Evaluating the Performance of WRF Urban Schemes and PBL Schemes over Dallas–Fort Worth during a Dry Summer and a Wet Summer

Jinxin Wang aCenter for Analysis and Prediction of Storms, University of Oklahoma, Norman, Oklahoma
bSchool of Meteorology, University of Oklahoma, Norman, Oklahoma
cNational Supercomputing Center, Wuxi, China

Search for other papers by Jinxin Wang in
Current site
Google Scholar
PubMed
Close
and
Xiao-Ming Hu aCenter for Analysis and Prediction of Storms, University of Oklahoma, Norman, Oklahoma
bSchool of Meteorology, University of Oklahoma, Norman, Oklahoma

Search for other papers by Xiao-Ming Hu in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

This study evaluated the Weather Research and Forecasting (WRF) Model sensitivity to different planetary boundary layer (PBL) schemes (the YSU and MYJ schemes) and urban schemes including the bulk scheme (BULK), single-layer urban canopy model (UCM), multilayer building environment parameterization (BEP) model, and multilayer building energy model (BEM). Daily reinitialization simulations were conducted over Dallas–Fort Worth during a dry summer month (July 2011) and a wet summer month (July 2015) with weaker (stronger) daytime (nocturnal) UHI in 2011 than 2015. All urban schemes overestimated the urban daytime 2-m temperature in both summers, but BEP and BEM still reproduced the daytime urban cool island in the dry summer. All urban schemes reproduced the nocturnal urban heat island, with BEP producing the weakest one due to its unrealistic urban cooling. BULK and UCM overestimated the urban canopy wind speed, while BEP and BEM underestimated it. The urban schemes showed prominent impact on daytime PBL profiles. UCM + MYJ showed a superior performance than other configurations. The relatively large (small) aspect ratio between building height and road width in UCM (BEM) was responsible for the overprediction (underprediction) of urban canopy temperature. The relatively low (high) building height in UCM (BEM) was responsible for the overprediction (underprediction) of urban canopy wind speed. Improving urban schemes and providing realistic urban parameters were critical for improving urban canopy simulation.

© 2021 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Corresponding author: Jinxin Wang, jinxin_wang@yeah.net

Abstract

This study evaluated the Weather Research and Forecasting (WRF) Model sensitivity to different planetary boundary layer (PBL) schemes (the YSU and MYJ schemes) and urban schemes including the bulk scheme (BULK), single-layer urban canopy model (UCM), multilayer building environment parameterization (BEP) model, and multilayer building energy model (BEM). Daily reinitialization simulations were conducted over Dallas–Fort Worth during a dry summer month (July 2011) and a wet summer month (July 2015) with weaker (stronger) daytime (nocturnal) UHI in 2011 than 2015. All urban schemes overestimated the urban daytime 2-m temperature in both summers, but BEP and BEM still reproduced the daytime urban cool island in the dry summer. All urban schemes reproduced the nocturnal urban heat island, with BEP producing the weakest one due to its unrealistic urban cooling. BULK and UCM overestimated the urban canopy wind speed, while BEP and BEM underestimated it. The urban schemes showed prominent impact on daytime PBL profiles. UCM + MYJ showed a superior performance than other configurations. The relatively large (small) aspect ratio between building height and road width in UCM (BEM) was responsible for the overprediction (underprediction) of urban canopy temperature. The relatively low (high) building height in UCM (BEM) was responsible for the overprediction (underprediction) of urban canopy wind speed. Improving urban schemes and providing realistic urban parameters were critical for improving urban canopy simulation.

© 2021 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Corresponding author: Jinxin Wang, jinxin_wang@yeah.net

1. Introduction

Cities are facing increasing challenges with growing global urbanization (Georgescu et al. 2014). Nowadays, more than half the world’s population lives in cities, and this is expected to reach two-thirds in the coming half century (United Nations Department of Economic and Social Affairs 2012). Urbanization alters the local weather and climate [e.g., urban heat island (UHI)] and threatens the comfort of the residents. Due to lack of dense surface and upper-level observation, numerical models are often utilized to better understand the urban impact on local weather and climate.

There are four urban schemes coupled with the Weather Research and Forecasting (WRF) Model. The first urban scheme is a bulk (BULK) scheme. It increases heat capacity and thermal conductivity to represent urban heat storage and enhances roughness to represent sink of momentum and generation of turbulence (Taha 1999; Liu et al. 2006). Its disadvantage is that it cannot represent the heterogeneity of urban variability among different neighborhoods (Salamanca et al. 2011). Reames and Stensrud (2017) separated single urban class into three urban classes to improve BULK’s performance. Since WRF version 3.7, detailed urban classification was implemented for BULK through incorporating the National Land Cover Database (NLCD) into the WRF Preprocessing System (WPS).

The second urban scheme is the single-layer Urban Canopy Model (UCM) (Kusaka et al. 2001; Kusaka and Kimura 2004a,b). It consists of two-dimensional symmetrical street canyons with infinite length and three urban surfaces (Kusaka et al. 2001). Though the geometry of the city is simplified, the radiation treatment in urban canopy is three-dimensional (Kusaka et al. 2001). UCM uses three different urban classes with different thermal and roughness properties to represent urban heterogeneity. UCM uses approximately 30 input parameters for the thermal properties. Among these parameters, urban fraction plays a dominant role in modulating latent heat flux (Nemunaitis-Berry et al. 2017) because urban fraction determines the relative contribution of urbanized and vegetated landscapes to the total flux.

The third urban scheme is the multilayer building environment parameterization (BEP) model. BEP introduces a scheme to represent the impact of urban buildings on resolved airflow (Martilli et al. 2002). BEP can reproduce the increase of Reynolds stress profiles with height, store more energy in the urban fabric during the day, and reduce the turbulence intensity below the average roof height by changing the length scale (Martilli 2002; Martilli et al. 2002).

The fourth urban scheme is multilayer building energy model (BEM). BEM is added to BEP to reproduce the generation of heat in buildings and the way heat is exchanged with the exterior flow (Salamanca et al. 2009). BEM considers heat diffusion through walls, roofs, and floors; natural ventilation; the radiation exchange between indoor surfaces; the generation of heat due to occupants and equipment; and the consumption of energy due to air conditioning (AC) systems (Salamanca et al. 2009). An offline evaluation of BEM against BEP showed that BEM satisfactorily produced the urban fluxes (Salamanca and Martilli 2009). Consequently, BEM might lead to a better simulation of UHI and urban boundary layer (Salamanca and Martilli 2009).

Efforts have been paid to evaluate the four schemes’ performance within the WRF Model. BULK and UCM can be coupled with any planetary boundary layer (PBL) schemes in WRF, while BEP and BEM are currently coupled only with two local PBL schemes, the Mellor–Yamada–Janjić (MYJ) (Janjić 1990, 1994) and the Bougeault–Lacarrére (BouLac) (Bougeault and Lacarrere 1989). Hendricks et al. (2020) achieved positive outcomes when coupling BEP/BEM with modified YSU scheme in their experiments. Salamanca et al. (2011) found that all urban schemes except BULK underestimate the air temperature, and UCM overestimated the wind speed while BEP and BEM underestimated it. Liao et al. (2014) concluded that the BULK scheme was suitable for real-time weather forecast and a multilayer urban scheme was beneficial for studying the impact of urbanization on regional climate. Gutiérrez et al. (2015) found that BULK tended to overestimate daytime wind speed and daytime temperature. Jänicke et al. (2017) concluded that BULK performed better than the UCM and BEP, and the BouLac performed better than MYJ. Sharma et al. (2017) found BEM with High-Resolution Land Data Assimilation System (HRLDAS) had quite low model bias and root-mean-square error for both during day and night.

Despite these previous evaluations, the performance of urban and PBL schemes and their influence factors remain inconclusive. Observational climatology studies appeared to suggest that the UHI intensity had a close relation with humidity, with strong UHI in dry scenarios and weak UHI in wet scenarios (Winguth and Kelp 2013; Hu et al. 2016). However, the background mechanism remains elusive. In this study, we aimed to evaluate the urban and PBL schemes’ performance in different humidity scenarios and attempted to explain the reason for their performance. To this end, we conducted daily simulations over a dry summer month (July 2011) and a wet summer month (July 2015). In addition to examining model sensitivity to different PBL and urban schemes, different characteristics of urban effects and their causative factors during the two months were also investigated.

In the following, section 2 describes selected episodes, observational datasets, and model configurations; section 3 evaluates model performance against surface observation and sounding, particularly in terms of model capability to simulate urban effects including both thermal and dynamic effects; section 4 gives the conclusion.

