In this study, the effect of anthropogenic heat release (AHR) on meteorological variables and atmospheric diffusion capability and implication for haze pollution in the Beijing-Tianjin-Hebei region in January 2013 were investigated by using Weather Research and Forecasting (WRF) model with an urban canopy model (UCM) and an AHR scheme. The comparison with observation demonstrated the WRF/UCM model taking AHR into account apparently improved meteorological prediction, especially for surface air temperature at 2 m (T2). The model also exhibited a better performance for planetary boundary layer (PBL) height. This study revealed that AHR from cities exerted a significant impact on meteorology by generally increasing surface air temperature and wind speed, decreasing relative humidity, and elevating PBL height and near surface turbulent kinetic energy (TKE), which could consequently reduce surface pollutant concentration and mitigate haze pollution by enhancing atmospheric instability and turbulent mixing and reducing aerosol hygroscopic growth.
In recent decades, large scale urbanization has developed rapidly, resulting in significant changes in local/regional environment and climate. Urbanization brings about changes in underlying surface and releases of anthropogenic heat and anthropogenic pollutants into the atmosphere, consequently altering urban air quality and boundary layer meteorology. Urban heat island (UHI) can be generated due to changes in land use (including land types and characteristics) and land surface processes during urbanization [
Anthropogenic heat release (AHR) is produced by human activities and spreads to surrounding atmosphere. AHR is generated from many kinds of sources, with major sources from human metabolism, vehicle, and energy consumption in buildings including electricity and heating fuels [
The anthropogenic heat flux depends on climate, population density, and intensity of industrial and commercial activities [
Because of the rapid economic and industrial development in China, urbanization has been accelerating over the past 30 years which draws increasing attentions of scientific community. Feng et al. [
Anthropogenic heat release in China reaches maximum in winter due to increasing energy consumption; meanwhile, haze pollution occurs most frequently in winter due to the combined effects of larger emission amount and stronger atmospheric stability. So, it is valuable to explore the impact of urban AHR on meteorology and turbulent diffusion, which plays an important role in haze formation and evolution. This study aims to investigate the effect of AHR on meteorology in the Beijing-Tianjin-Hebei region in January 2013 by using the WRF/UCM together with AHR parameterization. The new aspect in this study is to further explore the changes induced by AHR in atmospheric diffusion and mixing capacity during haze event and its implication for haze pollution.
The meteorological model used in this study is the Weather Research and Forecasting (WRF) model version 3.5 with the ARW dynamic core [
(a) The two nested domains (d01, d02) used for this study, (b) terrain height and cross section location denoted by red line AB in d02, (c) land use types and locations of observation sites in d02 (1: Beijing, 2: Baodi, 3: Tangshan, 4: Tanggu, and 5: Tianjin).
As to AHR emission in China, Feng et al. [
(a) Diurnal hourly scaling factors of AHR in Beijing. (b) Spatial distributions of AHR at 12:00 LST (units: W m−2 for AHR).
To investigate the effect of AHR on meteorological variables, two sensitivity model simulations are conducted. One is the base case without AHR; the other is case 1 by considering AHR in urban areas. The difference between the two cases reflects the AHR-induced meteorological changes. The observed meteorological data in January 2013 is derived from China Meteorological Administration (
The day to day variations of the observed and simulated meteorological variables in the two cases for January 2013 in Beijing are presented in Figures
Comparisons of simulated and observed daily mean surface meteorological variables in Beijing in January 2013. (a) Air pressure, (b) 2 m air temperature (T2), (c) 10 m wind speed (WS10), (d) 2 m relative humidity (RH2), and (e) PM2.5 (base case: black dotted line, case 1: red dotted line, and observation: black solid line).
The statistical comparison for T2, WS10, and RH2 at the 5 sites (Figure
Statistics for the observed and simulated daily mean wind speed at 10 m (WS10), temperature at 2 m (T2), and relative humidity at 2 m (RH2) in Beijing (BJ), Tianjin (TJ), Tangshan (TS), Tanggu (TG), and Baodi (BD) during January 2013. COR: correlation coefficient; MB: mean bias; NMB: normalized mean bias; bc: base case; c1: case 1.
City | T2 | WS10 | RH2 | ||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Obs (°C) | COR | MB (°C) | NMB (%) | Obs (m s−1) | COR | MB (m s−1) | NMB (%) | Obs (%) | COR | MB (%) | NMB (%) | ||||||||||
bc | c1 | bc | c1 | bc | c1 | bc | c1 | bc | c1 | bc | c1 | bc | c1 | bc | c1 | bc | c1 | ||||
BJ | −4.7 | 0.84 | 0.84 | −2.5 | −0.7 | 59 | 19 | 1.8 | 0.71 | 0.68 | −0.1 | −0.04 | −7 | −2 | 60.4 | 0.82 | 0.83 | 4.3 | −4.0 | 11 | −3 |
TJ | −4.7 | 0.90 | 0.90 | −1.7 | 0.1 | 36 | −1 | 2.2 | 0.74 | 0.72 | −0.5 | −0.5 | −23 | −22 | 67.1 | 0.64 | 0.66 | 4.5 | −3.8 | 7 | −6 |
TS | −6.9 | 0.94 | 0.92 | −0.4 | 0.9 | 6 | −12 | 2.1 | 0.70 | 0.69 | −0.2 | −0.1 | −8 | −6 | 74.6 | 0.74 | 0.73 | −8.6 | −14.6 | −12 | −20 |
TG | −4.0 | 0.87 | 0.87 | −1.7 | −0.3 | 42 | 7 | 2.3 | 0.74 | 0.70 | 0.2 | 0.3 | 9 | 13 | 65.9 | 0.86 | 0.87 | 4.4 | −2.4 | 7 | −4 |
BD | −5.8 | 0.90 | 0.88 | −1.8 | −0.6 | 32 | 10 | 1.6 | 0.80 | 0.77 | 0.3 | 0.2 | 20 | 14 | 66.6 | 0.65 | 0.66 | 3.1 | −3.5 | 5 | −5 |
All | −5.2 | 0.84 | 0.85 | −1.6 | −0.1 | 32 | 3 | 2.0 | 0.69 | 0.68 | −0.1 | −0.02 | −3 | −1 | 66.9 | 0.72 | 0.73 | 1.5 | −5.7 | 3 | −8 |
The above comparison and statistics indicate that the meteorological prediction can be significantly improved by considering AHR in urban areas of north China, especially for near surface air temperature, suggesting the importance to incorporate AHR in weather/climate models to represent meteorology and human activity more realistically.
