Fhdef) strongly increases the low-level moisture gradient at the dryline (x = 260 km) up to a height ranging from 2 to 2.5 km AGL (Fig. 16, top panels). The maximum horizontal BL wind along the cross section (shown schematically in top panels) is ±3 to 4 m s−1 concentrated below 1 km AGL within 100 km either side of the dryline for the LSM simulations. The contribution from tilting is also largest for the LSM simulations, concentrated at elevations ∼1–3 km AGL (Fig. 16, middle). The EDAS simulation has the largest tilting-based forcing in connection with stronger differential vertical velocity (strongest updrafts ∼1.1 m s−1 and downdrafts ∼−0.4 m s−1). The elevated frontogenetic/frontolytic tilting–forcing couplet from x = 260 to 310 km for the LSM simulations coincides with the elevated moisture bulge discussed previously, indicating the importance of the updraft/downdraft couplet in turning the vapor gradient into the vertical and maintaining the sharp gradients. The resulting total frontogenesis for CONTROL, GEM, and EDAS (Fig. 16, bottom) shows strong boundary layer frontogenetic forcing due to convergence along the dryline, resulting in the significant scale contraction evident in the moisture gradient. The tilting term dominates total forcing at and above the BL top, with the only other significant forcing outside the 50-km dryline zone evident due to tilting at the BL top.

The SLAB simulation is significantly different in its frontogenetic characteristics. At the moisture gradient at x = 260 km, Fhdef is virtually nonexistent in the BL and only weakly frontogenetic at the BL top (∼1 km AGL), several orders of magnitude less than CONTROL. Correspondingly, there is little BL convergence with regions of maximum BL wind displaced more than 100 km from the dryline and weaker (∼2 to 3 m s−1) than the LSM simulations. Here Ftilt is at least one order of magnitude smaller than CONTROL, concentrated at the BL top, and strongest at the dryline and to the west (x = 75 to 200 km), but with much less vertical extent than in CONTROL due to the lack of an elevated moisture bulge. Likewise the differential vertical velocity is reduced compared to LSM simulations. The resulting SLAB total forcing resembles the nonclassical dryline characteristics described in Segal and Arritt (1992). Neither convergent forcing nor tilting is present to maintain a sharp moisture gradient.

Figure 17 illustrates the land–atmosphere feedback of EDAS, GEM, and SLAB relative to CONTROL averaged over a 1-h period from 2000 to 2100 UTC for a 60 km × 60 km region near Shamrock, Texas (box S in Fig. 8a). The purpose of this comparison is to emphasize the relative importance of land–vegetative processes under similar synoptic forcing (i.e., clear sky and prefrontal). For the comparison time and for all simulations, this region is south of the front and there is no precipitation. The surface is a mixture of grassland and bare ground. While some of the differences may still be a function of synoptic situation, careful choice of the area should ensure that most of the differences are related to differences in the treatment of the surface variables.

As shown in Fig. 17, the canopy resistance for GEM is almost 500% larger than CONTROL and almost 50% larger for EDAS (SLAB does not include vegetation response). Larger vegetation resistances correlate directly with less transpiration and indirectly with soil temperature/moisture changes. In GEM the larger canopy resistance reduces transpiration, reducing the release of moisture to the atmosphere and thus increasing soil moisture. For EDAS the soil moisture/soil temperature change can be considered as the direct effect and the associated changes in the transpiration/canopy resistance as feedback. EDAS soil moisture is 7% less than CONTROL, which contributes to warmer soil (via emissivity and albedo feedbacks in the model parameterization). The combined effect of the soil moisture/soil temperature and canopy resistance changes contributes to the overall reduction in transpiration. For this region, the small vegetation cover (42%) reduces the importance of transpiration relative to evaporation from the soil. Thus, for GEM the nearly 60% reduction in transpiration translates to only a 36% reduction in latent heat flux. However, the 6% reduction in EDAS transpiration corresponds to a 7% reduction in latent heat flux because of the additional effect of lowered soil moisture. For SLAB the latent heat flux changes very little (1%) compared to CONTROL. For sensible heat flux the most significant change is the almost 25% increase for GEM.

The changes in the surface layer response propagate into the BL. For GEM and EDAS the 2-m air temperatures are much warmer and the mixing ratios much drier than CONTROL (0.3 K and 0.4 K %, an increase of ∼1 K; −12% and −11%, a decrease of ∼1.5 g kg−1). The SLAB air temperature is cooler by 0.43 K % (more than 1 K) in response to a shallower BL depth (by 36%) that entrains less warm air from above. The BL is deeper by 5%–10% in GEM and EDAS, with warming and drying near the surface correlated with a deeper BL. Mesoscale BL vertical velocities show the largest reduction for SLAB (43%), supporting a generally shallower, moister, and less energetic BL. CAPE is also the smallest for SLAB (−28%), though each of the simulations has sufficient CAPE to support convection (values range form 1900 to 2600 J kg−1), owing to the synoptic forcing. It is the increase in CIN for SLAB (113%) that precludes the development of convection. The LSM simulations show similar values of CIN, with GEM the smallest. Boundary layer cloud cover is less than 4% coverage for all simulations for this relatively cloud free, prefrontal region.

In summary, changes in canopy resistance affect transpiration, which in turn modulates the water loss from the surface (and hence soil moisture). The changes in soil moisture affect the emissivity and albedo and can impact soil temperature. Indeed, the canopy resistance and transpiration depend on the soil temperature and soil moisture. As discussed in Niyogi et al. (2002), the soil–vegetation coupling is relative to the soil moisture availability (i.e., larger soil moisture availability results in greater interaction between vegetation and soil, and hence systematic transpiration and soil moisture changes). Changes in the surface characteristics alter the surface fluxes for sensible and latent heat. This in turn modifies the air temperature and moisture content of the surface layer/lower boundary layer. The response propagates upward through the boundary layer via turbulent transport and affects boundary layer growth. In turn, the growth of the boundary layer leads to engulfment of warm and dry air above the boundary layer and at the same time dilutes the effect of surface heating (since the same amount of heating spreads through a larger depth; see, e.g., LeMone et al. 2000). These processes change the CAPE and CIN, and in conjunction with other mesoscale feedbacks contribute to the frontogenetic forcing, boundary layer clouds, and eventually timing and amount of precipitation at a particular location.

