Abstract

Together with economic development and accelerated urbanization, the urban population in China has been increasing rapidly, and anthropogenic heat released by large-scale energy consumption in cities is expected to be a vital factor affecting the climate. In this paper, the Weather Research and Forecasting (WRF) model coupled with the Urban Canopy Model (UCM) is employed to simulate the regional impacts on climate under the two scenarios: the underlying surface changes due to urbanization (USCU) and anthropogenic heat release (AHR). Three experiments were performed from December 2006 to December 2008. With respect to the USCU, the surface albedo and the available surface soil water decrease markedly. With the inclusion of AHR, the two scenarios give rise to increased surface temperatures over most areas of China. Especially in the urban agglomeration area of the Yangtze River delta, the combination of USCU and AHR could result in an increase of 2°C in the surface air temperature. The influence of AHR on surface air temperature in winter is greater than the influence of USCU without considering any extra sources of heat, but the opposite is found in summer. The combination of USCU and AHR leads to changes in the surface energy budget. They both increase sensible heat flux, but USCU decreases latent heat flux significantly, and AHR increases latent heat flux slightly. Nevertheless, under the influence of these two scenarios, the precipitation increases in some areas, especially in the Beijing–Tianjin–Hebei region, while it decreases in other areas, most notably the Yangtze River delta.

1. Introduction

Greenhouse gas emissions are considered to be the main cause of climate warming (Core Writing Team 2007). However, large-scale urbanization has developed rapidly in recent decades. The underlying surface changes due to urbanization (USCU) thereby change the thermal and dynamical properties of land surfaces. Meanwhile, massive urban energy consumption directly releases heat into atmosphere. The USCU and the anthropogenic heat release (AHR) associated with human social and economic activities have been recognized as important factors that have serious impacts on climate and environment at local and regional scales.

The world’s population has become increasingly urbanized. In the year 2000, 45% of the world’s population lived in urban areas, with an even greater fraction of 75% in more developed countries. According to the data provided by the Population Reference Bureau (2010), half of the earth’s population is composed of city dwellers; most of this additional urbanization occurs in developing countries. The expansion of urban areas and increased urban activities have serious impacts on urban environments. These impacts are mainly represented by anthropogenic heat emission and massive changes of land cover (Ichinose et al. 1999). There are many previous works referring to the impact of urban activities (Oke 1995; Li et al. 2004; Roth 2007).

Previous modeling studies of the urban heat island (UHI) effect mainly focus on the city scale. Based on a modified version of the fifth-generation Pennsylvania State University–National Center for Atmospheric Research Mesoscale Model (MM5), numerical simulations were performed for an extended area of Greece. A strengthening of the nocturnal urban heat island and a decrease in turbulence and fluxes during the daytime appeared in the urban areas (Dandou et al. 2005). The amplitude of the urban heat island at Barrow, Alaska, was most pronounced in winter months, with temperatures in urban areas averaging 2°C warmer than in the surrounding tundra and occasionally exceeding 6°C (Hinkel and Nelson 2007). The Community Land Model, in which an urban parameterization is designed to represent the urban energy balance, was employed to simulate the UHI effect in Mexico City, Mexico, and Vancouver, British Columbia, Canada. The results indicated that the daily minimum temperatures increased more than daily maximum temperatures in urban areas (Oleson et al. 2008). The expansion of urban areas caused a series of changes, including increases in the sensible heat fluxes and urban temperatures, a decrease in evaporation, and even time-specific variations in regional atmospheric stratification stability (Jiang et al. 2009; Zhang et al. 2009).

Waste heat produced by human activities is one contributor to urban heat islands, and it is likely that anthropogenic heat sources will play an important role in achieving an urban surface energy balance in the future (Allen et al. 2010). The anthropogenic heat flux varies spatially and temporally. Under certain conditions, it can exceed the energy receipt from net all-wave radiation (Lee et al. 2009). As suggested by Sailor (2011), energy consumption in the urban environment impacts the urban surface energy budget and leads to the emission of anthropogenic sensible heat and moisture into the atmosphere. Meanwhile, the author provides a historical perspective on the development of models of energy consumption in the urban environment and the associated anthropogenic impacts on the urban energy balance. In many urban areas, a positive feedback cycle has developed, whereby higher temperatures result from greater amounts of energy used for air cooling. This in turn adds to the total heat emissions into the atmosphere, and air temperature is further increased (Crutzen 2004). The results of an investigation into the influence of AHR on atmospheric temperature in a Japanese megacity (Keihanshin district) indicated that the quantity of heat released was lower at night than during the day, but the extent of the nocturnal temperature increase was nearly threefold greater (Narumi et al. 2009).

Some reports describe AHR and its climate effect on the regional scale and the urban agglomeration scale. Pigeon et al. (2007) investigated the AHR in the European agglomeration of Toulouse in France. They found that vehicle traffic was the major source of the AHR during summer. During winter, the AHR reached 100 W m−2 in the densest areas, whereas it ranged between 5 and 25 W m−2 in suburban areas. Block et al. (2004) investigated the principal effect of AHR on regional climate conditions in central Europe. The results revealed that temperature effects not only depended on the amount of added heat but also on orographic factors.

Recently, several correlated works also focused on the global scale. Judging from the model simulations of the large-scale urban consumption of energy (LUCY), the global mean urban AHR had a diurnal range of 0.7–3.6 W m−2 and was greater on weekdays than during weekends. The heat released from buildings was the largest contributor (89%–96%) to heat emissions globally (Allen et al. 2010). Equilibrium climate experiments by GCMs using developed present and future global inventories of anthropogenic heat flux and parameterizations derived for seasonal and diurnal flux cycles show statistically significant continental-scale surface warming (0.4°–0.9°C) produced by one 2100 AHR scenario (Flanner 2009).

All of these reports point to the importance of anthropogenic heat emissions in the urban environment. At present, the numerical simulations of the effects of AHR and USCU on temperature or other meteorological factors mainly focus on a single city or smaller spatial scales, and their time scales are mostly restricted to seasonal changes or even diurnal changes. There are few simulation studies that address the effect of AHR on climate on a regional scale. Therefore, in this paper, the WRF model coupled with UCM is employed to study the climate effects of the underlying surface changes due to urbanization and the anthropogenic heat release in China on regional and urban agglomeration scales. A description of the model and the details of the experiments are presented in section 2. The results from the climate simulation experiments are presented in section 3. Discussions and conclusions are provided in section 4.

