• Arnfield, A., 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, 126.

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

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

    • Search Google Scholar
    • Export Citation
  • Chen, F., , and Y. Zhang, 2009: On the coupling strength between the land surface and the atmosphere: From viewpoint of surface exchange coefficients. Geophys. Res. Lett., 36, L10404, doi:10.1029/2009GL037980.

    • Search Google Scholar
    • Export Citation
  • Chen, F., and Coauthors, 2011a: The integrated WRF/urban modeling system: Development, evaluation, and applications to urban environmental problems. Int. J. Climatol., 31, 273288, doi:10.1002/joc.2158.

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

    • Search Google Scholar
    • Export Citation
  • Davis, C., , B. Brown, , and R. Bullock, 2006: Object-based verification of precipitation forecasts. Part I: Methodology and application to mesoscale rain areas. Mon. Wea. Rev., 134, 17721784.

    • Search Google Scholar
    • Export Citation
  • Davis, C., and Coauthors, 2008: Prediction of landfalling hurricanes with the Advanced Hurricane WRF model. Mon. Wea. Rev., 136, 19902005.

    • Search Google Scholar
    • Export Citation
  • Dixon, P. G., , and T. L. Mote, 2003: Patterns and causes of Atlanta’s urban heat island initiated precipitation. J. Appl. Meteor., 42, 12731284.

    • Search Google Scholar
    • Export Citation
  • Done, J., , C. Davis, , and M. Weisman, 2004: The next generation of NWP: Explicit forecasts of convection using the Weather Research and Forecasting (WRF) model. Atmos. Sci. Lett., 5, 110117.

    • Search Google Scholar
    • Export Citation
  • Du, Y., , Z. Xie, , Y. Zeng, , Y. Shi, , and J. Wu, 2007: Impact of urban expansion on regional temperature change in the Yangtze River Delta (in Chinese). J. Geogr. Sci., 17, 387398.

    • 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.

    • Search Google Scholar
    • Export Citation
  • Erell, E., , and T. Williamson, 2007: Intra-urban differences in canopy layer air temperature at a mid-latitude city. Int. J. Climatol., 27, 12431255.

    • Search Google Scholar
    • Export Citation
  • Ferguson, G., , and A. Woodbury, 2007: Urban heat island in the subsurface. Geophys. Res. Lett., 34, L23713, doi:10.1029/2007gl032324.

  • Fujibe, F., , N. Yamazaki, , M. Katsuyama, , and K. Kobayashi, 2005: The increasing trend of intense precipitation in Japan based on four-hourly data for a hundred years. SOLA, 1, 4144, doi:10.2151/sola.2005-012.

    • Search Google Scholar
    • Export Citation
  • Goswami, B. N., , M. S. Madhusoodanan, , C. P. Neema, , and D. Sengupta, 2006: A physical mechanism for North Atlantic SST influence on the Indian summer monsoon. Geophys. Res. Lett., 33, L02706, doi:10.1029/2005GL024803.

    • Search Google Scholar
    • Export Citation
  • Grell, G. A., , J. Dudhia, , and D. Stauffer, 1994: A description of the fifth-generation Penn State/NCAR Mesoscale Model (MM5). NCAR Tech. Note NCAR/TN-398+STR, 117 pp.

  • Grell, G. A., , S. E. Peckham, , R. Schmitz, , S. A. McKeen, , G. Frost, , W. C. Skamarock, , and B. Eder, 2005: Fully coupled “online” chemistry within the WRF model. Atmos. Environ., 39, 69576975.

    • 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.

    • Search Google Scholar
    • Export Citation
  • Houghton, J. T., , Y. Ding, , D. J. Griggs, , M. Noguer, , P. J. van der Linden, , X. Dai, , K. Maskell, , and C. A. Johnson, Eds., 2001: Climate Change 2001: The Scientific Basis. Cambridge University Press, 881 pp.

  • Inoue, E., 1963: On the turbulent structure of airflow within crop canopies. J. Meteor. Soc. Japan,41, 317–325.

  • Kain, J., , and J. Fritsch, 1993: Convective parameterization for mesoscale models: The Kain–Fritsch scheme. The Representation of Cumulus Convection in Numerical Models, Meteor.Monogr., No. 46, Amer. Meteor. Soc., 165–170.

  • Kain, J., , S. Weiss, , J. Levit, , M. Baldwin, , and D. Brigh, 2006: Examination of convection-allowing configurations of the WRF model for the prediction of severe convective weather: The SPC/NSSL Spring Program 2004. Wea. Forecasting, 21, 167181.

    • Search Google Scholar
    • Export Citation
  • Kalnay, E., , and M. Cai, 2003: Impacts of urbanization and land-use change on climate. Nature, 423, 528531.

  • Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc.,77, 437–471.

  • Karl, T. R., , and R. W. Knight, 1998: Secular trend of precipitation amount, frequency, and intensity in the United States. Bull. Amer. Meteor. Soc., 79, 231242.

    • Search Google Scholar
    • Export Citation
  • Kaufmann, R., , K. Seto, , A. Schneider, , Z. Liu, , L. Zhou, , and W. Wang, 2007: Climate response to rapid urban growth: Evidence of a human-induced precipitation deficit. J. Climate, 20, 22992306.

    • Search Google Scholar
    • Export Citation
  • 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, 6780.

    • Search Google Scholar
    • Export Citation
  • Kusaka, H., , H. Kando, , 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.

    • Search Google Scholar
    • Export Citation
  • Liang, J., , and S. Wu, 1999: Climatological diagnosis of winter temperature variations in Guangdong (in Chinese). J. Trop. Meteor., 15, 221229.

    • Search Google Scholar
    • Export Citation
  • Liao, J., , X. Wang, , Y. Li, , and B. Xia, 2011: An analysis study of urbanization on precipitation in Guangzhou (in Chinese). J. Meteor. Sci., 31, 384390.

    • Search Google Scholar
    • Export Citation
  • Lin, C.-Y., , W.-C. Chen, , S. C. Liu, , Y. Liou, , G. R. Liu, , and T. H. Lin, 2008: Numerical study of the impact of urbanization on the precipitation over Taiwan. Atmos. Environ., 42, 29342947.

    • Search Google Scholar
    • Export Citation
  • Lin, W., , L. Zhang, , D. Du, , and L. Yang, 2009: Quantification of land use/land cover changes in Pearl River Delta and its impact on regional climate in summer using numerical modeling. Reg. Environ. Change, 9, 7582.

    • Search Google Scholar
    • Export Citation
  • Lin, Y.-L., , R. D. Farley, , and H. D. Orville, 1983: Bulk parameterization of the snow field in a cloud model. J. Climate Appl. Meteor., 22, 10651092.

    • Search Google Scholar
    • Export Citation
  • Liu, X.-H., , Y. Zhang, , K.-M. Olsen, , W.-X. Wang, , B. A. Do, , and G. M. Bridgers, 2010: Responses of future air quality to emission controls over North Carolina, Part I: Model evaluation for current-year simulations. Atmos. Environ., 44, 24432456.

    • Search Google Scholar
    • Export Citation
  • Miao, S., , and F. Chen, 2008: Formation of horizontal convective rolls in urban areas. Atmos. Res., 89, 298304.

  • Miao, S., , F. Chen, , M. LeMone, , M. Tewari, , Q. Li, , and Y. Wang, 2009a: An observational and modeling study of characteristics of urban heat island and boundary layer structures in Beijing. J. Appl. Meteor. Climatol., 48, 484501.

    • Search Google Scholar
    • Export Citation
  • Miao, S., , F. Chen, , Q. Li, , and S. Fan, 2009b: Impacts of urbanization on a summer heavy rainfall in Beijing. Proc. Seventh Int. Conf. on Urban Climate, Yokohama, Japan, Int. Association for Urban Climate, B12-1.

  • Misenis, C., , and Y. Zhang, 2010: An examination of sensitivity of WRF/Chem predictions to physical parameterizations, horizontal grid spacing, and nesting options. Atmos. Res., 97, 315334.

    • 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 (D14), 16 66316 682.

    • Search Google Scholar
    • Export Citation
  • Montaveza, J., , J. Gonzalez-Rouco, , and F. Valero, 2008: A simple model for estimating the maximum intensity of nocturnal urban heat island. Int. J. Climatol., 28, 235242.

    • Search Google Scholar
    • Export Citation
  • Oke, T. T., 1982: The energetic basis of the urban heat island. Quart. J. Roy. Meteor. Soc., 108, 124.