This analysis supports the premise that including advanced photosynthetic processes in a mesoscale model will produce stronger coupling between the surface and the overlying surface and boundary layers. On the other hand, EDAS results are responsive to changes in the soil moisture but these changes do not necessarily produce as strong a boundary layer response. The EDAS surface feedback is somewhat limited because even though the surface has altered soil moisture/soil temperature, the response to the surface layer and the atmosphere has to be through the vegetation/transpiration scheme, which in this case is relatively less interactive with fewer variables as compared to the photosynthesis-based GEM.

6. Conclusions

Numerical model simulations are conducted to understand the effect of land–atmosphere interactions on a mesoscale convective event over the southern Great Plains during IHOP_2002 characterized by strong dryline synoptic forcing in conjunction with a quasi-stationary cold-frontal system. Variations to the specification of land surface model (LSM), canopy resistance formulation, and type and resolution of soil assimilation system are examined. Each of the LSM simulations develops convection 2–3 h after observed, as synoptic-scale forcing determines the location and timing of the frontal boundaries on the large scale. Simulations with the LSM develop convection in approximately the correct location and much earlier than the simulation using a simpler slab surface model. The slab model also has larger low-level temperature and moisture biases and root-mean-square errors computed from mesonet data over Oklahoma and west Texas. Thus, the physical parameterizations in the slab model are insufficient to properly account for land–vegetative processes such as those occurring along the dryline and frontal boundaries in this case.

Coarser-resolution soil data (EDAS) that is generally drier and warmer than the high-resolution land data assimilation system (HRLDAS) used in the control simulation provides an environment more conducive for convection (larger CAPE and less CIN). Thus, the development of convection in association with the dryline is typically more extensive for EDAS than CONTROL. However, statistics computed from mesonet data show soil moisture and temperature biases (as well as air temperature and moisture during the prefrontal period) are larger using the coarser-resolution EDAS data compared to HRLDAS. Thus, land–vegetative processes in EDAS are forced by anomalously warmer and drier conditions than observed.

An advanced representation of photosynthesis-based evapotranspiration shows improvements in predictive skill for 2-m air temperature and moisture. This is because model soil moisture changes by themselves (such as those tested by using a different soil assimilation system like EDAS) do not directly affect the coupled land–atmosphere response. Rather, the atmosphere responds to changes in soil moisture via latent heat flux, boundary layer growth, heating/cooling, CIN, and CAPE. This manifestation of the changes in the surface/subsurface details on the soil moisture/temperature is more effectively achieved by enhancing the vegetation/transpiration scheme (as in GEM). This is because transpiration is the most efficient means of water vapor exchange from the surface to the atmosphere.

Acknowledgments

The research was supported by the Program Element 0602435N of the Naval Research Laboratory Base Program Project Number BE-435-003; NSF-ATM 0233780 (Dr. S. Nelson), the NASA–THP (NNG04GI84G, Dr. J. Entin), and the NASA-IDS (NNG04GL61G, Drs. J. Entgin and G. Gutman). The IHOP_2002 data collection and processing were supported by the NCAR Water Cycle Initiatives and by NSF/NCAR USWRP funds. The first author would like to thank James Doyle for many fruitful discussions and suggestions for improving the manuscript. Also thanks to Joseph Alfieri and Steve Williams for help in processing IHOP_2002 data and Dr. Stan Trier for an insightful review.

REFERENCES

  • Atkins, N. T., R. M. Wakimoto, and C. L. Ziegler, 1998: Observations of the finescale structure of a dryline during VORTEX95. Mon. Wea. Rev, 126 , 525550.

    • Search Google Scholar
    • Export Citation
  • Baker, N. L., 1992: Quality control for the navy operational atmospheric database. Wea. Forecasting, 7 , 250261.

  • Ball, J., I. Woodrow, and J. Berry, 1987: A model predicting stomatal resistance and its contribution to the control of photosynthesis under different environmental conditions. Progress in Photosynthesis Research, J. Biggins, Ed., Vol. IV, Martinus Nijhoff, 221–224.

    • Search Google Scholar
    • Export Citation
  • Barker, E. H., 1992: Design of the navy's multivariate optimum interpolation analysis system. Wea. Forecasting, 7 , 220231.

  • Calvet, J-C., J. Noilhan, J. Roujean, P. Bessemoulin, M. Cabelguenne, A. Olioso, and J. Wigneron, 1998: An interactive vegetation SVAT model tested against data from six contrasting sites. Agric. For. Meteor, 92 , 7395.

    • Search Google Scholar
    • Export Citation
  • Campbell, G. S., and J. M. Norman, 1998: An Introduction to Environmental Biophysics. 2d ed. Springer, 312 pp.

  • Chang, J-T., and P. J. Wetzel, 1991: Effects of spatial variations of soil moisture and vegetation on the evolution of a prestorm environment: A case study. Mon. Wea. Rev, 119 , 13681390.

    • Search Google Scholar
    • Export Citation
  • Chen, F., and J. Dudhia, 2001: Coupling an advanced land surface/hydrology model with the Penn State–NCAR MM5 modeling system. Part I: Model implementation and sensitivity. Mon. Wea. Rev, 129 , 569585.