2. Model details and designs of the experiments

The regional meteorological model used in this study is the Weather Research and Forecasting (WRF) model with Advanced Research WRF (ARW) dynamic core version 3.1 (Skamarock et al. 2008). In terms of physical options, we use the WRF single-moment three-class scheme (WSM3) microphysical parameterization (Hong et al. 2004), the new Kain–Fritsch convective parameterization (Kain 2004), the Dudhia shortwave radiation (Dudhia 1989) and Rapid Radiative Transfer Model (RRTM) longwave radiation (Mlawer et al. 1997), the Yonsei University (YSU) planetary boundary layer (PBL) scheme (Hong et al. 2006), and the Noah land surface model (Chen and Dudhia 2001; Ek et al. 2003) with the Urban Canopy Model (UCM) (Kusaka et al. 2001; Kusaka and Kimura 2004).

Initial and boundary conditions (ICBCs) for the large-scale atmospheric fields, sea surface temperature (SST), and initial soil parameters (i.e., soil water, moisture, and temperature) are given by the 1° × 1° NCEP Global Final Analysis (FNL) 6-h data. The model domain is centered at 37°N, 102.5°E, with 175 × 151 horizontal grid points and a resolution of 30 km. The Lambert conformal conic projection is used for the model horizontal coordinates, with the standard parallel at 105°E. With respect to the vertical coordinates, 28 terrain-following eta levels from the surface to 50 mb are used. The simulation period is from 1 December 2006 to 31 December 2008. The model outputs results every 6 h. These results are averaged into daily data, which are then used to obtain the monthly averages for analyses.

Experimental design schemes are shown in Table 1, and three experiments are included. The first experiment is the control run (Ctr); it uses U.S. Geological Survey (USGS) 24-category data, but all of the urban types are replaced by nearby vegetation types according to the nearest-neighbor interpolation method, and no AHR is included. The second and third experiments are the sensitivity runs carried out by WRF coupled with UCM using the USGS 33-category dataset, where the urban category is divided into industrial/commercial, high-intensity residential, and low-intensity residential categories. The first sensitivity experiment (S1) treats three different urban types without considering AHR. The second sensitivity experiment (S2) is based on the S1 with AHR added. For each experiment, the corresponding model parameters and settings are selected based on its respective urban categories.

Table 1.

Experimental design.

Experimental design.
Experimental design.

Generally, simulations for high-resolution urban areas require more accurate urban classifications based on actual internal urban layouts. However, in this study, because of the large simulation region, the horizontal resolution of 30 km is not adequate to describe many cities. Meanwhile, in the Noah land surface model (LSM), the method that calculates the physical parameters of land surfaces according to the percentage of land cover over each grid cell is not supported. Rather, a semivirtualization method is adopted, such that if the urban area in a grid cell reaches a certain threshold, the grid is then assigned an urban type. In this study, to demonstrate the urban effect on climate and include more cities as much as possible, the threshold value is set to 36 km2 (i.e., an urban fraction of 4% in a grid cell). This threshold value may appear loose compared to the area of one grid cell, but it is still valuable for simulating the relative effects of USCU and AHR on the climate at a regional scale.

The population and energy data for the cities used in the experiments are taken from the annual total statistics included in the 2008 versions of the China Energy Statistical Yearbook and China City Statistical Yearbook (China Statistics Press 2008a,b). The data for the populations and areas of the cities used in this paper include more than 600 cities in China. The data on energy consumption focus on various energy sources, including coal, oil, and natural gas. These data are also obtained for more than 600 cities in 2008. According to the calorific values and utilization efficiency of each kind of energy, the annual mean anthropogenic heat release of each city is calculated, and then combined with the number for the urban population. Cities are then divided into three types: industrial/commercial and high- and low-density residential. Cities with populations larger than 1.5 million and AHRs larger than 40 W m−2 would be treated as industrial/commercial, and cities with populations smaller than 0.5 million and AHRs smaller than 15 W m−2 are treated as low-density residential. The rest of the cities are treated as high-density residential (Fig. 1a).

Fig. 1.

(a) City classifications (LC is industrial/commercial, MC is high-intensity residential, and SC is low-intensity residential) and (b) regional divisions (1, northeast China; 2, northwestern China; 3, Beijing–Tianjin–Hebei region; 4, Yangtze River delta region; and 5, southern China).

Fig. 1.

(a) City classifications (LC is industrial/commercial, MC is high-intensity residential, and SC is low-intensity residential) and (b) regional divisions (1, northeast China; 2, northwestern China; 3, Beijing–Tianjin–Hebei region; 4, Yangtze River delta region; and 5, southern China).

In current version of the WRF/UCM ensemble, each urban type can only accept one AHR value, but the AHR can have a prescribed diurnal change for three urban categories. The diurnal AHR profiles in summer and winter are likely different. However, in the study, only annual mean energy consumption data were collected; thus the calculation of the AHR is based on the annual average data, so the AHR profiles for summer periods and winter periods in the experiments are from the same data, and the diurnal variation of the AHR is based on the default diurnal coefficient of the AHR in the WRF/UCM. Because of the lack of detailed daily and diurnal energy data, the daily maximal AHR data are calculated using the annual mean data and the default diurnal coefficient.

According to the calculated annual mean AHR of each city, based on the above urban classification method, the mean AHR values for the three urban types are computed as approximately 50, 28, and 11 W m−2. The diurnal values at different times are determined by the coefficients that have peak values at 0800 and 1700 local time. The values at other times are relatively minor. The average coefficient is 0.56, and the daily maximal AHR values for three urban types are 90, 50, and 20 W m−2.

At present, there is a lack of parameters suitable for the urban class of China, so for each urban class, the parameters stored in the predefined urban table of the WRF model are used in the experiments. In the WRF/UCM model, the anthropogenic heat is added in the form of sensible heat. However, the net radiation and the AHR can be partitioned to sensible heat and latent heat.

The static data, including the various spatial resolution terrestrial and USGS land-use data, are provided in the WRF model. These static data essentially describe the real terrestrial and land cover characteristics of the regions of interest. This paper mainly focuses on a comparative analysis of the regional climates in two urban agglomeration regions: Beijing–Tianjin–Hebei and the Yangtze River delta area.