  • Ren, G., , Y. Zhou, , Z. Chu, , J. Zhou, , A. Zhang, , J. Guo, , and X. Liu, 2008: Urbanization effects on observed surface air temperature trends in north China. J. Climate, 21, 13331348.

    • Search Google Scholar
    • Export Citation
  • Rizwan, A., , Y. Dennis, , and C. Liu, 2008: A review on the generation, determination and mitigation of urban heat island. J. Environ. Sci., 20, 120128.

    • Search Google Scholar
    • Export Citation
  • Rozoff, C. M., , W. R. Cotton, , and J. O. Adegoke, 2003: Simulation of St. Louis, Missouri, land use impacts on thunderstorms. J. Appl. Meteor., 42, 716738.

    • Search Google Scholar
    • Export Citation
  • Shem, W., , and M. Shepherd, 2008: On the impact of urbanization on summertime thunderstorms in Atlanta: Two numerical model case studies. Atmos. Res., 92, 172–189, doi:10.1016/j.atmosres.2008.09.013.

    • Search Google Scholar
    • Export Citation
  • Tomita, T., , H. Kusaka, , R. Akiyoshi, , and Y. Imasato, 2007: Thermal and geometric controls on the rate of surface air temperature changes in a medium-sized, midlatitude city. J. Appl. Meteor. Climatol., 46, 241247.

    • Search Google Scholar
    • Export Citation
  • Trier, S., , F. Chen, , K. Manning, , M. A. LeMone, , and C. Davis, 2008: Sensitivity of the simulated PBL and precipitation to land surface conditions for a 12-day warm-season convection period in the central United States. Mon. Wea. Rev., 136, 23212343.

    • Search Google Scholar
    • Export Citation
  • Wang, X. M., , W. S. Lin, , L. M. Yang, , R. R. Deng, , and H. Lin, 2007: A numerical study of influences of urban land-use change on ozone distribution over Pearl River Delta region, China. Tellus, 59B, 633641.

    • Search Google Scholar
    • Export Citation
  • Wang, X. M., , F. Chen, , Z. Wu, , M. Zhang, , M. Tewari, , A. Guenther, , and C. Wiedinmyer, 2009a: Impacts of weather conditions modified by urban expansion on surface ozone over the Pearl River Delta and Yangtze River Delta regions, China. Adv. Atmos. Sci., 26, 962972, doi:10.1007/s00376-009-8001-2.

    • Search Google Scholar
    • Export Citation
  • Wang, X. M., , Z. Wu, , and G. Liang, 2009b: WRF/CHEM modeling of impacts of weather conditions modified by urban expansion on secondary organic aerosol formation over Pearl River Delta. Particuology, 7, 384391.

    • Search Google Scholar
    • Export Citation
  • Wiedinmyer, C., , M. Barlage, , M. Tewari, , and F. Chen, 2012: Meteorological impacts of forest mortality due to insect infestation in Colorado. Earth Interact.,16,doi:10.1175/2011EI419.1.

  • Wu, Z., and Coauthors, 2011: Evaluating the calculation of dry deposition velocities for reactive nitrogen oxides and ozone from two community dry deposition models for a temperate deciduous forest. Atmos. Environ., 45, 26632674.

    • Search Google Scholar
    • Export Citation
  • Xiong, Y., , Z. Zhang, , and F. Chen, 2010: Estimation of urbanization effect on climatic warming over the recent 30 years in Guangzhou, south China. Proc. 18th Int. Conf. on Geoinformatics, Beijing, China, IEEE, 1–6.

  • Zhang, C., , F. Chen, , S. Miao, , Q. Li, , A. Xia, , and C. 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, H., , N. Sato, , T. Izumi, , K. Hanaki, , and T. Aramaki, 2008: Modified RAMS-urban canopy model for heat island simulation in Chongqing, China. J. Appl. Meteor. Climatol., 47, 509524.

    • Search Google Scholar
    • Export Citation
  • Zhang, N., , Z. Gao, , X. Wang, , and Y. Chen, 2010a: Modeling the impact of urbanization on the local and regional climate in Yangtze River Delta, China. Theor. Appl. Climatol., 102, 331342.

    • Search Google Scholar
    • Export Citation
  • Zhang, Y., , X. Wen, , and C. Jang, 2010b: Simulating chemistry–aerosol–cloud–radiation–climate feedback over the continental U.S. using online-coupled Weather Research Forecasting Model with chemistry (WRF/Chem). Atmos. Environ., 44, 35683582.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    (a) Terrain height of southern China; the red box denotes the PRD. (b) An enlarged view of the PRD region with its 11 cities.

  • View in gallery

    Modeling domains used in WRF simulations with terrain heights. (a) Domains D01 and D02, and (b) D02 with locations of cities GZ, DG, ZS, ZH, and SZ in red. (c),(d) The 1992 USGS land-use data and 2004 MODIS land-use data used for D02, respectively.

  • View in gallery

    Model vs observation comparison of the five site-average diurnal variation of (top to bottom in each panel set) T2, RH2, and WS10. OBS indicates the observations (dashed line), SIM the simulations (solid line), and OBS±SD the ±1 standard deviation of OBS (gray area): (a) January and (b) July.

  • View in gallery

    Time series of daily precipitation (mm day−1) for (a) January and (b) July from observations (light gray) and model simulations (dark gray) for the cities (top to bottom in each panel set) GZ, DG, SZ, ZS, and ZH.

  • View in gallery

    Difference in the seasonal-average T2 between the URB and NOURB cases for (a) MAM, (b) JJA, (c) SON, and (d) DJF. Differences that were not significant at the 95% confidence level (Student’s t test) have been masked out.

  • View in gallery

    Seasonal wind direction frequency (%) in the PRD for 2008: MAM (blue), JJA (red), SON (green), and DJF (black). Circles indicate increasing wind direction percent frequency with an interval of 5%; and N = northern wind, E = eastern wind, S = southern wind, and W = western wind.

  • View in gallery

    As in Fig. 5, but for the diurnal temperature (°C) range.

  • View in gallery

    Diurnal variation of TSK for URB (solid line) and NOURB (dash line) cases over urbanized areas during the four seasons.

  • View in gallery

    Annual average WS10 (m s−1) for (a) the URB cases and (b) the difference between the URB and NOURB cases. Differences from (b) that were not significant at the 95% confidence level (Student’s t test) have been masked out.

  • View in gallery

    As in Fig. 5, but for the 2-m water vapor mixing ratio (g kg−1).

  • View in gallery

    Differences between the URB and NOURB cases for (a) annual rainy days and (b) the total precipitation ratio (%). Differences that were not significant at the 95% confidence level (Student’s t test) have been masked out.

  • View in gallery

    Differences in rain days between the URB and NOURB cases for (a) light, (b) moderate, (c) heavy, and (d) extreme rainfall days. Differences that were not significant at the 95% confidence level (Student’s t test) have been masked out.

  • View in gallery

    Vertical cross section at 23.1°N showing the differences between the URB and NOURB cases for (a) wind speed (cm s−1) and (b) water vapor mixing ratio (g kg−1) during the rainy season (the horizontal red bars represent urbanized areas).

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 102 102 20
PDF Downloads 89 89 13

A Numeric Study of Regional Climate Change Induced by Urban Expansion in the Pearl River Delta, China

View More View Less
  • 1 School of Environmental Science and Engineering, Sun Yat-sen University, Guangzhou, China
  • 2 School of Atmospheric Sciences, Nanjing University, Nanjing, China
© Get Permissions
Full access

Abstract

The Pearl River Delta region has experienced rapid urbanization and economic development during the past 20 years. To investigate the impacts of urbanization on regional climate, the Advanced Research core of the Weather Research and Forecasting (ARW-WRF) model is used to conduct a pair of 1-yr simulations with two different representations of urbanization. Results show that the reduction in vegetated and irrigated cropland due to urban expansion significantly modifies the near-surface temperature, humidity, wind speed, and regional precipitation, which are obtained based on the significance t test of the differences between two simulations with different urbanization representations at the 95% level. Urbanization causes the mean 2-m temperature over urbanized areas to increase in all seasons (from spring to winter: 1.7° ± 0.7°C, 1.4° ± 0.3°C, 1.3° ± 0.3°, and 0.9° ± 0.4°C, respectively) and the urban diurnal temperature range decreases in three seasons and increases in one (from spring to winter: −0.5° ± 0.3°C, +0.6° ± 0.3°C, −0.4° ± 0.2°C, and −0.8° ± 0.2°C, respectively). Urbanization reduces near-surface water vapor (1.5 g kg−1 in summer and 0.4 g kg−1 in winter), 10-m wind speed (37% independent of season), and annual total precipitation days (approximately 6–14 days). However, the total rainfall amount increases by approximately 30%, since the decrease in the number of days with light rain (8–12) is overcome by the increase in the number of days of heavy or extreme rain (3–6), suggesting that urbanization induces more heavy rain events over the urban areas. Overall, the effect of urbanization on regional climate in the Pearl River Delta is found to be significant and must be considered in any broader regional climate assessment.