    • Search Google Scholar
    • Export Citation
  • Chen, F., and Coauthors, 1996: Modeling of land-surface evaporation by four schemes and comparison with FIFE observations. J. Geophys. Res, 101 , 72517268.

    • Search Google Scholar
    • Export Citation
  • Chen, F., K. W. Manning, D. N. Yates, M. A. LeMone, S. B. Trier, R. Cuenca, and D. Niyogi, 2004: Development of a High Resolution Land Data Assimilation System (HRLDAS). Preprints, 16th Conf. on Numerical Weather Prediction, Seattle, WA, Amer. Meteor. Soc., CD-ROM, 22.3.

  • Clark, C. A., and R. W. Arritt, 1995: Numerical simulations of the effect of soil moisture and vegetation cover on the development of deep convection. J. Appl. Meteor, 34 , 20292045.

    • Search Google Scholar
    • Export Citation
  • Collatz, G. J., J. Ball, C. Grivet, and J. Berry, 1991: Physiological and environmental regulation of stomatal conductance, photosynthesis, and transpiration: A model that includes a laminar boundary layer. Agric. For. Meteor, 54 , 107136.

    • Search Google Scholar
    • Export Citation
  • Collatz, G. J., M. Ribas-Carbo, and J. Berry, 1992: Coupled photosynthesis–stomatal conductance model for leaves of C4 plants. Aust. J. Plant Physiol, 19 , 519538.

    • Search Google Scholar
    • Export Citation
  • Doran, J. C., and S. Zhong, 1995: Variations in mixed-layer depths arising from inhomogeneous surface conditions. J. Climate, 8 , 19651973.

    • Search Google Scholar
    • Export Citation
  • Ek, M. B., K. E. Mitchell, Y. Lin, E. Rogers, P. Grunmann, V. Koren, G. Gayno, and J. D. Tarpley, 2003: Implementation of Noah land surface model advances in the National Centers for Environmental Prediction operational mesoscale Eta Model. J. Geophys. Res, 108 .8851, doi:10.1029/2002JD003296.

    • Search Google Scholar
    • Export Citation
  • Findell, K. L., and E. A. B. Eltahir, 2003: Atmospheric controls on soil moisture–boundary layer interactions. Part II: Feedbacks within the continental United States. J. Hydrometeor, 4 , 570583.

    • Search Google Scholar
    • Export Citation
  • Grasso, L. D., 2000: A numerical simulation of dryline sensitivity to soil moisture. Mon. Wea. Rev, 128 , 28162834.

  • Hodur, R. M., 1997: The Naval Research Laboratory's Coupled Ocean/Atmosphere Mesoscale Prediction System (COAMPS). Mon. Wea. Rev, 125 , 14141430.

    • Search Google Scholar
    • Export Citation
  • Jacquemin, B., and J. Noilhan, 1990: Sensitivity study and validation of a land surface parameterization using the HAPEX-MOBILHY data set. Bound.-Layer Meteor, 52 , 93134.

    • Search Google Scholar
    • Export Citation
  • Jarvis, P. G., 1976: The interpretation of the variations in leaf water potential and stomatal conductance found in canopies in the field. Philos. Trans. Roy. Soc. London, B273 , 593610.

    • Search Google Scholar
    • Export Citation
  • LeMone, M. A., and Coauthors, 2000: Land–atmosphere interaction research, early results, and opportunities in the Walnut River Watershed in southeast Kansas: CASES and ABLE. Bull. Amer. Meteor. Soc, 81 , 757779.

    • Search Google Scholar
    • Export Citation
  • Mahfouf, J-F., E. Richard, and P. Mascart, 1987: The influence of soil and vegetation on the development of mesoscale circulations. J. Climate Appl. Meteor, 26 , 14831495.

    • Search Google Scholar
    • Export Citation
  • Mahrt, L., and M. Ek, 1984: The influence of atmospheric stability on potential evaporation. J. Climate Appl. Meteor, 23 , 222234.

  • Mahrt, L., and H. L. Pan, 1984: A two-layer model of soil hydrology. Bound.-Layer Meteor, 29 , 120.

  • McCumber, M. C., and R. A. Pielke, 1981: Simulation of the effects of surface fluxes of heat and moisture in a mesoscale numerical model. Part I: Soil layer. J. Geophys. Res, 86 , 99299938.

    • Search Google Scholar
    • Export Citation
  • McGuire, E. L., 1962: The vertical structure of three drylines as revealed by aircraft traverses. National Severe Storms Project Rep. 7, 11 pp. [Available from NCAR, P.O. Box 3000, Boulder, CO 80307.].

  • Miller, J. E., 1948: On the concept of frontogenesis. J. Meteor, 5 , 169171.

  • Miller, R. C., 1967: Notes on analysis and severe-storm forecasting procedures of the Military Weather Warning Center Tech. Rep. 200, U.S. Air Force Air Weather Service, Scott Air Force Base, IL, 170 pp.

  • Niyogi, D. S., 2000: Biosphere–atmosphere interactions coupled with carbon dioxide and soil moisture changes. Ph.D. dissertation, North Carolina State University, 509 pp.

  • Niyogi, D. S., and S. Raman, 1997: Comparison of four different stomatal resistance schemes using FIFE observations. J. Appl. Meteor, 36 , 903917.

    • Search Google Scholar
    • Export Citation
  • Niyogi, D. S., K. Alapaty, and S. Raman, 1998: Comparison of four different stomatal resistance schemes using FIFE observations. Part II: Analysis of terrestrial biospheric–atmospheric interactions. J. Appl. Meteor, 37 , 13011320.

    • Search Google Scholar
    • Export Citation
  • Niyogi, D. S., Y-K. Xue, and S. Raman, 2002: Hydrological land surface response in a tropical regime and a midlatitudinal regime. J. Hydrometeor, 3 , 3956.

