含固定效应的多形立地指数模型构建与应用研究
发布时间：2018-10-29 10:17
【摘要】：立地指数模型是评价森林立地质量的核心和关键。鉴于现有的立地质量评价研究,没有全面的考虑立地类型和林分类型对立地质量的影响,从而导致同一立地类型、同一树种在相同的条件下得到不同的生长过程,使得建立的立地质量评价模型难以推广、限制了其适用性,且现有的立地质量评价研究对天然混交林研究较少。通过研究,以期使建立的立地指数模型容易推广、适用性强,解决区域性兼容问题,并尝试探索解决天然林立地质量评价问题。本文以湖南省杉木人工林和亚热带栎类天然混交林为研究对象,利用调查数据,从基于树高-年龄分级的多形立地指数模型和基于立地类型聚类的多形立地指数模型两方面入手,深入研究了含固定效应的多形立地指数模型在湖南省杉木人工林和亚热带栎类天然混交林上的构建与应用。主要研究内容和结论如下:(1)通过树高-年龄分级构建固定效应的方法,建立了基于树高-年龄分级的杉木人工林多形立地指数模型。指出龄阶间距为5年,树高分9级,其所得的分级结果最佳,与基础模型相比,将模型的确定系数从0.79964增至0.95303,大大提高了模型的精度。同时也验证了树高-年龄分级的方法应用于人工林立地质量评价种是正确、可行的。(2)通过立地类型聚类构建固定效应的方法,建立了基于立地类型聚类的杉木人工林多形立地指数模型。将606个样本中立地因子相同的进行组合,划分为240个初始立地类型,并将初始立地类型作为固定效应加在参数a上进行含固定效应的非线性回归模型模拟,得到240个初始立地类型所对应的系数值;用R软件中k-means聚类,将240个不同的立地类型聚为12类,合并后的因子水平信息包含合并前的因子水平信息的99%。将各样本的立地类型转换为对应的类,并将12个类作为固定效应加在基础模型上进行非线性拟合回归分析,模型的确定系数R2=0.98811,进一步提高了模型的精度,证实了立地类型聚类构建固定效应的思维和方法应用于人工林立地质量评价中是具有可行性的。(3)将两种立地质量评价方法应用于栎类天然混交林中进行探索研究,建立了两种不同的栎类天然混交林多形立地指数模型:1)采用树高-年龄分级构造固定效应建立多形立地指数模型的方法,建立了栎类天然林多形立地指数曲线模型,确定系数R2= 0.9807,模型精度较高,且建模过程无需添加立地因子,操作简单。2)采用立地类型聚类构造固定效应建立多形立地指数模型的方法,建立了栎类天然林多形立地指数曲线模型,确定系数R2= 0.87584,模型精度较高,且能够保证同一立地类型、同一树种在相同的条件下得到几乎相同的生长过程。3)相对于同形立地指数模型无法拟合栎类天然林生长过程的情况,两种方法均能适用于栎类天然林立地质量评价,且建立的模型精度高、适用性强,可解决天然林立地质量评价问题,为天然林立地质量评价提供了一种新思路和参考方法。(4)研究提出的两种多形立地指数模型构建方法,既适用于人工林又适用于天然林立地质量评价。建立的模型适用性强,可解决区域性兼容问题和天然林立地质量评价问题。含固定效应的多形立地指数模型为我国森林立地质量评价提供了一种新思路和方法,不仅适用于杉木人工林和栎类天然混交林,也可借鉴于其它林分类型的立地质量评价中。
[Abstract]:Site index model is the core and key point of evaluating forest site quality. In view of the existing site quality evaluation research, there is no comprehensive consideration of the influence of site type and forest sub-type on the quality, thus leading to the same site type, and the same tree species are subjected to different growth processes under the same conditions, so that the established site quality evaluation model is difficult to popularize, The applicability of the study was limited, and the existing site quality assessment was less than that of the natural mixed forest. Through the research, the establishment of the site index model is easy to popularize, the applicability is strong, the regional compatibility problem is solved, and the problem of the quality evaluation of the natural forest land is attempted to be solved. Based on the investigation data, this paper starts with the multi-site index model based on tree height-age classification and the multi-site index model based on site type based on the investigation data. In this paper, the construction and application of multi-site index model with fixed effect on the natural mixed forest of Chinese fir plantation and sub-tropical oak forest were studied. The main research contents and conclusions are as follows: (1) The multi-shape site index model of Chinese fir plantation based on tree height-age classification is established by constructing the fixed effect by tree height-age classification. It is pointed out that the age-order spacing is 5 years and the tree high score is 9. The obtained grading results are the best. Compared with the basic model, the determination coefficient of the model is increased from 0. 79964 to 0. 95303, which greatly improves the accuracy of the model. At the same time, it is proved that the method of tree height-age classification is correct and feasible. (2) A multi-shape site index model of Chinese fir plantation based on site type is established by constructing fixed effect by site type poly. combining 606 sample neutral factors into 240 initial site types, adding the initial site type as a fixed effect and a nonlinear regression model containing a fixed effect on the parameter a to obtain the coefficient values corresponding to 240 initial site types; Using the k-means poly in R software, 240 different site types are grouped into 12 classes, and the combined factor level information contains 99% of the factor level information before consolidation. the site conditions of each sample are classified into corresponding classes, and 12 classes are added as a fixed effect to carry out nonlinear fitting regression analysis on the basic model, the determination coefficient R2 of the model is 0.98811, and the accuracy of the model is further improved, It is proved that the method of thinking and method for constructing fixed effect of site type polytype is feasible in the quality evaluation of artificial forest land. (3) the two site quality evaluation methods are applied to the exploration and research of the oak-type natural mixed forest, and two different kinds of natural mixed forest multi-shape site index models are established: 1) a method for establishing a multi-shape site index model by adopting a tree height-age hierarchical structure fixing effect is established; A multi-site index curve model of oak forest is established, and the coefficient R2 = 0.9807, the model precision is high, and the modeling process does not need to add site factors, and the operation is simple. The multi-site index curve model of oak forest is established, and the coefficient R2 = 0.987584, the model precision is high, and the same site type can be ensured. the same tree species are almost the same under the same conditions. 3) the condition that the growth process of the oak forest natural forest cannot be fitted with respect to the homomorphic site index model, both methods can be applied to the quality evaluation of the oak natural forest, and the established model is high in accuracy and strong in applicability, The method can solve the problem of quality evaluation of the natural forest, and provides a new idea and a reference method for the quality evaluation of the natural forest. (4) The construction method of two kinds of site index models proposed in this paper is suitable for the quality evaluation of natural forest. The established model has strong applicability, can solve the regional compatibility problem and the quality evaluation problem of natural forest land. The multi-site index model with fixed effect provides a new idea and method for forest site quality evaluation in China. It is not only suitable for Chinese fir plantation and oak-type natural mixed forest, but also can be used for site quality evaluation of other forest types.
