疏伐度對於臺灣3種造林木胸高直徑生長影響之研究

森林經營的主要目的是在可持續經營的基礎上提高林木的生長量，改進林木的形質，森林經營規劃的目的是以森林資源為對象，進行合理的時空布局以達到經營者所設定之經營目標，而透過疏伐作業改變林分密度，調整成為合理的林分結構可以提高人工林生產力，是管理森林生長與生產量的有效方法。當需要對一個地區的森林資源量進行評估時，其中的一種方式是估算森林蓄積量，因為森林蓄積不僅反映了一個地區森林資源的整體水平，也是衡量森林生態系統狀況的重要依據。進行評估時，如何簡化調查過程、降低調查成本，並且準確測算森林蓄積量一直是林業人員和管理單位的目標。本研究以疏伐對臺灣杉人工林直徑生長，疏伐對大葉桃花心木胸徑之生長與相對變化，以及疏伐對於楓香人工林胸高形數與形率之關係等三個面向進行研究，各研究內容分述如下：在疏伐對臺灣杉人工林直徑生長之研究中，本研究以林業試驗所六龜研究中心臺灣杉人工林14年生時進行之下層疏伐試驗區為研究對象，為了解疏伐後造林木的生長隨時間推移的變化，探討1.林分胸徑定期平均生長量的變化，以及2.不同直徑階生長量於疏伐強度間的比較。疏伐採逢機完全區集設計，分為強、中、弱度疏伐與對照組，並於伐後第2、5、7、13、17年進行5次調查。結果顯示，在林分胸徑定期平均生長量上，經疏伐處理之留存木於5次調查時，其與對照組均具極顯著差異(P<0.01)，各疏伐強度以疏伐後2-5年差異較大，之後隨伐後年期增加而漸不明顯；疏伐強度越高，促進生長的效果越好，維持生長的時間越長，各直徑階定期平均生長量相對百分率於伐後5-7年達高峰；各疏伐度間，胸徑20 cm以下的留存木因林分密度的下降程度不同，於伐後第7年生長開始衰退時，各疏伐度間的生長下降程度不一，而胸徑20-30 cm的留存木對於強度、中度疏伐生長促進表現相似，顯示疏伐強度對於優勢木胸徑生長促進效果有一定上限；弱度疏伐於伐後第5年到達高峰後穩定維持1.4倍的生長。在疏伐對大葉桃花心木直徑生長與林分結構變化之影響研究中，桃花心木為台灣中南部低海拔地區重要造林樹種，亦具熱帶林永續生產之潛力，為了瞭解疏伐對大葉桃花心木的胸徑生長及林分結構變化情形，於2011年在臺灣台南市關廟地區的14年生大葉桃花心木人工林進行下層疏伐試驗，設置強、中、弱度疏伐與對照組合計4區，於2011年、2012年、2013年、2015年、2017年進行DBH調查，進行1.簡單線性迴歸式的連續變動斜率值觀察樣區DBH變動趨勢，2.以Weibull機率密度函數配適徑級分布，3.比較不同處理下的平均胸徑生長量。結果顯示，留存小徑木於伐後仍然生長緩慢，而中-強度的疏伐可以在一定的時間內使較大直徑的留存木有相似的DBH增加量；疏伐後單木DBH定期平均生長量於第2-6年與對照組具顯著差異(p<0.05)，但疏伐處理間差異不顯著，中度疏伐與強度疏伐對於促進單木定期平均生長量效果相近。在花蓮大農大富地區楓香人工林胸高形數與形率之關係探討研究中，林木胸高形數與形率是立木材積計算上為重要的參數之一，然而在資料取得上需透過大量伐倒林木並逐步累積數據後才能獲得，本研究嘗試透過非破壞性的作業，以地面光達技術探討林木胸高形數(F1.3)與胸高形率(Qn)兩者間的關連性並建立轉換模式，以取代傳統胸高形數獲取方式。結果顯示，低強度疏伐樣區林木 F1.3平均值為0.40， Qn平均值為0.56；高強度疏伐樣區林木F1.3平均值則為0.48，Qn平均值為0.68。在推估形率轉換形數模式方面，以公式F1.3 = a + b × Qn2 + c × (Qn × H)-1 在2種林分密度皆有最佳的擬合結果，由Wilcoxon檢定結果，形率轉換式所估算的材積相較於一般通用形數差異較為平均，可以快速可靠的估算平均材積，也能減少單木材積的計算誤差。