3. Subregional division and climatic background

The climatic characteristics of different regions are very diverse, and there are different levels of urban density in China. Therefore, to research the effects of climate change caused by USCU and AHR in different areas with different local climates, the cities in China are divided into five regions based on the densities of the cities (Fig. 1b): Beijing–Tianjin–Hebei, the Yangtze River delta area, and southern, northwestern, and northeastern China. Most of the subregions are located in the plains. However, the northwest is mainly located in the Loess Plateau and an arid region. The Yangtze River delta is influenced by strong East Asian monsoons, which bring rich rainfall and high temperatures in the summer. The Beijing–Tianjin–Hebei area lies in the north of China, which is also influenced by the monsoon but is drier than the Yangtze River delta. Both the above-mentioned regions include numerous developed cities, and they represent two large urban agglomerations.

In this study, the simulation duration comprises 2007 and 2008. Here, temperature and precipitation data from the University of East Anglia’s Climatic Research Unit (CRU) are used to demonstrate the climate during 2007–08 relative to historical averages. Figure 2 shows the interannual variations of the anomaly of annual average temperature and precipitation from 1990 to 2009 over the whole of China and the Beijing–Tianjin–Hebei and Yangtze River delta regions. Figure 2 indicates that the temperature has a high positive anomaly value in 2007 and an approached historical value in 2008 over all regions. The precipitation has a negative anomaly value in 2007 and 2008 over the whole of China, but it is less anomalous in 2007 than in 2008. It approached the historical value in 2007 over both the Beijing–Tianjin–Hebei region and the Yangtze River delta region, but it has a positive anomaly in the Beijing–Tianjin–Hebei region and a negative anomaly in the Yangtze River delta region in 2008.

Fig. 2.

Interannual variation of (left) annual average temperature (°C) and (right) precipitation (%) for (a),(d) China; (b),(e) Beijing–Tianjin–Hebei; and (c),(f) Yangtze River delta region.

Fig. 2.

Interannual variation of (left) annual average temperature (°C) and (right) precipitation (%) for (a),(d) China; (b),(e) Beijing–Tianjin–Hebei; and (c),(f) Yangtze River delta region.

4. Results

a. Validation of temperature and precipitation

A comparison was made between the observed data and the WRF simulation data in the average annual temperature and average summer precipitation. The observed temperature is mainly taken from the station data in China, and the observed precipitation is based on Tropical Rainfall Measuring Mission (TRMM) data. The simulation results come from the control run.

The spatial patterns of the observed and simulated annual average temperatures are shown in Figs. 3a and 3b. The temperature figures show that the reproduced temperature distributions are essentially consistent with the observed distributions, though there are low temperature areas in the Yangtze River delta region. East China is one of the world’s most highly populated agricultural regions, and it is dominated by the well-known East Asian monsoon. One of the most prominent features of the precipitation is that it is concentrated in summer. Figures 3c and 3d show the spatial patterns of the observed and simulated summer precipitation over the modeling domain. In general, the model reproduces well the spatial distribution of precipitation. The observed precipitation pattern for June–August (JJA; see Fig. 3c) is characterized by less precipitation in northern China and more precipitation in the southern China and Yangtze River delta regions. For the most part, the reproduced precipitation data capture the main patterns and magnitudes (Fig. 3d), though they overestimate the precipitation in southwestern China.

Fig. 3.

Comparison of annual mean temperature (°C) and summer precipitation (mm day−1) between observations and simulations for (a) observed temperature, (b) simulated temperature, (c) observed precipitation, and (d) simulated precipitation.

Fig. 3.

Comparison of annual mean temperature (°C) and summer precipitation (mm day−1) between observations and simulations for (a) observed temperature, (b) simulated temperature, (c) observed precipitation, and (d) simulated precipitation.

b. Influence on temperature and surface energy balance

The surface air temperature and its variations in a given region are influenced by several factors, such as incoming shortwave radiation, sensible and latent heat fluxes at the surface, and advection (Block et al. 2004). The calculations of the surface air temperatures, as well as the humidity and near-surface wind speed, are based on the Monin–Obukhov theory of the atmospheric boundary layer in the WRF/UCM simulations.

Figure 4 shows the spatial distribution of the surface air temperatures of the control run and the differences between the sensitivity run and the control run during the summer and winter. Figs. 4d and 4h show the spatial distribution of the control runs in summer and winter, including the distribution characteristics of the surface air temperature. Figures 4a and 4e show the differences between the S1 sensitivity run and the control run, and they represent the surface temperature changes caused by USCU throughout China. Figures 4b and 4f show the temperature differences between the results from the two sensitivity runs (S2 and S1), revealing the influence of AHR. Figures 3c and 3g show the simulation differences between the S2 sensitivity run and the control run, which reflect the total effect of both the USCU and AHR. As shown in Fig. 4, after considering the combination of USCU and AHR, the surface temperature in most regions of China increases during the summer, with an amplitude of 1°C appearing in northern China and the Yangtze River delta. In particular, the temperature in the Yangtze River delta region increases by 2°C. The areas with added temperature amplitudes above 1°C during winter have somewhat expanded, while no area has an amplitude greater than 2°C. Overall, the effect of USCU on summer temperatures is quite obvious, while the contribution of AHR to winter warming is significant. Judging from the annual average simulation results (not shown), after considering both the factors, the temperature increases in most areas of China, especially in the Yangtze River delta region, where the amplitude attains 2°C.

Fig. 4.

Temperature of the control run and the temperature differences between the control run and sensitivity runs for (a)–(d) summer and (e)–(h) winter (°C): (a) S1-Ctr, (b) S2-S1, (c) S2-Ctr, (d) Ctr, (e) S1-Ctr, (f) S2-S1, (g) S2-Ctr, and (h) Ctr.

Fig. 4.

Temperature of the control run and the temperature differences between the control run and sensitivity runs for (a)–(d) summer and (e)–(h) winter (°C): (a) S1-Ctr, (b) S2-S1, (c) S2-Ctr, (d) Ctr, (e) S1-Ctr, (f) S2-S1, (g) S2-Ctr, and (h) Ctr.

