臺灣中部塔塔加地區臺灣雲杉人工林之樹幹解析研究

Abstract

本研究以塔塔加地區臺灣雲杉（Picea morrisonicola）36年生人造林之林木為材料，計數與量測其不同高度斷面圓盤的樹輪數與每一年的樹輪寬度資料，並縱向地解剖林木樹幹，量測其每一年頂芽生長停止的位置，以建立樹高生長歷程，並以之檢測Graves、Carmean、ISSA與TARG等四種年齡─樹高關係估算法在分段長度分別為100 cm、200 cm與300 cm時其估算之精確性。 研究的結果顯示臺灣雲杉37年生的人工造林木，其直徑的生長已經趨於緩和，樹高的生長仍有持續的增加之趨勢，然材積之連年生長曲線於30年生時達最大生長，而後下降。在分段長度處理為100 cm與200 cm時，Carmean法為四種估算法中精確性最高者，ISSA法可以在分段長度為100 cm時有良好的估算表現。所有的樹高估算法在分段長度為300 cm時皆無法精確地推估樹高與年齡之關係。Graves之傳統樹高估算法則是在三個不同長度分段下皆顯著地高估樹高的生長。TARG樹高估算法在三個不同長度分段下皆顯著地低估樹高的生長。依胸高直徑區分為最優勢木，次優勢木與被壓木後所進行的統計分析獲得的結果，Carmean方法、ISSA方法與Graves方法不適合用於建立被壓木之樹齡─樹高曲線，三種方法皆會出現顯著地高估。在建立被壓木之樹齡─樹高上TARG方法有良好之結果。在每年樹高增長的估算上，則以ISSA估算法為最佳。

Based on 28 trees from a 36-year Taiwan spruce (Picea morrisonicola) plantation at the Tatachia area, central Taiwan, the objectives of this study were to (i) use both cross- and longitudinal- sectionings to reconstruct the height growth trajectory of each tree and (ii) evaluate the accuracies of four methods (Graves’ method, Carmean’s method, ISSA method, and TARG method) in establishing tree age-height curves at three cross-sectioning lengths (100, 200, and 300 cm) based on the findings from (i). For each tree, ring numbers and widths from a series of 10-cm cross-sectioned discs taken at 1-m interval were first determined, and when applicable the location of apical bud at the end of each growing season (nodal diaphragm) within each disc was determined. Each tree’s height growth trajectory was then constructed based on the locations of nodal diaphragms and was used to test the accuracies of the four methods for both accumulative height growth and annual height increment. The results showed that the average height growth of the plantation could still increase, however individual tree volume already peaked at age 30. Among the four methods examined, Carmean’s method had the highest accuracies for both 100- and 200-cm cross-section lengths, and the ISSA method worked well only at 100-cm length. All methods performed poorly at 300-cm length. Graves’ method significantly overestimated and TARG method significantly underestimated the true tree height for all three lengths. The ISSA method had the highest accuracy in estimating annual height increment.