透過您的圖書館登入
IP:18.116.239.195
  • 期刊

應用福衛二號影像與地面樣區資料於光蠟樹平地造林地之碳吸存推估

Applying FORMOSAT-2 Image and Field Survey Data to Estimate Carbon Sequestration in Fraxinus Farm Afforestation

摘要


面對全球暖化日趨嚴重及後京都議定書時代來臨,大面積人工造林地之林木生物量具有相當的碳吸存能力,如何運用遙測影像結合地面調查資料精確估算人工林生物量對碳吸存量,已成為一個國際性的重要課題。本研究以花蓮縣光復鄉台糖大農及大富農場2002年至2009年平地造林之光蠟樹林地為試驗區,使用2009年2月13日拍攝之福衛二號衛星影像及試區造林樹種分區地理資訊。應用福衛二號光譜輻射常態化處理後的土壤調整植生指標SAVI,對佔有17%試區、總面積達168ha的光蠟樹純林區塊進行SAVI分級。參酌SAVI分級分布圖,以分層目的取樣法,於試區內共設計並調查了70個地面樣區,選用適用的光蠟樹材積式,計算樣區光蠟樹的蓄積量。結合統計資料探索過程及空間相關分析指標,最後以60個地面樣區的單位材積蓄積與SAVI建立推估迴歸式,以IPCC碳吸存公式及光蠟樹轉換係數,推估試區光蠟樹碳吸存量。研究結果顯示經本研究所得最適直線迴歸式可解釋的變異量達75%與國外相關研究相當,比在台灣山區人工林研究成果可靠度提升很多。估計平均每ha光蠟林蓄積量為16.93±1.16m^3/ha,平均每ha吸存的碳量為7.55±0.52ton/ha,全區的光蠟林總碳吸存量為1,276.40±87.28ton,所得成果可用空間分布圖展示。此外,使用迴歸推估法優於直接使用樣區推估法、本研究最適地面樣區數目為40個,一併在文中加以探討。

並列摘要


Due to afforested plantations have a considerable amount of biological carbon sequestration capacity, the government has been aiming at large area of plantations on the agricultural lands to relieve the pressure of global warming and face the coming of Post Kyoto Protocol. How to use satellite images combined with ground survey data to accurately estimate the biomass of plantation carbon sequestration has become an international issue. In this study, the afforested lands between 2002 and 2009 belonging to Taiwan Sugar Company's Dai-Lung and Dai-Fun farms at Guangfu township, Hualien county were used as experimental area. The FORMOSAT- 2 (FS2) satellite image of this area acquired on February 13, 2009 and plantations maps of these two farms were collected as geographic information for biomass estimation. After a procedure of radiation normalization for the FS2 image, we applied soil adjusted vegetation index (SAVI) of FS2 focused on plantations of Fraxinus griffithii (Fg) that occupied 17%, a total area of 168 hectares distributed in the area, to classifying the SAVI of them to 5 separated grades. Considering the spatial heterogeneity of SAVI on graded patches of plantations, a stratified purposive sampling method investigated a total of 70 ground plots (20 m x 20 m) , in which per-tree field survey was made, was conducted for the study. After filtering out 10 plots through data exploration procedure and spatial correlation analysis, the optimal linear regression model for the estimates of volume per hectare was established by linking mean SAVI and average timber volume per hectare of 60 plots. Referencing to IPCC formula and optimal conversion factors from domestic researches, the amount of carbon sequestration of Fg at each plantation patch and in the whole experimental area was calculated and evaluated. The results show that the optimal regression model derived from the methods can explain the variation amounted to 75%, which is comparable to some best results of oversea researches and enhanced much more reliability when compared with the past studies conducted at mountain areas in Taiwan. The estimated average Fg stock volume is 16.93 ± 1.16 m^3ha^(-1), average amount of carbon sequestration is 7.55 ± 0.52 ton ha^(-1) and the total is 1,276.40 ± 87.28 ton on the whole experimental area. In addition, the discussion of this study explores that the use of regression estimation method is superior to direct estimation method from the sample plots and the optimal number of ground plots is 40.

被引用紀錄


張起華(2015)。整合地面樣區調查資料與空間遙測資訊推估森林蓄積量〔碩士論文,國立屏東科技大學〕。華藝線上圖書館。https://doi.org/10.6346/NPUST.2015.00217
徐健榮(2014)。合歡山地區臺灣冷杉枯立木空間分布型態與枯死之影響因素〔碩士論文,國立屏東科技大學〕。華藝線上圖書館。https://doi.org/10.6346/NPUST.2014.00240
魏浚紘(2014)。應用光達技術於人工林之經營與監測〔博士論文,國立屏東科技大學〕。華藝線上圖書館。https://doi.org/10.6346/NPUST.2014.00169
黃翊瑋(2017)。氣候變遷對台灣高山群系之衝擊風險評估〔碩士論文,國立臺灣大學〕。華藝線上圖書館。https://doi.org/10.6342/NTU201703744

延伸閱讀