Corresponding author address: Xuemei Wang, School of Environmental Science and Engineering, Sun Yat-sen University, 135 Xingang West Road, Guangzhou 510275, China. E-mail: eeswxm@mail.sysu.edu.cn

Abstract

The Pearl River Delta region has experienced rapid urbanization and economic development during the past 20 years. To investigate the impacts of urbanization on regional climate, the Advanced Research core of the Weather Research and Forecasting (ARW-WRF) model is used to conduct a pair of 1-yr simulations with two different representations of urbanization. Results show that the reduction in vegetated and irrigated cropland due to urban expansion significantly modifies the near-surface temperature, humidity, wind speed, and regional precipitation, which are obtained based on the significance t test of the differences between two simulations with different urbanization representations at the 95% level. Urbanization causes the mean 2-m temperature over urbanized areas to increase in all seasons (from spring to winter: 1.7° ± 0.7°C, 1.4° ± 0.3°C, 1.3° ± 0.3°, and 0.9° ± 0.4°C, respectively) and the urban diurnal temperature range decreases in three seasons and increases in one (from spring to winter: −0.5° ± 0.3°C, +0.6° ± 0.3°C, −0.4° ± 0.2°C, and −0.8° ± 0.2°C, respectively). Urbanization reduces near-surface water vapor (1.5 g kg−1 in summer and 0.4 g kg−1 in winter), 10-m wind speed (37% independent of season), and annual total precipitation days (approximately 6–14 days). However, the total rainfall amount increases by approximately 30%, since the decrease in the number of days with light rain (8–12) is overcome by the increase in the number of days of heavy or extreme rain (3–6), suggesting that urbanization induces more heavy rain events over the urban areas. Overall, the effect of urbanization on regional climate in the Pearl River Delta is found to be significant and must be considered in any broader regional climate assessment.

Corresponding author address: Xuemei Wang, School of Environmental Science and Engineering, Sun Yat-sen University, 135 Xingang West Road, Guangzhou 510275, China. E-mail: eeswxm@mail.sysu.edu.cn

1. Introduction

Land-use/land-cover change modifies land surface physical properties, such as albedo, roughness length, and so on, which in turn affect climate, air quality, human health, and urban development. Tall buildings in cities form the urban canopy layer and modify the surface energy balance and dynamic characteristics of the surface layer. These changes influence the exchange of the heat, water vapor, and momentum between the atmosphere and the surface and, therefore, change the local and regional weather and climate, and impact the transport and dispersion of pollutants and air quality (Zhang et al. 2010b; Wang et al. 2007, 2009a,b; Arnfield 2003; Zhang et al. 2008; Liu et al. 2010; Miao et al. 2009a,b; Chen et al. 2011b).

Climatological data indicate that the mean December–February (DJF) temperature of southern China has increased by approximately 0.0326°C yr−1 for the past 37 yr (Liang and Wu 1999). Changes in extreme temperatures in megacities are driven not only by the large-scale or regional warming, but also by the urban heat island (UHI) effect (Houghton et al. 2001). In many cities, the increases in extremely high temperatures are more greatly affected by the UHI effect than the global warming signal. Based on global reanalysis data from 1950 to 1999, Kalnay and Cai (2003) suggested that half of the observed decrease in the diurnal temperature range in the continental United States (0.27°C century−1 of surface warming) was caused by urban and other land-use changes.

Recently, numerical models have been used to identify the contribution of land-cover change to climate variability with a focus on UHI (Erell and Williamson 2007; Ferguson and Woodbury 2007; Montaveza et al. 2008; Rizwan et al. 2008; Tomita et al. 2007; Wang et al. 2009a; Zhang et al. 2008), local and regional warming and circulation changes induced by urbanization (Xiong et al. 2010; Lin et al. 2008, 2009; Zhang et al. 2010a; Wang et al. 2007, 2009a; Chen et al. 2011b), and impacts of urbanization on precipitation and water resources (Arnfield 2003; Kaufmann et al. 2007; Shem and Shepherd 2008; Lin et al. 2008; Miao et al. 2009b; Zhang et al. 2009). For instance, Miao et al. (2009b) showed that intense urbanization contributed to enhancing a heavy rainfall event in Beijing on 1 August 2006, and Zhang et al. (2009) demonstrated that augmenting Beijing urban green vegetation could reduce extreme rainfall events in summer. Therefore, it is possible that intense urban development might have exacerbated these and other recent strong Beijing thunderstorms. Furthermore, rapid urban expansion of the Yangtze River Delta in eastern China and of the Pearl River Delta (PRD) in southern China has been shown to change the local climate, increasing the surface temperature and boundary layer mixing (Wang et al. 2009a).

The Pearl River Delta region, located in a coastal zone at the center of Guangdong Province in southern China, has experienced remarkable economic development and urbanization in the last 30 years. Today, the urban area accounts for more than 60% in the total land cover, which is 2 times higher than the Chinese national average level. Observational data indicate that the surface temperature is increasing by 0.44°C decade−1 in Guangzhou, as compared to 0.11°C decade−1 in northern China (Xiong et al. 2010; Ren et al. 2008). Moreover, a data analysis by Liao et al. (2011) demonstrates that urbanization contributes to an increase in heavy rain days and precipitation amounts in Guangzhou by 2.8 days decade−1 and 2.4% decade−1, respectively. Furthermore, Lin et al. (2009) simulate the effects of urban expansion on monthly climate in the PRD region and find an increase in the UHI (0.9°C) and a decrease in the wind velocity (20%).

However, these studies focus on a single weather event or only cover a limited period of time, such as a single season of the year. This study presents a systematic analysis of the annual variability of meteorological conditions influenced by land-use change in the PRD. The Weather Research and Forecast (WRF) model coupled to a new generation of urban canopy schemes (Chen et al. 2011a) is used to conduct a 1-yr simulation for 2008 under two land-use and land-cover scenarios in the PRD. This scheme is shown to be capable of capturing the impacts of urbanization on near-surface meteorological conditions and the evolution of atmospheric boundary layer structures in cities (Miao et al. 2009a,b; Wang et al. 2009a; Chen et al. 2011a,b). Furthermore, the WRF urban-canopy model is enhanced with high-resolution land-use data to allow for a more realistic simulation of the underlying surface properties and the evolution of the atmospheric boundary layer.

Analysis in this study focuses on seasonal changes in temperature, humidity, wind speed, boundary layer depth, and precipitation extremes. The next section contains a detailed description of the WRF urban model and numerical experiments; section 3 evaluates model simulations, observations, and the impacts of urbanization; and section 4 summarizes and discusses the results.

2. Model description and simulation design

a. Model description

For this study, we have chosen to use the Advanced Research core of the Weather Research and Forecasting (WRF) model system version 3.1.1, a next-generation mesoscale weather and regional climate numerical simulation system (e.g., Done et al. 2004; Grell et al. 1994, 2005; Davis et al. 2006, 2008; Kain et al. 2006; Misenis and Zhang, 2010; Chen et al. 2011b; Wu et al. 2011; Wiedinmyer et al. 2012). The modeling system is coupled to a new generation of urban canopy schemes (Chen et al. 2011a). This combination is shown to be capable of capturing the impacts of urbanization on near-surface meteorological conditions and the evolution of atmospheric boundary layer structures in cities (Miao et al. 2009a,b; Wang et al. 2009a; Chen et al. 2011a,b). Finally, it is possible to alter the land surface component of the WRF model with high-resolution land-use data, making it a good fit for the project at hand. Overall, it has been shown that the WRF model performs well and is suitable for regional weather or climate phenomenon simulations (http://www.wrf-model.org).

b. Pearl River Delta region

The PRD region is in the center of Guangdong Province in southern China, as shown in Fig. 1, and contains the nine cities: Guangzhou, Shenzhen, Foshan, Dongguan, Zhongshan, Zhuhai, Huizhou, Jiangmen, and Zhaoqing. It has a monsoonal climate, with an annual average temperature of 21.0°–23.0°C. The rainy season spans from April to September, and the dry season from October to March. The annual average precipitation in the PRD is about 1600 mm over 130 rainy days per year.