2. Model setup and data

a. Model configuration

The WRF Model (version 3.8) was employed to conduct the simulation over the two months. Four two-way nested domains were at spatial resolutions of 27 km (D1, 190 × 160), 9 km (D2, 190 × 160), 3 km (D3, 190 × 160), and 1 km (D4, 205 × 235) (Fig. 1a), with 48 levels extending from the surface to 100 hPa. The heights of the lowest 27 levels are listed in Table 1. The innermost domain, D4, covers both the Dallas–Fort Worth (DFW) urban region and its surrounding rural region. All the following model analysis was based on the simulation result of D4. The initial and boundary conditions were from North American Regional Reanalysis (NARR) dataset. All experiments were initialized at 1200 UTC [0600 central standard time (CST)] every day from 30 June to 30 July and simulated for 47 h. The boundary conditions were updated every 3 h. The first 23 h were discarded as model spinup, the remaining 24 h were used for analysis. The physics schemes used for this study included WRF single-moment 6-class (WSM6) microphysics scheme (Hong et al. 2004), the Rapid Radiative Transfer Model longwave radiation scheme (Mlawer et al. 1997), the Dudhia shortwave radiation scheme (Dudhia 1989), the Eta surface layer scheme (Janjić 1994), the MYJ PBL scheme (Janjić 1994), the Noah Land Surface Model (Chen and Dudhia 2001), and the four urban schemes. The Kain–Fritsch cumulus scheme (Kain 2004), was applied for the D1 and D2. One additional set of simulations with UCM coupled with the YSU PBL scheme (Hong et al. 2006) was conducted, comparing with UCM + MYJ to evaluate the PBL schemes’ impact on urban atmosphere simulation. In simulation of real cases, BEM usually worked with BEP, which was named BEP + BEM in previous studies, but in this study, we named it BEM for simplicity.

Fig. 1.
Fig. 1.

(a) Configuration of the four two-way nested domains for simulation. Shading is terrain height in meters. (b) NLCD 40-class land-cover maps for D4. (c) Updated representative diurnal profiles of AH for three urban land-use categories (black dots and lines denote IC, purple dots and lines denote HIR, and red dots and lines denote LIR).

Citation: Journal of Applied Meteorology and Climatology 60, 6; 10.1175/JAMC-D-19-0195.1

Table 1.

The heights of the lowest 27 model levels.

Table 1.

b. Land-use category update

The heterogeneity of land-use category is one of the reasons for the heterogeneity of near-surface meteorological variables. Therefore, an up-to-date high-resolution land-cover dataset becomes important for reliable simulation of the urban atmosphere. The original 24-class USGS land-cover data with its finest spatial resolution 30 s (~1 km) was collected in the early 1990s, and much of the land cover has measurably changed after two decades. The 20-class NLCD 2011 with 30-m spatial resolution is an up-to-date high-resolution land-cover database over the United States. By integrating the 20-class NLCD 2011 and the 20-class MODIS (Wickham et al. 2015), the U.S. Environmental Protection Agency produces the 40-class NLCD 2011 (Table 2; see the NLCD over the DFW region in Fig. 1b). The percentage of the three urban classes (high, medium, and low intensity) were also derived from NLCD 2011, which are 0.1, 0.16, and 0.74 over DFW.

Table 2.

Description of land-use categories and their corresponding percentages in D4. Asterisks indicate that the land-use category only exists in the polar region.

Table 2.

The National Urban Database and Access Portal Tool (NUDAPT) provides additional information with a resolution of 1 km2 for urban parameterization (Ching et al. 2009). The NUDAPT contains the urban morphology of 44 cities in the United States.

c. Anthropogenic heat update for UCM

There are two major energy sources over the urban land surface: net radiation and anthropogenic heat (AH). In UCM, the diurnal AH profile has a close relation with urban density. The default values of maximum hourly AH in WRF are 90, 50, and 20 W m−2, respectively, for the industry or commercial (IC), high-intensity residential (HIR), and low-intensity residential (LIR) urban land-use categories. Different cities have different AH profiles. Efforts for creating proper AH profiles for major cities have been made in recent years. Based on U.S. National Emission Inventory year 2005 data, Lee et al. (2014) reported that the summertime maximum AH over 4 km × 4 km grid cells of DFW was 65.9 W m−2 and the city-scale averaged maximum AH of DFW was 17.4 W m−2. With consideration of multiple datasets, including electricity data and transportation data, Sailor et al. (2015) reported that the summertime city-scale averaged maximum summer AH in DFW was 13.99 W m−2, and provided an updated hourly AH profile. In this study, based on Sailor et al. (2015), a simple approach was designed to derive the diurnal maximum AH, with the ratio of maximum AH of high-, medium-, low-intensity urban area assumed to be 9:5:2 and citywide averaged AH to be 13.99 W m−2. With these assumptions, one could derive that the diurnal maximum AH for the three urban types is 39.4, 21.9, and 8.8 W m−2, respectively. The updated AH profiles for the three urban classes were shown in Fig. 1c. The default AH parameters in BEP and BEM were also recommended to be updated (Salamanca et al. 2011; Liao et al. 2014; Gutiérrez et al. 2015; Salamanca et al. 2015; Sharma et al. 2017). However, because of the lack of official document about the DFW urban parameters for BEP and BEM, the parameters in BEP and BEM kept as default in this study.

d. Sensitivity study of urban geometry parameters

Previous studies have confirmed that urban geometry had significant effect on urban energy budget (Oke 1982; Stewart and Oke 2012). The urban geometry can be described by a few variables including sky view factor (SVF) and aspect ratio (AR) (Stewart and Oke 2012), which can be further derived from building height (BH), road width (RDW), and roof width (RFW).

The urban geometry parameters in UCM and BEM adopted in the control experiments (UCM-CTRL and BEM-CTRL) were listed in Table 3 and Table 4, respectively. The default BH in UCM-CTRL was lower than that in BEM-CTRL, which was suspected as the reason for the larger canopy wind speed in UCM-CTRL than that in BEM-CTRL. For studying the impact of BH on canopy wind speed, UCM-HBH adopted higher building height (HBH) and BEM-LBH adopted lower building height (LBH). The BH in BEM-LBH was same as the BH in UCM-CTRL, while the BH in UCM-HBH was lower than the BH in BEM-CTRL because UCM was not allowed to adopt a BH higher than the first model level. For studying the impact of RDW and RFW on urban canopy atmosphere, BEM-LBH-NRDW based on BEM-LBH adopted narrower RDW (same to the RDW in UCM-CTRL) and BEM-LBH-NRDW-NRFW based on BEM-LBH-NRDW adopted narrower RFW (same to the RFW in UCM-CTRL). Therefore, the urban geometry parameters in BEM-LBH-NRDW-NRFW were same as those in UCM-CTRL. See Table 5 for all parameters for the sensitivity simulations. All experiments in this part were initialized at 1200 UTC (0600 CST) 30 June 2011 and ended at 1100 UTC (0500 CST) 1 July 2011. The boundary conditions were updated every 3 h. The first 23 h were discarded as model spinup, the remaining 24 h were used for analysis. All other physics schemes remained same.

Table 3.

Urban geometry parameters (building height, road width, and roof width) used for three urban categories.

Table 3.
Table 4.

As in Table 3, but used in BEM. The building height in BEM is not fixed as in UCM but classified with different percentages.

Table 4.
Table 5.

The urban geometry parameters in the sensitivity simulations.

Table 5.

e. Observation data

Texas Commission on Environmental Quality (TCEQ) observation network was used to evaluate simulated near-surface variables. TCEQ data were proven reliable for UHI analysis (Winguth and Kelp 2013; Hu and Xue 2016). There were 26 (40) stations available during July 2011 (2015) (Table 6; see Fig. 2 for their locations). According to the NLCD land-use category, 8 (16) urban stations were within DFW urban area during July 2011 (2015) (Figs. 2a,b). However, the land-use category is not the only reference to identify the station type. Thermal inertia is another key factor to distinguish urban sites from rural sites. During the early evening transition period, the urban cooling rates were lower than the rural cooling rates, and the cooling rates difference between the two summers was also noticeable (Figs. 2c,d). The cooling rates during the wet summer were lower than the cooling rates during the dry summer (Figs. 2c,d), which meant the near-surface atmosphere had larger thermal inertia during the wet summer than during the dry summer because of the soil moisture difference between the two summers. With consideration of both land-use category and thermal inertia, five TCEQ urban stations were chosen for model verification. All the five stations had a record of 2-m temperature and 10-m wind speed, but only two stations, Dallas Hinton and Fort Worth Northwest, have records of temperature of dewpoint (TD).

Table 6.

Information of the TCEQ stations in the DFW metropolitan area in 2011, including their full names, station number, short names (STN), land-use categories (LULC), urban fraction/surface imperviousness (IMP), elevation (HGT), and availability of records of temperature, wind, dewpoint, and relative humidity.

Table 6.
Fig. 2.
Fig. 2.

The land-use categories of TCEQ stations in (a) July 2011 and (b) July 2015 and the spatial distribution of early evening transition cooling rate of TCEQ stations in (c) July 2011 and (d) July 2015 overlaid on imperviousness (%) from NLCD 2011. The blue background is bodies of water. The blue star denotes the location of the Fort Worth radiosonde. The five red diamonds denote the locations of the selected urban stations, and the eight blue squares denote the locations of the selected rural stations.

