發(fā)布時間：2018-06-29 19:27

  本文選題：中亞熱帶天然闊葉林 + 最大受光面��； 參考：《中國林業(yè)科學(xué)研究院》2016年博士論文

【摘要】：中國中亞熱帶天然闊葉林是世界上較為罕見的植被類型,是我國植物資源最豐富的地區(qū)之一,研究其林層劃分和特征對該地區(qū)森林的保護(hù)與利用、經(jīng)營和管理具有重要的指導(dǎo)意義。為了揭示中亞熱帶天然闊葉林分自然分層規(guī)律,以典型和次典型中亞熱帶天然闊葉林為實驗對象進(jìn)行林層劃分,并在此基礎(chǔ)上探討各林層直徑分布、樹高胸徑關(guān)系、林分測樹因子特征和蓄積估計等。(1)林層劃分。本文在嘗試了剖面圖法、TSTRAT法、LMS法和聚類分析法后,根據(jù)中亞熱帶天然闊葉林喬木層的自然分異特征,提出林層定量劃分新方法-最大受光面法,揭示中亞熱帶天然闊葉林林層的自然分異規(guī)律。采用最大受光面法將典型林分(1-5號標(biāo)準(zhǔn)地)劃分為3個亞層,第Ⅱ亞層的下限值分別為17.0m,16.5m,17.0m,17.0m,16.0m;第Ⅰ亞層的下限值分別為25.0m,27.0m,25.0m,22.9m,25.0m。最大受光面法將次典型林分(6號和7號標(biāo)準(zhǔn)地)分為2個亞層,第Ⅰ亞層的下限值分別為17.0m和16.5m。7塊標(biāo)準(zhǔn)地林層劃分結(jié)果與剖面圖上判斷的林層分層結(jié)果相近,同時也符合國標(biāo)(GBT 26424-2010)中林層劃分規(guī)定的各項指標(biāo)。與剖面圖法、TSTRAT法、LMS法和聚類分析法相比,最大受光面法能較好的反映中亞熱帶天然闊葉林自然分層特征,分層結(jié)果與現(xiàn)地按林木能否接受直射光情況做出的林層歸屬初步判斷基本一致,也符合相關(guān)國家標(biāo)準(zhǔn)中的規(guī)定。該方法從是否接受直射光和接受直射光的高度差異為劃分依據(jù),反映林木之間對光照和空間資源競爭,具有一定的生物學(xué)意義。(2)典型林分直徑分布。劃分林層后,采用Shapiro-Wilk(S-W)檢驗直徑分布是否服從正態(tài)分布;偏度(SK)和峰度(KT)分析圖形形狀特征;Meyer負(fù)指數(shù)函數(shù)和Weibull分布函數(shù)分析林層直徑分布規(guī)律,并用卡方檢驗直徑分布是否服從Meyer負(fù)指數(shù)函數(shù)和Weibull分布函數(shù)。典型林分中,所有標(biāo)準(zhǔn)地的全林分、第Ⅲ亞層和第Ⅱ亞層直徑分布S-W檢驗的W值均小于0.05,說明全林分、第Ⅲ亞層和第Ⅱ亞層直徑分布均不服從正態(tài)分布;在第Ⅰ亞層中,1號-3號標(biāo)準(zhǔn)地的W值大于0.05,所以服從正態(tài)分布,而4號和5號標(biāo)準(zhǔn)地的W值小于0.05,所以不服從正態(tài)分布;所有標(biāo)準(zhǔn)地內(nèi)S-W檢驗的W值隨亞層平均高的增加而增大,表明亞層直徑分布隨亞層高度的增大而趨向正態(tài)分布。各亞層直徑分布的偏度(sk)和峰度(kt)隨亞層高度的增大而減小,5個標(biāo)準(zhǔn)地第Ⅰ亞層的偏度和峰度的絕對值最小,說明直徑分布圖形在向正態(tài)分布過渡,驗證了s-w的檢驗結(jié)果�？ǚ綑z驗結(jié)果表明,全林分直徑分布中,除2號標(biāo)準(zhǔn)地服從meyer負(fù)指數(shù)分布函數(shù)外,其余4塊標(biāo)準(zhǔn)地均不服從;除1號標(biāo)準(zhǔn)地外,其余4塊標(biāo)準(zhǔn)地均服從weibull分布函數(shù),形狀參數(shù)c1,表明全林分直徑分布為反“j”型曲線;第Ⅲ亞層直徑分布中3號-5號標(biāo)準(zhǔn)地服從meyer負(fù)指數(shù)函數(shù)分布,5塊標(biāo)準(zhǔn)地第Ⅲ亞層直徑分布均服從weibull函數(shù)分布;第Ⅲ亞層直徑分布與全林分直徑分布類似,不同的是第Ⅲ亞層徑階跨度較小,分布像是截尾的反“j”型曲線。第Ⅱ亞層直徑分布中,2號和4號標(biāo)準(zhǔn)地服從meyer負(fù)指數(shù)函數(shù)分布,1號、3號和5號不服從;5塊標(biāo)準(zhǔn)地第Ⅱ亞層直徑分布均服從weibull分布函數(shù),其形狀參數(shù)c處于1-3.6之間,表明其為右偏山狀曲線;5個標(biāo)準(zhǔn)地的第Ⅰ亞層直徑分布均服從meyer負(fù)指數(shù)函數(shù)和weibull分布函數(shù)�？傮w上看,weibull分布函數(shù)在擬合中亞熱帶天然闊葉林各林層直徑分布時具有更好的適應(yīng)性。(3)次典型林分直徑分布。在次典型林分中(6號和7號標(biāo)準(zhǔn)地),6號標(biāo)準(zhǔn)地的全林分直徑分布較為復(fù)雜,像是反“j”型曲線和右偏山狀曲線的結(jié)合,其不服從正態(tài)分布、meyer負(fù)指數(shù)函數(shù)和weibull分布函數(shù);7號標(biāo)準(zhǔn)地為典型的反“j”型曲線,服從meyer負(fù)指數(shù)函數(shù)和weibull分布函數(shù),不服從正態(tài)分布。兩個標(biāo)準(zhǔn)地的第Ⅰ亞層不服從正態(tài)分布和meyer負(fù)指數(shù)函數(shù),服從weibull分布函數(shù),參數(shù)c處于1-3.6之間,說明其為右偏山狀曲線。6號標(biāo)準(zhǔn)地第Ⅱ亞層不服從正態(tài)分布和meyer負(fù)指數(shù)分布,服從weibull分布函數(shù);7號標(biāo)準(zhǔn)地第Ⅱ亞層服從meyer負(fù)指數(shù)函數(shù)和weibull分布函數(shù),不服從正態(tài)分布。s-w檢驗的w值與典型林分類似,隨亞層平均直徑的增加而增大。(4)典型和次典型林分樹高胸徑關(guān)系。選擇schumacher式(簡稱s式)和curtis式(簡稱c式)對各林層樹高胸徑進(jìn)行擬合,結(jié)果表明c式具有較高的r2和較低的mase、amr,故采用c式分析各林層樹高直徑。用全林分樹高胸徑模型估算第Ⅰ、Ⅱ亞層的樹高并用曲線散點圖分析,結(jié)果發(fā)現(xiàn)在典型林分中,第Ⅰ、Ⅱ亞層林木的樹高胸徑關(guān)系不是很顯著,很難用普通模型來表現(xiàn),如果采用全林分樹高胸徑模型擬合典型林分第Ⅰ、Ⅱ亞層中林木產(chǎn)生amr比采用各亞層單獨的樹高胸徑曲線得到的大。證明了如果單純使用全林分樹高模型來估計3層結(jié)構(gòu)的典型林分中的上層林木樹高,都將會產(chǎn)生較大誤差。而對于只有2層結(jié)構(gòu)的次典型試驗林分,全林分樹高曲線估計不同亞層樹高的誤差較小,基本可用全林分樹高來研究各亞層的樹高。(5)典型和次典型林分主要測樹因子。采用標(biāo)準(zhǔn)差和變異系數(shù)研究各林層平均胸徑、平均高,計算各亞層株數(shù)和蓄積有全林分的比重。結(jié)果表明全林平均胸徑與第Ⅱ亞層接近,全林分的平均胸徑和平均高的變異系數(shù)都顯著高于各亞層。各亞層平均胸徑的變異系數(shù)隨亞層高度的減小總體上略有下降,第Ⅰ、Ⅱ亞層平均高的變異系數(shù)相似且小于第Ⅲ亞層。林分中第Ⅰ、Ⅱ亞層株數(shù)之和只占全林分的20%-30%,但蓄積卻是全林分的90%,采用亞層平均高或亞層中值高代替樹高計算全林分蓄積。采用相對誤差分析和方差分析驗證這三種方法得出的蓄積,結(jié)果表明采用亞層平均高計算各林層蓄積的相對誤差均在5%以內(nèi),層中值高全林分和第Ⅰ、Ⅱ亞層的相對誤差在5%以內(nèi),第Ⅲ亞層的誤差在10%以內(nèi),方差分析表明這三種方法結(jié)果之間沒有顯著差異,證明采用各亞層平均高或亞層中值高代替林木樹高計算林分蓄積是可行的。在次典型林分中,與典型林分相似包括各林層平均胸徑和平均高的標(biāo)準(zhǔn)差和變異系數(shù)、蓄積結(jié)構(gòu)。不同點包括平均胸徑不與各亞層相近;各亞層的株數(shù)比例,6號標(biāo)準(zhǔn)地第Ⅰ亞層(受光層)的株數(shù)比例超出全林分50%,與典型林分差異較大,7號標(biāo)準(zhǔn)的第Ⅰ亞層株數(shù)比例為25%,與典型林分相似。
