بررسی پارامترسازی عمق لایه پایدار شبانه و تاثیر آن در آلودگی هوای یک منطقه شهری با توپوگرافی پیچیده (تهران)

Mixing height of the atmospheric boundary depends on the vertical variation of temperature in the atmosphere which includes temperature inversion (including surface and elevated inversions) that has a significant effect on air quality. The mixing height like some other meteorological variables has diurnal variations. The reason for that is its dependence on some other basic meteorological parameters e.g. surface turbulent fluxes (momentum and heat fluxes), wind speed, temperature stratification. In urban area surface roughness and topography also affect the mixing height. Since mixing height cannot be observed by standard measurements, it is not routinely measured in meteorological stations especially for urban area. Therefore it should be estimated indirectly from vertical profile of some meteorological fields such as wind, potential temperature or must be parameterized. Due to the importance of mixing height for air pollution calculations, numerous algorithms were prepared to estimate it. It is noticeable that most of these algorithms are practical for rural area. Hence applying these algorithms for urban area may lead to large errors; therefore they are needed to be validated for each terrain individually. Previous studies have shown that for day time, results of these algorithms for urban area have good agreement with observations but at night they lead to large errors. An increased error at night is due to the error in heat flux estimation in urban area compare to rural area. Over urban areas nocturnal heat flux includes two different parts. One is a downward heat flux due to the surface cooling (similar to what happened for rural area) and the other is an upward heat flux due to the urban effect (which is not included in rural models). Recently many efforts have been done to evaluate applicability of different models and parameterization for urban areas. They were performed to estimate stable boundary height using two methods namely using surface fluxes or Richardson number (Vicker and Mahrt, 2004). Tehran, the capital of Iran (35° 42? N, 51° 25?E) is one of the the polluted city in the world. It suffers from poor air quality most of the year. E.g. Zawar-Reza et al (2008) indicated that analysis of data from permanent air quality monitoring stations showed that Tehran regularly exceeds the WHO guideline of 50µg/m3 for PM10. There are inter-annual differences in percentage of times that this condition occurs, but in general this occurs more than 70% of times. Moreover than manmade sources, Tehran location has significant influence on its air quality. It has an area of 700 km2 and is situated in a semi-enclosed basin in the south of Alborz mountain range, down to the low lying flat terrain. The mountainous nature of Tehran has a dominating control on low level wind climatology and also air quality. Tehran has an unusual location and an average low annual rainfall of approximately 230 mm. Its precipitation is mostly in autumn and winter months. Local precipitation is absent for 6 months of the year on the low lying areas that may lead to some air pollution episodes. Due to the high elevation (approximately 1140 m), aridity and latitude, the city experiences four seasons. Climate can be extremely hot in the summer (with midday temperatures ranging between 30 to 40°C), and cold in winter when night time temperatures can be below the freezing point. Another characteristic of this city is its heat island effect, for e.g. horizontal temperature difference between different parts of that varies from 2-4 °C in different seasons. The stability parameter (z/L, z=15m) range varies from -1 to 1(Pegahfar et al., 2006). Hence Tehran experiences a larger stablility range for weather condition compare with Moriwaki et al. (2002) results for Tokyo ([-0.2, 0.2]) and Haizhen et al. (2006) results ([-0.1, 0.1]) for idealized simulations in numerical experiments on the east coast of Vancouver Island. The static stability may be affected by mountains surroundings the city that also prevent pollution ventilation. This stable stratification brings worth air quality for this area. Another characteristic of Tehran is its inhomogeneous surface that gives large value of surface roughness in various directions. North-easterly wind experience the most roughness ( ) comparing to southerly wind ( ) which shows the complexity of the terrain ( represent roughness length and zero displacement height respectively). This non homogenous surface may affect horizontal mixing height which has a significant effect on air pollution. During this study according to the seasonal variations of the wind profile and turbulent structure, mixing height varies from 554 to1739m in day time periods, and stable boundary height varies from 17 to 500m at night. In this research data from a Radiosond for the 9 first months of 2009 at 0000 UTC were used to calculate stable boundary layer height and to validate some parameterizations applied in dispersion models for such urban area with complex topography. Radiosond measures vertical profile of wind speed and direction, temperature, pressure, humidity and geo-potential height with 2 second time step. Also data from a surface station at the Geophysics Institute from the University of Tehran were used to calculate PM10 and CO concentrations in this period. Diurnal variations of CO and PM10 demonstrate that their pattern have two maximum which is affected by traffic peak and transition times. Transition time effect on pollution concentrations is due to the sharp variations of meteorological fields such as surface fluxes, wind speed and direction and temperature. In this paper we demonstrated that PM10 concentrations changes linearly versus stable boundary layer height. According to the importance of stable boundary layer height in dispersion models we decided to find the best parameterization scheme for Tehran. Hence 10 parameterizations schemes that were more applicable for urban area with complex topography were validated for stable boundary layer height estimation. These relations were defined using various meteorological parameters such as friction velocity, Monin-Obukhov length, Brunt-Vaisala frequency and Coriolis parameter. Since some of the meteorological parameters were not measured directly, we used some relationships to calculate them, e.g. , and for friction velocity, for Monin-Obukhov length, and ( ) for Brunt-Vaisala frequency. The first parameterization scheme was introduced by Zilitinkevich (1972): (1) The best linear fit analysis was achieved using that is more than those obtained by other researches for rural area. This relation estimates the stable boundary layer height for Tehran with . Benkly and Scholman (1979) calculated stable boundary layer height using wind speed at 10m height and a constant equals to 125. This relation for Tehran leads to over estimation that may be the cause of complicate wind structure for this terrain and it's seasonally variations. It is noticeable that this formula for rural area produces up to 264m error (Baklanov, 2001). Other parameterization such as , , , , , , defined by Venkatram (1980), Arya (1981, after Zilitinkevich, 1972), Arya (1981), Niewstadt (1981), Mahrt (1082), Niewstadt (1984), Niewstadt (1984) lead to , , , , , , respectively. The last parameterization for stable boundary layer estimation was defined by Zilitinkevich and Miranov (1996): (2) In this relation that was derived from equation of turbulent kinetic energy budget, 5 length scales were included. Three lengths for rotation effects (including Ekman layer and free atmosphere) and two lengths for Monin-Obukhov length and non rotating fluid were used. Recently this relation was applied in numerous boundary layer algorithms and air pollution models. Performing least square analysis indicated that the results of the last parameterization have good agreements with observations. It produced and for Tehran. Results indicated that applying relations defined according to the individual meteorological parameters may lead to less accuracy for Tehran. In this research Zilitinkevich and Miranov (1996) parameterization compare to the other relations, includes more physical and thermo dynamical parameters for surface and free atmosphere, so it can estimate the stable boundary layer height with more consistency to the observations for urban areas with complex topography