Influence of Soil Moisture on Urban Microclimate and Surface-Layer Meteorology in Oklahoma City

Syed Zahid Husain Meteorological Research Division, Environment Canada, Dorval, Quebec, Canada

Search for other papers by Syed Zahid Husain in
Current site
Google Scholar
PubMed
Close
,
Stéphane Bélair Meteorological Research Division, Environment Canada, Dorval, Quebec, Canada

Search for other papers by Stéphane Bélair in
Current site
Google Scholar
PubMed
Close
, and
Sylvie Leroyer Meteorological Research Division, Environment Canada, Dorval, Quebec, Canada

Search for other papers by Sylvie Leroyer in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

The influence of soil moisture on the surface-layer atmosphere is examined in this paper by analyzing the outputs of model simulations for different initial soil moisture configurations, with particular emphasis on urban microclimate. In addition to a control case, four different soil moisture distributions within the urban and surrounding rural areas are considered in this study. Outputs from the Global Environmental Multiscale atmospheric model simulations are compared with observations from the Joint Urban 2003 experiment held in Oklahoma City, Oklahoma, and the relevant conclusions drawn in this paper are therefore valid for similar medium-size cities. In general, high soil moisture is found to be associated with colder near-surface temperature and lower near-surface wind speed, whereas drier soil resulted in warmer temperatures and enhanced low-level wind. Relative to urban soil moisture content, rural soil conditions are predicted to have larger impacts on both rural and urban surface-layer meteorological conditions. Dry rural and wet urban soil configurations are shown to have a strong influence on the urban–rural temperature contrast and resulted in city-induced secondary circulations that considerably affect the near-surface wind speed. Dry rural soil in particular is found to intensify the nocturnal low-level jet and significantly affect the thermal stability of nocturnal near-neutral urban surface layer by altering both thermal and mechanical generation of turbulence.

Corresponding author address: Syed Zahid Husain, Atmospheric Numerical Weather Prediction Research Section, Environment Canada, 2121 TransCanada Highway, Dorval, QC H9P 1J3, Canada. E-mail: syed.husain@ec.gc.ca

Abstract

The influence of soil moisture on the surface-layer atmosphere is examined in this paper by analyzing the outputs of model simulations for different initial soil moisture configurations, with particular emphasis on urban microclimate. In addition to a control case, four different soil moisture distributions within the urban and surrounding rural areas are considered in this study. Outputs from the Global Environmental Multiscale atmospheric model simulations are compared with observations from the Joint Urban 2003 experiment held in Oklahoma City, Oklahoma, and the relevant conclusions drawn in this paper are therefore valid for similar medium-size cities. In general, high soil moisture is found to be associated with colder near-surface temperature and lower near-surface wind speed, whereas drier soil resulted in warmer temperatures and enhanced low-level wind. Relative to urban soil moisture content, rural soil conditions are predicted to have larger impacts on both rural and urban surface-layer meteorological conditions. Dry rural and wet urban soil configurations are shown to have a strong influence on the urban–rural temperature contrast and resulted in city-induced secondary circulations that considerably affect the near-surface wind speed. Dry rural soil in particular is found to intensify the nocturnal low-level jet and significantly affect the thermal stability of nocturnal near-neutral urban surface layer by altering both thermal and mechanical generation of turbulence.

Corresponding author address: Syed Zahid Husain, Atmospheric Numerical Weather Prediction Research Section, Environment Canada, 2121 TransCanada Highway, Dorval, QC H9P 1J3, Canada. E-mail: syed.husain@ec.gc.ca

1. Introduction

With continuing urbanization the amount of pervious natural surface is increasingly being replaced by impervious built-up surfaces. As a consequence, the amount of natural surfaces (vegetation and soil) exposed to interact with the adjacent atmosphere is declining. This leads to modifications in the aerodynamic and thermodynamic structures of the boundary layer through changes in the effective surface roughness and surface–atmosphere heat exchange characteristics. Changing landscape and the associated modifications in surface characteristics therefore can affect the evolution of vertical structures of the meteorological fields in the lower part of the atmospheric boundary layer, known as the surface layer, which is strongly influenced by the surface fabric. Since urban regions are now occupied by the majority of the world’s population, studying the forcing mechanisms that can potentially affect the evolution and structure of the urban surface layer is an imperative.

Sophisticated parameterization schemes have been devised by meteorological researchers to suitably represent the evolution of boundary layer wind, temperature, moisture, and turbulence in atmospheric models. Such schemes usually allow multiple surface energy budgets for the different surface components, for example, soil, vegetation, and urban canopy (Bélair et al. 2003; Masson 2000), and often permit surface–atmosphere interactions at multiple vertical atmospheric model levels that intersect vegetation and manmade structures (Husain et al. 2013; Dupont et al. 2004; Martilli et al. 2002). Despite the capabilities of sophisticated land surface schemes for significantly improving the representation of the surface effects, there are other aspects associated with numerical weather prediction (NWP) systems that are considered to be just as important. Examples of critical factors for acceptable weather prediction include initial conditions and parameterization of various physical processes including radiation, boundary layer turbulence, clouds, and precipitation.

One of the important preconditions for an atmospheric model to accurately predict the surface-layer meteorology is related to its ability of correctly estimating the soil moisture content (Jacobson 1999). Soil moisture plays a fundamental role in determining the evolution of soil temperature by affecting the exchange of latent heat and strongly influences the overall surface layer characteristics through changes in evaporation (Jacobson 1999; McCorcle 1988; Fast and McCorcle 1990). It has been found to affect the contrast between urban and rural near-surface temperatures generally referred to as the urban heat island (UHI) effects (Hafner and Kidder 1999) and is also observed to impact the frequency of UHI occurrences (Ahmad and Hashim 2007).

Jacobson’s (1999) numerical experiment, using an Eulerian air pollution model, revealed a strong influence of soil moisture on near-surface wind speed in addition to temperature and ozone. This influence of soil moisture on the surface layer can therefore induce changes in boundary layer turbulence as well as thermal stability characteristics of the lowest atmosphere. Studies have also shown considerable positive feedback between soil moisture and rainfall over areas such as Illinois (Findell and Eltahir 2003) and Florida (Baker et al. 2001). Using coupled land–atmosphere model runs, Ek and Holtslag (2004) have shown that in some cases a decrease in soil moisture is associated with increase in boundary layer clouds. Moreover, numerical experiments have established a strong correlation between soil moisture and the intensity of the nocturnal low-level jets (LLJs), a meteorological phenomenon commonly observed during summer in the Great Plains of the United States (McCorcle 1988; Fast and McCorcle 1990).

Land surface parameterization schemes employed by many of the atmospheric models are capable of predicting the evolution of both the near-surface and root-zone soil moisture contents. Similar to any other prognostic variables in a NWP system, accurate forecast of the evolution of soil moisture critically depends on its initial estimation. Inaccurate initialization of the soil moisture field can therefore result in incorrect prediction of the surface-layer vertical profiles of the meteorological variables and the associated thermal stability characteristics. As microscale dispersion models are often driven by outputs from mesoscale atmospheric models, this may lead to unacceptable and erroneous forecast for dispersion of pollutants released to the atmosphere. In the context of simulating a sea-breeze event in southwestern Western Australia, Kala et al. (2010) have shown that an increase in initial values of soil moisture results in weaker sea breeze with reduced duration and delayed onset.

The primary objective of the present work is to study the influence of soil moisture on urban-induced mesoscale features and understand the underlying mechanisms as well as to investigate the effects of rural and urban soil moisture on urban microclimate. Information on the influence of soil moisture with similar perspectives is very limited as far as the existing literature is concerned. A relevant numerical study was conducted by Chen et al. (2011), who investigated the interactions between soil moisture and sea-breeze circulations in the context of their effects on stagnation within an urban region. The outcome of this study shows that extremely dry soil conditions leads to stagnant wind with longer duration in the Houston–Galveston area during the evening hours, and increase the possibility of nighttime air pollution. The existence of city-induced mesoscale circulation, known as urban breeze, resulting from the differences in surface heat fluxes between an urban area and its surroundings has also been confirmed by analyzing observational data and through numerical simulations (Hidalgo et al. 2008a,b).