The regional average temperatures for annual, summer, and winter changes in China are shown in Table 2. The results show that the summer temperature is most affected by USCU, particularly in the Yangtze River delta region, with an increase of 1.44°C; southern China shows the second-highest impact, and the lowest impact is in northwestern China. AHR has the greatest impact on winter temperatures; the most striking change occurs again in the Yangtze River delta region, with an increase in amplitude of 1.25°C, followed by southern China and the Beijing–Tianjin–Hebei region. AHR produces the lowest impact on summer temperature in northeastern China. After considering the combined effect of USCU and AHR, the temperature change in the Yangtze River delta region is the most significant, with an increase in amplitude of 2.10°C during the summer and winter temperature increasing by 1.55°C. Summer temperatures in the Beijing–Tianjin–Hebei region increase by 0.22°C, while winter temperatures increase by 0.63°C.

Table 2.

The regional average temperature change (°C).

The regional average temperature change (°C).
The regional average temperature change (°C).

Regarding the regional average effect over the whole of China, USCU increases the annual average temperature by 0.13°C. This change is more obvious during summer than winter. However, the magnitude of the change in the annual average temperature due to AHR is 0.15°C, and this change is more evident in winter than in summer. Overall, the amplitudes of the average annual temperature change due to USCU and AHR are roughly equal.

The Yangtze River delta has experienced rapid urbanization in the last 30 yr and is the largest urban agglomeration in China. In addition, there are large areas of irrigated cropland in the Yangtze River delta area. This region is influenced by strong East Asian monsoon with warm and wet climate characteristics. The Beijing–Tianjin–Hebei region is located at a higher latitude and is drier than the Yangtze River delta. It represents another urban agglomeration in China, but the density of its cities is lower. For cities with higher levels of urbanization and more intensive energy use, the change in regional temperature is significant. The comprehensive effects of USCU and AHR on the regional temperatures are different for different parts of China. For example, the increased amplitude of the temperatures in the Yangtze River delta is higher in summer than in winter, while the increase in amplitude in the Beijing–Tianjin–Hebei region is more obvious in winter than in summer. However, for regions with a low level of urbanization, such as northwestern China, there is little change in the temperature.

The monthly changes in temperature due to USCU and AHR in each region are shown in Fig. 5. As observed from the graph, in China, USCU plays a leading role in determining the temperatures during spring and summer; however, in the fall and winter, AHR has the dominant impact. The temperature curves for the Yangtze River delta region and northeastern and southern China show a pattern of change similar to that for the regional average across all of China. In northwestern China and the Beijing–Tianjin–Hebei region, AHR plays a key role during most of the year.

Fig. 5.

The monthly temperature change in the 2-yr average under the USCU and AHR scenarios over each region (°C): (a) China, (b) northeastern China, (c) northwestern China, (d) Beijing–Tianjin–Hebei region, (e) Yangtze River delta region, and (f) southern China.

Fig. 5.

The monthly temperature change in the 2-yr average under the USCU and AHR scenarios over each region (°C): (a) China, (b) northeastern China, (c) northwestern China, (d) Beijing–Tianjin–Hebei region, (e) Yangtze River delta region, and (f) southern China.

As suggested by Arnfield (2003) and Oke (1995), several interrelated causes can be invoked to explain the warmer temperatures observed in urban settings. For example, the complex three-dimensional urban geometry exhibits a much rougher surface, disrupting wind patterns and trapping energy within the urban canopy.

The above results regarding the temperature changes imply that AHR plays a vital role in urban area heating. In the WRF/UCM model, the AHR is added in the form of a sensible heat flux. However, from the view of energy balance, in the urban environment, the equation of surface energy balance can be expressed as follows:

 
formula

Unlike in the traditional equation, a term for anthropogenic heat QF is included in the left-hand part of the equation. In the equation, RN is the net radiation flux at the surface, including solar radiation and longwave radiation. In addition, QH is the sensible heat flux, QE is the latent heat flux, and ΔQS is the net heat storage flux in the surface layer of the soil and conducted into the soil layer. In the urban environment, the net heat storage includes the energy storage within the buildings, the roads, and the underlying soil. The AHR can directly heat the atmosphere and then increase the surface temperature and low-level air temperature. When the AHR is added to the model, as with net radiation, the anthropogenic heat QF could be reallocated to the sensible heat, latent heat, and soil heat.

As is known, the surface albedo is defined as the ratio of the reflected solar radiation to the total incoming solar radiation. The surface albedo directly controls the net solar radiation absorbed at the surface and thus the surface energy balance. In the process of exchanging the flux within the earth’s land–atmosphere system, the surface albedo has an impact on the surface radiation and the energy balance. Figure 6a shows the differences in the annual average surface albedo for the control run and the sensitivity runs. Figure 6b, which depicts the control run, shows that the surface albedo is small in eastern and southeastern China, but the values are high in northern and northwestern China. In addition, the figure shows that USCU reduces the surface albedo value in most areas; the values decrease markedly in eastern China, where the distribution of cities is concentrated. Because the AHR does not directly change the surface properties, the annual average change of albedo due to AHR is far smaller than that attributed to USCU (not shown).

Fig. 6.

Annual mean surface albedo of the control run and the differences between the control run and the sensitivity run: (a) S1-Ctr and (b) Ctr.

Fig. 6.

Annual mean surface albedo of the control run and the differences between the control run and the sensitivity run: (a) S1-Ctr and (b) Ctr.

Under the USCU, the latent heat flux decreases in most areas, especially in the Yangtze River delta region, where the amplitude is 10 W m−2. In contrast, the AHR increases the latent heat flux in most areas, but the magnitude is small (Fig. 7). Both USCU and AHR increase the sensible heat flux prominently, especially in the Beijing–Tianjin–Hebei and Yangtze River delta regions, where the magnitude reaches approximately 20 W m−2 (Fig. 8). The sensible heat flux is much more strongly affected by AHR than by USCU.

Fig. 7.

Annual mean latent heat flux of the control run and the differences between the control run and the sensitivity runs (W m−2): (a) S1-Ctr, (b) S2-S1, (c) S2-Ctr, and (d) Ctr.

Fig. 7.

Annual mean latent heat flux of the control run and the differences between the control run and the sensitivity runs (W m−2): (a) S1-Ctr, (b) S2-S1, (c) S2-Ctr, and (d) Ctr.

Fig. 8.

Annual mean sensible heat flux of the control run and the differences between the control run and the sensitivity runs (W m−2): (a) S1-Ctr, (b) S2-S1, (c) S2-Ctr, and (d) Ctr.

