| Chinese loess plateau’s land surface process,which is unique over global, plays a veryimportant role in the environment of our country zoology pattern formation and the East AsiaClimate and atmospheric circulation. Previous workers have done some preliminary analysisof land surface process over the area, but understanding is also very limited. Therefore, on onehand, firstly instrumental accuracy and observational error were analyzed by using dataobserved parallely in Dingxi comprehensive observation station during2009in LOPEX;secondly the influences of climate conditions on land surface water in Gansu Loess Plateau areanalyzed by using the historical observation data from Semi-Arid Climate and EnvironmentObservatory (SACOL) in LOPEX. Observation provides reliable data for land surface processstudy, but often the length and continuity of its time can not meet the demand, so land surfacemodel plays an irreplaceable role in the land surface process study. On the other hand, firstlyEvaluate land surface models CLM, Mosaic, Noah (which is driven by GLDAS atmosphericforcing data respectively) and CABLE (which is driven by NCC by data driven) in semi-aridareas using the observation data from Semi-Arid Climate and Environment Observatory(SACOL); secondly, time variation and spatial distribution characteristics of loess plateau aresimulated by CLM. The main conclusions are as follows:(1) Instrumental accuracy and observational error in “LOPEX”: In most cases, thedeviation of wind speed (by A301P), temperature (by HMP45D) and humidity (by HMP45D)were no more than±0.10m/s,±0.16oC and±1.00%. Soil temperature was measured by107and109, whose debiation were no more than±0.1oC and±0.03oC respectively. And thedebiation of soil heat flux (by HFP01), sensible heat, latent heat and momentum flux (byCSAT3+Li7500) were less than±11W/m2,±8W/m2,±8W/m2and±0.02kg/(ms2), respectively.As to some meteorological element, its data were consistent with each other, whether it wasmeasured by the same type or different types of instruments. The results can basically meet therequirements of land surface observations.(2) Characteristics of land surface water: In dry years, annual trends of Land surfaceevapotranspiration and precipitation are very consistent, and evapotranspiration is most sensitive to precipitation; but in wet years, changes of evapotranspiration and precipitation areconsistent only in dry seasons. Land surface evapotranspiration shows strong positivecorrelation with temperature. That is, high temperature, large evapotranspiration and lowtemperature, small evapotranspiration. However, the combined effect of precipitation andtemperature make greater impact on evapotranspiration than precipitation or temperatureaffects evapotranspiration separately. Among this four years, land surface water imbalance isvery remarkable in2007, and relative value of annual total imbalance can arrive at24.8%.It’spossible that precipitation infiltrates into soil rather than evaporates rapidly. As regards soilmoisture, the upper soil moisture keeps low during dry season; with rainfall increasing insummer, the upper soil moisture increases rapidly and keeps high throughout rainy seasonuntil late autumn; whereas year changes of the deep soil moisture are relatively flat. Soilmoisture is seen to be more affected by precipitation in summer half year; but the combinedeffect of precipitation and temperature make greater impact on soil moisture in winter halfyear.(3) Evaluate simulation data: The two kinds of models can well simulate the annual andseasonal cycle of the fluxes between land surface and atmosphere in this region. But there arestill differences among results simulated by different models, which is because that differentphysical process and soil layers, or different forcing data. By comparing the performance ofCLM, Mosaic and Noah, we find that in addition to the simulation of reflection radiation,CLM is after Noah; for the simulations of the ground surface temperature, ground emittedlongwave radiation, net radiation, sensible heat and latent heat, CLM is better than Mosaic andNoah. Compared with above three models, CABLE is close to observation except for thesimulation of sensible heat. As several physical quantities of the processing method in CABLEand CLM is similar, different simulation results maybe due to different forcing data.Additionally, the simulation of winter sensible heat by CABLE is wrong, which may be soilfreezing process of CABLE is described inaccurately.(4) Characteristics of land surface process simulated by CLM: For the interannualvariations of surface fluxes, the climatic value of sensible heat flux is almost two times of thatof latent heat flux, so sensible heat flux plays a leading role in the area of surface heat balance.The interannual changes of spring, summer sensible heat flux are similar, that is the sensibleheat flux has been small until the end of nineteen90’s; the autumn sensible heat flux has noobvious regularity; the winter sensible heat flux is relatively large in the beginning of nineteen80’s, and has been small from the metaphase of nineteen80’s. For the Spatial variation ofsurface fluxes, latent heat flux from southeast to northwest, decreases with increase in latitude,and the sensible heat flux from southwest to northeast increases with increase in latitude; inthe other hand, latent heat flux increases with longitude increasing in the east of105.5oE, and the sensible heat flux decreases with longitude increasing the east of108.5oE. For therelationship between precipitation and surface fluxes, the spatial correlation is more consistentthan interannual correlation. |
PDF Full Text Request