单形和多形地位指数模型的对比研究

Comparison for Single and Polymorphic Site Index Models

本文以杉木为例,根据在南岭山地收集的建模和检验两套样本,以Richards函数为基本模型,对单形和多形地位指数模型进行了对比研究。结果表明:①由于不同地位指数优势高的生长过程曲线簇存在多形特性,多形模型的预估精度明显高于单形模型,而且单形可视为多形的一种特例,因此,在一般情况下,应以建立多形模型为宜;②本文给出的简捷迭代法,只需具备可编程序计算器(如SHARPEL-5100S),便可按现实林分的年龄和优势高实测值,根据所建立的多形地位指数模型,迅速求解地位指数值,既能提高预估精度,便于野外应用,又便于输入计算机处理;③将单形导向曲线方程辰开成地位指数表的几种现行方法,其结果差异较大,而且编成地位指数表应用,有损预估精度和增加计算机处理的麻烦。因此,即使在可以采用单形导向曲线法的情况下,亦应加以改进,以采用生长函数为模型,建立以地位指数为因变量的单形地位指数模型为宜。

The practicability of the single and polymorphic site index models is researched by taking Chinese fir as an example, adapting Richards growth function as basic model and using the two sets of samples of fitting and testing which were collected in Mount Nanling. The results show that (1) the predicted precision of the polymorphic sits index model is obviously higher than that of the single's because of the polymorphic characts of the dominant heigh growth curves in different site indexes. Actually, the single is a kind of special form of the polymorphic, so it is better to develop the polymorphic site index model in common condition. (2) It can quickly solve the site index value based on the stand age and mean dominant heigh by using a simple iteration on a calculator such as SHARP EL-5100S. It can improve the precision as well as convenience for the field work and computer processing. (3) There are several current methods of developing site index table by adopting guide curve equation, but their results are different. And the application of the developing site index equation into site index table can decrease the predicted precision as well as increase the trouble of the computer processing. Therefore, it is better to choose a growth function as basic model and develop it by taking site index as dependent varible even if the single guide curve equation can be adapted in some locales.