Fig. 8.

Annual mean sensible heat flux of the control run and the differences between the control run and the sensitivity runs (W m−2): (a) S1-Ctr, (b) S2-S1, (c) S2-Ctr, and (d) Ctr.

The net radiation could be dissipated as sensible heat, latent heat, and soil heat fluxes. Generally, evaporation is mainly affected by surface water, wind speed, temperature, and so on. If considering only the USCU, in the urbanization processes, the underlying surface changes from that of a nonurban area to a waterproof surface mostly made up of cement. USCU decreases the surface albedo, leading to increased incoming solar radiation at the urban surface. The impervious and hard urban surface reduces the available surface moisture significantly, and the increased building height leads to increased surface roughness and a decrease in the near-surface wind, so the latent heat flux decreases accordingly. Urban buildings and urban surfaces have higher heat storage capacities and greater heat transport to deep soil or building interiors. This heat storage can be released at night, heating the surface atmosphere. All of these factors impact the heat flux partitions in urbanized areas, with an increase in the sensible heat flux and a decrease in the latent heat flux. Thus, the sensible heat flux is markedly larger than the latent heat flux over urban areas.

The AHR is added to the UCM in the form of sensible heat, but the AHR can also be partitioned to sensible heat and latent heat through the adjustment of the surface energy balance. Although AHR mainly increases the sensible heat, because the moisture at the surface is essentially invariant in the AHR scenario, the increase in the temperature may cause an increase in the latent heat flux, though the magnitude is very small relative to the increase in the sensible heat.

The temperature change results show that the influence of USCU is greater in summer than in winter in most urban agglomeration areas, especially in the Yangtze River delta. Zhang et al. (2010) also suggested a stronger UHI effect in summer in the Yangtze River delta. This result could be explained by the following mechanism. In the surface energy budget, the solar radiation is the main energy source. The incoming solar radiation has a pronounced seasonal variation. The albedo of most nonurban surfaces also has a seasonal variation in the WRF/Noah LSM, but the urban surface albedo is largely invariant. The incoming solar radiation is the largest in summer and the smallest in winter. For the underlying vegetative surface, the albedo is smaller in summer than in winter. Based on these factors, the urban surface absorbs the stronger solar radiation in summer more so than in winter. Meanwhile, the urban heat storage is greater in summer than in winter. For vegetative surfaces, the soil is wet and evapotranspiration is active, making the surface latent heat flux greater than the sensible heat flux. With USCU, the available soil moisture becomes very low. Thus, the sensible heat flux and ground heat flux increase, and the surface temperature increases accordingly. In winter, the vegetative cover is lower and the soil is drier than in summer. The available soil water is sparse, and evapotranspiration is weak, such that the sensible heat flux is greater than the latent heat flux in the surface energy balance. Although USCU also increases the sensible heat flux and decreases the latent heat flux, the magnitude of the heat flux change from nonurban to urban areas is smaller in winter than in summer.

In our study, if only considering the USCU, the summer temperature over the Beijing–Tianjin–Hebei area shows only a slight decrease. That is, the urban heat island effect is not observed if the AHR is not considered in summer. This is likely associated with the drier nonurban surface in the Beijing–Tianjin–Hebei area. In addition, the coarse resolution in the simulations and sparse urban point in this area might also contribute to a weak UHI effect.

Compared with the seasonal change in the solar radiation, the AHR is assumed to be stable. In summer, the shortwave radiation is strong, and thus the influence of AHR is relatively small. In winter, the shortwave radiation is weak, and thus the influence of anthropogenic heat is relatively large. This is the main reason that the effect of the AHR is larger in winter than in summer.

c. Influence on precipitation

The spatial distributions of summer precipitation differences across different experiments in China are shown in Fig. 9. When USCU and AHR are taken into consideration, the summer precipitation increases in some areas and decreases in others, with the most obvious decrease in the Yangtze River delta region. The change in the winter precipitation is much smaller than that observed for the summer precipitation, and it increases slightly in most regions of China (not shown). Because monsoonal precipitation in much of China is predominant in summer, the spatial distribution of the annual average precipitation (not shown) shows that, after accounting for USCU and AHR, the changes in the annual average precipitation follow a similar spatial pattern as the changes observed during the summer. The increase in precipitation in the Beijing–Tianjin–Hebei region and the decrease in precipitation in the Yangtze River delta region are both noteworthy.

Fig. 9.

Summer precipitation of the control run and the differences between the control run and sensitivity runs (mm day−1): (a) S1-Ctr, (b) S2-S1, (c) S2-Ctr, and (d) Ctr.

Fig. 9.

Summer precipitation of the control run and the differences between the control run and sensitivity runs (mm day−1): (a) S1-Ctr, (b) S2-S1, (c) S2-Ctr, and (d) Ctr.

Table 3 shows the regional average percentage change in annual, winter, and summer precipitation in China. According to the average across China, USCU reduces precipitation, but there is an opposite effect attributed to AHR. At the regional level, USCU has the greatest impact on the summer precipitation in the Beijing–Tianjin–Hebei region, which has an amplitude value of 15.5%, while it has the least impact on winter precipitation in northwestern China. AHR increases the annual precipitation in the Beijing–Tianjin–Hebei region by 6.6%, while it reduces precipitation in the Yangtze River delta region by 2.7%. If USCU and AHR are both taken into account, the summer precipitation in the Beijing–Tianjin–Hebei region shows the largest percentage change, with an amplitude of 26.9%, while the smallest percentage change is observed in the southern regions.

Table 3.

The regional average precipitation change (%).

The regional average precipitation change (%).
The regional average precipitation change (%).

Figure 10 is similar to Fig. 5, but it includes data for precipitation. This graph shows that AHR increases precipitation over China, while USCU reduces precipitation. Similar regional average results are observed in northeastern and southern China. In the Yangtze River delta and northwestern China, the combined influence of both factors plays a leading role during late spring and summer, but it has little impact on precipitation during other seasons. In addition, AHR in northwestern China and the Yangtze River delta region almost has an opposite impact on precipitation. Precipitation in the Beijing–Tianjin–Hebei region is affected by USCU during summer and autumn, but in spring, AHR is dominant. So, in urban areas, such as the densely populated Yangtze River delta and Beijing–Tianjin–Hebei regions, USCU has a much greater influence than AHR; meanwhile, in relatively sparse urban areas, AHR plays the leading role.

