台灣雲杉造林木單木葉部生物量之估算

Abstract

本研究目的為估算塔塔加地區台灣雲杉人造林之單木葉部生物量。除期望能準確與精密地估算葉生物量外，並希望透過數種估算方法的嘗試及比較，建立適當且有效率的取樣方法，以供往後相關研究之參考。 本研究取樣方法為分層二階段取樣法，樣本以「與樣本單位大小成比例」（PPS）的不等機率取樣法抽出。分析資料時，除使用屬於以取樣設計為基準的估算外，亦使用以模式為基準的估算。以模式為基準的估算包括比例估算與迴歸估算。其中迴歸估算的模式為基本的allometric模式，並利用對數轉換及加權非線性迴歸兩種方式估算其參數。 整體而言，29株雲杉的葉部生物量隨著胸徑增加而有增加的趨勢，唯PPS估算所得結果較其他方式不同，也凸顯了以取樣設計為基準的估算及以模式為基準的估算兩大研究方法的差異。若將樹冠分上、中、下層來看，可發現上、下層各方法所推估的葉生物量非常一致；然而中層葉量較多，以對數轉換法所得的生物量有明顯高估的傾向。 PPS估算法雖然比簡單逢機取樣法費時，但其精密度並不差。在考量枝條抽取數目的取樣設計上，透過蒙地卡羅檢定，每枝一級枝條僅需抽取兩個二級枝條便可，其精密度並不比抽取三或四個枝條為差。

As part of a research on net primary productivity of Taiwan spruce （Picea morrisonicola Hayata） in Tatachia Long-Term Ecological Research, the objective of this thesis was to estimate foliar biomass of individual planted Taiwan spruce with different approaches. In addition, estimations from different approaches were also compared in order to establish a proper and efficient method for future references. The basic sampling design of this study was a stratified two-stage probability proportional to size （PPS） sampling. Based on tree crown length, the crown of each tree was divided into equal thirds, representing the upper, middle, and the bottom layers. Both design- and model-based estimations were applied. For model-based estimation, the standard allometric function was used. Both log-log transformation and nonlinear weighted least squares were used to estimate the parameters in the standard allometric models. All approaches showed that the foliar biomass increased with DBH, but PPS estimate had slightly different results from the model-based estimations. All approaches yield similar estimation results for upper and lower layers. However, likely due to higher foliar biomass content in the middle layer, the log-log transformed allometric model had the tendency of overestimating in this layer. Although PPS sampling required more time to collect data than simple random sampling, the precision of PPS was higher. Also, Monte Carlo simulations showed that it was not necessary to collect more than two sampling units to reach a higher efficiency and precision.