石栎—青冈林土壤有机碳贮量和钾含量的空间分布格局研究

发布时间：2018-05-08 08:52

本文选题：地统计学 + 空间异质性　；参考：《中南林业科技大学》2015年硕士论文

【摘要】：森林土壤是维持森林生态系统可持续性的重要组成部分,由于受到成土母质、地形、气候等自然因素和人为因素的影响,土壤在空间尺度上表现为非均质体,高度空间变异性是土壤的重要自然属性之一。土壤养分在其中显现出复杂的镶嵌性,对其含量、分布等空间特征及影响因素进行研究,对于了解土壤活动、土壤肥力评价、施肥控制、植被更新过程、林木的产量分析和环境保护等具有重要的参考价值,并有助于外推预测采样区域外的土壤特征。应用地统计学的方法对土壤空间异质性的研究相对于传统的统计方法更为准确、直观,本研究以湘中石栎(Lithocarpus glaber)-青冈(Cyclobalanopsis glauca)常绿阔叶林为研究对象,使用地统计学的半方差函数方法和基于区域化变量理论的地质统计软件GS+对目标区域内的100个10 m 10 m样地取得的不同深度的300份土壤样本进行了碳贮量和全钾的空间特征进行研究,同时对可能引起空间变异的影响因素如地形因子(坡度、坡向、海拔和凸凹度)、凋落物量等进行了统计学分析,为揭示其影响因素及作用机理,了解该林型的土壤养分特征并在此基础上合理进行林地管理和保护生态环境提供科学参考。结果表明：青冈-石砾常绿阔叶林0-30 cm土壤碳贮量平均值为21.21t·hm-2,达到了中等变异强度,土壤碳贮量均值随土层深度增加而减小,3个层次土壤碳贮量均符合指数模型分布,具有明显的空间异质性。通过半函数方差分析,在10-20 cm土层土壤碳贮量的空间结构比C/(Co+C)大于0.7,表明此土层碳贮量具有中等强度的空间自相关性,0-10 cm、20-30 cm和0-30 cm层土壤碳贮量的空间结构比C/C0+C)大于0.9,表明表层土壤和深层土壤的C贮量具有较弱的空间自相关性。各土层土壤碳贮量与海拔高度、坡度和凹凸度均呈负相关关系,但未达到显著水平,说明土壤C贮量的变化受地形因子影响较小。0-30 cm土壤钾含量平均值为28.04 g·kg-1,为中等变异强度,具有明显的空间异质性。除20-30 cm土层中K含量符合线性模型外,0-10 cm和10-20 cm土层和0-30 cm土层K含量符合球状模型分布,各土层土壤K含量具有较强的空间自相关性。0-10 cm、10-20 cm、20-30 cm和0-30 cm土层中K含量的空间自相关变程范围分别为37.51 m、25.38 m、39.78 m和28.92 m,土壤K含量有较小的空间异质性尺度。各土层K含量随地形的变化趋势较为明显,表现为斑块状空间分布,低值出现在中下部的山脊,高值出现在上部山脊和西南角的部分沟谷。土壤K含量与海拔高度、土壤粘粒含量呈显著的正相关,K含量的空间异质性主要受成土母质和地形等结构因素影响。
[Abstract]:Forest soil is an important component of maintaining the sustainability of forest ecosystem. Due to the influence of natural and human factors such as parent material, topography, climate and so on, the soil is heterogeneous on spatial scale. High spatial variability is one of the important natural attributes of soil. Soil nutrients show complex mosaic. The spatial characteristics and influencing factors, such as content, distribution and so on, are studied to understand soil activity, soil fertility evaluation, fertilization control, vegetation regeneration process, etc. The analysis of forest yield and environmental protection has important reference value and is helpful to extrapolate and predict the soil characteristics outside the sampling area. The study of soil spatial heterogeneity by geostatistics was more accurate and intuitionistic than traditional statistical methods. The evergreen broad-leaved forest of Lithocarpus glaberus-Cyclobalanopsis glauca was studied in this study. Using the semi-variance function method of geostatistics and the geostatistics software GS based on regionalization variable theory, the carbon storage and total potassium of 300 soil samples of different depths obtained from 100 10 m 10 m sample plots in the target area were studied. To study the spatial characteristics of, At the same time, the influence factors of spatial variation, such as slope, slope direction, elevation and concavity, litter quantity and so on, were analyzed statistically, in order to reveal the influencing factors and mechanism. To understand the soil nutrient characteristics of this forest type and to provide scientific reference for forest land management and ecological environment protection. The results showed that the average value of soil carbon storage in the evergreen broad-leaved forest was 21.21 t 路m ~ (-2), and the mean value of soil carbon storage decreased with the increase of soil depth, and the soil carbon storage of the three layers was in accordance with the exponential model distribution. It has obvious spatial heterogeneity. Through the analysis of variance of half function, The spatial structure of soil carbon storage in 10-20 cm soil layer is greater than that of C/(Co C, which indicates that the spatial structure of carbon storage in this soil layer is of moderate spatial correlation (0-10 cm ~ (-1) 20-30 cm and 0-30 cm ~ (-1), which indicates that the spatial structure of carbon storage in this soil layer is greater than that in C/C0 _ (C) layer. The C storage of soil and deep soil has weak spatial autocorrelation. Soil carbon storage in each soil layer was negatively correlated with elevation, slope and concavity, but not significant, indicating that the variation of soil C storage was less affected by topographic factors. The average value of soil potassium content was 28.04 g 路kg ~ (-1), which was a moderate variation intensity. It has obvious spatial heterogeneity. K content in 0-10 cm, 10-20 cm and 0-30 cm soil layers was in accordance with the distribution of spherical model except that K content in 20-30 cm soil layer was in accordance with linear model. The spatial autocorrelation range of K content in 0-10 cm ~ (10) cm ~ (-1) ~ (-20) cm ~ (-1) and 0-30 cm soil layer was 37.51 m ~ 25.38 m ~ (39.78 m) and 28.92 m, respectively. The soil K content had a small spatial heterogeneity scale. The variation trend of K content in each soil layer with topography is obvious, showing patch spatial distribution, the low value appears in the middle and lower part of the ridge, the high value appears in the upper ridge and part of the valley in the southwest corner. The spatial heterogeneity of soil K content and soil clay content was significantly positively correlated with the elevation of soil. The spatial heterogeneity of soil K content was mainly affected by soil parent material and topography and other structural factors.
【学位授予单位】：中南林业科技大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：S714

【参考文献】