Citation: Journal of Applied Meteorology and Climatology 60, 6; 10.1175/JAMC-D-19-0195.1

The 1-s resolution Stratosphere–Troposphere Process and Their Role in Climate (SPARC) sounding at Fort Worth upper-air station (32.83500°N, 97.29806°W) (blue star in Fig. 2) during July 2011 and July 2015 was used for verifying simulated profiles. The SPARC soundings were launched twice each day, one from 0500 to 0700 CST and another from 1700 to 1900 CST.

3. Results

a. Diurnal variation of urban canopy temperature, wind speed, RH, and TD

All urban schemes over predicted the daytime 2-m temperature in both summers (Figs. 3a,b). Among the four urban schemes, BULK gave the largest overprediction of 2-m temperature at both daytime and nighttime in the two summers (Figs. 3a,b). In dry summer, UCM gave the second-largest overprediction of daytime 2-m temperature (Fig. 3a), while in wet summer, UCM simulated similar daytime 2-m temperature to BEP and BEM. In both summers, the performance of BEP was almost same to that of BEM at daytime, but BEP cooled the city much faster at nighttime (Figs. 3a,b). This was because the indoor temperature of BEP was fixed to 298 K (25°C), which was lower than the outdoor temperature in summer. As a result, the building became a thermal sink and produced toward building heat flux at night. We can imagine that BEP would overpredict urban air temperature when the indoor temperature would be higher than the outdoor temperature. UCM + YSU produced higher nocturnal near-surface temperature than UCM + MYJ (Figs. 3a,b) because of stronger PBL vertical mixing by YSU, which transported the warmer upper air down to the surface.

Fig. 3.
Fig. 3.

Observed and simulated mean diurnal variation of (a),(b) 2-m temperature, (c),(d) 10-m wind speed, (e),(f) 2-m relative humidity, and (g),(h) 2-m dewpoint at five urban stations during (left) July 2011 and (right) July 2015. The five urban stations are denoted by red diamonds in Fig. 2.

Citation: Journal of Applied Meteorology and Climatology 60, 6; 10.1175/JAMC-D-19-0195.1

BULK and UCM predicted stronger wind speed than observation, while BEP and BEM predicted more realistic wind speed (Figs. 3c,d). This is because BULK and UCM use a roughness length that is not directly dependent on urban morphology, while BEP and BEM estimate the momentum sink with a drag force that depends on urban morphology (Salamanca et al. 2011). By adjusting the roughness length for the urban category, Reames and Stensrud (2017) improved BULK’s performance on simulating urban wind speed. Hu et al. (2016) contributed the overestimation of urban wind speed in UCM to the overestimation of vertical transport of momentum. BULK + MYJ produced larger nocturnal near-surface urban wind speed than UCM + MYJ (Figs. 3c,d). UCM + YSU produced larger nocturnal near-surface wind speed than UCM + MYJ (Figs. 3c,d) because of stronger vertical mixing by YSU. The wind speed simulation contrast between urban schemes and PBL schemes suggested that the vertical transport of momentum does take effect on urban wind simulation. However, improving urban schemes would bring more significant effect on urban wind simulation. Barlage et al. (2016) pointed out the dynamic reason why BEP and BEM performed better than BULK and UCM in urban canopy wind simulation: BULK and UCM were separated from the atmosphere above and communicated with the above atmosphere only with bulk surface flux and gridscale roughness length, while BEP and BEM allowed buildings to protrude into the lower atmosphere and were more direct and realistic in capturing building drag effect.

Because of the significant negative relationship between RH and temperature, the performance of the four urban schemes on RH prediction was reversed to that on temperature prediction. All urban schemes tend to underpredict the daytime RH, especially in wet summer (Figs. 3e,f). As TD is less sensitive to temperature, the simulation of TD was evaluated to better understand the urban schemes’ performance on urban humidity prediction. In both summers, all models tend to overpredict the TD at daytime and underpredict the TD at nighttime (Figs. 3g,h). In dry summer, UCM + YSU predicted the lowest daytime TD. In wet summer, BULK + MYJ predicted the lowest daytime TD, while UCM + MYJ predicted the highest daytime TD (Figs. 3g,h). The TD predicted by BEP and BEM were almost the same (Figs. 3g,h). The TD diurnal profiles by all models deviated much more away from observation in wet summer than in dry summer because the low-resolution WRF initialization data from NARR could not resolve the heterogeneity of soil moisture in wet summer. In dry summer, the simulated TD diurnal profiles were closer to observation because it was too dry within the city and the urban soil moisture heterogeneity was reduced. Some studies used the HRLDAS to improve the initial soil condition in the WRF Model (Nemunaitis-Berry et al. 2017; Reames and Stensrud 2017; Sharma et al. 2017). However, the soil moisture simulated by HRLDAS still deviated from observation, though the urban–rural contrast could be resolved (Reames and Stensrud 2017).

b. Diurnal variation of surface energy balance

According to Oke (1982) and Oke (1988), the surface energy balance equation is written as
Q*+QF=QH+QE+ΔQs+ΔQA,
where Q* is the net all-wave radiation, QF is anthropogenic energy, QH is sensible heat flux, QE is latent heat flux, ΔQs is stored energy, and ΔQA is net heat advection. The net all-wave radiation Q* received by the ground is defined as
Q*=(1α)S+ε(LσTskin4),
where α is surface albedo, ε is surface emissivity, σ is the Stefan–Boltzmann constant, S is downward shortwave radiation, L is downward longwave radiation, and Tskin is skin temperature in kelvin.

In UCM and BEM, QF is added to QH. The net heat advection ΔQA is suggested to be negligible when the measurement is accurate (Christen and Vogt 2004). It is reported that ΔQs has a positive correlation with Q* (Grimmond 1992; Christen and Vogt 2004). The positive (negative) value of ΔQs means there is storing (releasing) ground heat flux. The positive (negative) value of Q* means there is a net income (outcome) of net radiation for the surface.

The net radiation amount received by the city were similar for all urban schemes and PBL schemes in both summers (Figs. 4a,b). The energy partition was more sensitive to urban schemes than to PBL schemes (Figs. 4c–f). The energy partition was almost unchanged for BULK between the two summers (Figs. 4c–h) because BULK treated all urban grids as concrete surface (assuming 100% urban fraction) with very high resistance to evapotranspiration (Figs. 4e,f). As a result, in BULK, most energy was partitioned into sensible heat flux, which was more apparent in wet summer (Figs. 4c,d). The weakest latent heat flux by BULK resulted in the lowest TD. BULK stored the most ground heat flux at daytime and released them as sensible heat flux at nighttime (Figs. 4g,h), which explained why BULK predicted the highest nocturnal 2-m temperature (Figs. 3a,b). Reames and Stensrud (2017) modified the scaling factors that modulated evapotranspiration in response to insolation and vapor pressure deficit and achieved some improvement to BULK’s performance. The weak evapotranspiration in BULK is actually caused by the underestimation of subgrid-scale variability of land surface characteristics (Li et al. 2013). Li et al. (2013) pointed out that the subgrid-scale variability of land surface characteristics continued to be important and should be accounted for, even if the resolution of numerical simulation was as high as 1 km.

Fig. 4.
Fig. 4.

Simulated mean diurnal variation of (a),(b) net radiation, (c),(d) sensible heat flux, (e),(f) latent heat flux, and (g),(h) ground heat flux at five urban stations during (left) July 2011 and (right) July 2015.

Citation: Journal of Applied Meteorology and Climatology 60, 6; 10.1175/JAMC-D-19-0195.1

UCM stored the second most ground heat flux at daytime and released the second-most ground heat flux at nighttime in Figs. 4g and 4h, which resulted in the second-highest 2-m temperature at nighttime. The latent heat flux predicted by UCM was same to that predicted by BEP and BEM (Figs. 4e,f). Nemunaitis-Berry et al. (2017) concluded that the urban fraction was the only UCM parameter to significantly affect the latent heat flux. The urban fraction defined in the lookup table in the WRF Model is shared by UCM, BEP, and BEM, which is why the latent heat flux is same for the three urban schemes. The sensible heat flux predicted by UCM was least among these urban schemes (Figs. 4c,d). The predicted net radiation, latent heat flux, and ground heat flux were same between BEP and BEM, except the predicted sensible heat flux. Because BEM considered the impact of AC, it produced more sensible heat flux than BEP. BEP and BEM both stored the least ground heat flux at daytime and released the least ground heat flux at nighttime, therefore the near-surface atmosphere simulated by BEP and BEM cooled faster than that simulated by BULK and UCM.

c. Evaluation of schemes’ performance at daytime

Daytime urban cool island (UCI) and UHI were observed in dry summer and wet summer, respectively (Figs. 5a,b). In dry summer, because of urban irrigation, urban soil was wetter than rural soil, it was likely that urban surface released less sensible heat flux than rural surface and UCI formed. On the contrary, in wet summer, the surface humidity was decided by stored precipitation rather than irrigation because of urban impervious surface, urban soil was drier than rural soil, and UHI formed. In observation, the urban wind speed is usually weaker than the rural wind speed (Hu et al. 2016). The observed daytime wind speed within the city was weaker than that in the upstream urban edge and rural area, and the observed daytime wind speed in wet summer was generally stronger than that in dry summer (Figs. 6a,b).

