Responses of the land-surface process and its parameters over the natural vegetation underlying surface of the middle of Gansu in loess plateau to precipitation fluctuation

The Loess Plateau is well known as a specific region sensitive to global climate change, and thus its land-surface process is significantly influenced by climatic fluctuation. Up to now, the land-surface physical process over the Loess Plateau has been basically understood under a specific climatic condition, but the dynamic variation regularity of land-surface process in different climatic states is still lacking in its knowledge. Utilizing the continuous five-year data collected at the Semi-Arid Climate and Environment Observatory of Lanzhou University (SACOL) supported by the program 'the Loess Plateau Land-surface Process Experiment (LOPEX)', in this paper, we systematically analyze the regularity of responses of factors including land-surface water and energy budget as well as surface albedo and roughness to climatic fluctuation over the natural vegetation surface of Loess Plateau. The results show that the land-surface process and its relevant parameters are very sensitive to the fluctuations of annual precipitation. Besides, they respond markedly to annual rainfall amount and are also closely related to the nature of rainfall. Soil moisture fluctuales not fully with total amount of rainfall but rises with the increase of effective precipitation with which land-surface water and energy budget also vary. In addition, the vertical sensible heat advection goes up with rainfall increasing, while the trend of surface energy imbalance is opposite. The averaged albedo in summer half-year comes down with the increase of the ratio of effective precipitation to the climatic average, but yearly averaged albedo is evidently affected by the snow-accumulating winter time and rises with the increase of snow-accumulating hours. Soil thermal conductivity and surface roughness both increase with annual effective precipitation increasing to the climatic average, but they are more sensitive to the precipitation fluctuation under a low precipitation condition than under the normal precipitation condition. However, land-surface process parameters over the natural vegetable surface of Loess Plateau basically fluctuate between those of deserts and farmland, and have not yet broken the climatic restrictions. Fitting relationship is used to quantitatively reflect the regularity of response of the parameters to rainfall fluctuations so as to evidently eliminate possible bias brought to numerical models and microclimatic analysis by the previous fixed and un-dynamic changeable land-surface parameters.