The Study On Turbulence Characteristics And Methods For Estimating The Fluxes Over A Typical Steppe Prairie In China North

The turbulence characteristic of atmospheric boundary layer is one of the basic research contents of micro-meteorology. As one of the main terrestrial ecosystems, the grassland have been facing the desertification, continuous drought, water shortage and a series of major environmental issues. The study on turbulence characteristics and the methods of turbulence fluxes for the underlying surface of the typical grassland ecological system in Northern China would contribute to understand the material and energy cycle, and have important conductive role for the further research on global change.In this thesis, the micro-meteorology data obtained from the experiment, based on the underlying surface of typical grassland at Hebei Guyuan National Grassland Ecosystem Research Station as the research object, was used to analyze the surface turbulent motion characteristics and the methods to measure turbulence fluxes for underlying surface of the typical grassland.The main conclusions are as follows:(1)Micrometeorology and turbulence characteristics.During June to August, the prevailing wind is from southwest at Guyuan station. The average wind speed is2m/s. Temperature difference between day and night is approximately20℃. The averaged relative humidity is64%, easily condensate in the morning. The typical daily meteorological elements profiles follow a logarithmic law. The atmospheric stability is distributed within±0.5. The averaged friction velocity has the diurnal variation rule. The turbulence intensity meets the general rules of Iu>Iv>Iw. The standard deviation of wind speed meets1/3power law, and w direction fits best. The standard deviation of temperature satisfies-1/3law. Wind power spectral inertial meets-2/3rule, which matches fm(w)>fm(v)>fm(u); Co-spectrum between vertical wind speed and temperature meets-4/3law; under the stable conditions, the power spectrum and co-spectral peaks both exhibit a tendency to move towards the higher frequency region. The three components of the wind power spectrum match Sn (v)/Sn (u)≈4/3, and Sn (w)/Sn (u)<4/3. The constant frequency of the horizontal direction is lower than that of the vertical direction. The average scale of turbulence is7mm.(2) Turbulent fluxNet radiation, sensible heat, latent heat and soil heat flux all showed obviously diurnal variation; latent heat, sensible heat and soil heat flux proportion was64%,20%and14%, respectively. The energy closure rate is91.3%during the experiment time, and the energy closure error is9.7%, with an R2of0.88. The averaged CO2flux was-0.145mg/m2/s, expressed as the sink of CO2. Under the unstable conditions, it needed about thirty minutes averagely to capture all the basic flux; under the stable conditions, sampling time needed to be extended to capture all the basic flux. Under the unstable and general stable stratification conditions, the source of eddy covariance was distributed in a relatively small measuring region, but it will exceed the measurement area under extremely stable conditions. (3) Methods for esmating the fluxes and impact factorsThe fit slope and R2between the results got from Eddypro and Edire are both higer than0.97. Considering the eddy covariance measurements as a reference, the results of the aerodynamics measurement for direct or iterative methods were poor; no transposition Bowen ratio method had a little higher R2; the transposition Bowen ratio method and BLS450large aperture scintillometer had much better results. BLS450had the highest R2, with the larger source area than the eddy covariance system, and the spatial representation is much better.(4) Large aperture scintillometer impact factorsThe meteorological conditions could lead to the received signal weakened, and the water vapor absorption and flashing saturation would affect Cn2calculations. Compared with the single channel scintillometer, BLS900had lower measurement error. There would have water vapor absorption effect when the large aperture scintillometer used the880nm or940nm near-infrared light. Both the actual signal power spectrum analysis and the analysis of large aperture scintillometer theoretical spectra met-8/3slope at the low-frequency part, constant at intermediate frequency, and-12/3slope at high-frequency portion. The part which is under switching frequency is the contribution of water vapor absorption, needed to be filtered. The constant of power spectrum increased with the decreasing wind speed, but the corner frequency decreased when the wind speed decreased. The lateral wind change would alter the spectral shape of the spectrum but did not lead to reduced spectrum area. When using fixed threshold filter to filter water vapor absorption, low wind conditions could cause the results underestimated. In actual observations, the impact of water vapor absorption is relatively higher in the evening, with the greatest impact up to80%which needed the water vapor absorption correction. Both software filter and the related dual-channel signals are two effective methods to remove the contribution produced by water vapor absorption. Bowen ratio coefficient at Guyuan station is typically less than0.3; meanwhile, Ct2would be affected by Cq2and Crq, so Bowen ratio must be used to amend the results, and the impact of buoyancy on the stability should be taken into consideration. The correlation between CT2obtained from corrected large aperture scintillometer and the CT2calculated by the acquisition function method, similarity theory method or spectrum method performed excellent, with an R2larger than0.88. Because of its strong spatial representation, large aperture scintillometer is more applicable for larger scale sensible heat flux measurement.