Study On The Spatial Heterogeneity Of Soil Physical And Chemical Properties Of Primary Tropical Montane Rainforest In Jianfengling, Hainan Island

The spatial heterogeneity of soil physical and chemical properties can strongly affect the patterns and dynamics of biogeochemical cycling in tropical moist rainforest, and also create potential niche differentiations, so promoting coexistence of species, and therefore, maintaining a high degree of plant and soil organisms species richness in local scales. The geostatistics method can quantitatively real the spatial structure of soil properties with the range and spatial denpendence, and also the mian aims of soil heterogeneity studying. The information of soil saptial structure can give scitific basis for field soil sampling.In this study, we use the semivariogram amd kriging interpolation of geostatistics to quantitatively analyses the spatial heterogeneity of21soil properties in a1hm2tropical primary montane rainforest in Jianfengling, Hainan island. It is the fist time to investigate the saptial structure of the soil properties with intensively sampling efforts in this aereas. The main results we got as follows:(1) The basic characteristics of the soil properties in primary tropical rain forestAll the8soil physical properties and13soil chemical properties of interest are changing with the soil depth increasing, which denotes the soil heterogeneity in the vertical profile. In general, with the depth increment, the soil bulk density increasing, and the soil water content, soil porosity and soil moisture holding capacity decreasing; the soil pH becoming alkalescence, and the soil nutrient lowering, especially the available one. The results show that the soil properties becoming less favarable for plants with soil depth increasing in this pramary tropical rainforest.(2) Different soil properties do not have the similar correlation between the different soil layers.All the soil properties in this study have somewhat correlation between different soil layers. In general, the middle layer versus the deep layer have the most stronger relationships than the others layers, especially the upper layer has the weaken correlations with the inner layers within the soil profile. The results here indicate that different ecological processes control the soil properties at different soil depth.(3) All the21soil properties of interest in this study have the definite spatial structure, nor randomly distribution.The CV of the classic statistcs denotes the variability of the soil properties within a study areas. In my study plot, the CV ranges from0.03to1.35with all21soil properties. The pH, soil bulk density, soil capillary and total porosity have lower variation, and soil organic matter content, total nitrogen, available nitrogen, total phosphorus, CEC, soil capillary moisture holding capacity, soil minimum and maximum moisture holding capacity have the middle level variation, campared with the soil non-capillary porosity, soil available phosphorus, total and available K, the total exchangeable base cations, Ca, Mg and Na which have largest varaition;Unfortunately, the CV can not reveal the spatial structure of soil properties, but the semivariagram is a powerfull tool to unvover it. Using this tool, we find the ranges of the soil properties autocorrelations ranges from10.81m to45.50m with all variables of interests. And most soil properties have the ranges around20m and30m, but the soil water content and soil capillary porosity have the largest ranges exceeding40m, campared with the soil non-capillary porosity, the soil phosphorus at the upper layers and the soil pH at the middle layers having the samllest ranges below20m.Although, all the soil variables in this study has the spatial structure, the spatial dependence is not same for different soil properties at different layers with the ranges of C0/(C+Co) from10.2%to87.8%. Soil bulk density, avabiliable P and exchangeable Na at B layes, soil total porosity at C layers, the TK at the A layes have the higest saptial dependence; and the soil total and capillary porosity at A layers, the maximum water holding capacity and phosphorus in B layes have the lowest spatial denpendence compared with the others.(4) All the21soil properties have different spatial distribution patterning with different patchness in the study plot. The kriging interpolation provied the optimal estimating of the unsampled location in our studying plot. Based on the estimation, we can generate the spatial distribution of the soil properties in the study areas.In our study, we find different soil properies having different spatial patterning, but for a particular properties, the different layers have some simialr patchness, however, somewhat different.(5) The topography is a factor that can affect the spatial distribution of soil properties in this tropical rainforest.Using the topography digitized mehods, we analyses the relationships between the topography in all plot and the soil properties we investigated. The results show that the topography have different influences on every soil properties.