[Abstract]:The natural broad-leaved forest in the middle and subtropical regions of China is one of the most rare vegetation types in the world. It is one of the most abundant plant resources in China. It is of great guiding significance to study the division and characteristics of the forest layer and the management and management of the forest in the region. A typical medium subtropical natural broad-leaved forest is divided into the forest layer, and on this basis, the distribution of the diameter of each forest layer, the relationship between the height of the tree, the characteristics of the tree factor and the estimation of the accumulation of the trees. (1) the forest layer is divided. In this paper, the natural broad-leaved forest in the middle subtropics is based on the section drawing, the TSTRAT method, the LMS method and the cluster analysis. The natural differentiation characteristics of the arbor layer, a new method of the forest layer quantitative division - the maximum light surface method, is proposed to reveal the natural differentiation of the natural broad-leaved forest in the middle subtropics. The typical stand (No. 1-5 standard place) is divided into 3 sublayers by the maximum light surface method. The lower limit of the second sublayer is 17.0m, 16.5m, 17.0m, 17.0m, 16.0m, and the first sublayer. The lower limits are 25.0m, 27.0m, 25.0m, 22.9m, and the maximum light surface method of 25.0m. is divided into 2 sublayers (No. 6 and No. 7). The lower limit of the first sublayer is 17.0m and 16.5m.7, respectively, the result is similar to that of the forest layer, which is judged by the section map, and also conforms to the national standard (GBT 26424-2010) in the forest layer. Compared with the profile method, the TSTRAT method, the LMS method and the cluster analysis, the maximum light surface method can reflect the natural stratification characteristics of the natural broad-leaved forest in the middle subtropics, and the stratification results are basically the same as that in the first step of the forest layer attribution according to whether the trees can receive direct light, but also in accordance with the relevant national standards. The method is based on whether the height difference between receiving direct light and receiving direct light is divided, which reflects the biological significance of light and space resources competition between trees. (2) the distribution of typical stand diameter. After the forest layer is divided, Shapiro-Wilk (S-W) is used to test whether the diameter distribution obeys the normal distribution; the bias (SK) and the peak Degree (KT) analysis of graphic shape characteristics; Meyer negative exponential function and Weibull distribution function analysis of the forest layer diameter distribution law, and using chi square test whether the diameter distribution is subordinate to the Meyer negative exponential function and Weibull distribution function. In the typical stand, all the total stand, the third sublayer and the second sublayer diameter distribution S-W test are all less than 0 of the W values in the typical stand. .05 shows that the total stand, the third sublayer and the second sublayer diameter distribution do not obey the normal distribution; in the first sublayer, the W value of No. 1 -3 is greater