Fig. 1.
Fig. 1.

(a) Terrain height of southern China; the red box denotes the PRD. (b) An enlarged view of the PRD region with its 11 cities.

Citation: Journal of Applied Meteorology and Climatology 53, 2; 10.1175/JAMC-D-13-054.1

This region has undergone rapid expansion and today contributes roughly 70% of the province’s gross domestic product (GDP). The region accounts for 23.4% of the total area of Guangdong Province and contains 31.4% of its total population. The PRD has witnessed rapid urbanization over the past decade, with a 72.7% urbanization fraction (urban land use) as of 2010. Furthermore, the absolute population has increased from 24.0 million in 1990 to 56.1 million. Additionally, both the urbanization fraction and population are expected to continue to increase in the future.

c. Experimental design

For this study, the WRF model is configured with two nested domains with grid spacings of 12 km × 12 km and 4 km × 4 km, respectively; both domains are centered at 23°N and 113°E, and 27 sigma levels up to 100 hPa in the vertical. The gridpoint dimensions are 130 × 110 × 27 and 151 × 109 × 27, respectively. A representation of the two domains is shown in Fig. 2. The outer domain (D01) comprises southern China and the South China Sea, while the inner domain (D02) comprises a large portion of Guangdong Province centered on the PRD region.

Fig. 2.
Fig. 2.

Modeling domains used in WRF simulations with terrain heights. (a) Domains D01 and D02, and (b) D02 with locations of cities GZ, DG, ZS, ZH, and SZ in red. (c),(d) The 1992 USGS land-use data and 2004 MODIS land-use data used for D02, respectively.

Citation: Journal of Applied Meteorology and Climatology 53, 2; 10.1175/JAMC-D-13-054.1

The simulation period covers the entire year of 2008, which was chosen because there were heavy rain events during the summer. This allows the seasonal impacts induced by urban expansion on both the regional climate (i.e., surface temperature) and extreme events (i.e., rain) to be examined. Furthermore, Meteorological observation sites supervised by the Bureau of Meteorology have a complete record of hourly data during 2008, allowing for validation of the modeling results. The locations of these sites are given in Fig. 2b.

To better understand the influence of urbanization on regional climate change, two land-use scenarios for the PRD region are used here. To represent the pre-urbanization land-use conditions (referred to as NOURB hereafter), U.S. Geological Survey (USGS) global land-use types based on 1992–93 1-km Advanced Very High Resolution Radiometer data (Fig. 2c) are used in the simulation. The present-day representation of the urban distribution in the PRD region (referred to as URB hereafter) is based on 500-m Moderate Resolution Imaging Spectroradiometer (MODIS) data from 2004 (Fig. 2d). Comparisons of these two datasets clearly show that the PRD region has undergone a rapid urban expansion between 1992 and 2004, with the changes mainly centered in Guangzhou, Foshan, Shenzhen, and Dongguan.

The physical parameterization schemes used in WRF are designed to address the specifics of the problem at hand (see Table 1 for a summary). The initial and boundary conditions for D01 are interpolated from the 1° × 1° National Centers for Environmental Prediction–National Center for Atmospheric Research (NCEP–NCAR) global reanalysis dataset (Kalnay et al. 1996). The initial and boundary conditions for D02 are interpolated from D01. The Kain–Fritsch cumulus scheme (KF; Kain and Fritsch 1993) is employed in D01, while no cumulus parameterization scheme is used in D02 because convection is assumed to be reasonably resolved by the explicit microphysics (Trier et al. 2008). The Noah land surface scheme (Chen and Dudhia 2001; Chen and Zhang 2009) provides to the WRF, surface sensible and latent heat fluxes along with surface skin temperatures for the lower boundary conditions.. To represent the thermal and dynamic effects of urban areas, the single-layer urban canopy model (Kusaka et al. 2001; Kusaka and Kimura. 2004) was coupled to Noah (Chen et al. 2011a), and accounts for the urban geometry’s effects on the surface energy budget and wind shear calculations (Miao and Chen 2008). This model is able to represent seven distinct effects including 1) 2D street canyons on urban heat distribution; 2) shadowing and reflection from buildings on solar radiation; 3) canyon orientation and the diurnal cycle of the solar azimuth angle; 4) man-made surfaces; 5) canopy flows (Inoue, 1963); 6) the multilayer heat equation for the roof, wall, and road interior temperatures; and 7) a very thin bucket model for evaporation and runoff from road surfaces. The impacts of aerosols are not included, as these may lead to additional model uncertainties in the simulations.

Table 1.

The physical parameterization schemes used in the model.

Table 1.

The simulated regional influence of urbanization can be assessed by the effect index (EI), described by Zhang et al. (2010a) as
e1
where x can be any meteorological parameter (such as temperature or wind speed), Achange(x) is the area over which the variable x differs between simulations, and Aurban is the area that has had its land-use modified from other land-use types to urban cover. If the absolute difference of the domain average value between the URB and NOURB is larger than 0.2°C for air temperature, 0.2 m s−1 for wind speed, or 0.2 g kg−1 for water mixing ratio, then these changes are considered as having been caused by land-use change (Zhang et al. 2010b). An EI(x) of 1 indicates that only the urbanized area is affected, an EI(x) less than 1 indicates that only part of the urbanized area is affected, and an EI(x) greater than 1 means the nonurbanized area in addition to the urbanized area will be impacted.

3. Results and analysis

a. Observational verification of model performance

To evaluate model performance, hourly results are compared from the URB simulation with observation station data from each of five cities: Guangzhou (GZ), Shenzhen (SZ), Dongguan (DG), Zhongshan (ZS), and Zhuhai (ZH) (see Fig. 2b). Statistical comparisons between these data and the 2008 model results quantify the mean bias (MB), root-mean-square error (RMSE), and correlation coefficient (Corr) for 2-m temperature (T2), 2-m relative humidity (RH2), and 10-m wind speed (WS10) (Wang et al. 2009a). The WRF model simulates the annual T2 reasonably well (Table 2) with an MB of generally less than 1.5°C [1.2°C from Lin et al. (2009)], an RMSE of 1.1°–1.9°C, and a Corr of 0.80–0.91. The annual RH2 has an MB of less than 8.0% at four sites and an RMSE of 14.4%. The annual WS10 MB is generally less than 1.0 m s−1, and has an RMSE of less than 2.0 m s−1. The general overestimation of WS10 occurs since it is more affected by local underlying surface characteristics than are the other meteorological parameters studied here (Zhang et al. 2010b).

Table 2.

The annual prediction statistics of the WRF model for select major cities in the PRD.a

Table 2.

Representations of seasonal results are created using hourly URB results from January and July. The two months are representative of the relatively cold and dry season of the year, and the relatively hot and wet season of the year, respectively. Data from five observation sites are compared with model average T2, RH2, and WS10, as shown in Fig. 3, and calculated evaluation statistics for hourly data and model comparisons in January and July are given in Table 3. Comparing the simulated diurnal variations of T2, RH2, and WS10 with observations shows that there is a better match in January than that in July. In January, the simulated T2 biases are 0.2°–0.7°C from 0000 local standard time (LST) to 1700 LST, and only 0.1°–0.2°C from 1800 to 2300 LST. This differs with July, where the model overestimates the T2 throughout the whole day, with an MB of 0.2°–3.1°C. In January, the modeled RH2 from 0000 to 0600 is 2.3%–5.3% lower than the corresponding observations, while from 0700 to 2300 LST it is 0.2%–6.5% higher. In July, however, it is always 2.3%–12.4% lower than the corresponding observations. Furthermore, T2 has an MB of 0.2°–1.0°C and a correlation coefficient of 0.86–0.92 in January, while in July, the MB of T2 is larger than in January, where the MB values are 0.6°, 0.4°, 1.8°, 2.1°, and 1.4°C, respectively, at SZ, ZS, GZ, DG, and ZH. It should be noted that severe freezing weather occurred during January 2008 (Liu et al. 2010), and the model reproduces these low temperatures in January quite well. On the other hand, the larger biases in July may be due to the influence of the urban canopy. One such explanation is that solar radiation absorption in July is larger, and therefore more heat is stored in the canopy layer. This heat storage may reheat the air between the buildings in the street canyon and thus lead to a temperature increase (Kusaka and Kimura 2004).