Fig. 5.
Fig. 5.

Observed mean UHI intensity (colored dots) with respect to the average of the eight rural stations in the (a),(b) daytime (0800–1700 CST) and (c),(d) nighttime (2000–0500 CST) in (left) July 2011 and (right) July 2015. The gray–black background is imperviousness (%) from NLCD 2011, and the blue background is bodies of water.

Citation: Journal of Applied Meteorology and Climatology 60, 6; 10.1175/JAMC-D-19-0195.1

Fig. 6.
Fig. 6.

As in Fig. 5, but for wind speed (colored dots) and wind vectors (black arrows). The horizontal arrow in the bottom left of the panels denotes the reference vector (5 m s−1).

Citation: Journal of Applied Meteorology and Climatology 60, 6; 10.1175/JAMC-D-19-0195.1

In dry summer, all urban schemes erroneously produced daytime UHI (Figs. 7a–c), and the daytime UHI intensity produced by BEP and BEM was weaker than that by BULK and UCM (Figs. 7d,e). In wet summer, all urban schemes produced daytime UHI, and the simulated daytime UHI intensities from high to low were by BULK, UCM, BEM, and BEP (Figs. 7f–j).

Fig. 7.
Fig. 7.

Daytime averaged (0800–1700 CST) 2-m UHI in (top) July 2011 and (bottom) July 2015 simulated by (a),(f) BULK + MYJ, (b),(g) UCM + MYJ, (c),(h) UCM + YSU, (d),(i) BEP + MYJ, and (e),(j) BEM + MYJ. The UHI was calculated by the 2-m temperature minus the average of the eight rural stations.

Citation: Journal of Applied Meteorology and Climatology 60, 6; 10.1175/JAMC-D-19-0195.1

At daytime, UCM + YSU could not produce the weaker urban wind speed (Figs. 8c,h), while UCM + MYJ could produce weaker urban wind speed (Figs. 8b,g) because the vertical momentum transport in the MYJ’s local mixing was weaker than that in the YSU’s nonlocal mixing, which suggested that improving PBL schemes can help improving urban wind speed simulation. However, the urban wind speed was much better simulated by BEP and BEM, which suggested that improving the urban schemes was more effective than improving the PBL schemes because BEP and BEM estimated the momentum sink with a drag force that depended on urban morphology (Salamanca et al. 2011).

Fig. 8.
Fig. 8.

As in Fig. 7, but for 10-m wind speed.

Citation: Journal of Applied Meteorology and Climatology 60, 6; 10.1175/JAMC-D-19-0195.1

YSU predicted higher PBL height (PBLH) than MYJ (Figs. 9a,b) because of stronger mixing in the nonlocal PBL scheme (Hu et al. 2010b). The potential temperature θ profiles by YSU and MYJ showed more obvious difference in wet summer than in dry summer. The shape of θ profiles by MYJ was closer to observation than YSU. In both summers, the simulated θ was about 2 K higher than observation because of model error and unrealistic surface properties (Hu et al. 2010a). The daytime θ profiles were more sensitive to PBL schemes than to urban scheme. The θ profiles in dry summer were not as sensitive as the θ profiles in wet summer to the urban schemes.

Fig. 9.
Fig. 9.

Vertical profiles of (a),(b) potential temperature, (c),(d) wind speed, and (e),(f) water vapor at 1700 CST in (left) July 2011, and (right) July 2015 observed from SPARC sounding and WRF simulation result.

Citation: Journal of Applied Meteorology and Climatology 60, 6; 10.1175/JAMC-D-19-0195.1

For daytime PBL wind simulation, the PBL schemes were more dominant than the urban schemes (Figs. 9c,d). YSU predicted the weakest wind speed at 400–2000 m above ground, especially in wet summer. The urban schemes had obvious impacts on the daytime PBL wind profiles because of the strong daytime coupling between surface and PBL. The simulated daytime wind speed by all urban schemes was weaker than observation at levels higher than 400 m (200 m) above ground in dry (wet) summer. The daytime PBL wind speed simulated by BEP and BEM was weaker than that by BULK and UCM because BEP and BEM adopted a rougher urban surface.

The lapse rates and intensity of daytime water vapor (QV) were both not well simulated by all schemes (Figs. 9e,f). The errors of QV lapse rates simulation were caused by PBL schemes, while the errors of QV intensity simulation were caused by the model initialized soil data. In PBL schemes, all scalars, such as QV and θ, are treated in the same way. The simulation result showed that the QV profiles were not well simulated as θ profiles, which meant the QV and θ should be treated differently in PBL schemes. The urban schemes’ impact on QV profiles was noticeable. BULK produced the driest PBL because of its weakest evapotranspiration. YSU predicted drier PBL than MYJ because of its stronger vertical mixing. In wet summer, BULK + MYJ and UCM + YSU showed similar performance in simulating daytime QV profile in the lowest 1200 m of the PBL (Fig. 9f). Both YSU and MYJ could not well simulate the sharp lapse rate of QV in the lowest 200 m of the PBL.

d. Evaluation of schemes’ performance at nighttime

The observed nocturnal UHI in both summers did not show obvious difference in their intensity (Figs. 5c,d). The observed nocturnal wind speed within the city was weaker than that outside the city. The observed nocturnal wind speed in wet summer was not obviously stronger than that in dry summer (Figs. 6c,d).

All urban schemes produced nocturnal UHI in both summers (Fig. 10). The nocturnal UHI intensity was overpredicted by BULK (Figs. 10a,f) and underpredicted by BEP (Figs. 10d,i). The nocturnal UHI intensity produced by UCM and BEM were closer to observation (Figs. 10b,c,e,g,h,j). The nocturnal UHI simulation was more sensitive to urban schemes than to PBL schemes (Fig. 10).

Fig. 10.
Fig. 10.

As in Fig. 7, but for the nocturnal average (2000–0500 CST).

Citation: Journal of Applied Meteorology and Climatology 60, 6; 10.1175/JAMC-D-19-0195.1

Both BULK and UCM (no matter coupled with YSU or MYJ) predicted stronger urban wind speed than rural wind speed (Figs. 11a–c, f–h), which meant properly simulating urban nocturnal wind speed remained as a challenge for PBL schemes. Previous studies have contributed the overpredicted urban wind speed to the nonproper setting of urban roughness (Reames and Stensrud 2017) and imbalance between urban schemes and PBL schemes (Hu et al. 2016). However, BEP and BEM could reasonably reproduce weaker urban wind speed (Figs. 11d,e,i,j), which again proved that improving the urban schemes was more effective than improving the PBL schemes in urban wind speed simulation.

Fig. 11.
Fig. 11.

As in Fig. 8, but for the nocturnal average (2000–0500 CST).

Citation: Journal of Applied Meteorology and Climatology 60, 6; 10.1175/JAMC-D-19-0195.1

The nocturnal θ profiles showed apparent distinction among these urban schemes (Figs. 12a,b), which was caused by the different urban surface cooling rates. The patterns of nocturnal θ profiles by YSU were closer to observation, but the intensity was about 1–2 K larger than observation. BEP cooled the near-surface atmosphere fastest and produced the most stable nocturnal PBL. As BULK released sensible heat flux into atmosphere at nighttime, it produced the most unstable nocturnal PBL structure.

Fig. 12.
Fig. 12.

As in Fig. 9, but at 0500 CST.

Citation: Journal of Applied Meteorology and Climatology 60, 6; 10.1175/JAMC-D-19-0195.1

The simulated nocturnal PBL jet by MYJ had similar profiles to the observed PBL jet (Figs. 12c,d), with weak impact from urban schemes, because the nocturnal surface layer was basically decoupled from PBL. YSU predicted a weaker and lower nocturnal PBL jet in dry summer, while its performance was improved in wet summer. The lapse rates and intensity of nocturnal QV profiles were still not well simulated (Figs. 9e,f).

e. Urban geometry parameters’ impact on urban temperature and wind speed

The default urban geometry configurations between UCM and BEM showed significant distinction (Tables 3 and 4). The default BH in BEM, almost twice as large as that in UCM, was higher than the first model level and allowed the urban structure to interact with the lower atmosphere. In UCM, BH, RDW, and RFW all increased with urban residential intensity. However, in BEM, BH and RFW increased with urban residential intensity, while RDW decreased with urban residential intensity. In UCM (BEM), the ARs between default BH and RDW were 0.6 (0.33), 0.8 (0.6), and 1.0 (1.2) for low-intensity residential (LIR), high-intensity residential (HIR), and industrial/commercial (IC) areas, respectively. Considering most urban areas were LIR or HIR, the SVFs of UCM’s street canyons were generally smaller than that of BEM; therefore, radiation energy loss in UCM was less than that in BEM, which could result in higher urban temperature in UCM than in BEM (Figs. 13a,c).

Fig. 13.
Fig. 13.

The nocturnal averaged (2000–0500 CST) 2-m temperature simulated with different urban geometry parameters.

