【作者】 杨子清；

【导师】 陈平留； 刘健；

【作者基本信息】 福建农林大学 ， 森林经理学， 2012， 博士

【摘要】 森林在应对和减缓全球气候变化中的作用,已得到越来越多人的重视。在我国生态公益林因其特殊的属性,其生态服务功能强弱与人们生活质量息息相关,其在净化大气、固碳释氧等方面发挥着举足轻重的作用。然而长期以来我国生态公益林经营粗放和“一刀切”的管理模式,已经限制了生态公益林生态服务功能的发挥。为此,本文运用3S (GPS/RS/GIS)技术、采用方差分析、多重比较分析等数理统计方法,结合树种(组)、龄组、生境划分森林类型,研究各森林类型的适宜林分密度,通过景观调整和林分调整实现限制性利用(采伐／补植),旨在增强生态服务功能的前提下,优化树种配置,改善林分结构,在收获一定木材的同时提高林分质量。主要研究结论如下：(1)本文运用生态位模型MAXENT结合气候因子、地形因子、土壤养分预测树种(组)的潜在分布并划分生境等级。结果表明：杉木、马尾松、阔叶混交林、针阔混交林潜在分布的预测精度分别达到85.0%,85.1%,80.5%,81.1%。按树种(组)适生生境的面积大小排序为：马尾松>阔叶混>针阔混>杉木,说明马尾松的适生性最强生态位最宽,而阔叶混、针阔混交林次之,杉木适生性最弱对环境因子要求较高生态位最窄。根据树种(组)、龄组、生境划分出16种森林类型。(2)本文采用方差分析和多重比较法研究林分密度对林分生长、林下植被、土壤养分的影响,结果表明：①林分生长方面,林分密度对胸径有极显著影响(P<0.01),对树高、蓄积无显著影响(P>0.05)；②林下植被方面,林分密度对植被种类、灌木高度有极显著影响(P<0.01),对植被盖度有显著影响(P<0.05)；③土壤养分方面,林分密度对土壤有机质、全氮、全磷、全钾含量有显著影响(P<0.05),对pH值无显著影响(P>0.05)(3)本文运用3S技术,采用C-fix模型估算植被净初级生产力NPP；采用修正后的USLE通用土壤流失方程估算土壤侵蚀模数；采用地统计学Kriging法和RBF(径向基函数)法对采样点的土壤养分数据进行空间插值,以此为基础数据,依据《森林生态系统服务功能评估规范》、参考《中国森林生态服务功能评估》,从水源涵养、固碳释氧、保育土壤、净化大气4方面估算生态公益林的生态服务功能价值。结果表明：顺昌县生态公益林NPP (g C·m-2a-1)平均值为708.43,按树种(组)排序：阔叶混(764.08)>针阔混(590.05)>杉木(518.23)>马尾松(474.05)；按龄组排序：成年(744.73)>幼年(683.76)；按生境排序：适生生境(711.84)>一般生境(703.22)；按生态区位排序：一级保护(736.00)>二级保护(726.31)>三级保护(563.74)顺昌县生态公益林土壤侵蚀模数(t-hm-2·a-1)平均值为12.43,按树种(组)排序：针阔混(10.45)<阔叶混(10.96)<马尾松(12.59)<杉木(12.69)；按龄组排序：成年(11.42)<幼年(12.44)；按生境排序：适生生境(11.71)<一般生境(13.00)；按生态区位排序：一级保护(12.11)<二级保护(12.64)<三级保护(13.86)。土壤养分空间插值的RMS(均方根误差)：有机质为16.42,全氮为0.61,全磷为0.16,全钾为2.68。按全国第二次土壤普查土壤养分含量分级标准,顺昌县土壤有机质含量等级以丰富和很丰富为主(15～81g·kg-1),全氮含量等级以中等和丰富为主(0.09～0.29g·kg-1),全磷含量等级以偏低和中等为主(0.05～0.15g·kg-1),全钾含量等级以中等和偏低为主(14～24g·kg-1)。顺昌县生态公益林生态服务功能价值(元·hm-2·a-1)平均为37836.4,按树种(组)排序：针阔混(40185.5)>阔叶混(39173.4)>马尾松(30256.2)>杉木(25428.9)；按龄组排序：成年(33781.4)>幼年(33655.9)；按生境排序：适生生境(40251.0)>一般生境(40024.1)；按生态区位排序：一级保护(38689.5)>二级保护(38251.6)>三级保护(32509.2)。灰色关联度分析表明：净初级生产力、太阳辐射量和土壤侵蚀模数与生态服务功能价值有较强的关联度,是生态服务功能价值的主要影响指标,而土壤有机质和全磷含量对不同森林类型生态服务功能价值的影响相对较小(4)根据(2)和(3)的结论,生态公益林林分密度和树种配置与其生态服务功能价值关系密切。研究表明,各森林类型随着林分密度的增加,生态服务功能价值有先增加后减小的趋势。故本文采用二次多项式方程进行拟合,对其一阶求导即得到各森林类型的适宜林分密度。本文以“增加阔叶树比重、降低斑块破碎化、增强景观异质性”的原则,通过计算机模拟调整生态公益林景观格局优化树种配置,结果表明：适宜林分密度大小,按龄组排序：幼年>成年；按生境排序：一般生境>适生生境；按树种(组)排序(株.hm-2)：杉木(1395～2315)>马尾松(1525～2210)>针阔混(1190～2007)>阔叶混(1455～2105)。景观格局优化调整后,杉木、马尾松面积比重降低,阔叶混、针阔混比重增加(占景观面积的百分比PLAND),斑块破碎化程度明显降低(斑块密度PD、破碎化度FN),景观异质性略有提高(分维数FDI、景观形状指数LSI),斑块之间的连通性和团聚性明显增强(蔓延度CONTAG),说明优化调整后生态公益林景观格局明显改善,生态服务功能得到强化。(5)根据(4)的结论,若小班林分密度高于所对应森林类型的适宜林分密度则需要采伐,若低于则需补植,若相近则需保持。为此,本文以10个样地为例,选择竞争指数、混交度指数、聚集度指数作为林分结构调整(采伐木/补植区选择)的依据。采用角尺度比较林分结构调整前后的变化。根据林道修建费用模型,探讨了集材路线的选择；依据适地适树的原则,对补植树种的选择给出建议。结果表明：经采伐或补植后,角尺度分布以向左偏移为主即林木趋于均匀分布。更多还原