The present study is a continuation of the work by Husain et al. (2013) in which the authors developed a multilayer urban canopy scheme to improve surface–atmosphere interactions within urban regions. The experimental framework used in this study follows the one presented by Husain et al. (2013) and is based on comparison of model outputs with observations for urban microclimate as well as the vertical and horizontal structures of meteorological variables within the urban surface layer. Conclusions in this paper are derived through numerical experiments where model simulations are carried out for a period covered by an intensive observation period (IOP) of the Joint Urban 2003 (JU2003) experiment, held in Oklahoma City, Oklahoma (Allwine et al. 2004), for which the urban scheme developed by Husain et al. (2013) has already been evaluated.

2. Model description

The mesoscale version of the Global Environmental Multiscale (GEM) atmospheric model based on nonhydrostatic governing equations (Zadra et al. 2008; Yeh et al. 2002) is used in this study. The GEM model works by first solving a set of dynamical equations directly on the model grid. Solution of the dynamic components of the governing equations are then supplemented with tendencies from various parameterized subgrid-scale physical processes that include atmospheric radiation, fluxes from natural and built-up surfaces, boundary layer turbulent mixing, and gravity wave drag as well as clouds and precipitation (Zadra et al. 2008; Yeh et al. 2002).

Parameterization of surface–atmosphere interactions in the GEM model follows a tiles-based approach (Lemonsu et al. 2009; Leroyer et al. 2011). Each model grid cell is composed of five surface tiles representing the urban surface (roads and buildings), natural surface (soil and vegetation), water, continental ice, and sea ice. Fluxes of momentum, heat, and moisture associated with each tile are aggregated using a weighted average method based on the fractions of each surface type to determine the lower boundary conditions for the turbulent vertical diffusion scheme. In the present study, the effects of soil–vegetation and urban canopy are parameterized using the Interactions between Soil, Biosphere, and Atmosphere (ISBA; Bélair et al. 2003) and the Canadian Multilayer version of Town Energy Balance (CaM-TEB; Husain et al. 2013) surface schemes, respectively.

The version of ISBA (Bélair et al. 2003) used in GEM for this study is an improved edition of the original scheme developed by Noilhan and Planton (1989). The prognostic equations used in ISBA are based on the force-restore concept proposed by Deardorff (1978). The current implementation of ISBA is capable of the prognostic prediction of temperatures representative of the aggregate vegetated surfaces composed of soil and vegetation. Soil moisture content along with equivalent water content of snow reservoirs within a model grid cell is resolved prognostically. The hydrology scheme in ISBA considers one near-surface and one root-zone soil layer. Evolution of soil moisture contents for both of these layers is determined by a set of force-restore equations that take into account the effects of rainfall, evaporation, and snowmelt as well as surface runoff. Freezing and thawing of soil water and transpiration from roots are also included in the prognostic equation for the root-zone layer while drainage is controlled by keeping it proportional to the amount of water exceeding the field capacity. Interactions between GEM and ISBA take place through canopy-averaged surface fluxes of momentum, temperature, and moisture that are used in determining the lower boundary conditions for the turbulent vertical diffusion scheme.

The multilayer urban scheme (CaM-TEB) used in the present work to parameterize the effects of urban structures on the atmosphere is an extension of the single-layer TEB model originally developed by Masson (2000). The role of TEB in the multilayer approach is limited to determining temperatures of the characteristic canopy surfaces (roofs, roads, and walls) that are used to parameterize the heat and moisture fluxes associated with these surfaces. The urban scheme uses a simple three-dimensional definition of the urban canyon although buildings with different heights are allowed to exist within a grid cell. In contrast with TEB, where GEM’s first vertical level is located tens of meters above the roofs, the CaM-TEB scheme makes it possible to have a number of vertical levels in the surface layer including a few inside the urban canopy. Drag and friction induced by the canopy surfaces along with surface–atmosphere exchanges of heat and moisture are parameterized in CaM-TEB and are applied at the multiple model levels that are in contact with canopy. The mixing and dissipation length along with the prognostic equation for turbulent kinetic energy (TKE) are also modified in CaM-TEB to take the effects of the urban canopy into account. Unlike the standard implementation of TEB (Lemonsu et al. 2009), the multilayer CaM-TEB scheme permits a prognostic evolution of the meteorological fields of interest in the surface layer, particularly inside the urban canyons, which is critical for this study. Similar to ISBA, CaM-TEB also provides lower boundary conditions for the vertical diffusion scheme through the calculated surface fluxes that are only representative of the roads.

In general, natural surfaces for ISBA are composed of 26 subcategories that include different types of vegetation and surface states (ice, water, desert, etc.). On the other hand, the aggregate urban surface for a grid cell is constructed from 12 urban types, each having different fractions for built-up surfaces (roads and buildings), soil, and vegetation. The present implementation of CaM-TEB only works for the fraction of a grid cell that is covered by built-up surfaces. Any soil and vegetation within the urban fraction of a grid cell is added to the natural surfaces and is dealt with by ISBA. As a result, the evolution of soil moisture in urban areas and the associated moisture fluxes are determined using ISBA. A future version of CaM-TEB will, however, include a separate scheme to handle natural surfaces within the urban fraction of a grid cell (e.g., lawns, backyards, and vegetation within urban canyons).

3. The Joint Urban 2003 experiment

The Joint Urban 2003 experiment was held in Oklahoma City (OKC), Oklahoma, from 28 June to 31 July 2003. The principal objectives of the JU2003 field campaign were to document the urban microclimate, study transportation of contaminants, and investigate the dynamical and thermodynamical structures of the boundary layer, both upwind and downwind of OKC’s central business district (CBD; Allwine et al. 2004). To achieve these goals, networks of meteorological sensors were deployed in and around the city during the experiment that included observation sites composed of surface stations, radars, sodars, and soundings.

The near-surface rural wind speed and air temperature in this study are estimated by averaging the screen-level data (1.5 m above the ground) recorded at 11 surface stations located around the city that are part of the Oklahoma Mesonet (www.mesonet.org; see Fig. 1a). Their corresponding urban values are calculated by averaging the data recorded (8 m above street level) by 13 Portable Weather Information Display Systems (PWIDS) installed within the CBD (see Fig. 1b).

Fig. 1.
Fig. 1.

(a) The 2.5-km GEM–LAM simulation domain and the corresponding orography. The long-dashed and short-dashed rectangles represent the 1-km simulation domain (161 × 161 grid points, excluding the piloting zone for enforcing lateral boundary conditions) and the approximate bounds of OKC (30 × 41 grid points), respectively. The rural and urban stations are shown using green and black markers, respectively. (b) Distribution of urban fraction over OKC along with the positions of the ANL and PNNL sites with respect to the CBD. The PWIDS network of 13 stations in the CBD appears as a single point in both figures because of their close proximity to each other.

Citation: Journal of Applied Meteorology and Climatology 53, 1; 10.1175/JAMC-D-13-0156.1

To characterize the flow and pertinent structures of the boundary layer, balloon soundings were launched during a number of JU2003s day- and nighttime IOPs at two sites, located upwind and downwind of the CBD. One of these sites, about 5 km north of the CBD, was set up by the Argonne National Laboratory (ANL). The other site was installed by the Pacific Northwest National Laboratory (PNNL), located approximately 2 km south of the CBD (see Fig. 1b).

4. Description of simulations

The effects of soil moisture on urban microclimate and surface-layer meteorology are investigated in this study by comparing model simulations against observations from the JU2003 experiment. Simulation results used in the present work are obtained with the limited-area configuration of GEM known as GEM–Limited-Area Model (LAM) for a horizontal grid spacing of 1 km. The simulation domains for the present work along with the associated orography are shown in Fig. 1. The figure reveals a relatively flat orography for OKC, with elevation gradually increasing westward because of the presence of the Rocky Mountains approximately 700 km west of the city.