    • Search Google Scholar
    • Export Citation
  • Noilhan, J., and S. Planton, 1989: A simple parameterization of land surface processes for meteorological models. Mon. Wea. Rev, 117 , 536549.

    • Search Google Scholar
    • Export Citation
  • Ogura, Y., and Y. Chen, 1977: A life history of an intense mesoscale convective storm in Oklahoma. J. Atmos. Sci, 34 , 14581476.

  • Pan, H-L., and L. Mahrt, 1987: Interaction between soil hydrology and boundary-layer development. Bound.-Layer Meteor, 38 , 185202.

  • Pielke, R. A., 2001: Influence of the spatial distribution of vegetation and soils on the prediction of cumulus convective rainfall. Rev. Geophys, 39 , 151177.

    • Search Google Scholar
    • Export Citation
  • Rhea, J. O., 1966: A study of thunderstorm formation along drylines. J. Appl. Meteor, 5 , 5863.

  • Sanders, F., 1955: An investigation of the structure and dynamics of an intense surface frontal zone. J. Meteor, 12 , 542552.

  • Schaefer, J. T., 1986: The dryline. Mesoscale Meteorology and Forecasting, P. S. Ray, Ed., Amer. Meteor. Soc., 549–570.

  • Segal, M., and R. W. Arritt, 1992: Nonclassical mesoscale circulations caused by surface sensible heat flux gradients. Bull. Amer. Meteor. Soc, 73 , 15931604.

    • Search Google Scholar
    • Export Citation
  • Segal, M., W. E. Schreiber, G. Kallos, J. R. Garratt, A. Rodi, J. Weaver, and R. A. Pielke, 1989: The impact of crop areas in northeast Colorado on midsummer mesoscale thermal circulations. Mon. Wea. Rev, 117 , 809825.

    • Search Google Scholar
    • Export Citation
  • Segal, M., R. Arritt, C. Clark, R. Rabin, and J. Brown, 1995: Scaling evaluation of the effect of surface characteristics on potential for deep convection over uniform terrain. Mon. Wea. Rev, 123 , 383400.

    • Search Google Scholar
    • Export Citation
  • Sellers, P., S. O. Los, C. J. Tucker, C. O. Justice, D. A. Dazlich, G. J. Collatz, and D. A. Randall, 1996: A revised land surface parameterization (SiB2) for atmospheric GCMs. Part II: The generation of global fields of terrestrial biophysical parameters from satellite data. J. Climate, 9 , 706737.

    • Search Google Scholar
    • Export Citation
  • Shaw, B. L., R. A. Pielke, and C. L. Ziegler, 1997: A three-dimensional numerical simulation of a Great Plains dryline. Mon. Wea. Rev, 125 , 14891506.

    • Search Google Scholar
    • Export Citation
  • Trier, S. B., F. Chen, and K. W. Manning, 2004: A study of convection initiation in a mesoscale model using high-resolution land surface initial conditions. Mon. Wea. Rev, 132 , 29542976.

    • Search Google Scholar
    • Export Citation
  • Walko, R. L., and Coauthors, 2000: Coupled atmosphere–biophysics–hydrology models for environmental modeling. J. Appl. Meteor, 39 , 931944.

    • Search Google Scholar
    • Export Citation
  • Weckwerth, T. M., and Coauthors, 2004: An overview of the International H2O Project (IHOP_2002) and some preliminary highlights. Bull. Amer. Meteor. Soc, 85 , 253277.

    • Search Google Scholar
    • Export Citation
  • Weiss, C. C., and H. B. Bluestein, 2002: Airborne pseudo–dual Doppler analysis of a dryline–outflow boundary intersection. Mon. Wea. Rev, 130 , 12071226.

    • Search Google Scholar
    • Export Citation
  • Wilks, D. S., 1995: Statistical Methods in the Atmospheric Sciences. Academic Press, 467 pp.

  • Zhang, D., and R. A. Anthes, 1982: A high-resolution model of the planetary boundary layer—Sensitivity tests and comparison with SESAME-79 data. J. Appl. Meteor, 21 , 15941609.

    • Search Google Scholar
    • Export Citation
  • Ziegler, C. L., and C. E. Hane, 1993: An observational study of the dryline. Mon. Wea. Rev, 121 , 11341151.

  • Ziegler, C. L., W. J. Martin, R. A. Pielke, and R. L. Walko, 1995: A modeling study of the dryline. J. Atmos. Sci, 52 , 263285.

Fig. 1.
Fig. 1.

COAMPS model domain for (a) 12-km outer nest and the location of 4-km inner nest, and (b) inner nest terrain (shaded, interval of 200 m) and locations of mesonet surface observations (dots) for Oklahoma and west Texas. Amarillo and Shamrock, TX, are indicated by the A and S, respectively

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3057.1

Fig. 2.
Fig. 2.

COAMPS nest-2 static surface fields of (a) 24-category vegetation and (b) 16-category soil types

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3057.1

Fig. 3.
Fig. 3.

COAMPS 12-km analysis valid at 0000 UTC 24 May 2002 of (a) sea level pressure (interval 4 hPa), 10-m wind barbs (full barb = 5 m s−1), regions of surface moisture convergence greater than 7.5 × 10−7 g kg−1 s−1 (shaded), and estimated surface frontal position; (b) 500-hPa geopotential heights (interval 30 m) and wind barbs, and regions of 700-hPa moisture convergence greater than 7.5 × 10−7 g kg−1 s−1 (shaded); (c) and (d) same as (a) and (b) except for the COAMPS 24-h forecast valid at 0000 UTC 25 May 2002

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3057.1

Fig. 4.
Fig. 4.