Fig. 3.
Fig. 3.

Model vs observation comparison of the five site-average diurnal variation of (top to bottom in each panel set) T2, RH2, and WS10. OBS indicates the observations (dashed line), SIM the simulations (solid line), and OBS±SD the ±1 standard deviation of OBS (gray area): (a) January and (b) July.

Citation: Journal of Applied Meteorology and Climatology 53, 2; 10.1175/JAMC-D-13-054.1

Table 3.

The prediction statistics of the WRF model over major cities.a

Table 3.

The WS10 from the URB case is overestimated at all meteorological sites for both January and July; however, similar results from WRF simulations are also reported upon by other studies (Wang et al. 2009a; Liu et al. 2010; Zhang et al. 2010a). The average URB case MBs for WS10 are 0.6 m s−1 in January and 1.0 m s−1 in July. The model overestimates WS10 the most for both January and July during the day, with respective MBs of 0.9–1.1 and 0.2–1.8 m s−1.

Overall, the model captures the trends of RH2 from the observed data during most of January and July, but with an underestimation occurring mainly during the daytime. This may be due to an excess amount of water vapor decrease in the URB case. The maximum MB in RH2 is −6.5% in January, but reaches −14.7% in July.

The model generally captures the daily variations of precipitation in January and July with heavy rain events well reproduced, although there are a few discrepancies as shown for the URB case in Fig. 4. The modeled total precipitation amounts at the five sites are 479.4 mm in January and 1357.4 mm in July, which compares well with the observed values of 424.0 and 1278.1 mm, respectively. However, a day-to-day comparison of January precipitation reveals a 1-day lag of simulated precipitation, with the modeled heavy rain event of 26 January occurring a day later than the observations. On the other hand, in July the simulations do not exhibit a time lag and match the overall amount better, with the exception that the ZH site is underestimated from 6 to 15 July.

Fig. 4.
Fig. 4.

Time series of daily precipitation (mm day−1) for (a) January and (b) July from observations (light gray) and model simulations (dark gray) for the cities (top to bottom in each panel set) GZ, DG, SZ, ZS, and ZH.

Citation: Journal of Applied Meteorology and Climatology 53, 2; 10.1175/JAMC-D-13-054.1

Generally, simulations in this study compare reasonably well with the observations and other studies regarding model performance. Some of the biases are found in the simulation when compared with observations, which are possibly attributable to the initial meteorological and boundary conditions (Zhang et al. 2010b); however, the magnitude of these biases is such that they should not impact a long-running climate simulation. Additionally, there are limitations of the small-scale spatial schemes, like the planetary boundary layer (PBL) scheme, the land surface model, and the radiation scheme, which also affect the simulation quality in the current generation of mesoscale meteorological models (Wang et al. 2009a). It is for these reasons that experiments are designed to only look at the differences between land-use scenarios, which should reduce the impact that any bias by varying other model physics schemes may have on the results. Furthermore, the performance of the WRF model, as used in this study for this region, is generally consistent with other current studies undertaken using meteorological models. Therefore, for the purposes of analyzing the change in climate induced by a change in urbanization over this region of the world, we believe that this is a suitable and appropriate modeling tool.

b. Influence on surface temperature and diurnal range

The difference between the URB and NOURB seasonally averaged T2 values is shown in Fig. 5, with the nonsignificant differences (based on the 95% confidence level obtained using a Student’s t test) masked out. The warming areas are concentrated in the center of the PRD, where a large portion of the area is changed from nonurban to urban between the two scenarios. The annual average increase of T2 for URB minus NOURB is about 1.4° ± 0.8°C over the whole urbanized area in the PRD region. The March–May (MAM) and June–August (JJA) increases in T2 are 1.7° ± 0.7° and 1.4° ± 0.3°C, which are greater than the September–November (SON) and December–February (DJF) increases of 1.3° ± 0.3° and 0.9° ± 0.4°C. The influence of urbanization on T2 in JJA has an EI = 2.4, which is larger than in DJF, with an EI = 1.4. This is possibly due to greater radiation heating at the surface in the summer, coupled with more effective horizontal heat transport.

Fig. 5.
Fig. 5.

Difference in the seasonal-average T2 between the URB and NOURB cases for (a) MAM, (b) JJA, (c) SON, and (d) DJF. Differences that were not significant at the 95% confidence level (Student’s t test) have been masked out.

Citation: Journal of Applied Meteorology and Climatology 53, 2; 10.1175/JAMC-D-13-054.1

Other aspects related to changes in temperature include the PBL and prevailing winds. The top of the planetary boundary layer in JJA is typically higher (Wang et al. 2009b) due to the greater heating in the urban canopy enhancing thermal convection, which facilitates the transport of heat to higher layers of the atmosphere. The prevailing winds, as shown in Fig. 6, from the south enhance the horizontal heat and the vapor exchange between the urban and rural areas, which heat the atmosphere over downwind rural areas. The same phenomena also happen in SON and DJF where the north and northeast winds transport the heat to the south and southwest of the urban areas, which influences their regional temperature.

Fig. 6.
Fig. 6.

Seasonal wind direction frequency (%) in the PRD for 2008: MAM (blue), JJA (red), SON (green), and DJF (black). Circles indicate increasing wind direction percent frequency with an interval of 5%; and N = northern wind, E = eastern wind, S = southern wind, and W = western wind.

Citation: Journal of Applied Meteorology and Climatology 53, 2; 10.1175/JAMC-D-13-054.1

Urbanization has a significant influence on the difference between the maximum and minimum temperatures during a day [diurnal temperature range (DTR)], as shown by Du et al. (2007), Zhang et al. (2010a), and Lin et al. (2009). Modifications to the energy budget in the urban canopy are the main cause of these changes in the DTR. Also, modification of the land use to an urban type leads to changes the surface albedo through the shading effect, thereby influencing surface radiation absorption in the daytime. Furthermore, the radiation feedbacks in the urban canopy are important for the energy balance, with more solar radiation trapped by buildings and retained in street canyons, leading to the release of the heat during the nighttime. Although urbanization increases both the minimum and maximum temperatures (not shown), the decreased DTR reveals that the increase in the minimum temperature is greater than that of the maximum temperature, especially during DJF.

The spatial distribution of the DTR differences between the URB and NOURB cases for different seasons is shown in Fig. 7, where the differences that are not significant at the 95% confidence level (Student’s t test) are masked out. The DTRs over the five cities decrease by −0.5° ± 0.3°C (EI = 0.9) in MAM, −0.4° ± 0.2°C (EI = 0.9) in SON, and −0.8° ± 0.2°C (EI = 1.1) in DJF. Such a DTR change is much stronger in the PRD region when compared with the Yangtze River Delta (−0.13° ± 0.73°C in January) (Zhang et al. 2010a). It is demonstrated that the DTR decreases as a result of urbanization, similar to the finding of Zhang et al. (2010a).

Fig. 7.
Fig. 7.

As in Fig. 5, but for the diurnal temperature (°C) range.

Citation: Journal of Applied Meteorology and Climatology 53, 2; 10.1175/JAMC-D-13-054.1

However, the variation of the DTR in JJA is different from other seasons in the simulation. The DTR in JJA is higher in URB than NOURB, not only in the center of PRD but also in the rural area, with the maximum exceeding 0.6° ± 0.3°C (EI = 2.9), suggesting that the increase in the maximum temperature is greater than that of minimum temperature. The increase of the DTR difference in JJA is caused by the fact that the PRD is located in a subtropical region, where much stronger solar radiation reaches the urban canopy layer due to the higher elevation angle of the sun in JJA than in DJF. Therefore, more energy is gained in the daytime, which in turn, increases the daytime temperature (urbanized area maximum temperature is increased by 1.4°C in the daytime in this study). At night, air masses from the ocean cool the atmosphere (urbanized area minimum temperature is increased by 0.9°C during the nighttime in this study) and constrain the increase in temperature at night.