than 0.05, so it obeys the normal distribution, and the W value of 4 and 5 is less than 0.05, so the normal distribution is disobedient, and the W value of all the standard ground S-W tests increases with the average height of the sublayer. The distribution of sublayer diameter distribution (SK) and kurtosis (KT) decreases with the increase of sublayer height, and the absolute value of the bias and kurtosis of the 5 sublayer I sublayer is the smallest, indicating the transition of the diameter distribution pattern to the normal distribution, which verifies the test result of S-W. The results of the chi square test showed that, in the diameter distribution of the total stand, except for No. 2, the 4 standard sites were disobedient to the Meyer negative exponential distribution function, and the other 4 standards were all subject to the Weibull distribution function and the shape parameter C1, indicating that the diameter of the whole stand was divided into the anti "J" type curve, and the third sublayer diameter distribution was 3 - No. 5 is subject to the distribution of Meyer negative exponential function, and the distribution of the diameter of the third sublayer of the 5 standard ground sublayers obeys the distribution of the Weibull function; the third sublayer diameter distribution is similar to the total stand diameter distribution, and the third sublayer diameter is smaller, and the distribution is like the truncated anti "J" type curve. In the second sublayer diameter distribution, 2 and 4 standard sites The distribution of Meyer negative exponential function, No. 1, No. 3 and No. 5 are disobedient; the diameter distribution of the second sublayer of the 5 standard ground sublayer obeys the Weibull distribution function, and its shape parameter C is between 1-3.6, indicating that it is the right partial mountain curve; the distribution of the first sublayer diameter of the 5 standard land obeys the Meyer negative exponential function and the Weibull distribution function. In general, Weib The ull distribution function has better adaptability to the distribution of the diameter distribution of the subtropical natural broad-leaved forest. (3) the typical stand diameter distribution. In the sub typical stand (No. 6 and 7 standard), the total stand diameter distribution of the 6 standard land is more complex, such as the combination of the anti "J" curve and the right partial mountain curve, which does not obey the normal state. Distribution, Meyer negative exponential function and Weibull distribution function; No. 7 is a typical anti "J" type curve, obeying Meyer negative exponential function and Weibull distribution function, disobeying the normal distribution. The first sublayer of two standard areas does not obey the normal distribution and the negative exponential function of the Weibull, which is subordinate to the Weibull distribution function and the parameter C is 1-3.6 between the 1-3.6. The second sublayer of the right partial mountain curve.6 No. II sublayer does not obey the normal distribution and the Meyer negative exponential distribution, obeys the Weibull distribution function; the No. 7 sublevel of the standard site No. II obeys the Meyer negative exponential function and the Weibull distribution function. The W value of the.S-w test that does not obey the normal distribution is similar to that of the typical stand, and increases with the increase of the average diameter of the sublayer. (4) The high breast