Spatial Distribution Of Optimal Coverages For Two Plants Across Precipitation Gradient Of The Centre Loess Plateau

The Loess Plateau of China has severe soil erosion and fragile ecology system.Vegetative recovery is an effective method to control soil erosion and improve ecologicalenvironment. Plant coverage is different from climatic conditions. In the Loess Plateau, theannual precipitation ranges from almost200mm in the northwest to700mm in the southeast,while soils are sandier in texture in the northwest and gradually become more clayey in thesoutheast. The vegetation type changes from forest through forest-steppe to typical-steppe,desert-steppe, and then to steppe-desert zones. According to the characteristics of theprecipitation gradient, it is very important to determine the optimal plant coverage at differentsoil-vegetation-climate zones for accelerating vegetation recovery in the Loess Plateau.The Biome-BGC model uses general stand information and daily meteorological data tosimulate energy, carbon, water and nitrogen cycling. But the simple empirical relationships inwater sub-model of the Biome-BGC are not capable of simulating the water flow in theloessial soils. We compared the simulated evapotranspiration and biomass of Robiniapseudoacacia, Hippophae rhamnoides and Onobrychis viciaefolia with field data from theliterature for these species at three sites (Guyuan, Ziwuling and Changwu) to evaluate theapplicability of the modified Biome-BGC model in the study area. The assessed Biome-BGCmodel was used to simulate the spatial distribution characteristics of NPP (net primaryproductivity)、GPP (gross primary productivity)、ET (evapotranspiration) and LAI (leaf areaindex) for Robinia pseudoacacia and Hippophae rhamnoides to determine the optimal plantcoverages. The main results are as follows:1. Six published PTFs for predicting saturated hydraulic conductivity have beenevaluated and optimized using the measured soil data from the Loess Plateau. The improvedSaxton method yielded the best predictions and could be used to estimate Ksvalues forloessial soils in the Loess Plateau of China.2. We compared the field observation data and simulated evapotranspiration and biomassof Robinia pseudoacacia, Hippophae rhamnoides and Onobrychis viciaefolia at three sites(Guyuan, Ziwuling and Changwu). The modified Biome-BGC model was capable ofsimulating the NPP and ET dynamics in the study area and could be a useful tool for simulating the water and carbon cycling.3. The NPP, GPP, ET and plant coverage for Robinia pseudoacacia and Hippophaerhamnoides at three sites along the precipitation gradient were significantly different. Theydecreased gradually from Luochuan, Wuqi to Haiyuan in response to the decreasingprecipitation. At the same site, NPP, GPP and ET for Robinia pseudoacacia were greater thanthose for Hippophae rhamnoides while the plant coverage for Hippophae rhamnoides wasgreater than that for Robinia pseudoacacia. The significant correlations between NPP, ET andprecipitation during the August previous year to July were found in both plants.4. Plant optimal coverages for Robinia pseudoacacia and Hippophae rhamnoidespresented a distinct spatial variation characteristics and distribution patterns in the study area.Plant optimal coverage for both plants decreased gradually from southwest to northwest. Thedominant control factor is precipitation. The optimal plant coverage for Robinia pseudoacaciawas more than70%and for Hippophae rhamnoides more than80%in areas with annualprecipitation greater than550mm. In the areas with annual precipitation greater than450mm,optimal plant coverage for Robinia pseudoacacia was more than60%and for Hippophaerhamnoides more than70%. In the areas with annual precipitation less than300mm, optimalplant coverage for Robinia pseudoacacia was less than40%and for Hippophae rhamnoidesless than50%.