Fig. 10.

The monthly precipitation change in the 2-yr average under the USCU and AHR scenario over each region (mm day−1): (a) China, (b) northeastern China, (c) northwestern China, (d) Beijing–Tianjin–Hebei region, (e) Yangtze River delta region, and (f) southern China.

Fig. 10.

The monthly precipitation change in the 2-yr average under the USCU and AHR scenario over each region (mm day−1): (a) China, (b) northeastern China, (c) northwestern China, (d) Beijing–Tianjin–Hebei region, (e) Yangtze River delta region, and (f) southern China.

The land surface models use atmospheric information from the surface-layer scheme, radiative forcing from the radiation scheme, and precipitation forcing from the microphysics and convective schemes, together with internal information on the land’s state variables and land surface properties, to provide heat and moisture fluxes over land points. Urbanization may influence precipitation by changing the surface albedo, roughness, energy budgets, and water flows to the local atmosphere. Because the low-level moisture is one of the most important factors for UHI-induced precipitation, urbanization leads to reduced surface evaporation and a deficit of moisture available for convective precipitation formation. On the other hand, UHI can strengthen the vertical movement and convective activity, produce local circulation, and increase the convective available potential energy. In addition, changes in precipitation due to UHIs may be related to the interaction between the seasonal prevailing wind and the temperature gradient. The UCSU and AHR can change the partition of sensible heat and latent heat, and the heat impacts the precipitation.

Therefore, it is difficult to give a clear mechanism for the impact of urbanization on precipitation because of the complexity of the system. Kaufmann et al. (2007) suggested there is no causal relationship between urbanization and precipitation during the summer. Summer coincides with the rainy season, when the East Asian monsoon has a dominant effect at spatial scales far beyond urban areas. As such, the magnitude of this effect may overwhelm local urban impacts. The relationship between urbanization and precipitation may vary by location.

The turbulent activity in the boundary layer and the convection are weaker in winter than in summer. Although the magnitude of temperature increase attributed to AHR in winter is larger than in summer, the impacts of AHR and USCU on temperature and wind speed are mainly limited below 800 hPa in winter (Fig. 11). In the USCU scenario, the decreases in near-surface wind speeds are clear in both winter and summer. However, in summer, temperature effects and wind speed effects can attain higher levels. Based on these points, the precipitation effects of USCU and AHR are weaker in winter than summer.

Fig. 11.

Vertical profiles of (a)–(d) temperature and (e)–(h) wind speed for (top) summer and (bottom) winter over Beijing–Tianjin–Hebei and Yangtze River delta: (a),(c),(e),(g) Beijing–Tianjin–Hebei and (b),(d),(f),(h) Yangtze River delta.

Fig. 11.

Vertical profiles of (a)–(d) temperature and (e)–(h) wind speed for (top) summer and (bottom) winter over Beijing–Tianjin–Hebei and Yangtze River delta: (a),(c),(e),(g) Beijing–Tianjin–Hebei and (b),(d),(f),(h) Yangtze River delta.

d. Comparison of the results from 2007 and 2008

In the above analysis, the results from both years averaged together are analyzed. In fact, the climatic background is different between 2007 and 2008, so further comparison of the temperature and precipitation changes due to USCU and AHR in each year is necessary.

Tables 4 and 5 show the regional average annual temperature and precipitation changes due to USCU and AHR in 2007 and 2008, respectively. In 2007 and 2008, the USCU and AHR increase the temperature in all regions. The USCU and AHR increase the precipitation in 2007 and 2008 in the Beijing–Tianjin–Hebei region, but the situation is more complicated in the Yangtze River delta region. In the Yangtze River delta region, USCU decreases the precipitation in 2007 and increases it in 2008, while AHR increases precipitation in 2007 and decreases it in 2008. For the average of 2007 and 2008, the combined impact of USCU and AHR is a decrease in the precipitation in the Yangtze River delta region.

Table 4.

The regional average annual temperature change (°C).

The regional average annual temperature change (°C).
The regional average annual temperature change (°C).
Table 5.

The regional average annual precipitation change (%).

The regional average annual precipitation change (%).
The regional average annual precipitation change (%).

Figures 12 and 13 show the monthly temperature and precipitation changes during 2007 and 2008, respectively. In total, the patterns of monthly temperature changes due to USCU and AHR are similar between 2007 and 2008. Both USCU and AHR increase the temperature in all months over the whole of China and in the Yangtze River delta region. However, USCU shows more influence in the decreasing temperatures during late summer in 2008 than that in 2007 in the Beijing–Tianjin–Hebei region.

Fig. 12.

The monthly temperature changes in (left) 2007 and (right) 2008 under the USCU and AHR scenarios over each region (°C): (a),(d) China; (b),(e) Beijing–Tianjin–Hebei region; and (c),(f) Yangtze River delta region.

Fig. 12.

The monthly temperature changes in (left) 2007 and (right) 2008 under the USCU and AHR scenarios over each region (°C): (a),(d) China; (b),(e) Beijing–Tianjin–Hebei region; and (c),(f) Yangtze River delta region.

Fig. 13.

The monthly precipitation changes during (left) 2007 and (right) 2008 under the USCU and AHR scenarios over each region (mm day−1): (a),(d) China; (b),(e) Beijing–Tianjin–Hebei region; and (c),(f) Yangtze River delta region.

Fig. 13.

The monthly precipitation changes during (left) 2007 and (right) 2008 under the USCU and AHR scenarios over each region (mm day−1): (a),(d) China; (b),(e) Beijing–Tianjin–Hebei region; and (c),(f) Yangtze River delta region.

The agreement for the monthly precipitation change between 2007 and 2008 is weaker than that for the temperature. Over the whole of China, the USCU has little impact on the precipitation in 2007, but it decreased the precipitation in 2008, especially in the summer. In the Beijing–Tianjin–Hebei region, the USCU and AHR almost increased the precipitation in 2007 and 2008. In the Yangtze River delta region, there is a clear difference in the monthly precipitation change due to USCU and AHR between 2007 and 2008, and the magnitude of the precipitation change is larger in 2008 than in 2007. In a year with a strong monsoon, due to the control of large-scale circulation, the climate effect due to urbanization may be relatively weak. Contrarily, in a year with a weak monsoon, the climate effect due to urbanization may be relatively strong. The Yangtze River delta region is strongly controlled by the East Asian monsoon, and from the observed interannual variation of precipitation, 2008 is a year with a weak monsoon. This may be a main reason for the large differences observed in the precipitation changes due to USCU and AHR between 2007 and 2008.