(a) IHOP surface analysis valid at 0000 UTC 24 May 2002 over the COAMPS domain, (b) Geostationary Operational Environmental Satellite-8 (GOES-8) visible 1-km-resolution satellite image for 0008 UTC 24 May 2002, and (c) CONTROL analysis at 0000 UTC 24 May of 10-m wind barbs (full barb = 5 m s−1), 10-m mixing ratio (contour interval = 1 g kg−1), and vertically integrated total cloud fraction (shaded). The subjective locations of the cold front and dryline (dashed line) are also given. The dashed box in (c) is the location of the satellite image (b) on the COAMPS domain. The estimated observed surface frontal positions at 1800 UTC 24 May (label 18/24 May), 0000 UTC (label 00/25 May), and 0600 UTC (label 06/25 May) 25 May are given by the solid lines in (a)

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3057.1

Fig. 5.
Fig. 5.

Radar reflectivity (dBZ) valid at 0002 UTC 25 May for 2-km observations and corresponding 24-h COAMPS forecasts valid 0000 UTC 25 May. The box shows the estimated orientation of the observed cold-frontal precipitation band over Oklahoma and north-central Texas

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3057.1

Fig. 6.
Fig. 6.

Time series of (a) 2-m air temperature (°C) and (b) mixing ratio (g kg−1) statistics from 1500 UTC 24 May to 0600 UTC 25 May computed using the Oklahoma surface mesonet stations shown in Fig. 1b. The statistics are computed in the prefrontal region as shown in Fig. 4a

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3057.1

Fig. 7.
Fig. 7.

Time series of statistics similar to Fig. 6, except using the west Texas Mesonetstations from 1500 to 2100 UTC 24 May in the area of the dryline before passage of the front

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3057.1

Fig. 8.
Fig. 8.

The 21-h forecast valid 2100 UTC 24 May of 2-m mixing ratio (contour interval 2 g kg−1), 10-m winds (arrows every seventh grid point, m s−1), and boundary layer depth AGL (m, shaded) for (a) CONTROL and (b) SLAB. Cross section A–B is also indicated along with box S used for averaged fields given in Fig. 17

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3057.1

Fig. 9.
Fig. 9.

Vertical cross section of 21-h forecast valid at 2100 UTC 24 May along line A–B given in Fig. 8 of mixing ratio (g kg−1, shaded), virtual potential temperature (bold line, interval of 1 K), and vertical circulation (arrows) for (a) CONTROL and (b) SLAB. The 17–21-h average surface fluxes (W m−2) of sensible heat (CONTROL: bold dots; SLAB: bold line) and latent heat (CONTROL: thin dots; SLAB: thin line) along the cross section are also given

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3057.1

Fig. 10.
Fig. 10.

The 21-h forecast of 10-cm volumetric soil moisture valid at 2100 UTC 24 May for (a) CONTROL and (b) EDAS

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3057.1

Fig. 11.
Fig. 11.

Averaged 17–21-h sensible heat flux (W m−2) for (a) CONTROL and (b) EDAS. The contour is the BL depth (1200-m contour only). Point C is the location for the time–height cross section shown in Fig. 12

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3057.1

Fig. 12.
Fig. 12.

Time–height (AGL) cross section at point C shown in Fig. 11 in north-central Texas of virtual potential temperature (shaded) and mixing ratio (solid lines) for (a) CONTROL and (b) EDAS. The surface heat fluxes during the daytime (15–24 h) illustrate the dominance of sensible to latent for EDAS due to the drier soil conditions (EDAS 10-cm soil moisture ∼0.148 vs 0.239 for CONTROL). The evolution of the BL depth indicated by the dashed line likewise indicates more rapid deepening for EDAS

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3057.1

Fig. 13.
Fig. 13.

Time series of (a) 10-cm soil temperature (°C) and (b) 10-cm soil moisture (volumetric fraction) statistics from 1500 UTC 24 May to 0600 UTC 25 May computed using prefrontal Oklahoma surface mesonet stations similar to Fig. 6

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3057.1

Fig. 14.
Fig. 14.

The 21-h forecast valid 2100 UTC 24 May of canopy resistance (s m−1) for (a) CONTROL and (b) GEM

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3057.1

Fig. 15.
Fig. 15.

Averaged 17–21-h latent heat flux (W m−2) for (a) CONTROL and (b) GEM. The contours are for 9 (dashed) and 13 g kg−1 (solid) 2-m mixing ratio. Differences (model − obs) of mixing ratio for five selected mesonet stations are shown (solid circles) to illustrate the impact of latent heat flux differences

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3057.1

Fig. 16.
Fig. 16.

Vertical cross section of 2000–2100 UTC 24 May averaged forcing terms of mixing ratio frontogenesis (×10−7 g kg m−1 s−1) of (top) horizontal deformation, (middle) tilting, and (bottom) total for (a) CONTROL, (b) SLAB, (c) GEM, and (d) EDAS along line A–B. Dark regions are frontogenetic areas and light regions are frontolytic. Contours are maximum horizontal boundary layer wind (m s−1) along the cross section in (a), maximum vertical velocity (m s−1) in (b), and mixing ratio (interval of 2 g kg−1) in (c)

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3057.1

Fig. 16.
Fig. 16.

(Continued)

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3057.1

Fig. 17.
Fig. 17.

Percent change of quantities for simulations GEM, EDAS, and SLAB relative to CONTROL values averaged from 2000 to 2100 UTC 24 May 2002 over the 60 km × 60 km Shamrock, TX, subset region (given by box S in Fig. 8a). Positive percent changes indicate an increase relative to the CONTROL values. This time is considered prefrontal, with no precipitation for any simulation from 2000 to 2100 UTC

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3057.1

Table 1.

Description of COAMPS model simulations

Table 1.

1

COAMPS is a registered trademark of the Naval Research Laboratory.