The simulated values of near-surface wind speed and temperature for urban and rural areas correspond to the averages of their values predicted at the same locations used for estimating their observed averages. The temporal dimensions for all of the figures in this paper are based on the local standard time (LST), where LST = UTC − 6 h with UTC being the coordinated universal time. Analyses of observations and model results previously presented by Lemonsu et al. (2009) have shown the influence of the city on daytime thermodynamic structure of the boundary layer to be negligible. This study builds on the outcome of work by Husain et al. (2013) in which the performance of CaM-TEB was evaluated against a nighttime IOP, namely IOP9, which initiated at 2100 LST 26 July and ended at 0600 LST 27 July 2003. The observational records for this IOP point to clear-sky conditions along with the occurrence of southwesterly nocturnal LLJ. The present study primarily uses the observed nocturnal vertical profiles of the meteorological fields from this IOP of the JU2003 experiment to compare the simulation outcomes. Daytime model results are, however, also presented for the day preceding IOP9 although the observational data lack information to construct the pertinent vertical profiles. Overall, the protocols used in the present work to evaluate the model performance to analyze the role of soil moisture are based on those presented by Husain et al. (2013).

A set of simulations are carried out in this study to investigate the effects of both the near-surface and root-zone soil moisture with different initial conditions for the urban and rural areas. A first guess for the spatially variable soil moisture fields is obtained from operational analyses of the Canadian Meteorological Centre, performed on a 15-km North American grid and based on the assimilation of screen-level observations (air temperature and relative humidity) using an optimal interpolation technique (Bélair et al. 2003). These heterogeneous soil moisture fields (both near-surface and root-zone soil) are used to initialize the fields for a 15-km GEM–LAM simulation. It should be noted that the 15-km GEM–LAM simulations are driven using the outputs of the regional GEM model, which is based on a variable-resolution global grid with an effective horizontal grid spacing of 15 km over North America. The outputs of the 15-km simulation are iteratively improved by modifying the soil moisture distribution to produce the best correspondence between the predicted and observed values of near-surface meteorological variables for the rural stations using 2.5-km GEM–LAM forecasts. The selection of 15-km GEM–LAM simulations over the regional GEM model is primarily based on the motivation for improving the computational efficiency of the aforementioned strategy.

The corresponding best case values of the soil moisture fields are considered as the control case for this study. Four additional cases with varying initial values of soil moisture are investigated for the sensitivity analyses linked to the objectives of the present work. The relative differences between the soil moisture content of the additional cases and the control are presented in Table 1. Soil moisture level is increased or reduced by 33% for each of these cases to make the signals attributable to its changes on near-surface meteorology more apparent. Both the near-surface and root-zone values of soil moisture content are modified as required by the individual cases stated in the table, and the resulting values are used to initialize the additional GEM–LAM 2.5-km simulations. The outputs of 2.5-km model runs are used to initialize the 1-km LAM simulations to obtain the final results presented in this paper. The numerical setup used in this study for nesting from 15- to 1-km grid sizes is presented in Table 2.

Table 1.

Variation in the urban and rural soil moisture contents for both near-surface and root-zone soil layers in ISBA as compared to the control case. Soil moisture for the additional cases is kept below or equal to the field capacity.

Table 1.
Table 2.

Setup of the numerical experiments.

Table 2.

The maximum allowable soil moisture content for the additional cases is cut off at the field capacity for the given soil texture. But values below the wilting point are permitted. The rural and urban areas for the 2.5-km GEM–LAM simulation domain along with the near-surface and root-zone soil moisture content for the control case are highlighted in Figs. 2a and 2b.

Fig. 2.
Fig. 2.

Volumetric soil moisture content (m3 m−3) for (a) near-surface and (b) root-zone layers corresponding to the control case for 2.5-km GEM–LAM simulation domain valid at 0400 UTC 26 Jul 2003. The rectangle in (a) indicates the urban area.

Citation: Journal of Applied Meteorology and Climatology 53, 1; 10.1175/JAMC-D-13-0156.1

5. Results

Near-surface soil moisture content directly influences the evolution of surface-layer meteorological variables, particularly the daytime temperature. Increased values of available soil moisture result in enhanced evapotranspiration and greater use of daytime radiative energy for latent heat flux that eventually reduces sensible heat flux from the surface to the atmosphere. As a result, wet soil is generally linked to lower daytime heating of the near-surface atmosphere through release of sensible heat from the surface, and by extension to a lower daytime maximum temperature. Changes in near-surface temperature induced by available soil moisture can also affect the near-surface wind speed through modifications in the level of thermally induced turbulence within the surface layer. This study aims to add to this general understanding by investigating the impact of soil moisture levels in rural and urban areas on near-surface temperature and wind speed within the relevant areas in the context of OKC.

Figure 3 demonstrates the impact of soil moisture on near-surface temperature. Model predictions based on the control case can be seen to match closely with observations for both rural and CBD stations. As expected, the figure shows that for both rural and urban areas increasing soil moisture level [wet rural soil (RW) and wet urban soil (UW)] reduces the daytime maximum near-surface temperature, whereas decreasing soil moisture [dry rural soil (RD) and dry urban soil (UD)] has the opposite effect. The figure also reveals a larger impact of rural soil moisture level on rural and urban near-surface temperature compared to the effects of urban soil condition particularly in the rural zone. This is apparently a result of the small size of the urban area for Oklahoma City and is compounded by large urban fractions for most of the city (see Fig. 1b), which implies less pervious natural surface available to hold moisture. In addition, Fig. 3b shows that increased availability of soil moisture in rural areas affects nighttime temperature in CBD during the early phase of the night, although the nighttime minimum temperature remains effectively unchanged.

Fig. 3.
Fig. 3.

Evolution of screen-level air temperature (°C) in (a) rural and (b) urban areas for IOP9 obtained from observations (gray line with squares) and 1-km GEM–LAM simulations using five initial estimates for the soil moisture field, namely, control (black line with triangles), RW (blue line), RD (red line), UW (green line), and UD (orange line).

Citation: Journal of Applied Meteorology and Climatology 53, 1; 10.1175/JAMC-D-13-0156.1

Figure 4 demonstrates the contrast between urban and rural values of near-surface air temperature, generally referred to as the UHI effect. Overall, the figure shows a stronger impact of rural soil moisture level on UHI, whereas urban soil moisture content mostly affects the magnitude of daytime negative UHI. Wet rural soil is found to reduce the daytime negative UHI as the associated reduction of daytime maximum temperature is stronger in rural areas compared to the CBD. Similarly, RD leads to larger increase in daytime maximum temperature in rural areas and as a result produces stronger negative UHI effect during the day. Modifications in urban soil moisture level for a city of OKC’s size have considerable influence on the daytime heating rate within the urban near-surface atmosphere. The magnitude of daytime negative UHI is found to increase for UW and decrease for UD. In this respect, the daytime effects of wet and dry urban soil configurations are predicted to be similar to those for dry and wet rural soils respectively.

Fig. 4.
Fig. 4.

Comparison of urban heat island [UHI = (TCBDTRural)°C] computed from observations (gray line with squares) and 1-km GEM–LAM simulations using five initial soil moisture fields: control (black line with triangles), RW (blue line), RD (red line), UW (green line), and UD (orange line).

Citation: Journal of Applied Meteorology and Climatology 53, 1; 10.1175/JAMC-D-13-0156.1

During nighttime, RW is found to increase UHI intensity, whereas dry rural soil leads to a diminishing urban–rural temperature contrast. Increased rural soil moisture level significantly reduces the daytime heating rate in the rural region, leading to a cooler daytime temperature. This eventually results in a lower nighttime rural temperature that produces a stronger contrast with the warmer urban nighttime temperature, which explains the large magnitude of nighttime UHI associated with the RW case as shown in Fig. 4. Urban soil moisture conditions, on the other hand, are found to have negligible impact on nighttime UHI, primarily because of its minimum effects on urban and rural nighttime near-surface temperatures.

The evolution of near-surface wind speed in rural and urban areas is presented in Fig. 5. A consistent overestimation in the near-surface wind speed outputs for the initial hours of model simulations, presumably caused by inaccurate initial conditions, is revealed for all soil moisture configurations. For the later simulation hours, Fig. 5a shows that the evolution of daytime rural wind speed near the ground is minimally affected when subjected to any modification in rural or urban soil moisture contents. A strong nighttime influence of rural soil moisture is, however, clearly visible.