5. Discussions and conclusions

This paper has assessed the impact of USCU and AHR on regional climate in China while employing the WRF model coupled with the UCM. Based on this study, integrating previous reports, we recognize that urbanization and AHR can influence the exchange of turbulent momentum, turbulent heat, and water vapor between the land surface and the atmosphere and therefore change the fields of temperature and the local circulation. They lead to increases in temperature and turbulence energy, and this impact will be transported to the upper atmosphere due to the strong turbulence.

In this study, USCU decreases the surface albedo and the latent heat flux, but AHR increases the latent heat slightly. Both USCU and AHR lead to significant increases in the sensible heat flux, and the surface temperature generally increases. Considering the combined effects of USCU and AHR, the temperature increases in most areas, especially in the Yangtze River delta region, which shows an amplitude of more than 2°C. The increases in the regional average summer temperature affected by USCU and AHR attain 1.44° and 0.66°C in the Yangtze River delta region, respectively. USCU increases summer temperatures more effectively, while AHR has a more apparent impact on winter temperatures. In the Yangtze River delta and the Beijing–Tianjin–Hebei region, USCU and AHR increase the sensible heat flux with an amplitude of up to 20 W m−2, while the influence on sensible heat flux attributed to AHR is more striking than that for USCU. In the Yangtze River delta region, USCU decreases the latent heat flux with an amplitude of 10 W m−2.

Considering the combined effects of USCU and AHR, the precipitation increases in certain areas, especially in the Beijing–Tianjin–Hebei region, while it decreases in other areas, most notably the Yangtze River delta region. Overall, USCU reduces precipitation on average, while AHR increases precipitation, especially in the Beijing–Tianjin–Hebei region, with an increase of 6.6%. However, precipitation in the Yangtze River delta region is reduced by 2.7%.

In fact, there are still many uncertainties that may affect the simulation results, such as the selection of physical process, the configuration of the resolution, the urban classification, the use of urban energy data, and the complexities of monsoon climates.

The Core Writing Team (2007) noted that the global temperature increased 0.74°C in the past of 100 yr. It also indicates that greenhouse gas emissions are the main cause of global warming and that the climate effect of urbanization may be negligible on the global scale. In this study, we focus on the regional and local scales to evaluate the effects of urbanization on climate. Meanwhile, based on the semivirtualization design described, the urban area is exaggerated to some extent as compared with reality. Under these circumstances, the average temperature change over the whole of China is 0.29°C. Jones et al. (2008) assessed possible urban influences over the eastern China mainland using sea surface temperature (SST) datasets. The results showed that urban-related warming over China is approximately 0.1°C decade−1 over the period 1951–2004, with true climatic warming accounting for 0.81°C during the same period.

Urbanization is but a single special case of land-use change. There have been many studies of climate effects of land use and land cover change (LUCC), but it is difficult to obtain a specific quantitative temperature effect because of the complexity and uncertainty of LUCC impact on climate. Large areas of Amazonian deforestation resulted in a 2°–5°C increase of temperature, and a 30% decrease in precipitation (Nobre et al. 1991; Henderson-Sellers et al. 1993). Douglas et al. (2006) suggested that agricultural irrigation resulted in a regional precipitation increase and a temperature decrease. Over some areas such as California in the United States, northwestern India, and northeastern China, the cooling effect of agricultural irrigation is as significant as the warming effect of greenhouse gases. Findell et al. (2007) showed that although the anthropogenic land-use change produced an obvious impact on local and regional climates, the effect was very small on the global scale.

Therefore, in conclusion, as suggested by the Core Writing Team (2007), the climate effect of urbanization may be negligible on the global scale, but regional and local effects caused by some underlying surface changes and anthropogenic heat release cannot always be neglected.

Acknowledgments

This study was supported by the Innovation Key Program of the Chinese Academy of Sciences (Grants KGCX2-YW-356 and KZCX2-EW-202), the General Project of the National Natural Science Foundation of China (Grant 40975048), and the Strategic Priority Research Program—Climate Change: Carbon Budget and Relevant Issues of the Chinese Academy of Sciences (Grant XDA05090207).