Figure
AHR-induced differences in monthly mean meteorological variables. (a) 2 m air temperature (T2) with wind vector, (b) 10 m wind speed (WS10), (c) 2 m relative humidity (RH2), (d) planetary boundary layer (PBL) height, (e) sensible heat flux, and (f) latent heat flux (units: °C for T2, m s−1 for wind speed, % for RH2, m for PBL height, and W m−2 for sensible and latent heat flux).
AHR led to an increase of wind speed (Figure
The above analysis demonstrates that AHR significantly modifies surface energy balance and results in considerable changes in the distribution and magnitude of meteorological variables, which could consequently affect atmospheric diffusion and haze pollution.
Given the large impact of AHR on monthly mean meteorology discussed above, it is interesting to explore the effect of AHR on meteorology during haze episode and a severe haze episode was selected. Figure
The base case simulated daily mean surface wind field and relative humidity in the Beijing-Tianjin-Hebei region on (a) 11 and (b) 12 January 2013 (units: m s−1 for wind speed, % for relative humidity).
In terms of daily mean (figure not shown), AHR consistently caused an increase in T2 and a decrease in RH2 in Beijing and Tianjin on 12 January, with maximums up to 2.5°C and 13.0%, respectively. Wind speed generally increased in most urban areas of Beijing and Tianjin but decrease also occurred in some areas south and west of Beijing city due to interaction between background wind and AHR-induced wind. AHR caused a maximum PBL height increase of 210 m in downtown Beijing and Tianjin, well corresponding to the areas of maximum temperature increase.
Figure
AHR-induced differences in T2 and wind vector (a, b) and WS10 (c, d) during daytime (a, c) and nighttime (b, d) on 12 January 2013 (units: °C for T2, m s−1 for wind speed).
PBL height variation exerts a direct impact on distribution and level of surface air pollutant through mixing and dilution effect [
Figure
Model simulated hourly variation of planetary boundary layer (PBL) height in Beijing on 11–13 January 2013 (base case: black solid line, case 1: red solid line).
The vertical cross sections of the simulated TKE and air temperature along 39.9°N (Figure
Latitude-altitude cross section of air temperature (contours) and turbulent kinetic energy (TKE) (shaded) at 14:00 (a, c) and 02:00 (b, d) LST on 12 January 2013 along 39.9°N in base case (a, b) and case 1 (c, d) (units: °C for air temperature, m2 s−2 for TKE).
The above results suggest the AHR-induced changes in meteorology and turbulence activity are generally favorable for pollutant diffusion and mixing and for weakening of aerosol hygroscopic growth, which could mitigate haze pollution in the study domain.
In this paper, a scheme for anthropogenic heat release was incorporated into WRF/UCM and applied to investigate the effects of AHR on meteorological variables, diffusion, and mixing capability, which has important implication for haze pollution in the Beijing-Tianjin-Hebei region. By taking AHR into account, meteorological prediction, especially for surface air temperature, had been improved significantly, and the model performed better for PBL height, which was a key factor controlling pollutant diffusion. In terms of monthly mean, AHR tended to increase near surface air temperature and wind speed by 2.1°C and 0.25 m s−1, to decrease relative humidity by 9.0% in the cities of Beijing and Tianjin. For the severest haze episode on 12 January, AHR induced the increases in daily mean T2, WS10, and RH2 in Beijing city of 2.5°C, 0.2 m s−1, and −13.0%, respectively. It was noteworthy that, during haze episode, PBL height decreased to about 400 m, which was quite unfavorable for vertical diffusion. AHR tended to increase PBL height by about 44% during the daytime, suggesting an enhanced atmospheric instability and vertical mixing. Near the surface, TKE was much larger during the daytime than during the nighttime, with a maximum up to 0.32 m2 s−2. AHR led to TKE increase throughout the day, with percent change being about 50% and 33% during the daytime and nighttime, respectively. The results from this study demonstrated the significant effect of AHR on meteorological variables by increasing surface air temperature and wind speed, decreasing relative humidity, as well as enhancing PBL height and TKE, which were mainly favorable for pollutant diffusion and weakening of aerosol hygroscopic growth and could consequently lead to a decrease of surface pollutant level and an increase of visibility. We are aware that pollutants are released together with anthropogenic heat during energy consumption; this study demonstrated that the AHR induced changes in meteorology and turbulence were generally favorable for mitigation of haze pollution, which were rarely considered in previous air quality modeling. Besides, by considering AHR, the model represented urban meteorology and turbulent diffusion more realistically and accurately, suggesting the necessity to couple this process into weather/climate model.
The authors declare that there is no conflict of interests regarding the publication of this article.
This study was supported by the National Natural Science Foundation of China (no. 41375151) and the Jiangsu Collaborative Innovation Center for Climate Change.