| The basic characteristics and daily variations of temperature, relatively humidity, and wind speed, wind direction were analyzed, using the observational data collected at Semi-Arid Climate Observatory and Laboratory (SACOL) station. At meantime, using Weather Research and Forecast model (WRF), the regional atmospheric factors were simulated, and comparison between the simulation result and observations was made. The control test with the WRF model was taken under the condition of different inputs of underlying surface (mainly topography and land types), and the effects of changing underlying surface on basic atmospheric elements were compared. The main conclusions are as follows:(1) The annual mean temperature has obvious daily and monthly variations during the year of2007-2009, and the maximum value appeared at the16:00BJT, the minimum at8:00, and the highest temperature is in July and the lowest is in January. The diurnal variation of the annual average relative humidity is reverse to temperature. Both the seasonal average temperature and relative humidity represent the obvious seasonal difference, the temperature in summer is the highest, followed by that in spring and autumn, and in winter is lowest. The relative humidity in autumn is the highest, then in summer and winter, and spring. In2007and2008, the highest relative humidity appeared in September, and the lowest was in June and May separately. The highest RH appeared in October and the lowest in May in2009.(2) With WRF model, simulations of six cases in different times were made. The method of correlation coefficient, RMSE, ME and MB were used to analyze the differences of simulated result and observational data, the results show that the WRF model has the ability to simulate the underlying surface meteorological elements in the area. Among the results of simulation, the simulated temperature is the best, and followed by the relative humidity. There are differences among the simulation effect in different times, but the correlation coefficients of all the simulation results and observations are significant at level0.05. The result suggests that the WRF results are credible.(3) Designed several sensitivity tests, mainly focusing on the land types and topography height, containing that change the land types of research areas to Urban and Built-up Land, Dryland Cropland and Pasture, Barren or Sparsely’Vegetated; change topography height to1400m,3600m. By comparing the sensitivity tests and the control test know that:all tests can be divided to three categories, the first type called T1, contain CTL1(HGT=1400m), CTL3+CTL1(LU_INDEX=1and HGT=1400m), CTL4+CTL1(LU_INDEX=2and HGT=1400m) and CTL5+CTL1(LU_INDEX=19and HGT=1400); the second type is T2, include CTL2(HGT=3600), CTL3+CTL2(LU_INDEX=1and HGT=3600m), CTL4+CTL2(LU_INDEX=2and HGT=3600m) and CTL5+CTL2(LU_INDEX=19and HGT=3600); the third type is T3, cover the control test, CTL3(LU_INDEX=1), CTL4(LU_INDEX=2) and CTL5(LU_INDEX=19). In the sensitivity tests, if the underlying surface is covering by Barren or Sparsely Vegetated, meanly containing CTL5, the temperature is higher than the other tests; then if the underlying surface is covering by Dryland Cropland and Pasture, that is containing CTL4, the temperature is lower than the other tests. Change of land use type is almost not change the actual distribution of meteorological elements, this suggests that the variation of meteorological elements cause by changing land use type is very small; moreover, there have big differences between the simulation results of changing topography height sensitivity tests and the control test, that is to say, the extent of changing topography height in this paper has greatly influenced in meteorological elements.(4) Design another set of sensitivity tests, including increasing or decreasing200m to all terrain height, changing the land type from Urban and Built-up Land to Grassland, or from Grassland to Urban and Built-up land. Then comparative analysis the result of the sensitivity tests and control test, the results are as follows:The land type modification in the CTL3sensitivity test didn’t cover SACOL station, so the magnitude and variation of atmospheric elements is almost consistent with the control test. But in the CTL4, the Urban and Built-up land instead of SACOL’s land type, the simulation result shows that SACOL’s temperature is lower or higher than the control test in the lowest or highest value. The reason is that changing the land type is virtually changing the soil heat capacity, Urban and Built-up’s heat capacity is small, so warming faster under the radiation from the sun during the day, and on the contrary night heat faster instead; so is the variation of relative humidity; and the wind speed is small.Because the spatial distribution of meteorological elements is mainly affected by topography ups and downs. It is necessary to get rid of the influence of topography ups and downs. The spatial distribution of atmospheric elements is less affected by the the topography height modification, but its magnitude significantly. Increasing topography height, the temperature will be decreased; relative humidity and wind speed will be increased slightly. And in reverse, the temperature increased with terrain height reduced and relative humidity and wind speed will be decreased slightly. By compared CTL3with CTL4, the influences of modification land type on simulation are related to land type, and also related to the size of modified region:In CTL3, with changing Urban and Built-up land to Grassland, the temperature decreased slightly, and relative humidity increased, because the modification region small is, so the scope of the influence also small, just limited to the modification areas, and almost didn’t influence wind speed and the spatial distribution of wind direction; In CTL4, with changing Grassland to Urban and Built-up land, the relative humidity and wind speed decreased, temperature increased, because the modified areas is enlarged, then the scope of influence also big. |
PDF Full Text Request