Fig. 5.
Fig. 5.

As in Fig. 3, but for the evolution of near-surface wind speed (m s−1).

Citation: Journal of Applied Meteorology and Climatology 53, 1; 10.1175/JAMC-D-13-0156.1

For this particular case study, dry rural soil is found to result in stronger near-surface wind speed throughout the night, whereas wet soil is found to produce significant weakening of wind speed until midnight. Lowering soil moisture level leads to a warmer surface temperature during daytime that extends into the night, and eventually increases thermal turbulence induced by the surface–atmosphere heat exchanges. This generates enhanced vertical turbulent transport of horizontal momentum, resulting in increased mixing of faster wind aloft with the slower wind close to the ground, and is apparently one of the factors behind nighttime increase in near-surface wind speed for drier soil. Wet soil, on the other hand, reduces thermally driven turbulence and the associated vertical mixing, thereby resulting in lower nighttime near-surface rural wind speed. Figure 5a also reveals that the impact of urban soil moisture content on the near-surface wind speed within the surrounding rural region is inconsequential for a city like OKC.

Figure 5b shows a significant overestimation of near-surface wind speed within the CBD for all test configurations, which is a consequence of inaccurate wind speed prediction for the rural stations as has been shown in Fig. 5a. Aside from these large differences, wind speed near ground within the CBD shows a stronger daytime variability for the different soil moisture configurations, contrary to what has been found for the rural areas.

Near-surface wind speed within the CBD shows an interesting relationship with soil moisture variations in both rural and urban areas. Figure 5b illustrates that an increase in daytime urban wind speed occurs for wet rural and dry urban soils, whereas the reverse is found for dry rural and wet urban conditions. Considering both Figs. 4 and 5, a possible explanation for this could be related with the increase or decrease in the magnitude of negative daytime UHI (i.e., the city cooling effect), which is expected to have an impact on the intensity of secondary circulations generated by the relatively cool air over the city. In this regard, Lemonsu and Masson (2002) have shown that in Paris daytime urban breeze can be induced directly by the city as a result of the UHI effects. A study by Bornstein and Lin (2000) also demonstrates the intensification of local circulations over Atlanta caused by daytime UHI. In the present study, for the UW and RD experiments, the negative UHI or cooling effect over the city is increased, leading to an intensification of downward and divergent flow over the city, which has the effect of slowing down low-level winds. Alternately, for the UD and RW experiments, the temperature contrast between the city and the surrounding rural areas is reduced together with the intensity of the secondary circulations, and the resulting slowing down of the low-level winds is therefore insignificant.

In support of the aforementioned argumentation, Fig. 6 shows the reduction of near-surface wind speed within the CBD compared to its value averaged over the rural stations subjected to the different soil moisture configurations. As could be expected, the near-surface wind speed is found to be smaller than its rural counterpart throughout the entire simulation period, irrespective of the initial soil moisture distribution. The figure indicates a stronger daytime effect of soil moisture on urban wind speed reduction, whereas the nighttime influence is predicted to be negligible. Again a strong correlation can be seen between the dry urban and wet rural soil cases as anticipated, where both configurations lead to smaller reduction of urban near-surface wind during daytime. The wet urban and dry rural cases are correlated as well, with both cases resulting in greater reductions in daytime urban wind speed. It should also be noted that the dry rural case exhibits significant temporal variability, greater than for the other model configurations, and is in better agreement with observations. It is suggested here that these oscillations in wind speed may be related to the intensity of the secondary circulations, expected to be the largest for this experiment.

Fig. 6.
Fig. 6.

As in Fig. 4, but for comparison of urban wind speed reduction [UWR = (WSCBD − WSRural) m s−1].

Citation: Journal of Applied Meteorology and Climatology 53, 1; 10.1175/JAMC-D-13-0156.1

The daytime wind speed profiles at three different times as predicted by the model for the different soil moisture configurations at the ANL site are shown in Fig. 7. The figure reveals a systematic weakening of the urban near-surface wind speed with time for all initial soil moisture distributions. During midday (1300 LST), larger values of wind speed are predicted for the RW and UD cases, which is in agreement with Fig. 6 and the low-level cool air mechanism previously described. Later in the afternoon (1700 LST), the wind speed prediction for all the experiments is similar, except for the larger values exhibited by the RW case.

Fig. 7.
Fig. 7.

Comparison of simulated (control: black; RW: blue; RD: red; UW: green; and UD: orange) and observed (gray) daytime wind speed (m s−1) profiles within the lowest 500 m at times (a) 0900, (b) 1300, and (c) 1700 LST 26 Jul 2003 at ANL.

Citation: Journal of Applied Meteorology and Climatology 53, 1; 10.1175/JAMC-D-13-0156.1

The simulated and observed wind speed profiles during nighttime at ANL as presented in Fig. 8 demonstrate more interesting features. The near-surface wind speed early in the night (2100 LST) is seen to be smaller for RD compared to the results for RW by as much as 1.5 m s−1 (see Figs. 8a,d). This prediction of low near-surface wind at 2100 LST is consistent with the prediction of an overall weaker daytime near-surface wind in the CBD for dry rural soil as demonstrated in Fig. 5.

Fig. 8.
Fig. 8.

Comparison of simulated (control: black; RW: blue; RD: red; UW: green; UD: orange) and observed (gray) nighttime wind speed (m s−1) profiles within the lowest (top) 2500 and (bottom) 500 m at times (a),(d) 2100 LST 26 Jul; (b),(e) 0300; and (c),(f) 0600 LST 27 Jul 2003 at ANL.

Citation: Journal of Applied Meteorology and Climatology 53, 1; 10.1175/JAMC-D-13-0156.1

The observations at three different times during the night as shown in Fig. 8 confirm the existence of a LLJ, which is a common summertime nocturnal phenomenon for the southern Great Plains of the United States. The model simulations for all soil moisture configurations were able to predict the LLJ, although the simulated LLJs are considerably different from the observed one in terms of strength. As has been described in the existing literature, the generation of nocturnal LLJs in the southern Great Plains is triggered by a combination of several factors (Stensrud 1996; Jiang et al. 2007). The dynamical foundation of the nocturnal LLJ is most closely linked with the diurnal boundary layer wind oscillations across the Great Plains associated with the east–west ground elevation gradient (Holton 1967) and inertial oscillation produced by nighttime stabilization of the planetary boundary layer resulting from reduction in eddy viscosity (Blackadar 1957). The blocking effect of the Rockies is also assumed to play a role in the generation of the LLJs (Wexler 1961).

Distribution of soil moisture over a large area within the Great Plains is also expected to have a potential of influencing the onset and intensity of LLJ, based on McCorcle (1988), and Fast and McCorcle (1990). In this regard, Fig. 8 shows an overnight intensification of the low-level jet when simulations are carried out for dry rural soil conditions, whereas for all other configurations the effects of soil moisture on LLJ is predicted to be insignificant. Drier rural soil enhances thermal turbulence through increased sensible heat flux that raises the afternoon and early evening temperatures by almost 2°C in case of the present study (see Fig. 3a). Intensification of the southerly LLJ is thus suggested to be caused by the associated changes in pressure gradient across the region. The considerable difference in the simulated and observed intensities of the LLJ in Husain et al. (2013) can therefore be assumed to have been caused by too-large values of rural soil moisture content used as initial condition for the model simulations. Furthermore, intensification of the LLJ induced by dry rural soil is responsible for the noticeable increase in nighttime near-surface wind speed in both rural and urban areas as seen in Fig. 5.

Rural soil moisture content is also found to affect the predicted locations of maximum and minimum jet speeds. Dry soil is found to result in a higher location for the jet maximum and a lower location of the jet minimum, while the opposite is true for wet soil. As the size of the urban area considered for this study is insignificant compared to area covered by the rural surroundings, changes in urban moisture level naturally fail to exert any visible influence on the formation and evolution of LLJ.