關鍵字

疏伐；林分結構；定期生長量；種植密度； Weibull分布；胸高形率；形數

並列摘要

The main purpose of forest management is to increase the growth of afforestation trees and improve the shape and quality of trees on the basis of sustainable management. The purpose of planning is to take forest resources as the object, and to carry out reasonable time and space layout to achieve the management goals set by the operator. Performing thinning operations to change stand density and adjust to a reasonable stand structure can increase the productivity of plantations and is an effective way to manage forest growth and productivity. When it is necessary to assess the amount of forest resources in an area, one of the ways is to estimate the forest stock. Forest accumulation not only reflects the overall level of forest resources in a region, but also is an important basis for measuring the status of forest ecosystems. How to simplify the survey process, reduce survey costs, and accurately measure forest stock has always been the goal of forestry personnel and management units. This paper uses three studies for discussion, including "Effect of Thinning on the Diameter Increment Growth of Taiwan Cedar", "Effects of Thinning on the Growth and Relative Change in the Diameter of a Mahogany Plantation", and " The Relationship between Breast Height Form Factor and Form Quotient of Liquidambar formosana in the Eastern Part of Taiwan", the content of this study is described as follows: In the content of "Effect of Thinning on the Diameter Increment Growth of Taiwan Cedar". This study aims to explore the comparison of the mean annual DBH increment and DBH grade increment after thinning of 14-year-old stands, Director of the Taiwan Forestry Research Institute. Refinement followed a randomized complete block design with 4 replicates each for heavy, medium, light and no refinement and revisited after years 2, 5, 7, 13, and 17. The results showed that the average annual increment of each thinned forest stand in the five inventories was significantly larger than that of the non-thinned forest. The average annual increase of diameter at breast height of each thinned stand reached a peak after 2 to 5 years of thinning, and then gradually decreased. The comparison of the DBH grade distribution (5cm width) of the 5 stocks showed that the more intense the thinning, the greater the increment, and the longer the thinning effect lasted. The average annual increment for all DBH grades peaked 5-7 years after thinning. Due to the difference in thinning intensity and capping time, the increments of remaining small trees (diameter at breast height less than 20 cm) were different among treatments but reached the maximum at 7 years after thinning. In contrast, the remaining large trees (between 20 and 30 cm DBH) in heavy and medium sparse forests showed similar increases, suggesting that the effect of heavy sparseness on promoting diameter growth was limited, and they reached 2.4 maximum increment. After 5 years of thinning, the multiple without thinning can reach 1.8 times after 17 years of thinning. The remaining amount of large trees after mild thinning was 1.4 times that of non-thinned stands and reached the maximum after 5 years of thinning, and it influence on the DBH grade increment weakened with time. In the content of "Effects of Thinning on Relative Changes in Growth and Diameter of Redwood Plantations." Honduras mahogany (Swietenia macrophylla King) is an important forestry tree species in low-altitude areas in central and southern Taiwan, and has good potential for sustainable forestry in tropical regions. The study conducted understory thinning experiments in 2011 at a 14 year-old redwood plantation in Guanmiao, Tainan City, Taiwan. Four heavily, moderately, and mildly sparse plots and one control plot were conducted in 2011, 2012, 2013, 2015, and 2017 at tree ages of 14, 15, 16, 18, and 20 years old, respectively. DBH surveyed trends were observed using simple linear regression with continuous slope values, and the Weibull density function was used to match the distribution of diameter grades and compare mean DBH growth under different thinning treatments. At 16, 18, and 20 years, the mean cyclical increase in DBH after individual thinning was significantly different from that control group trees (p-value < 0.05). However, the difference between thinning treatments was not significant, and the effects of intermediate thinning and severe thinning on promoting the average cyclic growth of single wood were similar. In the content of " The Relationship between Breast Height Form Factor and Form Quotient of Liquidambar formosana in the Eastern Part of Taiwan", the breast height form factor and quotient of trees are important parameters to calculate the stumpage volume in forest management. However, many types of datasets can only be accumulated by cutting down many trees. This research studied different thinning intensities of Formosan sweet gum (Liquidambar formosana) in Danongdafu Forest Park, Guangfu Township, Hualien County, Taiwan. The relationship and transformation modes of the breast height form factor (F1.3) and the normal form quotient (Qn) were examined using a terrestrial laser scanning (TLS) system, which can replace the conventional method to obtain F1.3. The results demonstrated that the average F1.3 and Qn in the low intensity thinning area were 0.40 and 0.56, respectively. The average F1.3 and Qn in the high intensity thinning area were 0.48 and 0.68, respectively. Whether the stand density is low or high intensity can be determined statistically depending on the optimal results for F1.3. One mode of estimation when the stand density is 1,200 trees ha−1 ,another is 600 trees ha−1. The root mean square error of the equations used was 0.002, and the mean absolute error <0.02. The mean percentage error was ±2.3%, and absolute mean percentage error <5%. If the TLS system can obtain datasets of quotient and height by surveying in the future, this will enable to calculate of F1.3. The stock volume will be calculated precisely, and the calculation error will be improved.