【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】：2017
【分类号】：S757
，
本文编号：2297409
[Abstract]:Site index model is the core and key point of evaluating forest site quality. In view of the existing site quality evaluation research, there is no comprehensive consideration of the influence of site type and forest sub-type on the quality, thus leading to the same site type, and the same tree species are subjected to different growth processes under the same conditions, so that the established site quality evaluation model is difficult to popularize, The applicability of the study was limited, and the existing site quality assessment was less than that of the natural mixed forest. Through the research, the establishment of the site index model is easy to popularize, the applicability is strong, the regional compatibility problem is solved, and the problem of the quality evaluation of the natural forest land is attempted to be solved. Based on the investigation data, this paper starts with the multi-site index model based on tree height-age classification and the multi-site index model based on site type based on the investigation data. In this paper, the construction and application of multi-site index model with fixed effect on the natural mixed forest of Chinese fir plantation and sub-tropical oak forest were studied. The main research contents and conclusions are as follows: (1) The multi-shape site index model of Chinese fir plantation based on tree height-age classification is established by constructing the fixed effect by tree height-age classification. It is pointed out that the age-order spacing is 5 years and the tree high score is 9. The obtained grading results are the best. Compared with the basic model, the determination coefficient of the model is increased from 0. 79964 to 0. 95303, which greatly improves the accuracy of the model. At the same time, it is proved that the method of tree height-age classification is correct and feasible. (2) A multi-shape site index model of Chinese fir plantation based on site type is established by constructing fixed effect by site type poly. combining 606 sample neutral factors into 240 initial site types, adding the initial site type as a fixed effect and a nonlinear regression model containing a fixed effect on the parameter a to obtain the coefficient values corresponding to 240 initial site types; Using the k-means poly in R software, 240 different site types are grouped into 12 classes, and the combined factor level information contains 99% of the factor level information before consolidation. the site conditions of each sample are classified into corresponding classes, and 12 classes are added as a fixed effect to carry out nonlinear fitting regression analysis on the basic model, the determination coefficient R2 of the model is 0.98811, and the accuracy of the model is further improved, It is proved that the method of thinking and method for constructing fixed effect of site type polytype is feasible in the quality evaluation of artificial forest land. (3) the two site quality evaluation methods are applied to the exploration and research of the oak-type natural mixed forest, and two different kinds of natural mixed forest multi-shape site index models are established: 1) a method for establishing a multi-shape site index model by adopting a tree height-age hierarchical structure fixing effect is established; A multi-site index curve model of oak forest is established, and the coefficient R2 = 0.9807, the model precision is high, and the modeling process does not need to add site factors, and the operation is simple. The multi-site index curve model of oak forest is established, and the coefficient R2 = 0.987584, the model precision is high, and the same site type can be ensured. the same tree species are almost the same under the same conditions. 3) the condition that the growth process of the oak forest natural forest cannot be fitted with respect to the homomorphic site index model, both methods can be applied to the quality evaluation of the oak natural forest, and the established model is high in accuracy and strong in applicability, The method can solve the problem of quality evaluation of the natural forest, and provides a new idea and a reference method for the quality evaluation of the natural forest. (4) The construction method of two kinds of site index models proposed in this paper is suitable for the quality evaluation of natural forest. The established model has strong applicability, can solve the regional compatibility problem and the quality evaluation problem of natural forest land. The multi-site index model with fixed effect provides a new idea and method for forest site quality evaluation in China. It is not only suitable for Chinese fir plantation and oak-type natural mixed forest, but also can be used for site quality evaluation of other forest types.
【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】：2017
【分类号】：S757
，
本文编号：2297409
本文链接：https://www.wllwen.com/shoufeilunwen/zaizhiyanjiusheng/2297409.html