The predicted profiles of wind direction for different soil moisture distributions do not show any qualitative difference irrespective of the time of the day, and are therefore not presented in this paper. It should, however, be mentioned that the model outputs for all configurations predict reasonably well the rotation of LLJ from south to southwest direction with the progression of night as has been observed during IOP9.

Figure 9 shows vertical profiles of TKE for different soil moisture conditions at three different daytime hours at ANL. Early in the day, at 0900 LST, TKE predictions for all levels of initial soil moisture condition are found to be close. Figure 9a also shows a sharp increase of TKE within the first 20 m above ground with its maximum occurring at an altitude slightly above the average height of urban canopy. Although this is an aspect of nocturnal TKE profiles that is well simulated by the CaM-TEB scheme (Husain et al. 2013), Fig. 9a demonstrates the feature to persist through late morning hours until TKE generation does not continue to be dominated by mechanical production. Later in the day, TKE is primarily produced through thermal generation, and as a consequence wet urban soil is seen to result in the lowest TKE caused by significantly reduced thermal generation. For similar reasons, during midday (1300 LST) dry urban soil is found result in maximum TKE values. By afternoon, the profiles for all the conditions show a reduction in TKE values presumably caused by the combined effects of reduction in near-surface wind speed and temperature. The RW case experiences the least reduction in wind speed (see Fig. 7c), whereas the RD case leads to the least drop in temperature. As a result, TKE values at 1700 LST are found to be relatively larger for the RD and RW cases aided by sustained thermal and mechanical generations respectively.

Fig. 9.
Fig. 9.

Profiles of TKE (m2 s−2) predicted at three (a)–(c) different daytimes using the five initial fields for soil moisture: control (black line), RW (blue line), RD (red line), UW (green line), and UD (orange line) at ANL.

Citation: Journal of Applied Meteorology and Climatology 53, 1; 10.1175/JAMC-D-13-0156.1

Figure 10 shows the same type of vertical profiles for TKE, at the same location (ANL) but during nighttime. Similar to the late morning TKE profiles, the sharp change in TKE within the urban canopy layer is predicted to increase in magnitude throughout the night for all soil moisture conditions. At 2100 LST, the modified urban soil moisture levels reveal larger and smaller values of TKE to be associated with UW and UD, respectively, and are consistent with the relevant features of daytime near-surface wind. TKE predicted at 2100 LST is, however, found to be the lowest for RD. This prediction is consistent with the weaker daytime near-surface wind in the CBD for this case (Fig. 5b) leading to lower wind shear and eventually reduces TKE production.

Fig. 10.
Fig. 10.

As in Fig. 9, but with profiles of TKE (m2 s−2) predicted at three (a)–(c) different nighttimes.

Citation: Journal of Applied Meteorology and Climatology 53, 1; 10.1175/JAMC-D-13-0156.1

Figure 10 also reveals that during late night (0300 LST) and early morning (0600 LST) hours, TKE predictions for RD becomes substantially larger compared to the other cases apparently because of the intensification of the LLJ. Lack of soil moisture results in stronger wind shear close to the surface and consequently increases mechanical generation of TKE. Changes in the urban soil moisture level, however, exert a negligible effect on nighttime near-surface wind. As nocturnal TKE is predominantly generated by mechanical production induced by drag and shear from the canopy, the impact of urban soil moisture on TKE profiles is thus insignificant during late night and early morning hours.

The daytime temperature profiles predicted using the different soil moisture configurations are illustrated in Fig. 11. The profiles show a shallow thermally unstable layer close to the surface for all soil moisture configurations, and the temperatures associated with the RD and UD cases are seen to be the largest. As can be seen in Fig. 9, higher temperature resulting from dry soil (RD and UD) leads to larger TKE values through enhanced thermally generated turbulence, particularly during midday. At 1700 LST, wet rural soils lead to the coolest surface-layer temperature. The increased surface-layer turbulence during this time (Fig. 9c) thus seems to be a result of the minimum reduction in near-surface wind speed that results in sustained mechanical generation of turbulence, and not from buoyancy effects.

Fig. 11.
Fig. 11.

Comparison of observed (gray) and simulated (control: black; RW: blue; RD: red; UW: green; and UD: orange) profiles of potential temperature (°C) at three (a)–(c) different nighttimes. The solid and dashed lines correspond to ANL and PNNL sites, respectively.

Citation: Journal of Applied Meteorology and Climatology 53, 1; 10.1175/JAMC-D-13-0156.1

Observed and predicted vertical profiles of potential temperature for the control case at three different hours during the night of IOP9 are presented in Figs. 12a–c for both upwind and downwind stations PNNL and ANL. The observed profiles during late night (0300 LST) and early morning (0600 LST) hours reveal the existence of a near-neutral layer (NNL) close to the surface for both stations (ANL and PNNL) that are marked by lower thermal stability compared to the layers above them. Near-neutral thermal layers close to the surface occasionally form during night because of slow stabilization of the neutrally stable surface layer forming around dusk that is sustained by increased mechanical turbulence and lower than usual cooling at the surface over urban areas. The model outputs using the control case of initial conditions for soil moisture, as shown in Fig. 12, also predicts the existence of NNL at both ANL and PNNL sites.

Fig. 12.
Fig. 12.

Comparison of observed (gray) and simulated (black: control initial soil moisture fields only) profiles of potential temperature (°C) at three (a)–(c) different nighttimes. Temperature at PNNL in (b) corresponds to 0400 LST 27 Jul 2003 because of missing observational records at 0300 LST. All other aspects are the same as in Fig. 11.

Citation: Journal of Applied Meteorology and Climatology 53, 1; 10.1175/JAMC-D-13-0156.1

Figure 13 shows the temperature profiles at 0300 LST predicted by the model for different soil moisture configurations in addition to the observed and control profiles. Thermal stability of the NNL is related to the vertical gradient of potential temperature within the layer. In view of this, only the RD case is found to significantly affect the thermal stability of the NNL. For a better quantitative comparison, the vertical gradient of the NNL is evaluated following Husain et al. (2013) by estimating a straight line that approximately represents the NNL in the least squares sense such that
e1
where the line θ and the slope represent the near-neutral layer and its vertical gradient respectively, and where is a constant.
Fig. 13.
Fig. 13.

Comparison of observed (gray) and simulated profiles of potential temperature (°C) using initial soil moisture fields given by (a) RW: blue and RD: red, and (b) UW: green and UD: orange. Simulated profiles with initial soil moisture fields denoted by control (black) are shown in both figures for comparison. Temperature at ANL and PNNL correspond to 0300 and 0400 LST 27 Jul 2003, respectively. All other aspects are same as in Fig. 11.

Citation: Journal of Applied Meteorology and Climatology 53, 1; 10.1175/JAMC-D-13-0156.1

Figure 14 presents the evolution of the vertical gradient of potential temperature within the NNL obtained from the observations as well as the model outputs using different initial conditions for soil moisture. Figure 14a shows rapid stabilization of the NNL for the RW case during the early phase of the night for sites apparently because of the colder surface temperature, leading to reduced thermal generation of TKE and weaker vertical mixing. Radiative cooling at the surface slows down with the progress of the night while the intensifying LLJ increases mechanical generation of TKE. The combination of these effects leads to a decreased stabilization of the NNL past midnight for the wet rural soil case as seen in Fig. 14a. The forecasts of vertical temperature gradient using all other cases of initial soil moisture distributions are overall consistent with the observed values.

Fig. 14.
Fig. 14.

Evolution of simulated (control: black; RW: blue; RD: red; UW: green; and UD: orange) and observed vertical gradient of potential temperature (°C m−1) during IOP9 at ANL (simulation: solid lines and observations: solid squares) and PNNL (simulation: dashed lines and observations: hollow squares).