Citation: Journal of Applied Meteorology and Climatology 60, 6; 10.1175/JAMC-D-19-0195.1

In addition, in UCM (BEM), the ratios between default RFW and RDW were 1.0 (0.43), 1.0 (0.68), and 1.0 (1.0) for LIR, HIR, and IC, respectively. Therefore, the building area was more in UCM than in BEM. More radiation energy would be absorbed at daytime and released at nighttime by buildings in UCM than in BEM, which could be another reason of the higher nocturnal urban temperature in UCM than in BEM (Figs. 13a,c).

After increasing the BH in UCM, the building volume increased and absorbed more radiation, in the meanwhile, the SVF decreased and more radiation was trapped in the street canyon. Therefore, the nocturnal urban temperature in UCM increased (Figs. 13a,b). However, the increased BH gave few impacts on urban canopy wind speed in UCM (Figs. 14a,b) because the increased BH in UCM was still lower than the first model level and did not well interact with the lower atmosphere. On the contrary, the urban temperature in BEM decreased after adopting low BH (Figs. 13c,d) because total building volume decreased and SVF increased. Meanwhile, as the low BH in BEM reduced the interaction between building and atmosphere, the urban canopy wind speed increased (Figs. 14c,d).

Fig. 14.
Fig. 14.

As in Fig. 13, but for 10-m wind speed.

Citation: Journal of Applied Meteorology and Climatology 60, 6; 10.1175/JAMC-D-19-0195.1

When adopting narrow RDW in BEM, the SVF decreased and nocturnal radiation loss reduced. Meanwhile the portion of building area to the total urban area increased, the total building volume increased and absorbed more solar radiation at daytime. Both abovementioned effects could result in higher urban temperature (Figs. 13d,e). When further adopting narrow RFW in BEM, though the total building volume decreased, the number of street canyons increased and the total SVF in the city decreased, urban temperature increased (Figs. 13e,f). The narrow RDW and RFW reduced the urban canopy wind speed (Figs. 14d–f) because they increased urban roughness.

4. Discussion and conclusions

The performance of four urban schemes and two PBL schemes in the WRF Model was evaluated over the DFW urban area during a dry summer and a wet summer. All urban schemes overestimated the daytime 2-m temperature. BULK gave the largest overestimation of 2-m temperature during daytime. During nighttime, only BULK overestimated the 2-m temperature, and BEP gave the largest underestimation of 2-m temperature.

BULK predicted the strongest daytime UHI during the two summers, while BEP and BEM predicted the weakest daytime UHI (with daytime UCI in dry summer). All urban schemes produced the nocturnal UHI in the two summers, with overestimation in BULK, underestimation in BEP, and the best performance from UCM and BEM.

BULK and UCM overestimated 10-m wind speed, but BEP and BEM underestimated 10-m wind speed. Previous studies have attributed the overestimation of 10-m wind speed in BULK and UCM to either improper urban description in model (Reames and Stensrud 2017) or unbalances coupling between urban schemes and PBL schemes (Hu et al. 2016). When coupled with MYJ, BULK and UCM produce weaker urban wind speed than rural wind speed at daytime, while when coupled with YSU, UCM produced stronger urban wind speed than rural wind speed at daytime. However, at nighttime, no matter coupled with MYJ or YSU, both BULK and UCM produced erroneous stronger urban wind speed than rural wind speed. This meant that unbalanced coupling between urban schemes and PBL schemes was one reason for the overestimation of wind speed within the city and simulating nocturnal urban canopy wind speed was more challenging for urban schemes and PBL schemes. When using BEP and BEM, the urban wind speed was better reproduced at both daytime and nighttime, which meant using a proper urban scheme with an improved description of urban roughness could significantly improve the simulation of the urban wind field. Thus, we identify that both deficiencies should be responsible for model bias of urban wind speed, and the proper description of urban in model is more critical.

For urban boundary layer simulation, PBL schemes were more dominant than urban schemes, and MYJ was generally superior to YSU. At daytime, MYJ predicted PBLH and wind speed better than YSU did. At nighttime, UCM + MYJ predicted the θ profiles better than other configurations. In dry summer, MYJ predicted better nocturnal PBL jet strength and nose location than YSU, while in wet summer YSU’s performance improved but was still inferior to MYJ. The shapes of the simulated θ and QV profiles were close to observation, but the bias of both profiles was large probably because of soil moisture bias in NARR.

The nocturnal PBL stability simulated by the four urban schemes were different because of their different surface cooling rates, with BULK predicting the most unstable PBL and BEP predicting the most stable PBL. The urban schemes had obvious effects on the daytime PBL wind speed profiles because of strong daytime land–atmosphere coupling, while at nighttime urban schemes’ impact on PBL wind speed profiles was weak because PBL was decoupled from the surface.

Generally, UCM and BEM showed a better performance than BULK and BEP, though significant difference existed between UCM and BEM. For better understanding the difference between UCM and BEM, we examined model sensitivity to urban geometry parameters. The simulation results showed that urban geometry parameters BH, RDW, and RFW all play roles in determining urban temperature and wind speed. The three parameters affected urban temperature by determining the SVF and the total building volume in the city and affected urban wind speed by determining the depth and number of the street canyons. In general, the BH has a positive correlation with 2-m temperature and negative correlation with 10-m wind speed; RDW and RFW have a negative correlation with 2-m temperature and positive correlation with 10-m wind speed. The lower BH in UCM should be partially responsible for its overestimation of wind speed, and the larger SVF in BEM should be partially responsible for its underestimation of temperature.

The initial status of land surface, such as soil moisture and soil type, also affect model performance over urban areas. A few efforts have been attempted to improve soil moisture status, including HRLDAS (Chen et al. 2007) and implementing an urban hydrology scheme in UCM (Yang et al. 2014). HRLDAS is an offline land surface model to spin up the initial land surface property through long offline simulations. The performance of the UCM hydrology scheme has been actively evaluated (Yang et al. 2014, 2016; Brownlee et al. 2017).

HRLDAS treats the urban land surface as a concrete surface like the BULK land surface treatment does (Chen et al. 2007). The concrete urban surface in HRLDAS is impervious; thus, the urban soil simulated by HRLDAS is usually drier than the surrounding rural soil (Reames and Stensrud 2017; Sharma et al. 2017). In wet summer, urban irrigation effect is negligible compared to abundant rainfall, and less rainfall is absorbed by the impervious urban surface than the rural soil, so urban soil becomes drier than rural soil. However, in dry summer, the urban irrigation effect dominates the rainfall effect, and therefore urban soil is wetter than the rural soil, which HRLDAS cannot reproduce but UCM with urban irrigation scheme can (Yang et al. 2014, 2016). In summary, when urban soil is drier than rural soil, HRLDAS usually gives better performance in wet season/climate with a stronger UHI intensity due to its crude/inappropriate totally impervious urban treatment.

Improving WRF urban simulation has to be through better understanding of urban processes and better calibration of urban parameters with the improved understanding of urban processes. For the urban schemes, further improvement may be achieved through fixing the unrealistic treatment of momentum in UCM and providing realistic urban parameters in UCM and BEM for more cities.

Acknowledgments

The first author thanks Drs. Petra Klein and Ming Xue for their advisement on this study and the financial support provided by CAPS/OU where most of the research was carried out. The first author was partially supported by two grants from the National Key Research and Development Program of China (2018YFB1502803 and 2018YFB0505000). The second author was partially supported by NSF Grant AGS-1917701 and a DOE ASR project (DE-SC0021159).

REFERENCES

  • Barlage, M., S. G. Miao, and F. Chen, 2016: Impact of physics parameterizations on high-resolution weather prediction over two Chinese megacities. J. Geophys. Res. Atmos., 121, 44874498, https://doi.org/10.1002/2015JD024450.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bougeault, P., and P. Lacarrere, 1989: Parameterization of orography-induced turbulence in a mesobeta-scale model. Mon. Wea. Rev., 117, 18721890, https://doi.org/10.1175/1520-0493(1989)117<1872:POOITI>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Brownlee, J., P. Ray, M. Tewari, and H. Tan, 2017: Relative role of turbulent and radiative flux on the near-surface temperature in a single-layer urban canopy model over Houston. J. Appl. Meteor. Climatol., 56, 21732187, https://doi.org/10.1175/JAMC-D-17-0088.1.