The modified surface albedo and radiation characteristics also affect the surface skin temperature (TSK). The differences in TSK between URB and NOURB are 2.3°C (EI = 1.6), 1.8°C (EI = 1.2), 2.1°C (EI = 1.1), and 1.9°C (EI = 1.4) during MAM, JJA, SON, and DJF, respectively, and show a similar spatial distribution as for T2 (not shown). As shown in Fig. 8, the increase in urbanized-area-averaged maximum (minimum) TSKs are 0.2°C (0.3°C) in MAM, 2.8°C (1.6°C) in JJA, 0.3°C (0.2°C) in SON, and 1.2°C (0.2°C) in DJF. The statistics for the TSK change are similar to those of the T2 change, but on an absolute basis, the increase in TSK is slightly stronger than the increase in T2, which is consistent with the direct changes in surface radiation characteristics.

Fig. 8.
Fig. 8.

Diurnal variation of TSK for URB (solid line) and NOURB (dash line) cases over urbanized areas during the four seasons.

Citation: Journal of Applied Meteorology and Climatology 53, 2; 10.1175/JAMC-D-13-054.1

c. Impact on 10-m wind speed and boundary layer depth

Urbanization also increases the surface roughness because of the heterogeneous distribution of tall surface elements (i.e., buildings). Furthermore, the friction effect of these buildings leads to drag, decreasing the near-surface wind speed, as shown in Fig. 9. The results from the URB case (Fig. 9a) show that the annual averaged WS10 over the urban area is ~1 m s−1 lower than that in the rural area (with a maximum of 2 m s−1 over the high-density area). Compared with the NOURB case (Fig. 9b; differences that were not significant at the 95% confidence level using a Student’s t test have been masked out), it is clear that the WS10 over the urban area decreases by roughly 1.2–1.5 m s−1 over highly urbanized regions. Overall, changes in urbanization cause an approximately 37% decrease in the mean wind speed, which is compatible with the 50% wind speed loss due to urbanization over the Yangtze River Delta region reported in Zhang et al. (2010a).

Fig. 9.
Fig. 9.

Annual average WS10 (m s−1) for (a) the URB cases and (b) the difference between the URB and NOURB cases. Differences from (b) that were not significant at the 95% confidence level (Student’s t test) have been masked out.

Citation: Journal of Applied Meteorology and Climatology 53, 2; 10.1175/JAMC-D-13-054.1

Mixing and transport within the boundary layer are important mechanisms influencing air pollutants. The large increase in surface roughness due to urbanization reduces wind speeds, which in turn change the venting of air pollution. In this study, urbanization also increases the height of the PBL during both the daytime and nighttime, with the urbanized-area average PBL height increasing by 125 ± 75 m in the daytime and by 100 ± 50 m at night, over an area with a spatial distribution similar to T2. The influence of urbanization on the PBL height is the strongest during JJA (~200 m in daytime and ~150 m at night) and the weakest in DJF (~50 m both in daytime and nighttime). Urbanization leads to the increase of air temperature, resulting in a higher PBL height. The seasonal effects are due to the larger changes in air temperature in JJA as compared to in DJF.

d. Impacts on surface water vapor

Urbanized land types consist of roads and other impervious concrete surfaces, leading to less water being available for evaporation compared to natural land surfaces prior to urbanization. The seasonal variation of the water vapor mixing ratio is given in Fig. 10, where differences that are not significant at the 95% confidence level (Student’s t test) are masked out. The influence of urbanization on RH2 is the strongest during JJA, with the decrease found to be 0.8 g kg−1 (EI = 1.4) in MAM, 1.5 g kg−1 (EI = 2.1) in JJA, 0.9 g kg−1 (EI = 1.1) in SON, and 0.4 g kg−1 (EI = 1.3) in DJF. By comparison, in the Yangtze River Delta the decreases in RH2 are about 1.5 g kg−1 during JJA and 0.1 g kg−1 during DJF (Zhang et al. 2010a). In addition to the reduction in the supply of water vapor, moisture transport from outside the urban area is relatively higher in JJA and relatively lower in DJF, further leading to the larger reduction of water vapor observed during JJA.

Fig. 10.
Fig. 10.

As in Fig. 5, but for the 2-m water vapor mixing ratio (g kg−1).

Citation: Journal of Applied Meteorology and Climatology 53, 2; 10.1175/JAMC-D-13-054.1

e. Impacts on precipitation

Precipitation is an important component of urban climatology, and it has been demonstrated that urbanization has a significant influence on mesoscale circulations affecting regional convection and precipitation (Oke 1982; Bornstein and Lin 2000; Dixon and Mote 2003; Zhang et al. 2009). Increased precipitation downwind of the urban area due to stronger urban heat island effects has been shown by Lin et al. (2008), while the connection between urbanization and enhanced surface convergence has been shown by Rozoff et al. (2003).

Generally, the thermal and dynamic effects caused by urbanization play a significant role in regional precipitation formation. Urbanization impacts precipitation in several ways: 1) urban land creates a heat island, inducing thermal circulation, which may trigger dynamical convection; 2) changes in trapped urban pollutants may act as local cloud nuclei; 3) urbanization reduces local evaporation; and 4) anthropogenic heat release may enhance convection. However, it is difficult to differentiate among these varied but related types of influences. Therefore, in this paper, differences of precipitation obtained from the URB and NOURB simulations are compared to discuss the net effects of the urban land-cover-induced dynamics upon precipitation.

The differences in the annual rainfall number of days and total precipitation between URB and NOURB are shown in Fig. 11, where differences that are not significant at the 95% confidence level (using a Student’s t test) are masked out. The numbers of annual total precipitation days decrease by 6–14 days over the northern PRD region (Fig. 11a) and increase by 2–10 days over the southern coastal PRD region (e.g., SZ, ZS, and ZH). Furthermore, the southern PRD region has a 30%–40% increase in total precipitation due to the increase in the total number of rainfall days (Fig. 11b).

Fig. 11.
Fig. 11.

Differences between the URB and NOURB cases for (a) annual rainy days and (b) the total precipitation ratio (%). Differences that were not significant at the 95% confidence level (Student’s t test) have been masked out.

Citation: Journal of Applied Meteorology and Climatology 53, 2; 10.1175/JAMC-D-13-054.1

However, in eastern GZ, there is a decrease in the total number of rainfall days accompanied by an increase in total precipitation, implying that the occurrence of heavy rain events is more frequent. Some climatological studies of land-use change have similarly shown this same pattern: an increase in extreme rainfall events coupled with a decrease in light precipitation events over land areas (Karl and Knight 1998; Fujibe et al. 2005; Goswami et al. 2006). The rainfall events on a total daily basis are classified into four levels: (a) light rainfall, 0.5–10.0 mm; (b) moderate rainfall, 10.1–50.0 mm; (c) heavy rainfall, 50.1–100.0 mm; and (d) extreme rainfall, above 100 mm. The impacts of urbanization on different rain event levels are shown in Fig. 12, where differences that not significant at the 95% confidence level (using a Student’s t test) are masked out. The number of light rainfall days decreases by about 14 ± 6 over the urbanized area, which is similar to the finding of Liao et al. (2011). The number of moderate precipitation days in southern GZ and coastal cities increases by about 4–12, while in northern GZ there is a decrease of 4–6 days. The frequency of heavy rain increases by 3–6 days over the southern PRD but decreases by 4–8 days over northern GZ. The occurrence of extreme rainfall days increases in DG, SZ, and FS by about 3–6 and decreases over central GZ. Statistically significant changes are also found far away from the urban area. These changes are smaller in magnitude, and are outside of the region of interest in this study. However, we hypothesize that these differences are likely the result of higher-order nonlinearities dynamics induced by land-use change and should be considered for future study (e.g., Wang et al. 2009a,b; Chen and Zhang 2009).

Fig. 12.
Fig. 12.

Differences in rain days between the URB and NOURB cases for (a) light, (b) moderate, (c) heavy, and (d) extreme rainfall days. Differences that were not significant at the 95% confidence level (Student’s t test) have been masked out.

Citation: Journal of Applied Meteorology and Climatology 53, 2; 10.1175/JAMC-D-13-054.1

It is likely that rapid urbanization enhances the UHI intensity and boundary layer circulations, resulting in more water vapor transport to the upper atmosphere. Meanwhile, associated higher surface roughness in urbanized regions slows down the airflow and increases the residence time over the urbanized area. This combination leads to an increase in overall precipitation, due to a decrease in light rain events and a corresponding increase in heavy and extreme rain events. Given that these changes occur during the rainy season, there is an increase in the risk of potential flooding over the PRD region.

