:::

太魯閣國家公園植群永久樣區監測計畫 (113)

摘要

關鍵詞:植群、永久樣區、太魯閣國家公園

一、研究緣起

太魯閣國家公園自1992年開始,分別於高海拔小風口附近的冷杉林、中海拔碧綠附近的針闊葉混淆林、布洛灣的低海拔常綠闊葉林及和仁海岸的常綠闊葉林,設置植群永久樣區進行長期監測,藉此了解區內不同海拔主要森林植群在氣候變遷下的演替趨勢,作為相關保育措施擬訂之參考。

植群永久樣區設立迄今已逾30年,期間陸續針對高海拔冷杉林樣區進行相關研究或複查,並於設立20年之際進行全面性複查。近年由於全球氣候暖化的趨勢日益明顯,氣候變遷對生態系的影響日受重視,本計畫擬對各海拔的永久樣區進行全面複查,並對樣區植群進行動態變化分析,以了解30年來不同海拔森林植群的演替現狀與趨勢。

 

二、研究方法及過程

本計畫於今年完成太魯閣國家公園低至高海拔共3個永久樣區的複查工作,分別是和仁低地常綠闊葉林、碧綠上部山地針闊混淆林和小風口亞高山台灣冷杉林樣區。布洛灣的低地常綠闊葉林樣區因0403地震和隨後的颱風豪雨,毀損殆盡無法修復,只能放棄複查,他日另覓合適之地重設樣區。

 

三、重要發現

1. 和仁樣區因受土石流災害之故,樣區下方約1/3面積屬於干擾頻繁的次生林,目前僅剩咬人狗和蟲屎等幾株零星的喬木,灌木和草本層皆被土石覆蓋。樣區上方約2/3面積干擾輕微,已由2004年的蟲屎、血桐和構樹為優勢的次生林,演替成以澀葉榕、樹杞、白肉榕和幹花榕等榕屬植物為優勢的林分,有朝耐陰性樹種演替的趨勢。

2. 碧綠樣區的物種組成多樣性高,雖在喬木層有尖葉槭和台灣紅榨槭;灌木層有疏果海桐和台灣小檗有退出林分的趨勢,對整體物種組成的影響不大。32年來有大量的小徑木死亡,樣木密度雖然大幅降低,但底面積呈穩定增加的趨勢,目前仍是以台灣雲杉為針葉樹冠層;赤柯、狹葉櫟和薄葉虎皮難為闊葉樹冠層優勢的典型針闊葉混淆林。

3. 小風口的台灣冷杉林樣區32年來有5次的調查資料,樣木的存亡機制和演替趨勢逐漸浮現。更新林的自疏現象明顯,樣木密度持續降低,競爭態勢激烈。樣木的生長和存亡趨勢,與競爭對象和光照資源的多寡有密切相關性。競爭壓力、胸徑年平均生長量或平均樹高,皆可做為評估樣木存亡機率的參數。更新林外側樣木的生長模式,由初期的高生長轉為加粗胸徑的生長,因為內側樣木的樹高逼近,有再轉為高生長的趨勢。更新林內側的樣木,因為密度高且光線條件較差,生長較為緩慢且一直以高生長為趨勢。

4. 674個立霧溪、三棧溪和清水斷崖的植群樣區,藉由棲地適宜性模擬中的MaxEnt模型,求取物種和環境因子之間的數值關係,模擬20種生長在國家公園不同海拔或區位殼斗科植物的潛在分布圖。

 

四、建議事項

建議一:立即可行之建議 

  1. 冷杉林樣區的森林結構變化、競爭機制、自疏現象、傳播假說和環境相關性等結果,可以把它轉化成環境教育的資源,於合歡山管理站展出。
  2. 和仁低地常綠闊葉林樣區今年同受地震和颱風的干擾,其植被恢復的過程,可做為蘇花沿線原生植被復育的參考,建議近五年內進行密集的監測,以掌握崩塌地植被演替和外來種入侵的趨勢,可做為一個非常好的對照資料。

建議二:中長期建議 

  1. 於外太魯閣低海拔地區另覓典型的常綠闊葉林,劃設永久樣區進行監測,以掌握本區植被和生態系的長期變化,做為日後環境教育和經營管理的參考。
  2. 於天祥附近山區,針對內太魯閣特殊的太魯閣櫟、青剛櫟和落葉闊葉樹形成的下部山地半落葉闊葉林,劃設永久樣區進行長期監測。

---------------------------------------------------------------------------------------------------

Abstract

keywordsplant community, permanent plot, Taroko National Park

This research involved a re-survey of three permanent plots in 2024, including He-Ren (lowland evergreen broad-leaved forest), Bi-Lu (upper montane mixed needle-broad-leaved forest), and Xiao-Feng-Kou (subalpine needle-leaved forest), spanning across different elevations in Taroko National Park.

In He-Ren, a landslide destroyed approximately one-third of the area, which was previously a frequently disturbed secondary forest. Now, only a few scattered trees, such as Dendrocnide meyeniana and Melanolepis multiglandulosa, remain, while the shrubs and herbs are covered by debris. In contrast, the remaining two-thirds of the area were less affected by the landslide. By 2024, the secondary forest in this part is dominated by Melanolepis multiglandulosa, Macaranga tanarius, and Broussonetia papyrifera, and is transitioning toward shade-tolerant species, such as fig trees, including Ficus irisana, Ardisia sieboldii, Ficus benjamina, and Ficus variegata var. garciae.

In Bi-Lu, species diversity is higher compared to the other plots. Although species such as Acer morrisonense and Acer kawakamii in the tree layer, along with Pittosporum pentandrum and Berberis kawakamii in the shrub layer, are retreating from the community, the overall composition has been minimally impacted. Over the past 32 years, many small trees have died, significantly reducing density; however, the basal area has steadily increased. Currently, the plot represents a typical mixed needle-broad-leaved forest. The canopy is dominated by the conifer Picea morrisonicola, while broadleaf species, including Cyclobalanopsis morii, Cyclobalanopsis stenophylloides, and Daphniphyllum himalaense subsp. Macropodum, make up the broadleaf canopy.

In Xiao-Feng-Kou, five surveys have been conducted over 32 years. This comprehensive dataset reveals the survival mechanisms and successional trends of Taiwan firs. The self-thinning phenomenon in the regenerating forest is evident, with tree density continually decreasing due to intense competition. Tree growth and survival are strongly influenced by competition and light resource availability. Therefore, competition pressure, average annual diameter growth, and average diameter at breast height (DBH) can serve as useful indicators for evaluating tree survival probability. On the outer edge of the regenerating forest, the growth pattern has shifted from height growth to trunk thickening. However, in the inward sample trees, the growth pattern shifts back to height growth. In contrast, within the inner forest, trees maintain a height growth pattern but grow slowly due to high density and poor light availability.

This research also adopted the habitat suitability concept, employing the MaxEnt model to derive numerical relationships between species and environmental factors using data from 674 vegetation plots. The simulation produced potential distribution maps for 14 oak species, showing their growth at different elevations and locations within the national park, which can inform future management and conservation plans.