Save
  • Atkins, N. T., R. M. Wakimoto, and C. L. Ziegler, 1998: Observations of the finescale structure of a dryline during VORTEX95. Mon. Wea. Rev, 126 , 525550.

    • Search Google Scholar
    • Export Citation
  • Baker, N. L., 1992: Quality control for the navy operational atmospheric database. Wea. Forecasting, 7 , 250261.

  • Ball, J., I. Woodrow, and J. Berry, 1987: A model predicting stomatal resistance and its contribution to the control of photosynthesis under different environmental conditions. Progress in Photosynthesis Research, J. Biggins, Ed., Vol. IV, Martinus Nijhoff, 221–224.

    • Search Google Scholar
    • Export Citation
  • Barker, E. H., 1992: Design of the navy's multivariate optimum interpolation analysis system. Wea. Forecasting, 7 , 220231.

  • Calvet, J-C., J. Noilhan, J. Roujean, P. Bessemoulin, M. Cabelguenne, A. Olioso, and J. Wigneron, 1998: An interactive vegetation SVAT model tested against data from six contrasting sites. Agric. For. Meteor, 92 , 7395.

    • Search Google Scholar
    • Export Citation
  • Campbell, G. S., and J. M. Norman, 1998: An Introduction to Environmental Biophysics. 2d ed. Springer, 312 pp.

  • Chang, J-T., and P. J. Wetzel, 1991: Effects of spatial variations of soil moisture and vegetation on the evolution of a prestorm environment: A case study. Mon. Wea. Rev, 119 , 13681390.

    • Search Google Scholar
    • Export Citation
  • Chen, F., and J. Dudhia, 2001: Coupling an advanced land surface/hydrology model with the Penn State–NCAR MM5 modeling system. Part I: Model implementation and sensitivity. Mon. Wea. Rev, 129 , 569585.

    • Search Google Scholar
    • Export Citation
  • Chen, F., and Coauthors, 1996: Modeling of land-surface evaporation by four schemes and comparison with FIFE observations. J. Geophys. Res, 101 , 72517268.

    • Search Google Scholar
    • Export Citation
  • Chen, F., K. W. Manning, D. N. Yates, M. A. LeMone, S. B. Trier, R. Cuenca, and D. Niyogi, 2004: Development of a High Resolution Land Data Assimilation System (HRLDAS). Preprints, 16th Conf. on Numerical Weather Prediction, Seattle, WA, Amer. Meteor. Soc., CD-ROM, 22.3.

  • Clark, C. A., and R. W. Arritt, 1995: Numerical simulations of the effect of soil moisture and vegetation cover on the development of deep convection. J. Appl. Meteor, 34 , 20292045.

    • Search Google Scholar
    • Export Citation
  • Collatz, G. J., J. Ball, C. Grivet, and J. Berry, 1991: Physiological and environmental regulation of stomatal conductance, photosynthesis, and transpiration: A model that includes a laminar boundary layer. Agric. For. Meteor, 54 , 107136.

    • Search Google Scholar
    • Export Citation
  • Collatz, G. J., M. Ribas-Carbo, and J. Berry, 1992: Coupled photosynthesis–stomatal conductance model for leaves of C4 plants. Aust. J. Plant Physiol, 19 , 519538.

    • Search Google Scholar
    • Export Citation
  • Doran, J. C., and S. Zhong, 1995: Variations in mixed-layer depths arising from inhomogeneous surface conditions. J. Climate, 8 , 19651973.

    • Search Google Scholar
    • Export Citation
  • Ek, M. B., K. E. Mitchell, Y. Lin, E. Rogers, P. Grunmann, V. Koren, G. Gayno, and J. D. Tarpley, 2003: Implementation of Noah land surface model advances in the National Centers for Environmental Prediction operational mesoscale Eta Model. J. Geophys. Res, 108 .8851, doi:10.1029/2002JD003296.

    • Search Google Scholar
    • Export Citation
  • Findell, K. L., and E. A. B. Eltahir, 2003: Atmospheric controls on soil moisture–boundary layer interactions. Part II: Feedbacks within the continental United States. J. Hydrometeor, 4 , 570583.

    • Search Google Scholar
    • Export Citation
  • Grasso, L. D., 2000: A numerical simulation of dryline sensitivity to soil moisture. Mon. Wea. Rev, 128 , 28162834.

  • Hodur, R. M., 1997: The Naval Research Laboratory's Coupled Ocean/Atmosphere Mesoscale Prediction System (COAMPS). Mon. Wea. Rev, 125 , 14141430.

    • Search Google Scholar
    • Export Citation
  • Jacquemin, B., and J. Noilhan, 1990: Sensitivity study and validation of a land surface parameterization using the HAPEX-MOBILHY data set. Bound.-Layer Meteor, 52 , 93134.

    • Search Google Scholar
    • Export Citation
  • Jarvis, P. G., 1976: The interpretation of the variations in leaf water potential and stomatal conductance found in canopies in the field. Philos. Trans. Roy. Soc. London, B273 , 593610.

    • Search Google Scholar
    • Export Citation
  • LeMone, M. A., and Coauthors, 2000: Land–atmosphere interaction research, early results, and opportunities in the Walnut River Watershed in southeast Kansas: CASES and ABLE. Bull. Amer. Meteor. Soc, 81 , 757779.

    • Search Google Scholar
    • Export Citation
  • Mahfouf, J-F., E. Richard, and P. Mascart, 1987: The influence of soil and vegetation on the development of mesoscale circulations. J. Climate Appl. Meteor, 26 , 14831495.

    • Search Google Scholar
    • Export Citation
  • Mahrt, L., and M. Ek, 1984: The influence of atmospheric stability on potential evaporation. J. Climate Appl. Meteor, 23 , 222234.

  • Mahrt, L., and H. L. Pan, 1984: A two-layer model of soil hydrology. Bound.-Layer Meteor, 29 , 120.