Since the vertical distributions of some key variables are also important in revealing the changes between URB and NOURB, a cross section through 23.1°N was chosen to intersect the downtown regions of Guangzhou and Foshan. The difference of the mean vertical wind speed and water mixing ratio between the URB and NOURB cases during the rainy season are analyzed and shown in Fig. 13a. During the rainy season, the low-level (below 850 hPa) convection is enhanced due to intensification from the UHI effect and stronger upward wind speed. The vertical wind speed in the URB case is 1.2–1.6 cm s−1 larger than in the NOURB case. The low-level water-vapor mixing ratio over the urban areas (Fig. 13b) in the URB case is lower than the NOURB case by −0.4 g kg−1, but is higher by 0.2 g kg−1 from 850 to 700 hPa. The urbanized surface induces higher air temperatures, vertical wind speeds, and PBL heights. Meanwhile, increased surface roughness also enhances the mechanical turbulence, which leads to a stronger low-level convergence and increased convection. Together, these processes more effectively lift water vapor to the upper layers and, ultimately, lead to more intense precipitation.

Fig. 13.
Fig. 13.

Vertical cross section at 23.1°N showing the differences between the URB and NOURB cases for (a) wind speed (cm s−1) and (b) water vapor mixing ratio (g kg−1) during the rainy season (the horizontal red bars represent urbanized areas).

Citation: Journal of Applied Meteorology and Climatology 53, 2; 10.1175/JAMC-D-13-054.1

4. Summary and discussion

This study investigates the urbanization impact on regional climate change over China’s Pearl River Delta, using two simulations integrated for the entire year of 2008 from the ARW-WRF model. The two simulations use identical initial and boundary conditions but different spatial characteristics of urbanization. In the NOURB case, USGS land-use data are used to represent the pre-urban conditions, whereas updated 2004 USGS land-use data based on MODIS observations are used in the URB case to represent current-day land-use conditions. Model evaluation with the observed data from five meteorological stations for T2, RH2, and WS10 indicates that the URB case performs well with model bias and within the boundaries of other studies.

The conversion of rural land use into urban land use results in substantial changes of air temperature, wind speed, and precipitation. Urbanization increases T2 over the urbanized area by 1.7° ± 0.7°, 1.4° ± 0.3°, 1.3° ± 0.3°, and 0.9° ± 0.4°C for JJA, MAM, SON and DJF, respectively. The urban-averaged DTRs decrease by −0.5° ± 0.3°, −0.4° ± 0.2°, and −0.8° ± 0.2°C for MAN, SON, and DJF, respectively, and increase by 0.6° ± 0.3°C for JJA, as a result of more efficient radiation trapping and heat release in the urban canopy layer, which is most efficient and has the largest impact during JJA. The reduction of WS10 is mainly caused by increased surface roughness in urbanized areas, and it is found that the surface wind speed is approximately 37% lower over urbanized areas. Replacing vegetated surfaces by urban surfaces leads to a decrease in available soil moisture, surface evaporation, and surface water vapor for all seasons. The water mixing ratios decrease by ~0.8, ~1.5, ~0.9, and ~0.4 g kg−1 for MAM, JJA, SON, and DJF, respectively. The annual total precipitation increases by approximately 30%, mainly due to the increase in heavy and extreme rain events, implying an increased risk of flooding over the PRD region.

The impact of urbanization on local and regional climate change is a complex scientific issue. Numerous complicated effects (drag effect, shading effect, UHI effect, etc.) are involved in urbanization processes. In this study, the physical and dynamic influences of urbanization due to land-use changes are investigated. Two additional avenues are proposed that may lead to a deeper understanding of the effects of urbanization on climate change. The first is to compute how these changes in anthropogenic heat may impact air pollutants and circulation changes. The other would include a careful analysis of additional physical and dynamic processes, such as the use of a multiple-canopy parameterization.

Acknowledgments

This research was supported by the NSFC (Projects U0833001, 41275018, 41275100), the National Key Basic Research Development Program of China (2010CB428504 and 2010CB428503), the key program of the Natural Science Foundation of Guangdong Province under Grant S2012020011044, and National Key Technologies Research and Development program in the 12th Five Year Plan of China (2012BAH32B03).

REFERENCES

  • Arnfield, A., 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, 126.

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

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

    • Search Google Scholar
    • Export Citation
  • Chen, F., , and Y. Zhang, 2009: On the coupling strength between the land surface and the atmosphere: From viewpoint of surface exchange coefficients. Geophys. Res. Lett., 36, L10404, doi:10.1029/2009GL037980.

    • Search Google Scholar
    • Export Citation
  • Chen, F., and Coauthors, 2011a: The integrated WRF/urban modeling system: Development, evaluation, and applications to urban environmental problems. Int. J. Climatol., 31, 273288, doi:10.1002/joc.2158.

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

    • Search Google Scholar
    • Export Citation
  • Davis, C., , B. Brown, , and R. Bullock, 2006: Object-based verification of precipitation forecasts. Part I: Methodology and application to mesoscale rain areas. Mon. Wea. Rev., 134, 17721784.

    • Search Google Scholar
    • Export Citation
  • Davis, C., and Coauthors, 2008: Prediction of landfalling hurricanes with the Advanced Hurricane WRF model. Mon. Wea. Rev., 136, 19902005.

    • Search Google Scholar
    • Export Citation
  • Dixon, P. G., , and T. L. Mote, 2003: Patterns and causes of Atlanta’s urban heat island initiated precipitation. J. Appl. Meteor., 42, 12731284.

    • Search Google Scholar
    • Export Citation
  • Done, J., , C. Davis, , and M. Weisman, 2004: The next generation of NWP: Explicit forecasts of convection using the Weather Research and Forecasting (WRF) model. Atmos. Sci. Lett., 5, 110117.

    • Search Google Scholar
    • Export Citation
  • Du, Y., , Z. Xie, , Y. Zeng, , Y. Shi, , and J. Wu, 2007: Impact of urban expansion on regional temperature change in the Yangtze River Delta (in Chinese). J. Geogr. Sci., 17, 387398.

    • 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.

    • Search Google Scholar
    • Export Citation
  • Erell, E., , and T. Williamson, 2007: Intra-urban differences in canopy layer air temperature at a mid-latitude city. Int. J. Climatol., 27, 12431255.

    • Search Google Scholar
    • Export Citation
  • Ferguson, G., , and A. Woodbury, 2007: Urban heat island in the subsurface. Geophys. Res. Lett., 34, L23713, doi:10.1029/2007gl032324.

  • Fujibe, F., , N. Yamazaki, , M. Katsuyama, , and K. Kobayashi, 2005: The increasing trend of intense precipitation in Japan based on four-hourly data for a hundred years. SOLA, 1, 4144, doi:10.2151/sola.2005-012.

    • Search Google Scholar
    • Export Citation
  • Goswami, B. N., , M. S. Madhusoodanan, , C. P. Neema, , and D. Sengupta, 2006: A physical mechanism for North Atlantic SST influence on the Indian summer monsoon. Geophys. Res. Lett., 33, L02706, doi:10.1029/2005GL024803.

    • Search Google Scholar
    • Export Citation
  • Grell, G. A., , J. Dudhia, , and D. Stauffer, 1994: A description of the fifth-generation Penn State/NCAR Mesoscale Model (MM5). NCAR Tech. Note NCAR/TN-398+STR, 117 pp.

  • Grell, G. A., , S. E. Peckham, , R. Schmitz, , S. A. McKeen, , G. Frost, , W. C. Skamarock, , and B. Eder, 2005: Fully coupled “online” chemistry within the WRF model. Atmos. Environ., 39, 69576975.

    • 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.

    • Search Google Scholar
    • Export Citation
  • Houghton, J. T., , Y. Ding, , D. J. Griggs, , M. Noguer, , P. J. van der Linden, , X. Dai, , K. Maskell, , and C. A. Johnson, Eds., 2001: Climate Change 2001: The Scientific Basis. Cambridge University Press, 881 pp.

  • Inoue, E., 1963: On the turbulent structure of airflow within crop canopies. J. Meteor. Soc. Japan,41, 317–325.

  • Kain, J., , and J. Fritsch, 1993: Convective parameterization for mesoscale models: The Kain–Fritsch scheme. The Representation of Cumulus Convection in Numerical Models, Meteor.Monogr., No. 46, Amer. Meteor. Soc., 165–170.

  • Kain, J., , S. Weiss, , J. Levit, , M. Baldwin, , and D. Brigh, 2006: Examination of convection-allowing configurations of the WRF model for the prediction of severe convective weather: The SPC/NSSL Spring Program 2004. Wea. Forecasting, 21, 167181.