diameter relationship between typical and sub typical stand trees was chosen. Schumacher (s) and Curtis (referred to as C) were selected to fit the high DBH of each forest tree. The results showed that C had higher R2 and lower mase, AMR, so the height diameter of each forest tree was analyzed by C. The height of the first, second subtree was estimated with the high DBH model of the whole forest tree. In the analysis of curve scatter plot, it is found that in the typical stand, the height BBH relationship of the trees in the first and second sublayers is not very significant. It is difficult to use the common model. If the tree height DBH model is used to fit the typical stand, the tree in the sublayer of the sub layer is larger than the single tree height curve of each sublayer. If the total stand tree height model is used to estimate the height of the upper forest tree in the typical stand of the 3 layer structure, there will be great error. For the sub typical test stand with only 2 layer structure, the total stand tree height curve estimates the error of the height of the different subtrees, and the height of the subtree height can be basically used to study the height of each subtree. (5) The average DBH of each stand was studied by standard deviation and coefficient of variation. The average height of each tree was studied with the average height, the number of each sublayer and the proportion of the total stand were calculated. The results showed that the average diameter of the whole forest was close to the second sublayer, and the average DBH and the average height variation coefficient of the whole stand were significantly higher than those of the sublayers. The variation coefficient of the average DBH decreased slightly with the decrease of the sublayer height, and the average height variation coefficient of the first and second sublayers was similar and smaller than that of the third sublayer. The sum of the first and second sublayers in the forest was only 20%-30% of the whole stand, but the accumulation was 90% of the total stand, and the height of sublayer average height or the middle value of sublayer was replaced by the height of the tree. The relative error analysis and variance analysis were used to verify the accumulation of the three methods. The results showed that the relative error of the average height calculation of each forest layer was within 5%, the relative error of the middle layer and the first, the second sublayer was within 5%, the third sublayer was within 10%, and the variance analysis showed that this three was three. There is no significant difference between the results of the method. It is proved that it is feasible to use the average height of each sublayer or the middle sublayer to replace the height of the tree tree. In the sub typical stand, the standard deviation and variation coefficient of the average height and average height of each stand are similar to the typical stand. The proportion of the number of sublayers of each sublayer, the number ratio of the first sublayer (light layer) of the 6 standard area exceeded 50% of the whole stand, and the difference was larger than that of the typical stand. The number of the number I sublayer number of the 7 standard was 25%, similar to that of the typical stand.
【學(xué)位授予單位】：中國林業(yè)科學(xué)研究院
【學(xué)位級別】：博士
【學(xué)位授予年份】：2016
【分類號】：S718.5

【相似文獻(xiàn)】

相關(guān)博士學(xué)位論文 前1條

1 莊崇洋;中亞熱帶天然闊葉林林層特征研究[D];中國林業(yè)科學(xué)研究院;2016年

，

本文編號：2083120