  • McCumber, M. C., and R. A. Pielke, 1981: Simulation of the effects of surface fluxes of heat and moisture in a mesoscale numerical model. Part I: Soil layer. J. Geophys. Res, 86 , 99299938.

    • Search Google Scholar
    • Export Citation
  • McGuire, E. L., 1962: The vertical structure of three drylines as revealed by aircraft traverses. National Severe Storms Project Rep. 7, 11 pp. [Available from NCAR, P.O. Box 3000, Boulder, CO 80307.].

  • Miller, J. E., 1948: On the concept of frontogenesis. J. Meteor, 5 , 169171.

  • Miller, R. C., 1967: Notes on analysis and severe-storm forecasting procedures of the Military Weather Warning Center Tech. Rep. 200, U.S. Air Force Air Weather Service, Scott Air Force Base, IL, 170 pp.

  • Niyogi, D. S., 2000: Biosphere–atmosphere interactions coupled with carbon dioxide and soil moisture changes. Ph.D. dissertation, North Carolina State University, 509 pp.

  • Niyogi, D. S., and S. Raman, 1997: Comparison of four different stomatal resistance schemes using FIFE observations. J. Appl. Meteor, 36 , 903917.

    • Search Google Scholar
    • Export Citation
  • Niyogi, D. S., K. Alapaty, and S. Raman, 1998: Comparison of four different stomatal resistance schemes using FIFE observations. Part II: Analysis of terrestrial biospheric–atmospheric interactions. J. Appl. Meteor, 37 , 13011320.

    • Search Google Scholar
    • Export Citation
  • Niyogi, D. S., Y-K. Xue, and S. Raman, 2002: Hydrological land surface response in a tropical regime and a midlatitudinal regime. J. Hydrometeor, 3 , 3956.

    • Search Google Scholar
    • Export Citation
  • Noilhan, J., and S. Planton, 1989: A simple parameterization of land surface processes for meteorological models. Mon. Wea. Rev, 117 , 536549.

    • Search Google Scholar
    • Export Citation
  • Ogura, Y., and Y. Chen, 1977: A life history of an intense mesoscale convective storm in Oklahoma. J. Atmos. Sci, 34 , 14581476.

  • Pan, H-L., and L. Mahrt, 1987: Interaction between soil hydrology and boundary-layer development. Bound.-Layer Meteor, 38 , 185202.

  • Pielke, R. A., 2001: Influence of the spatial distribution of vegetation and soils on the prediction of cumulus convective rainfall. Rev. Geophys, 39 , 151177.

    • Search Google Scholar
    • Export Citation
  • Rhea, J. O., 1966: A study of thunderstorm formation along drylines. J. Appl. Meteor, 5 , 5863.

  • Sanders, F., 1955: An investigation of the structure and dynamics of an intense surface frontal zone. J. Meteor, 12 , 542552.

  • Schaefer, J. T., 1986: The dryline. Mesoscale Meteorology and Forecasting, P. S. Ray, Ed., Amer. Meteor. Soc., 549–570.

  • Segal, M., and R. W. Arritt, 1992: Nonclassical mesoscale circulations caused by surface sensible heat flux gradients. Bull. Amer. Meteor. Soc, 73 , 15931604.

    • Search Google Scholar
    • Export Citation
  • Segal, M., W. E. Schreiber, G. Kallos, J. R. Garratt, A. Rodi, J. Weaver, and R. A. Pielke, 1989: The impact of crop areas in northeast Colorado on midsummer mesoscale thermal circulations. Mon. Wea. Rev, 117 , 809825.

    • Search Google Scholar
    • Export Citation
  • Segal, M., R. Arritt, C. Clark, R. Rabin, and J. Brown, 1995: Scaling evaluation of the effect of surface characteristics on potential for deep convection over uniform terrain. Mon. Wea. Rev, 123 , 383400.

    • Search Google Scholar
    • Export Citation
  • Sellers, P., S. O. Los, C. J. Tucker, C. O. Justice, D. A. Dazlich, G. J. Collatz, and D. A. Randall, 1996: A revised land surface parameterization (SiB2) for atmospheric GCMs. Part II: The generation of global fields of terrestrial biophysical parameters from satellite data. J. Climate, 9 , 706737.

    • Search Google Scholar
    • Export Citation
  • Shaw, B. L., R. A. Pielke, and C. L. Ziegler, 1997: A three-dimensional numerical simulation of a Great Plains dryline. Mon. Wea. Rev, 125 , 14891506.

    • Search Google Scholar
    • Export Citation
  • Trier, S. B., F. Chen, and K. W. Manning, 2004: A study of convection initiation in a mesoscale model using high-resolution land surface initial conditions. Mon. Wea. Rev, 132 , 29542976.

    • Search Google Scholar
    • Export Citation
  • Walko, R. L., and Coauthors, 2000: Coupled atmosphere–biophysics–hydrology models for environmental modeling. J. Appl. Meteor, 39 , 931944.

    • Search Google Scholar
    • Export Citation
  • Weckwerth, T. M., and Coauthors, 2004: An overview of the International H2O Project (IHOP_2002) and some preliminary highlights. Bull. Amer. Meteor. Soc, 85 , 253277.

    • Search Google Scholar
    • Export Citation
  • Weiss, C. C., and H. B. Bluestein, 2002: Airborne pseudo–dual Doppler analysis of a dryline–outflow boundary intersection. Mon. Wea. Rev, 130 , 12071226.

    • Search Google Scholar
    • Export Citation
  • Wilks, D. S., 1995: Statistical Methods in the Atmospheric Sciences. Academic Press, 467 pp.

  • Zhang, D., and R. A. Anthes, 1982: A high-resolution model of the planetary boundary layer—Sensitivity tests and comparison with SESAME-79 data. J. Appl. Meteor, 21 , 15941609.