    • Search Google Scholar
    • Export Citation
  • Kalnay, E., , and M. Cai, 2003: Impacts of urbanization and land-use change on climate. Nature, 423, 528531.

  • Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc.,77, 437–471.

  • Karl, T. R., , and R. W. Knight, 1998: Secular trend of precipitation amount, frequency, and intensity in the United States. Bull. Amer. Meteor. Soc., 79, 231242.

    • Search Google Scholar
    • Export Citation
  • Kaufmann, R., , K. Seto, , A. Schneider, , Z. Liu, , L. Zhou, , and W. Wang, 2007: Climate response to rapid urban growth: Evidence of a human-induced precipitation deficit. J. Climate, 20, 22992306.

    • Search Google Scholar
    • Export Citation
  • 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, 6780.

    • Search Google Scholar
    • Export Citation
  • Kusaka, H., , H. Kando, , 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.

    • Search Google Scholar
    • Export Citation
  • Liang, J., , and S. Wu, 1999: Climatological diagnosis of winter temperature variations in Guangdong (in Chinese). J. Trop. Meteor., 15, 221229.

    • Search Google Scholar
    • Export Citation
  • Liao, J., , X. Wang, , Y. Li, , and B. Xia, 2011: An analysis study of urbanization on precipitation in Guangzhou (in Chinese). J. Meteor. Sci., 31, 384390.

    • Search Google Scholar
    • Export Citation
  • Lin, C.-Y., , W.-C. Chen, , S. C. Liu, , Y. Liou, , G. R. Liu, , and T. H. Lin, 2008: Numerical study of the impact of urbanization on the precipitation over Taiwan. Atmos. Environ., 42, 29342947.

    • Search Google Scholar
    • Export Citation
  • Lin, W., , L. Zhang, , D. Du, , and L. Yang, 2009: Quantification of land use/land cover changes in Pearl River Delta and its impact on regional climate in summer using numerical modeling. Reg. Environ. Change, 9, 7582.

    • Search Google Scholar
    • Export Citation
  • Lin, Y.-L., , R. D. Farley, , and H. D. Orville, 1983: Bulk parameterization of the snow field in a cloud model. J. Climate Appl. Meteor., 22, 10651092.

    • Search Google Scholar
    • Export Citation
  • Liu, X.-H., , Y. Zhang, , K.-M. Olsen, , W.-X. Wang, , B. A. Do, , and G. M. Bridgers, 2010: Responses of future air quality to emission controls over North Carolina, Part I: Model evaluation for current-year simulations. Atmos. Environ., 44, 24432456.

    • Search Google Scholar
    • Export Citation
  • Miao, S., , and F. Chen, 2008: Formation of horizontal convective rolls in urban areas. Atmos. Res., 89, 298304.

  • Miao, S., , F. Chen, , M. LeMone, , M. Tewari, , Q. Li, , and Y. Wang, 2009a: An observational and modeling study of characteristics of urban heat island and boundary layer structures in Beijing. J. Appl. Meteor. Climatol., 48, 484501.

    • Search Google Scholar
    • Export Citation
  • Miao, S., , F. Chen, , Q. Li, , and S. Fan, 2009b: Impacts of urbanization on a summer heavy rainfall in Beijing. Proc. Seventh Int. Conf. on Urban Climate, Yokohama, Japan, Int. Association for Urban Climate, B12-1.

  • Misenis, C., , and Y. Zhang, 2010: An examination of sensitivity of WRF/Chem predictions to physical parameterizations, horizontal grid spacing, and nesting options. Atmos. Res., 97, 315334.

    • 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 (D14), 16 66316 682.

    • Search Google Scholar
    • Export Citation
  • Montaveza, J., , J. Gonzalez-Rouco, , and F. Valero, 2008: A simple model for estimating the maximum intensity of nocturnal urban heat island. Int. J. Climatol., 28, 235242.

    • Search Google Scholar
    • Export Citation
  • Oke, T. T., 1982: The energetic basis of the urban heat island. Quart. J. Roy. Meteor. Soc., 108, 124.

  • Ren, G., , Y. Zhou, , Z. Chu, , J. Zhou, , A. Zhang, , J. Guo, , and X. Liu, 2008: Urbanization effects on observed surface air temperature trends in north China. J. Climate, 21, 13331348.

    • Search Google Scholar
    • Export Citation
  • Rizwan, A., , Y. Dennis, , and C. Liu, 2008: A review on the generation, determination and mitigation of urban heat island. J. Environ. Sci., 20, 120128.

    • Search Google Scholar
    • Export Citation
  • Rozoff, C. M., , W. R. Cotton, , and J. O. Adegoke, 2003: Simulation of St. Louis, Missouri, land use impacts on thunderstorms. J. Appl. Meteor., 42, 716738.

    • Search Google Scholar
    • Export Citation
  • Shem, W., , and M. Shepherd, 2008: On the impact of urbanization on summertime thunderstorms in Atlanta: Two numerical model case studies. Atmos. Res., 92, 172–189, doi:10.1016/j.atmosres.2008.09.013.

    • Search Google Scholar
    • Export Citation
  • Tomita, T., , H. Kusaka, , R. Akiyoshi, , and Y. Imasato, 2007: Thermal and geometric controls on the rate of surface air temperature changes in a medium-sized, midlatitude city. J. Appl. Meteor. Climatol., 46, 241247.

    • Search Google Scholar
    • Export Citation
  • Trier, S., , F. Chen, , K. Manning, , M. A. LeMone, , and C. Davis, 2008: Sensitivity of the simulated PBL and precipitation to land surface conditions for a 12-day warm-season convection period in the central United States. Mon. Wea. Rev., 136, 23212343.

    • Search Google Scholar
    • Export Citation
  • Wang, X. M., , W. S. Lin, , L. M. Yang, , R. R. Deng, , and H. Lin, 2007: A numerical study of influences of urban land-use change on ozone distribution over Pearl River Delta region, China. Tellus, 59B, 633641.

    • Search Google Scholar
    • Export Citation
  • Wang, X. M., , F. Chen, , Z. Wu, , M. Zhang, , M. Tewari, , A. Guenther, , and C. Wiedinmyer, 2009a: Impacts of weather conditions modified by urban expansion on surface ozone over the Pearl River Delta and Yangtze River Delta regions, China. Adv. Atmos. Sci., 26, 962972, doi:10.1007/s00376-009-8001-2.

    • Search Google Scholar
    • Export Citation
  • Wang, X. M., , Z. Wu, , and G. Liang, 2009b: WRF/CHEM modeling of impacts of weather conditions modified by urban expansion on secondary organic aerosol formation over Pearl River Delta. Particuology, 7, 384391.

    • Search Google Scholar
    • Export Citation
  • Wiedinmyer, C., , M. Barlage, , M. Tewari, , and F. Chen, 2012: Meteorological impacts of forest mortality due to insect infestation in Colorado. Earth Interact.,16,doi:10.1175/2011EI419.1.

  • Wu, Z., and Coauthors, 2011: Evaluating the calculation of dry deposition velocities for reactive nitrogen oxides and ozone from two community dry deposition models for a temperate deciduous forest. Atmos. Environ., 45, 26632674.

    • Search Google Scholar
    • Export Citation
  • Xiong, Y., , Z. Zhang, , and F. Chen, 2010: Estimation of urbanization effect on climatic warming over the recent 30 years in Guangzhou, south China. Proc. 18th Int. Conf. on Geoinformatics, Beijing, China, IEEE, 1–6.

  • Zhang, C., , F. Chen, , S. Miao, , Q. Li, , A. Xia, , and C. 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, H., , N. Sato, , T. Izumi, , K. Hanaki, , and T. Aramaki, 2008: Modified RAMS-urban canopy model for heat island simulation in Chongqing, China. J. Appl. Meteor. Climatol., 47, 509524.

    • Search Google Scholar
    • Export Citation
  • Zhang, N., , Z. Gao, , X. Wang, , and Y. Chen, 2010a: Modeling the impact of urbanization on the local and regional climate in Yangtze River Delta, China. Theor. Appl. Climatol., 102, 331342.

    • Search Google Scholar
    • Export Citation
  • Zhang, Y., , X. Wen, , and C. Jang, 2010b: Simulating chemistry–aerosol–cloud–radiation–climate feedback over the continental U.S. using online-coupled Weather Research Forecasting Model with chemistry (WRF/Chem). Atmos. Environ., 44, 35683582.

    • Search Google Scholar
    • Export Citation
Save