Citation: Journal of Applied Meteorology and Climatology 53, 1; 10.1175/JAMC-D-13-0156.1

Stability characteristics of the NNL for both the upwind and downwind sites using the dry rural case are predicted by the model to remain almost unchanged (see Fig. 14a) with the progression of night, and is in better agreement with the observations. This could be explained again by the more intense LLJ that leads to increased mechanical generation of TKE and therefore to lower static stability throughout the night. The UD case, on the other hand, exhibits a gradual stabilization of the NNL through the night, although for the most part of the night the corresponding NNL is less stable compared to that for wet urban soil (see Fig. 14b). Overall, soil moisture conditions in a rural environment are found to impact the stability properties of the near-neutral surface layer more strongly compared to urban conditions.

The observed vertical profiles of temperature presented in Fig. 12 also point to a considerable temperature difference within the NNLs between the sites upwind and downwind of the CBD. Husain et al. (2013) have shown that this horizontal temperature gradient is caused by the city as southwesterly wind from PNNL warms up as it flows over the city toward ANL. In this regard, results presented in Fig. 11 demonstrate that the horizontal temperature gradient across the city is insignificant during daytime for any soil moisture configuration, and by extension implies that the nocturnal gradient visible in Fig. 12 is a result of the UHI effects.

The effect of soil moisture on the horizontal temperature gradient can be further quantified by computing a mean horizontal gradient of potential temperature within the NNL for the different soil moisture configurations using the following relation:
e2
In the above relation, zb and zNNL denote the heights of the base and of the upper limit of the NNL above ground (Husain et al. 2013), and the subscripts A and P stand for ANL and PNNL, respectively. Figure 15 shows the evolution of the temperature gradient evaluated from the observations and model predictions using (2). For all soil moisture configurations except RW, the predicted mean difference is found to rapidly increase until midnight followed by a period of unchanged difference until 0400 LST and then ending with a sharp decrease by early morning. These results are in relatively good agreement with observations, except in the last portion of the night when the predicted horizontal temperature differences are much less than what is observed. For the RW case, the model outputs lead to a negligible horizontal temperature gradient until 0100 LST, with a rapid increase later in the night that makes the predicted gradient similar to the other cases, followed by a sharp decrease similar to the other cases during the early morning hours. The large discrepancies with the other configurations found in the first part of the night could be related to the stronger low-level winds that are produced in the RW run (see Figs. 6 and 7c), which could have reduced the effect of the city on low-level temperatures because of a shorter warming time (exposition) of air parcels traveling northward over the city.
Fig. 15.
Fig. 15.

Comparison of mean horizontal temperature difference (°C) within the surface layer between ANL and PNNL sites computed from observation (gray squares) and 1-km GEM–LAM simulations using five initial fields for soil moisture: control (black line), RW (blue line), RD (red line), UW (green line), and UD (orange line).

Citation: Journal of Applied Meteorology and Climatology 53, 1; 10.1175/JAMC-D-13-0156.1

6. Conclusions

The influence of soil moisture conditions for rural and urban areas has been investigated in the context of Oklahoma City and its surroundings, and the relevant outcomes are presented in this paper. In addition to a control case for the initial soil moisture content, four additional cases involving modified soil moisture fields have been studied by comparing the model outputs with the observations from the Joint Urban 2003 experiment. In this regard, conclusions drawn from the outcome of this study generally appertain to the medium-size North American cities that are typically surrounded by large rural areas. Overall, initial soil moisture content is found to have considerable influence on the surface-layer temperature and thereby on the magnitude of UHI. This implies that in future, climate change—which is projected to affect precipitation, evaporation, and thus soil moisture conditions—combined with growing urbanization may strongly affect the urban–rural temperature contrasts. As suggested by the results obtained during the present study, this may affect the intensity of UHI and lead to enhanced or reduced generation of secondary mesoscale circulations induced by UHI. Soil moisture also affects the near-surface wind speed by altering thermally generated turbulence, and as a consequence exerts significant influence on vertical mixing within the urban surface layer, particularly during night.

For a city of the size of OKC, results presented in this paper predict a negligible influence of urban soil moisture level on the surrounding rural atmosphere. The daytime impact of urban soil moisture level is found to be a minimum in general, except some influence on the negative UHI intensity. During nighttime, the state of urban soil moisture content is, however, found to affect the urban surface-layer static stability by influencing the generation of mechanical and thermal turbulence.

For a medium-size city, this study demonstrates a greater role played by the rural soil moisture level, compared to its urban counterpart, in determining the surface-layer atmosphere in both rural and urban environments. In general, rural soil moisture content is found to have a stronger impact on daytime near-surface wind. Reduction of rural soil moisture in particular is found to intensify the nocturnal low-level jet. The impact of soil moisture on LLJ is indicative of its impact on large-scale dynamics, and such a conclusion is consistent with the prior findings of other researchers. The nocturnal LLJs are therefore expected to be more intense in the future provided that the soil moisture level in the Great Plains of the United States is reduced because of climate change. The rural soil moisture level is also found to noticeably alter the stability characteristic of the nocturnal near-neutral surface layer within the urban areas. As high-resolution dispersion models for atmospheric pollutants are generally driven by outputs from mesoscale models, accurate initial conditions for soil moisture, particularly for the rural areas, are therefore critical for acceptable forecast accuracy.

REFERENCES

  • Ahmad, S., and N. M. Hashim, 2007: Effects of soil moisture on urban heat island occurrences: Case of Selangor, Malaysia. Humanity Soc. Sci. J., 2, 132138.

    • Search Google Scholar
    • Export Citation
  • Allwine, K. J., M. J. Leach, L. W. Stockham, J. S. Shinn, R. P. Hosker, J. F. Bowers, and J. C. Pace, 2004: Overview of Joint Urban 2003—An atmospheric dispersion study in Oklahoma City. Preprints, Eighth Symp. on Integrated Observing and Assimilation Systems for Atmosphere, Oceans, and Land Surface, Seattle, WA, Amer. Meteor. Soc., J7.1. [Available online at http://ams.confex.com/ams/pdfpapers/74349.pdf.]

  • Baker, R. D., B. H. Lynn, A. Boone, W.-K. Tao, and J. Simpson, 2001: The influence of soil moisture, coastline curvature, and land-breeze circulations on sea-breeze-initiated precipitation. J. Hydrometeor., 2, 193211.

    • Search Google Scholar
    • Export Citation
  • Bélair, S., L.-P. Crevier, J. Mailhot, B. Bilodeau, and Y. Delage, 2003: Operational implementation of the ISBA land surface scheme in the Canadian regional weather forecast model. Part I: Warm season results. J. Hydrometeor., 4, 352370.

    • Search Google Scholar
    • Export Citation
  • Blackadar, A. K., 1957: Boundary layer wind maxima and their significance for the growth of nocturnal inversions. Bull. Amer. Meteor. Soc., 38, 283290.

    • Search Google Scholar
    • Export Citation
  • Bornstein, R., and Q. Lin, 2000: Urban heat islands and summertime convective thunderstorm in Atlanta: Three case studies. Atmos. Environ., 34, 507516.

    • Search Google Scholar
    • Export Citation
  • Chen, F., S. Miao, M. Tewari, J.-W. Bao, and H. Kusaka, 2011: A numerical study of interactions between surface forcing and sea breeze circulations and their effects on stagnation in the greater Houston area. J. Geophys. Res., 116, D12105, doi:10.1029/2010JD015533.

    • Search Google Scholar
    • Export Citation
  • Deardorff, J. W., 1978: Efficient prediction of ground surface temperature and moisture with inclusion of a layer of vegetation. J. Geophys. Res., 83 (C4), 18891903.

    • Search Google Scholar
    • Export Citation
  • Dupont, S., T. L. Otte, and J. K. S. Ching, 2004: Simulation of meteorological fields within and above urban and rural canopies with a mesoscale model (MM5). Bound.-Layer Meteor., 113, 111158.

    • Search Google Scholar
    • Export Citation
  • Ek, M. B., and A. A. M. Holtslag, 2004: Influence of soil moisture on boundary layer cloud development. J. Hydrometeor., 5, 8699.

  • Fast, J. D., and M. D. McCorcle, 1990: A two-dimensional numerical sensitivity study of the Great Plains low-level jet. Mon. Wea. Rev., 118, 151163.