    • Crossref
    • 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, https://doi.org/10.1175/1520-0493(2001)129<0569:CAALSH>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chen, F., and Coauthors, 2007: Description and evaluation of the characteristics of the NCAR high-resolution land data assimilation system. J. Appl. Meteor. Climatol., 46, 694713, https://doi.org/10.1175/JAM2463.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ching, J., and Coauthors, 2009: National Urban Database and Access Portal Tool. Bull. Amer. Meteor. Soc., 90, 11571168, https://doi.org/10.1175/2009BAMS2675.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Christen, A., and R. Vogt, 2004: Energy and radiation balance of a central European city. Int. J. Climatol., 24, 13951421, https://doi.org/10.1002/joc.1074.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dudhia, J., 1989: Numerical study of convection observed during the winter monsoon experiment using a mesoscale two-dimensional model. J. Atmos. Sci., 46, 30773107, https://doi.org/10.1175/1520-0469(1989)046<3077:NSOCOD>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Georgescu, M., P. E. Morefield, B. G. Bierwagen, and C. P. Weaver, 2014: Urban adaptation can roll back warming of emerging megapolitan regions. Proc. Natl. Acad. Sci. USA, 111, 29092914, https://doi.org/10.1073/pnas.1322280111.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Grimmond, C. S. B., 1992: The suburban energy balance: Methodological considerations and results for a mid-latitude West Coast city under winter and spring conditions. Int. J. Climatol., 12, 481497, https://doi.org/10.1002/joc.3370120506.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gutiérrez, E., J. E. Gonzalez, A. Martilli, R. Bornstein, and M. Arend, 2015: Simulations of a heat-wave event in New York City using a multilayer urban parameterization. J. Appl. Meteor. Climatol., 54, 283301, https://doi.org/10.1175/JAMC-D-14-0028.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hendricks, E. A., J. C. Knievel, and Y. Wang, 2020: Addition of multilayer urban canopy models to a nonlocal planetary boundary layer parameterization and evaluation using ideal and real cases. J. Appl. Meteor. Climatol., 59, 13691392, https://doi.org/10.1175/JAMC-D-19-0142.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hong, S.-Y., J. Dudhia, and S.-H. Chen, 2004: A revised approach to ice microphysical processes for the bulk parameterization of clouds and precipitation. Mon. Wea. Rev., 132, 103120, https://doi.org/10.1175/1520-0493(2004)132<0103:ARATIM>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hong, S.-Y., Y. Noh, and J. Dudhia, 2006: A new vertical diffusion package with an explicit treatment of entrainment processes. Mon. Wea. Rev., 134, 23182341, https://doi.org/10.1175/MWR3199.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hu, X.-M., and M. Xue, 2016: Influence of synoptic sea-breeze fronts on the urban heat island intensity in Dallas–Fort Worth, Texas. Mon. Wea. Rev., 144, 14871507, https://doi.org/10.1175/MWR-D-15-0201.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hu, X.-M., F. Zhang, and J. W. Nielsen-Gammon, 2010a: Ensemble-based simultaneous state and parameter estimation for treatment of mesoscale model error: A real-data study. Geophys. Res. Lett., 37, L08802, https://doi.org/10.1029/2010GL043017.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hu, X.-M., J. W. Nielsen-Gammon, and F. Q. Zhang, 2010b: Evaluation of three planetary boundary layer schemes in the WRF Model. J. Appl. Meteor. Climatol., 49, 18311844, https://doi.org/10.1175/2010JAMC2432.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hu, X.-M., M. Xue, P. M. Klein, B. G. Illston, and S. Chen, 2016: Analysis of urban effects in Oklahoma City using a dense surface observing network. J. Appl. Meteor. Climatol., 55, 723741, https://doi.org/10.1175/JAMC-D-15-0206.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Jänicke, B., F. Meier, D. Fenner, U. Fehrenbach, A. Holtmann, and D. Scherer, 2017: Urban-rural differences in near-surface air temperature as resolved by the Central Europe Refined analysis (CER): Sensitivity to planetary boundary layer schemes and urban canopy models. Int. J. Climatol., 37, 20632079, https://doi.org/10.1002/joc.4835.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Janjić, Z. I., 1990: The step-mountain coordinate: Physical package. Mon. Wea. Rev., 118, 14291443, https://doi.org/10.1175/1520-0493(1990)118<1429:TSMCPP>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Janjić, Z. I., 1994: The step-mountain Eta coordinate model: Further developments of the convection, viscous sublayer, and turbulence closure schemes. Mon. Wea. Rev., 122, 927945, https://doi.org/10.1175/1520-0493(1994)122<0927:TSMECM>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kain, J. S., 2004: The Kain–Fritsch convective parameterization: An update. J. Appl. Meteor., 43, 170181, https://doi.org/10.1175/1520-0450(2004)043<0170:TKCPAU>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kusaka, H., and F. Kimura, 2004a: Coupling a single-layer urban canopy model with a simple atmospheric model: Impact on urban heat island simulation for an idealized case. J. Meteor. Soc. Japan, 82, 6780, https://doi.org/10.2151/jmsj.82.67.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kusaka, H., and F. Kimura, 2004b: Thermal effects of urban canyon structure on the nocturnal heat island: Numerical experiment using a mesoscale model coupled with an urban canopy model. J. Appl. Meteor., 43, 18991910, https://doi.org/10.1175/JAM2169.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kusaka, H., H. Kondo, Y. Kikegawa, and F. Kimura, 2001: A simple single-layer urban canopy model for atmospheric models: Comparison with multi-layer and slab models. Bound.-Layer Meteor., 101, 329358, https://doi.org/10.1023/A:1019207923078.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lee, S. H., S. A. McKeen, and D. J. Sailor, 2014: A regression approach for estimation of anthropogenic heat flux based on a bottom-up air pollutant emission database. Atmos. Environ., 95, 629633, https://doi.org/10.1016/j.atmosenv.2014.07.009.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Li, D., E. Bou-Zeid, M. Barlage, F. Chen, and J. A. Smith, 2013: Development and evaluation of a mosaic approach in the WRF-Noah framework. J. Geophys. Res. Atmos., 118, 11 91811 935, https://doi.org/10.1002/2013JD020657.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Liao, J. B., T. J. Wang, X. M. Wang, M. Xie, Z. Q. Jiang, X. X. Huang, and J. L. Zhu, 2014: Impacts of different urban canopy schemes in WRF/Chem on regional climate and air quality in Yangtze River Delta, China. Atmos. Res., 145–146, 226243, https://doi.org/10.1016/j.atmosres.2014.04.005.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Liu, Y., F. Chen, T. Warner, and J. Basara, 2006: Verification of a mesoscale data-assimilation and forecasting system for the Oklahoma City area during the Joint Urban 2003 Field Project. J. Appl. Meteor. Climatol., 45, 912929, https://doi.org/10.1175/JAM2383.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Martilli, A., 2002: Numerical study of urban impact on boundary layer structure: Sensitivity to wind speed, urban morphology, and rural soil moisture. J. Appl. Meteor., 41, 12471266, https://doi.org/10.1175/1520-0450(2002)041<1247:NSOUIO>2.0.CO;2.

    • Crossref
    • 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, https://doi.org/10.1023/A:1016099921195.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mlawer, E. J., S. J. Taubman, P. D. Brown, M. J. Iacono, and S. A. Clough, 1997: Radiative transfer for inhomogeneous atmospheres: RRTM, a validated correlated-k model for the longwave. J. Geophys. Res., 102, 16 66316 682, https://doi.org/10.1029/97JD00237.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Nemunaitis-Berry, K. L., P. M. Klein, J. B. Basara, and E. Fedorovich, 2017: Sensitivity of predictions of the urban surface energy balance and heat island to variations of urban canopy parameters in simulations with the WRF Model. J. Appl. Meteor. Climatol., 56, 573595, https://doi.org/10.1175/JAMC-D-16-0157.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Oke, T. R., 1982: The energetic basis of the urban heat island. Quart. J. Roy. Meteor. Soc., 108 (455), 124, https://doi.org/10.1002/qj.49710845502.

    • Search Google Scholar
    • Export Citation
  • Oke, T. R., 1988: The urban energy balance. Prog. Phys. Geogr., 12, 471508, https://doi.org/10.1177/030913338801200401.

  • Reames, L. J., and D. J. Stensrud, 2017: Sensitivity of simulated urban–atmosphere interactions in Oklahoma City to urban parameterization. J. Appl. Meteor. Climatol., 56, 14051430, https://doi.org/10.1175/JAMC-D-16-0223.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sailor, D. J., M. Georgescu, J. M. Milne, and M. A. Hart, 2015: Development of a national anthropogenic heating database with an extrapolation for international cities. Atmos. Environ., 118, 718, https://doi.org/10.1016/j.atmosenv.2015.07.016.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Salamanca, F., and A. Martilli, 2009: A new building energy model coupled with an urban canopy parameterization for urban climate simulations—Part II. Validation with one dimension off-line simulations. Theor. Appl. Climatol., 99, 345356, https://doi.org/10.1007/s00704-009-0143-8.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Salamanca, F., A. Krpo, A. Martilli, and A. Clappier, 2009: A new building energy model coupled with an urban canopy parameterization for urban climate simulations—Part I. Formulation, verification, and sensitivity analysis of the model. Theor. Appl. Climatol., 99, 331344, https://doi.org/10.1007/s00704-009-0142-9.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Salamanca, F., A. Martilli, M. Tewari, and F. Chen, 2011: A study of the urban boundary layer using different urban parameterizations and high-resolution urban canopy parameters with WRF. J. Appl. Meteor. Climatol., 50, 11071128, https://doi.org/10.1175/2010JAMC2538.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Salamanca, F., M. Georgescu, A. Mahalov, and M. Moustaoui, 2015: Summertime response of temperature and cooling energy demand to urban expansion in a semiarid environment. J. Appl. Meteor. Climatol., 54, 17561772, https://doi.org/10.1175/JAMC-D-14-0313.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sharma, A., H. J. S. Fernando, A. F. Hamlet, J. J. Hellmann, M. Barlage, and F. Chen, 2017: Urban meteorological modeling using WRF: A sensitivity study. Int. J. Climatol., 37, 18851900, https://doi.org/10.1002/joc.4819.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Stewart, I. D., and T. R. Oke, 2012: Local climate zones for urban temperature studies. Bull. Amer. Meteor. Soc., 93, 18791900, https://doi.org/10.1175/BAMS-D-11-00019.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Taha, H., 1999: Modifying a mesoscale meteorological model to better incorporate urban heat storage: A bulk-parameterization approach. J. Appl. Meteor., 38, 466473, https://doi.org/10.1175/1520-0450(1999)038<0466:MAMMMT>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • United Nations Department of Economic and Social Affairs, 2012: World urbanization prospects, the 2011 revision. United Nations, 318 pp., https://www.un.org/en/development/desa/population/publications/pdf/urbanization/WUP2011_Report.pdf.