    • Search Google Scholar
    • Export Citation
  • Ziegler, C. L., and C. E. Hane, 1993: An observational study of the dryline. Mon. Wea. Rev, 121 , 11341151.

  • Ziegler, C. L., W. J. Martin, R. A. Pielke, and R. L. Walko, 1995: A modeling study of the dryline. J. Atmos. Sci, 52 , 263285.

  • Fig. 1.

    COAMPS model domain for (a) 12-km outer nest and the location of 4-km inner nest, and (b) inner nest terrain (shaded, interval of 200 m) and locations of mesonet surface observations (dots) for Oklahoma and west Texas. Amarillo and Shamrock, TX, are indicated by the A and S, respectively

  • Fig. 2.

    COAMPS nest-2 static surface fields of (a) 24-category vegetation and (b) 16-category soil types

  • Fig. 3.

    COAMPS 12-km analysis valid at 0000 UTC 24 May 2002 of (a) sea level pressure (interval 4 hPa), 10-m wind barbs (full barb = 5 m s−1), regions of surface moisture convergence greater than 7.5 × 10−7 g kg−1 s−1 (shaded), and estimated surface frontal position; (b) 500-hPa geopotential heights (interval 30 m) and wind barbs, and regions of 700-hPa moisture convergence greater than 7.5 × 10−7 g kg−1 s−1 (shaded); (c) and (d) same as (a) and (b) except for the COAMPS 24-h forecast valid at 0000 UTC 25 May 2002

  • Fig. 4.

    (a) IHOP surface analysis valid at 0000 UTC 24 May 2002 over the COAMPS domain, (b) Geostationary Operational Environmental Satellite-8 (GOES-8) visible 1-km-resolution satellite image for 0008 UTC 24 May 2002, and (c) CONTROL analysis at 0000 UTC 24 May of 10-m wind barbs (full barb = 5 m s−1), 10-m mixing ratio (contour interval = 1 g kg−1), and vertically integrated total cloud fraction (shaded). The subjective locations of the cold front and dryline (dashed line) are also given. The dashed box in (c) is the location of the satellite image (b) on the COAMPS domain. The estimated observed surface frontal positions at 1800 UTC 24 May (label 18/24 May), 0000 UTC (label 00/25 May), and 0600 UTC (label 06/25 May) 25 May are given by the solid lines in (a)

  • Fig. 5.

    Radar reflectivity (dBZ) valid at 0002 UTC 25 May for 2-km observations and corresponding 24-h COAMPS forecasts valid 0000 UTC 25 May. The box shows the estimated orientation of the observed cold-frontal precipitation band over Oklahoma and north-central Texas

  • Fig. 6.

    Time series of (a) 2-m air temperature (°C) and (b) mixing ratio (g kg−1) statistics from 1500 UTC 24 May to 0600 UTC 25 May computed using the Oklahoma surface mesonet stations shown in Fig. 1b. The statistics are computed in the prefrontal region as shown in Fig. 4a

  • Fig. 7.

    Time series of statistics similar to Fig. 6, except using the west Texas Mesonetstations from 1500 to 2100 UTC 24 May in the area of the dryline before passage of the front

  • Fig. 8.

    The 21-h forecast valid 2100 UTC 24 May of 2-m mixing ratio (contour interval 2 g kg−1), 10-m winds (arrows every seventh grid point, m s−1), and boundary layer depth AGL (m, shaded) for (a) CONTROL and (b) SLAB. Cross section A–B is also indicated along with box S used for averaged fields given in Fig. 17

  • Fig. 9.

    Vertical cross section of 21-h forecast valid at 2100 UTC 24 May along line A–B given in Fig. 8 of mixing ratio (g kg−1, shaded), virtual potential temperature (bold line, interval of 1 K), and vertical circulation (arrows) for (a) CONTROL and (b) SLAB. The 17–21-h average surface fluxes (W m−2) of sensible heat (CONTROL: bold dots; SLAB: bold line) and latent heat (CONTROL: thin dots; SLAB: thin line) along the cross section are also given

  • Fig. 10.

    The 21-h forecast of 10-cm volumetric soil moisture valid at 2100 UTC 24 May for (a) CONTROL and (b) EDAS

  • Fig. 11.

    Averaged 17–21-h sensible heat flux (W m−2) for (a) CONTROL and (b) EDAS. The contour is the BL depth (1200-m contour only). Point C is the location for the time–height cross section shown in Fig. 12

  • Fig. 12.

    Time–height (AGL) cross section at point C shown in Fig. 11 in north-central Texas of virtual potential temperature (shaded) and mixing ratio (solid lines) for (a) CONTROL and (b) EDAS. The surface heat fluxes during the daytime (15–24 h) illustrate the dominance of sensible to latent for EDAS due to the drier soil conditions (EDAS 10-cm soil moisture ∼0.148 vs 0.239 for CONTROL). The evolution of the BL depth indicated by the dashed line likewise indicates more rapid deepening for EDAS

  • Fig. 13.

    Time series of (a) 10-cm soil temperature (°C) and (b) 10-cm soil moisture (volumetric fraction) statistics from 1500 UTC 24 May to 0600 UTC 25 May computed using prefrontal Oklahoma surface mesonet stations similar to Fig. 6

  • Fig. 14.

    The 21-h forecast valid 2100 UTC 24 May of canopy resistance (s m−1) for (a) CONTROL and (b) GEM

  • Fig. 15.

    Averaged 17–21-h latent heat flux (W m−2) for (a) CONTROL and (b) GEM. The contours are for 9 (dashed) and 13 g kg−1 (solid) 2-m mixing ratio. Differences (model − obs) of mixing ratio for five selected mesonet stations are shown (solid circles) to illustrate the impact of latent heat flux differences