    • Search Google Scholar
    • Export Citation
  • Findell, K. L., and E. A. B. Eltahir, 2003: Atmospheric controls on soil moisture-boundary layer interactions: Three-dimensional wind effects. J. Geophys. Res., 108, 8385, doi:10.1029/2001JD001515.

    • Search Google Scholar
    • Export Citation
  • Hafner, J., and S. Q. Kidder, 1999: Urban heat island modeling in conjunction with satellite-derived surface/soil parameters. J. Appl. Meteor., 38, 448465.

    • Search Google Scholar
    • Export Citation
  • Hidalgo, J., G. Pigeon, and V. Masson, 2008a: Urban-breeze circulation during the CAPITOUL experiment: Observational data analysis approach. Meteor. Atmos. Phys., 102, 223241.

    • Search Google Scholar
    • Export Citation
  • Hidalgo, J., G. Pigeon, and V. Masson, 2008b: Urban-breeze circulation during the CAPITOUL experiment: Numerical simulations. Meteor. Atmos. Phys., 102, 243262.

    • Search Google Scholar
    • Export Citation
  • Holton, J. R., 1967: The diurnal boundary layer wind oscillation above sloping terrain. Tellus, 19, 199205.

  • Husain, S. Z., S. Bélair, J. Mailhot, and S. Leroyer, 2013: Improving the representation of the nocturnal near-neutral surface layer in the urban environment with a mesoscale atmospheric model. Bound.-Layer Meteor., 147, 525551.

    • Search Google Scholar
    • Export Citation
  • Jacobson, M. Z., 1999: Effects of soil moisture on temperatures, winds, and pollutant concentrations in Los Angeles. J. Appl. Meteor., 38, 607616.

    • Search Google Scholar
    • Export Citation
  • Jiang, X., N.-C. Lau, I. M. Held, and J. J. Ploshay, 2007: Mechanisms of the Great Plains low-level jet as simulated in an AGCM. J. Atmos. Sci., 64, 532547.

    • Search Google Scholar
    • Export Citation
  • Kala, J., T. J. Lyons, D. J. Abbs, and U. S. Nair, 2010: Numerical simulations of the impacts of land-cover change on a southern sea breeze in south-west Western Australia. Bound.-Layer Meteor., 135, 485503.

    • Search Google Scholar
    • Export Citation
  • Lemonsu, A., and V. Masson, 2002: Simulation of a urban summer breeze over Paris. Bound.-Layer Meteor., 104, 463490.

  • Lemonsu, A., S. Bélair, and J. Mailhot, 2009: The new Canadian urban modeling system: Evaluation for two cases from the Joint Urban 2003 Oklahoma City Experiment. Bound.-Layer Meteor., 133, 4770.

    • Search Google Scholar
    • Export Citation
  • Leroyer, S., S. Bélair, J. Mailot, and I. B. Strachan, 2011: Microscale numerical prediction over Montreal with the Canadian External Urban Modeling System. J. Appl. Meteor. Climatol., 50, 24102428.

    • Search Google Scholar
    • Export Citation
  • Martilli, A., A. Clappier, and M. W. Rotach, 2002: An urban surface exchange parameterisation for mesoscale models. Bound.-Layer Meteor., 104, 261304.

    • Search Google Scholar
    • Export Citation
  • Masson, V., 2000: A physically-based scheme for the urban energy budget in atmospheric models. Bound.-Layer Meteor., 94, 357397.

  • McCorcle, M. D., 1988: Simulation of surface-moisture effects on the Great Plains low-level jet. Mon. Wea. Rev., 116, 17051720.

  • 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
  • Stensrud, D. J., 1996: Importance of low-level jets to climate: A review. J. Climate, 9, 16981711.

  • Wexler, H., 1961: A boundary layer interpretation of the low-level jet. Tellus, 13, 368378.

  • Yeh, K.-S., J. Côté, S. Gravel, A. Methot, A. Patoine, M. Roch, and A. Staniforth, 2002: The CMC-MRB global environmental multiscale (GEM) model. Part III: Nonhydrostatic formulation. Mon. Wea. Rev., 130, 339356.

    • Search Google Scholar
    • Export Citation
  • Zadra, A., D. Caya, J. Cote, B. Dugas, C. Jones, R. Laprise, K. Winger, and L. P. Caron, 2008: The next Canadian regional climate model. Phys. Canada, 64, 7583.

    • Search Google Scholar
    • Export Citation
Save
  • Ahmad, S., and N. M. Hashim, 2007: Effects of soil moisture on urban heat island occurrences: Case of Selangor, Malaysia. Humanity Soc. Sci. J., 2, 132138.

    • Search Google Scholar
    • Export Citation
  • Allwine, K. J., M. J. Leach, L. W. Stockham, J. S. Shinn, R. P. Hosker, J. F. Bowers, and J. C. Pace, 2004: Overview of Joint Urban 2003—An atmospheric dispersion study in Oklahoma City. Preprints, Eighth Symp. on Integrated Observing and Assimilation Systems for Atmosphere, Oceans, and Land Surface, Seattle, WA, Amer. Meteor. Soc., J7.1. [Available online at http://ams.confex.com/ams/pdfpapers/74349.pdf.]

  • Baker, R. D., B. H. Lynn, A. Boone, W.-K. Tao, and J. Simpson, 2001: The influence of soil moisture, coastline curvature, and land-breeze circulations on sea-breeze-initiated precipitation. J. Hydrometeor., 2, 193211.

    • Search Google Scholar
    • Export Citation
  • Bélair, S., L.-P. Crevier, J. Mailhot, B. Bilodeau, and Y. Delage, 2003: Operational implementation of the ISBA land surface scheme in the Canadian regional weather forecast model. Part I: Warm season results. J. Hydrometeor., 4, 352370.

    • Search Google Scholar
    • Export Citation
  • Blackadar, A. K., 1957: Boundary layer wind maxima and their significance for the growth of nocturnal inversions. Bull. Amer. Meteor. Soc., 38, 283290.

    • Search Google Scholar
    • Export Citation
  • Bornstein, R., and Q. Lin, 2000: Urban heat islands and summertime convective thunderstorm in Atlanta: Three case studies. Atmos. Environ., 34, 507516.

    • Search Google Scholar
    • Export Citation
  • Chen, F., S. Miao, M. Tewari, J.-W. Bao, and H. Kusaka, 2011: A numerical study of interactions between surface forcing and sea breeze circulations and their effects on stagnation in the greater Houston area. J. Geophys. Res., 116, D12105, doi:10.1029/2010JD015533.

    • Search Google Scholar
    • Export Citation
  • Deardorff, J. W., 1978: Efficient prediction of ground surface temperature and moisture with inclusion of a layer of vegetation. J. Geophys. Res., 83 (C4), 18891903.

    • Search Google Scholar
    • Export Citation
  • Dupont, S., T. L. Otte, and J. K. S. Ching, 2004: Simulation of meteorological fields within and above urban and rural canopies with a mesoscale model (MM5). Bound.-Layer Meteor., 113, 111158.

    • Search Google Scholar
    • Export Citation
  • Ek, M. B., and A. A. M. Holtslag, 2004: Influence of soil moisture on boundary layer cloud development. J. Hydrometeor., 5, 8699.

  • Fast, J. D., and M. D. McCorcle, 1990: A two-dimensional numerical sensitivity study of the Great Plains low-level jet. Mon. Wea. Rev., 118, 151163.

    • Search Google Scholar
    • Export Citation
  • Findell, K. L., and E. A. B. Eltahir, 2003: Atmospheric controls on soil moisture-boundary layer interactions: Three-dimensional wind effects. J. Geophys. Res., 108, 8385, doi:10.1029/2001JD001515.

    • Search Google Scholar
    • Export Citation
  • Hafner, J., and S. Q. Kidder, 1999: Urban heat island modeling in conjunction with satellite-derived surface/soil parameters. J. Appl. Meteor., 38, 448465.

    • Search Google Scholar
    • Export Citation
  • Hidalgo, J., G. Pigeon, and V. Masson, 2008a: Urban-breeze circulation during the CAPITOUL experiment: Observational data analysis approach. Meteor. Atmos. Phys., 102, 223241.