  • Wickham, J., C. A. Barnes, M. S. Nash, and T. G. Wade, 2015: Combining NLCD and MODIS to create a land cover-albedo database for the continental United States. Remote Sens. Environ., 170, 143152, https://doi.org/10.1016/j.rse.2015.09.012.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Winguth, A. M. E., and B. Kelp, 2013: The urban heat island of the north-central Texas region and its relation to the 2011 severe Texas drought. J. Appl. Meteor. Climatol., 52, 24182433, https://doi.org/10.1175/JAMC-D-12-0195.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yang, J., Z.-H. Wang, F. Chen, S. Miao, M. Tewari, J. A. Voogt, and S. Myint, 2014: Enhancing hydrologic modelling in the coupled Weather Research and Forecasting–urban modelling system. Bound.-Layer Meteor., 155, 87109, https://doi.org/10.1007/s10546-014-9991-6.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yang, J., Z.-H. Wang, M. Georgescu, F. Chen, and M. Tewari, 2016: Assessing the impact of enhanced hydrological processes on urban hydrometeorology with application to two cities in contrasting climates. J. Hydrometeor., 17, 10311047, https://doi.org/10.1175/JHM-D-15-0112.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
Save
  • Barlage, M., S. G. Miao, and F. Chen, 2016: Impact of physics parameterizations on high-resolution weather prediction over two Chinese megacities. J. Geophys. Res. Atmos., 121, 44874498, https://doi.org/10.1002/2015JD024450.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bougeault, P., and P. Lacarrere, 1989: Parameterization of orography-induced turbulence in a mesobeta-scale model. Mon. Wea. Rev., 117, 18721890, https://doi.org/10.1175/1520-0493(1989)117<1872:POOITI>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Brownlee, J., P. Ray, M. Tewari, and H. Tan, 2017: Relative role of turbulent and radiative flux on the near-surface temperature in a single-layer urban canopy model over Houston. J. Appl. Meteor. Climatol., 56, 21732187, https://doi.org/10.1175/JAMC-D-17-0088.1.

    • Crossref
    • 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, https://doi.org/10.1175/1520-0493(2001)129<0569:CAALSH>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chen, F., and Coauthors, 2007: Description and evaluation of the characteristics of the NCAR high-resolution land data assimilation system. J. Appl. Meteor. Climatol., 46, 694713, https://doi.org/10.1175/JAM2463.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ching, J., and Coauthors, 2009: National Urban Database and Access Portal Tool. Bull. Amer. Meteor. Soc., 90, 11571168, https://doi.org/10.1175/2009BAMS2675.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Christen, A., and R. Vogt, 2004: Energy and radiation balance of a central European city. Int. J. Climatol., 24, 13951421, https://doi.org/10.1002/joc.1074.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dudhia, J., 1989: Numerical study of convection observed during the winter monsoon experiment using a mesoscale two-dimensional model. J. Atmos. Sci., 46, 30773107, https://doi.org/10.1175/1520-0469(1989)046<3077:NSOCOD>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Georgescu, M., P. E. Morefield, B. G. Bierwagen, and C. P. Weaver, 2014: Urban adaptation can roll back warming of emerging megapolitan regions. Proc. Natl. Acad. Sci. USA, 111, 29092914, https://doi.org/10.1073/pnas.1322280111.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Grimmond, C. S. B., 1992: The suburban energy balance: Methodological considerations and results for a mid-latitude West Coast city under winter and spring conditions. Int. J. Climatol., 12, 481497, https://doi.org/10.1002/joc.3370120506.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gutiérrez, E., J. E. Gonzalez, A. Martilli, R. Bornstein, and M. Arend, 2015: Simulations of a heat-wave event in New York City using a multilayer urban parameterization. J. Appl. Meteor. Climatol., 54, 283301, https://doi.org/10.1175/JAMC-D-14-0028.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hendricks, E. A., J. C. Knievel, and Y. Wang, 2020: Addition of multilayer urban canopy models to a nonlocal planetary boundary layer parameterization and evaluation using ideal and real cases. J. Appl. Meteor. Climatol., 59, 13691392, https://doi.org/10.1175/JAMC-D-19-0142.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hong, S.-Y., J. Dudhia, and S.-H. Chen, 2004: A revised approach to ice microphysical processes for the bulk parameterization of clouds and precipitation. Mon. Wea. Rev., 132, 103120, https://doi.org/10.1175/1520-0493(2004)132<0103:ARATIM>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hong, S.-Y., Y. Noh, and J. Dudhia, 2006: A new vertical diffusion package with an explicit treatment of entrainment processes. Mon. Wea. Rev., 134, 23182341, https://doi.org/10.1175/MWR3199.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hu, X.-M., and M. Xue, 2016: Influence of synoptic sea-breeze fronts on the urban heat island intensity in Dallas–Fort Worth, Texas. Mon. Wea. Rev., 144, 14871507, https://doi.org/10.1175/MWR-D-15-0201.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hu, X.-M., F. Zhang, and J. W. Nielsen-Gammon, 2010a: Ensemble-based simultaneous state and parameter estimation for treatment of mesoscale model error: A real-data study. Geophys. Res. Lett., 37, L08802, https://doi.org/10.1029/2010GL043017.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hu, X.-M., J. W. Nielsen-Gammon, and F. Q. Zhang, 2010b: Evaluation of three planetary boundary layer schemes in the WRF Model. J. Appl. Meteor. Climatol., 49, 18311844, https://doi.org/10.1175/2010JAMC2432.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hu, X.-M., M. Xue, P. M. Klein, B. G. Illston, and S. Chen, 2016: Analysis of urban effects in Oklahoma City using a dense surface observing network. J. Appl. Meteor. Climatol., 55, 723741, https://doi.org/10.1175/JAMC-D-15-0206.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Jänicke, B., F. Meier, D. Fenner, U. Fehrenbach, A. Holtmann, and D. Scherer, 2017: Urban-rural differences in near-surface air temperature as resolved by the Central Europe Refined analysis (CER): Sensitivity to planetary boundary layer schemes and urban canopy models. Int. J. Climatol., 37, 20632079, https://doi.org/10.1002/joc.4835.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Janjić, Z. I., 1990: The step-mountain coordinate: Physical package. Mon. Wea. Rev., 118, 14291443, https://doi.org/10.1175/1520-0493(1990)118<1429:TSMCPP>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Janjić, Z. I., 1994: The step-mountain Eta coordinate model: Further developments of the convection, viscous sublayer, and turbulence closure schemes. Mon. Wea. Rev., 122, 927945, https://doi.org/10.1175/1520-0493(1994)122<0927:TSMECM>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kain, J. S., 2004: The Kain–Fritsch convective parameterization: An update. J. Appl. Meteor., 43, 170181, https://doi.org/10.1175/1520-0450(2004)043<0170:TKCPAU>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kusaka, H., and F. Kimura, 2004a: Coupling a single-layer urban canopy model with a simple atmospheric model: Impact on urban heat island simulation for an idealized case. J. Meteor. Soc. Japan, 82, 6780, https://doi.org/10.2151/jmsj.82.67.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kusaka, H., and F. Kimura, 2004b: Thermal effects of urban canyon structure on the nocturnal heat island: Numerical experiment using a mesoscale model coupled with an urban canopy model. J. Appl. Meteor., 43, 18991910, https://doi.org/10.1175/JAM2169.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kusaka, H., H. Kondo, Y. Kikegawa, and F. Kimura, 2001: A simple single-layer urban canopy model for atmospheric models: Comparison with multi-layer and slab models. Bound.-Layer Meteor., 101, 329358, https://doi.org/10.1023/A:1019207923078.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lee, S. H., S. A. McKeen, and D. J. Sailor, 2014: A regression approach for estimation of anthropogenic heat flux based on a bottom-up air pollutant emission database. Atmos. Environ., 95, 629633, https://doi.org/10.1016/j.atmosenv.2014.07.009.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Li, D., E. Bou-Zeid, M. Barlage, F. Chen, and J. A. Smith, 2013: Development and evaluation of a mosaic approach in the WRF-Noah framework. J. Geophys. Res. Atmos., 118, 11 91811 935, https://doi.org/10.1002/2013JD020657.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Liao, J. B., T. J. Wang, X. M. Wang, M. Xie, Z. Q. Jiang, X. X. Huang, and J. L. Zhu, 2014: Impacts of different urban canopy schemes in WRF/Chem on regional climate and air quality in Yangtze River Delta, China. Atmos. Res., 145–146, 226243, https://doi.org/10.1016/j.atmosres.2014.04.005.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Liu, Y., F. Chen, T. Warner, and J. Basara, 2006: Verification of a mesoscale data-assimilation and forecasting system for the Oklahoma City area during the Joint Urban 2003 Field Project. J. Appl. Meteor. Climatol., 45, 912929, https://doi.org/10.1175/JAM2383.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Martilli, A., 2002: Numerical study of urban impact on boundary layer structure: Sensitivity to wind speed, urban morphology, and rural soil moisture. J. Appl. Meteor., 41, 12471266, https://doi.org/10.1175/1520-0450(2002)041<1247:NSOUIO>2.0.CO;2.