REFERENCES

REFERENCES
Allen
,
L.
,
F.
Lindberg
, and
C. S. B.
Grimmond
,
2010
:
Global to city scale urban anthropogenic heat flux: Model and variability
.
Int. J. Climatol.
,
31
,
1990
2005
,
doi:10.1002/joc.2210
.
Arnfield
,
A. J.
,
2003
:
Two decades of urban climate research: A review of turbulence, exchanges of energy and water, and the urban heat island
.
Int. J. Climatol.
,
23
,
1
26
.
Block
,
A.
,
K.
Keuler
, and
E.
Schaller
,
2004
:
Impacts of anthropogenic heat on regional climate patterns
.
Geophys. Res. Lett.
,
31
,
L12211
,
doi:10.1029/2004GL019852
.
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
,
569
585
.
China Statistics Press
,
2008a
:
China City Statistics Yearbook. China Statistics Press, 480 pp
.
China Statistics Press
,
2008b
:
China Energy Statistics Yearbook. China Statistics Press, 284 pp
.
Core Writing Team
,
2007
:
Climate Change 2007: Synthesis Report. Cambridge University Press, 104 pp
.
Crutzen
,
P. J.
,
2004
:
New directions: The growing urban heat and pollution “island” effect - Impact on chemistry and climate
.
Atmos. Environ.
,
38
,
3539
3540
.
Dandou
,
A.
,
M.
Tombrou
,
E.
Akylas
,
N.
Soulakellis
, and
E.
Bossioli
,
2005
:
Development and evaluation of an urban parameterization scheme in the Penn State/NCAR Mesoscale Model (MM5)
.
J. Geophys. Res.
,
110
,
D10102
,
doi:10.1029/2004JD005192
.
Douglas
,
E. M.
, and
Coauthors
,
2006
:
Changes in moisture and energy fluxes due to agricultural land use and irrigation in the Indian monsoon belt
.
Geophys. Res. Lett.
,
33
,
L14403
,
doi:10.1029/2006GL026550
.
Dudhia
,
J.
,
1989
:
Numerical study of convection observed during the Winter Monsoon Experiment using a mesoscale two-dimensional model
.
J. Atmos. Sci.
,
46
,
3077
3107
.
Ek
,
M. B.
,
K. E.
Mitchell
,
Y.
Lin
,
E.
Rogers
,
P.
Grunmann
,
V.
Koren
,
G.
Gayno
, and
J. D.
Tarpley
,
2003
:
Implementation of Noah land surface model advances in the National Centers for Environmental Prediction operational mesoscale Eta Model
.
J. Geophys. Res.
,
108
,
8851
,
doi:10.1029/2002JD003296
.
Findell
,
K. L.
,
E.
Shevliakova
, and
P. C. D.
Milly
,
2007
:
Modeled impact of anthropogenic land cover change on climate
.
J. Climate
,
20
,
3621
3634
.
Flanner
,
M. G.
,
2009
:
Integrating anthropogenic heat flux with global climate models
.
Geophys. Res. Lett.
,
36
,
L02801
,
doi:10.1029/2008GL036465
.
Henderson-Sellers
,
A.
,
R. E.
Dickinson
,
T. B.
Durbridge
,
P. J.
Kennedy
,
K.
McGuffie
, and
A. J.
Pitman
,
1993
:
Tropical deforestation: Modeling local- to regional-scale climate change
.
J. Geophys. Res.
,
98
(
D4
),
7289
7315
.
Hinkel
,
K. M.
, and
F. E.
Nelson
,
2007
:
Anthropogenic heat island at Barrow, Alaska, during winter: 2001–2005
.
J. Geophys. Res.
,
112
,
D06118
,
doi:10.1029/2006JD007837
.
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
,
103
120
.
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
,
2318
2341
.
Ichinose
,
T.
,
K.
Shimodozono
, and
K.
Hanaki
,
1999
:
Impact of anthropogenic heat on urban climate in Tokyo
.
Atmos. Environ.
,
33
,
3897
3909
.
Jiang
,
W. M.
,
Y. W.
Wang
, and
N.
Zhang
,
2009
:
Research on urban land-surface process and boundary layer structure (in Chinese)
.
Adv. Earth. Sci.
,
24
,
411
419
.
Jones
,
P. D.
,
D. H.
Lister
, and
Q.
Li
,
2008
:
Urbanization effects in large-scale temperature records, with an emphasis on China
.
J. Geophys. Res.
,
113
,
D16122
,
doi:10.1029/2008JD009916
.
Kain
,
J. S.
,
2004
:
The Kain–Fritsch convective parameterization: An update
.
J. Appl. Meteor.
,
43
,
170
181
.
Kaufmann
,
R. K.
, and
Coauthors
,
2007
:
Climate response to rapid urban growth: Evidence of a human-induced precipitation deficit
.
J. Climate
,
20
,
2299
2306
.
Kusaka
,
H.
, and
F.
Kimura
,
2004
:
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
,
67
80
.
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
,
329
358
.
Lee
,
S.-H.
,
C.-K.
Song
,
J.-J.
Baik
, and
S.-U.
Park
,
2009
:
Estimation of anthropogenic heat emission in the Gyeong-Inregion of Korea
.
Theor. Appl. Climatol.
,
96
,
291
303
.
Li
,
Q.
, and
Coauthors
,
2004
:
Urban heat island effect on annual mean temperature during the last 50 years in China
.
Theor. Appl. Climatol.
,
79
,
165
174
.
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
(
D14
),
16 663
16 682
.
Narumi
,
D.
,
A.
Kondo
, and
Y.
Shimoda
,
2009
:
Effects of anthropogenic heat release upon the urban climate in a Japanese megacity
.
Environ. Res.
,
109
,
421
431
.
Nobre
,
C. A.
,
P. J.
Sellers
, and
J.
Shukla
,
1991
:
Amazonian deforestation and regional climate change
.
J. Climate
,
4
,
957
988
.
Oke
,
T. R.
,
1995
:
The heat island of the urban boundary layer-characteristics, causes and effects
.
Nat. Adv. Sci. Inst. Ser.
,
277
,
81
107
.
Oleson
,
K. W.
,
G. B.
Bonan
,
J.
Feddema
, and
M.
Vertenstein
,
2008
:
An urban parameterization for a global climate model. Part II: Sensitivity to input parameters and the simulated urban heat island in offline simulations
.
J. Appl. Meteor. Climatol.
,
47
,
1061
1076
.
Pigeon
,
G.
,
D.
Legain
,
P.
Durand
, and
V.
Masson
,
2007
:
Anthropogenic heat release in an old European agglomeration (Toulouse, France)
.
Int. J. Climatol.
,
27
,
1969
1981
.
Population Reference Bureau
, cited
2010
:
World population data sheet. [Available online at http://www.prb.org/Publications/Datasheets/2010/2010wpds.aspx.]
Roth
,
M.
,
2007
:
Review of urban climate research in (sub) tropical regions
.
Int. J. Climatol.
,
27
,
1859
1873
.
Sailor
,
D.
,
2011
:
A review of methods for estimating anthropogenic heat and moisture emissions in the urban environment
.
Int. J. Climatol.
,
31
,
189
199
.
Skamarock
,
W. C.
, and
Coauthors
,
2008
:
A description of the Advanced Research WRF version 3. NCAR Tech. Note NCAR/TN-475+STR, 125 pp. [Available online at http://www.mmm.ucar.edu/wrf/users/docs/arw_v3.pdf.]
Zhang
,
C. L.
,
F.
Chen
,
S. G.
Miao
,
Q. C.
Li
,
X. A.
Xia
, and
C. Y.
Xuan
,
2009
:
Impacts of urban expansion and future green planting on summer precipitation in the Beijing metropolitan area
.
J. Geophys. Res.
,
114
,
D02116
,
doi:10.1029/2008JD010328
.
Zhang
,
N.
,
Z. Q.
Gao
,
X. M.
Wang
, and
Y.
Chen
,
2010
:
Modeling the impact of urbanization on the local and regional climate in Yangtze River delta, China
.
Theor. Appl. Climatol.
,
102
(
3–4
),
331
342
.