    • Search Google Scholar
    • Export Citation
  • Hidalgo, J., G. Pigeon, and V. Masson, 2008b: Urban-breeze circulation during the CAPITOUL experiment: Numerical simulations. Meteor. Atmos. Phys., 102, 243262.

    • Search Google Scholar
    • Export Citation
  • Holton, J. R., 1967: The diurnal boundary layer wind oscillation above sloping terrain. Tellus, 19, 199205.

  • Husain, S. Z., S. Bélair, J. Mailhot, and S. Leroyer, 2013: Improving the representation of the nocturnal near-neutral surface layer in the urban environment with a mesoscale atmospheric model. Bound.-Layer Meteor., 147, 525551.

    • Search Google Scholar
    • Export Citation
  • Jacobson, M. Z., 1999: Effects of soil moisture on temperatures, winds, and pollutant concentrations in Los Angeles. J. Appl. Meteor., 38, 607616.

    • Search Google Scholar
    • Export Citation
  • Jiang, X., N.-C. Lau, I. M. Held, and J. J. Ploshay, 2007: Mechanisms of the Great Plains low-level jet as simulated in an AGCM. J. Atmos. Sci., 64, 532547.

    • Search Google Scholar
    • Export Citation
  • Kala, J., T. J. Lyons, D. J. Abbs, and U. S. Nair, 2010: Numerical simulations of the impacts of land-cover change on a southern sea breeze in south-west Western Australia. Bound.-Layer Meteor., 135, 485503.

    • Search Google Scholar
    • Export Citation
  • Lemonsu, A., and V. Masson, 2002: Simulation of a urban summer breeze over Paris. Bound.-Layer Meteor., 104, 463490.

  • Lemonsu, A., S. Bélair, and J. Mailhot, 2009: The new Canadian urban modeling system: Evaluation for two cases from the Joint Urban 2003 Oklahoma City Experiment. Bound.-Layer Meteor., 133, 4770.

    • Search Google Scholar
    • Export Citation
  • Leroyer, S., S. Bélair, J. Mailot, and I. B. Strachan, 2011: Microscale numerical prediction over Montreal with the Canadian External Urban Modeling System. J. Appl. Meteor. Climatol., 50, 24102428.

    • Search Google Scholar
    • Export Citation
  • Martilli, A., A. Clappier, and M. W. Rotach, 2002: An urban surface exchange parameterisation for mesoscale models. Bound.-Layer Meteor., 104, 261304.

    • Search Google Scholar
    • Export Citation
  • Masson, V., 2000: A physically-based scheme for the urban energy budget in atmospheric models. Bound.-Layer Meteor., 94, 357397.

  • McCorcle, M. D., 1988: Simulation of surface-moisture effects on the Great Plains low-level jet. Mon. Wea. Rev., 116, 17051720.

  • 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
  • Stensrud, D. J., 1996: Importance of low-level jets to climate: A review. J. Climate, 9, 16981711.

  • Wexler, H., 1961: A boundary layer interpretation of the low-level jet. Tellus, 13, 368378.

  • Yeh, K.-S., J. Côté, S. Gravel, A. Methot, A. Patoine, M. Roch, and A. Staniforth, 2002: The CMC-MRB global environmental multiscale (GEM) model. Part III: Nonhydrostatic formulation. Mon. Wea. Rev., 130, 339356.

    • Search Google Scholar
    • Export Citation
  • Zadra, A., D. Caya, J. Cote, B. Dugas, C. Jones, R. Laprise, K. Winger, and L. P. Caron, 2008: The next Canadian regional climate model. Phys. Canada, 64, 7583.

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

    (a) The 2.5-km GEM–LAM simulation domain and the corresponding orography. The long-dashed and short-dashed rectangles represent the 1-km simulation domain (161 × 161 grid points, excluding the piloting zone for enforcing lateral boundary conditions) and the approximate bounds of OKC (30 × 41 grid points), respectively. The rural and urban stations are shown using green and black markers, respectively. (b) Distribution of urban fraction over OKC along with the positions of the ANL and PNNL sites with respect to the CBD. The PWIDS network of 13 stations in the CBD appears as a single point in both figures because of their close proximity to each other.

  • Fig. 2.

    Volumetric soil moisture content (m3 m−3) for (a) near-surface and (b) root-zone layers corresponding to the control case for 2.5-km GEM–LAM simulation domain valid at 0400 UTC 26 Jul 2003. The rectangle in (a) indicates the urban area.

  • Fig. 3.

    Evolution of screen-level air temperature (°C) in (a) rural and (b) urban areas for IOP9 obtained from observations (gray line with squares) and 1-km GEM–LAM simulations using five initial estimates for the soil moisture field, namely, control (black line with triangles), RW (blue line), RD (red line), UW (green line), and UD (orange line).

  • Fig. 4.

    Comparison of urban heat island [UHI = (TCBDTRural)°C] computed from observations (gray line with squares) and 1-km GEM–LAM simulations using five initial soil moisture fields: control (black line with triangles), RW (blue line), RD (red line), UW (green line), and UD (orange line).

  • Fig. 5.

    As in Fig. 3, but for the evolution of near-surface wind speed (m s−1).

  • Fig. 6.

    As in Fig. 4, but for comparison of urban wind speed reduction [UWR = (WSCBD − WSRural) m s−1].

  • Fig. 7.

    Comparison of simulated (control: black; RW: blue; RD: red; UW: green; and UD: orange) and observed (gray) daytime wind speed (m s−1) profiles within the lowest 500 m at times (a) 0900, (b) 1300, and (c) 1700 LST 26 Jul 2003 at ANL.

  • Fig. 8.

    Comparison of simulated (control: black; RW: blue; RD: red; UW: green; UD: orange) and observed (gray) nighttime wind speed (m s−1) profiles within the lowest (top) 2500 and (bottom) 500 m at times (a),(d) 2100 LST 26 Jul; (b),(e) 0300; and (c),(f) 0600 LST 27 Jul 2003 at ANL.

  • Fig. 9.

    Profiles of TKE (m2 s−2) predicted at three (a)–(c) different daytimes using the five initial fields for soil moisture: control (black line), RW (blue line), RD (red line), UW (green line), and UD (orange line) at ANL.

  • Fig. 10.

    As in Fig. 9, but with profiles of TKE (m2 s−2) predicted at three (a)–(c) different nighttimes.

  • Fig. 11.

    Comparison of observed (gray) and simulated (control: black; RW: blue; RD: red; UW: green; and UD: orange) profiles of potential temperature (°C) at three (a)–(c) different nighttimes. The solid and dashed lines correspond to ANL and PNNL sites, respectively.

  • Fig. 12.

    Comparison of observed (gray) and simulated (black: control initial soil moisture fields only) profiles of potential temperature (°C) at three (a)–(c) different nighttimes. Temperature at PNNL in (b) corresponds to 0400 LST 27 Jul 2003 because of missing observational records at 0300 LST. All other aspects are the same as in Fig. 11.

  • Fig. 13.

    Comparison of observed (gray) and simulated profiles of potential temperature (°C) using initial soil moisture fields given by (a) RW: blue and RD: red, and (b) UW: green and UD: orange. Simulated profiles with initial soil moisture fields denoted by control (black) are shown in both figures for comparison. Temperature at ANL and PNNL correspond to 0300 and 0400 LST 27 Jul 2003, respectively. All other aspects are same as in Fig. 11.

  • Fig. 14.

    Evolution of simulated (control: black; RW: blue; RD: red; UW: green; and UD: orange) and observed vertical gradient of potential temperature (°C m−1) during IOP9 at ANL (simulation: solid lines and observations: solid squares) and PNNL (simulation: dashed lines and observations: hollow squares).

  • Fig. 15.

    Comparison of mean horizontal temperature difference (°C) within the surface layer between ANL and PNNL sites computed from observation (gray squares) and 1-km GEM–LAM simulations using five initial fields for soil moisture: control (black line), RW (blue line), RD (red line), UW (green line), and UD (orange line).

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 1028 257 11
PDF Downloads 812 183 4