    • Crossref
    • 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, https://doi.org/10.1023/A:1016099921195.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mlawer, E. J., S. J. Taubman, P. D. Brown, M. J. Iacono, and S. A. Clough, 1997: Radiative transfer for inhomogeneous atmospheres: RRTM, a validated correlated-k model for the longwave. J. Geophys. Res., 102, 16 66316 682, https://doi.org/10.1029/97JD00237.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Nemunaitis-Berry, K. L., P. M. Klein, J. B. Basara, and E. Fedorovich, 2017: Sensitivity of predictions of the urban surface energy balance and heat island to variations of urban canopy parameters in simulations with the WRF Model. J. Appl. Meteor. Climatol., 56, 573595, https://doi.org/10.1175/JAMC-D-16-0157.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Oke, T. R., 1982: The energetic basis of the urban heat island. Quart. J. Roy. Meteor. Soc., 108 (455), 124, https://doi.org/10.1002/qj.49710845502.

    • Search Google Scholar
    • Export Citation
  • Oke, T. R., 1988: The urban energy balance. Prog. Phys. Geogr., 12, 471508, https://doi.org/10.1177/030913338801200401.

  • Reames, L. J., and D. J. Stensrud, 2017: Sensitivity of simulated urban–atmosphere interactions in Oklahoma City to urban parameterization. J. Appl. Meteor. Climatol., 56, 14051430, https://doi.org/10.1175/JAMC-D-16-0223.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sailor, D. J., M. Georgescu, J. M. Milne, and M. A. Hart, 2015: Development of a national anthropogenic heating database with an extrapolation for international cities. Atmos. Environ., 118, 718, https://doi.org/10.1016/j.atmosenv.2015.07.016.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Salamanca, F., and A. Martilli, 2009: A new building energy model coupled with an urban canopy parameterization for urban climate simulations—Part II. Validation with one dimension off-line simulations. Theor. Appl. Climatol., 99, 345356, https://doi.org/10.1007/s00704-009-0143-8.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Salamanca, F., A. Krpo, A. Martilli, and A. Clappier, 2009: A new building energy model coupled with an urban canopy parameterization for urban climate simulations—Part I. Formulation, verification, and sensitivity analysis of the model. Theor. Appl. Climatol., 99, 331344, https://doi.org/10.1007/s00704-009-0142-9.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Salamanca, F., A. Martilli, M. Tewari, and F. Chen, 2011: A study of the urban boundary layer using different urban parameterizations and high-resolution urban canopy parameters with WRF. J. Appl. Meteor. Climatol., 50, 11071128, https://doi.org/10.1175/2010JAMC2538.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Salamanca, F., M. Georgescu, A. Mahalov, and M. Moustaoui, 2015: Summertime response of temperature and cooling energy demand to urban expansion in a semiarid environment. J. Appl. Meteor. Climatol., 54, 17561772, https://doi.org/10.1175/JAMC-D-14-0313.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sharma, A., H. J. S. Fernando, A. F. Hamlet, J. J. Hellmann, M. Barlage, and F. Chen, 2017: Urban meteorological modeling using WRF: A sensitivity study. Int. J. Climatol., 37, 18851900, https://doi.org/10.1002/joc.4819.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Stewart, I. D., and T. R. Oke, 2012: Local climate zones for urban temperature studies. Bull. Amer. Meteor. Soc., 93, 18791900, https://doi.org/10.1175/BAMS-D-11-00019.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Taha, H., 1999: Modifying a mesoscale meteorological model to better incorporate urban heat storage: A bulk-parameterization approach. J. Appl. Meteor., 38, 466473, https://doi.org/10.1175/1520-0450(1999)038<0466:MAMMMT>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • United Nations Department of Economic and Social Affairs, 2012: World urbanization prospects, the 2011 revision. United Nations, 318 pp., https://www.un.org/en/development/desa/population/publications/pdf/urbanization/WUP2011_Report.pdf.

  • Wickham, J., C. A. Barnes, M. S. Nash, and T. G. Wade, 2015: Combining NLCD and MODIS to create a land cover-albedo database for the continental United States. Remote Sens. Environ., 170, 143152, https://doi.org/10.1016/j.rse.2015.09.012.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Winguth, A. M. E., and B. Kelp, 2013: The urban heat island of the north-central Texas region and its relation to the 2011 severe Texas drought. J. Appl. Meteor. Climatol., 52, 24182433, https://doi.org/10.1175/JAMC-D-12-0195.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yang, J., Z.-H. Wang, F. Chen, S. Miao, M. Tewari, J. A. Voogt, and S. Myint, 2014: Enhancing hydrologic modelling in the coupled Weather Research and Forecasting–urban modelling system. Bound.-Layer Meteor., 155, 87109, https://doi.org/10.1007/s10546-014-9991-6.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yang, J., Z.-H. Wang, M. Georgescu, F. Chen, and M. Tewari, 2016: Assessing the impact of enhanced hydrological processes on urban hydrometeorology with application to two cities in contrasting climates. J. Hydrometeor., 17, 10311047, https://doi.org/10.1175/JHM-D-15-0112.1.

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

    (a) Configuration of the four two-way nested domains for simulation. Shading is terrain height in meters. (b) NLCD 40-class land-cover maps for D4. (c) Updated representative diurnal profiles of AH for three urban land-use categories (black dots and lines denote IC, purple dots and lines denote HIR, and red dots and lines denote LIR).

  • Fig. 2.

    The land-use categories of TCEQ stations in (a) July 2011 and (b) July 2015 and the spatial distribution of early evening transition cooling rate of TCEQ stations in (c) July 2011 and (d) July 2015 overlaid on imperviousness (%) from NLCD 2011. The blue background is bodies of water. The blue star denotes the location of the Fort Worth radiosonde. The five red diamonds denote the locations of the selected urban stations, and the eight blue squares denote the locations of the selected rural stations.

  • Fig. 3.

    Observed and simulated mean diurnal variation of (a),(b) 2-m temperature, (c),(d) 10-m wind speed, (e),(f) 2-m relative humidity, and (g),(h) 2-m dewpoint at five urban stations during (left) July 2011 and (right) July 2015. The five urban stations are denoted by red diamonds in Fig. 2.

  • Fig. 4.

    Simulated mean diurnal variation of (a),(b) net radiation, (c),(d) sensible heat flux, (e),(f) latent heat flux, and (g),(h) ground heat flux at five urban stations during (left) July 2011 and (right) July 2015.

  • Fig. 5.

    Observed mean UHI intensity (colored dots) with respect to the average of the eight rural stations in the (a),(b) daytime (0800–1700 CST) and (c),(d) nighttime (2000–0500 CST) in (left) July 2011 and (right) July 2015. The gray–black background is imperviousness (%) from NLCD 2011, and the blue background is bodies of water.

  • Fig. 6.

    As in Fig. 5, but for wind speed (colored dots) and wind vectors (black arrows). The horizontal arrow in the bottom left of the panels denotes the reference vector (5 m s−1).

  • Fig. 7.

    Daytime averaged (0800–1700 CST) 2-m UHI in (top) July 2011 and (bottom) July 2015 simulated by (a),(f) BULK + MYJ, (b),(g) UCM + MYJ, (c),(h) UCM + YSU, (d),(i) BEP + MYJ, and (e),(j) BEM + MYJ. The UHI was calculated by the 2-m temperature minus the average of the eight rural stations.

  • Fig. 8.

    As in Fig. 7, but for 10-m wind speed.

  • Fig. 9.

    Vertical profiles of (a),(b) potential temperature, (c),(d) wind speed, and (e),(f) water vapor at 1700 CST in (left) July 2011, and (right) July 2015 observed from SPARC sounding and WRF simulation result.

  • Fig. 10.

    As in Fig. 7, but for the nocturnal average (2000–0500 CST).

  • Fig. 11.

    As in Fig. 8, but for the nocturnal average (2000–0500 CST).

  • Fig. 12.

    As in Fig. 9, but at 0500 CST.

  • Fig. 13.

    The nocturnal averaged (2000–0500 CST) 2-m temperature simulated with different urban geometry parameters.

  • Fig. 14.

    As in Fig. 13, but for 10-m wind speed.