【Abstract】 Dealing with and mitigating global climate change, the forest has been paid more and more attention. Because the ecological property, the ecosystem services of China’s ecological forest are closely related to the quality of people’s life. And ecological forest plays a pivotal role in the following aspects such as air purification, carbon fixation and oxygen release, et al. But, for a long time, extensive management of ecological forest and "one size fits all" management mode, has limited ecological forest ecosystem services. Therefore, this paper, based on3S (GPS/RS/GIS), used mathematical statistical method such as variance analysis and multiple comparisons analysis to divide forest type according to tree species (group), age group and niche; analyzed the optimal stand density of forest type, technology and proposed the restrictive utilization (thinning/replanting) of ecological forest through landscape adjustment and stand adjustment, aimed at enhancing the ecosystem services, optimizing the species configuration, improving stand structure and harvesting a certain amount of timber while improving stand quality. The main conclusions are as follows:(1) In this paper, the ecological niche model MAXENT, combined climatic factors, topographic factors, soil nutrient factors, was used to predict the potential distribution of tree species (group) niches and the division of the niches suitability. The results showed that:the prediction accuracy of potential distribution of C. lanceolata, P. massoniana, broadleaved forest, mixed wood was85.0%,85.1%,80.5%,81.1%, respectively. The descending order of suitable niche area:P. massoniana> broadleaved forest> mixed wood> C. lanceolata, indicating that P. massoniana’s adaptability is strongest and its niche width is widest, following by broadleaved forest and mixed wood. C. lanceolata has weakest adaptability and its niche is narrowest. Forest type was classified as16types, according to tree species (group), age group and niche.(2) This paper used variance analysis and multiple comparisons to analyze the impact of stand density on stand growth, understory vegetation and soil nutrients. The results showed that:①In the aspect of stand growth, stand density had significant effect on DBH (P<0.01), though had no significant effect on tree height, stand volume (P>0.05);②In the aspect of understory vegetation, stand density have significant effect on vegetation types, shrub height (P<0.01), and on vegetation coverage (P<0.05);③In the aspect of soil nutrients, stand density have a significant effect on soil organic matter, total nitrogen, total phosphorus, total potassium content (P<0.05), though had no significant effect on pH (P>0.05).(3) This paper, based on3S technology, used C-fix model to estimate vegetation net primary productivity(NPP); and used the revised Universal Soil Loss Equation(USLE) to estimate soil erosion modulus; and made spatial interpolation of the sampling points of soil nutrient by geostatistical Kriging method and radial basis function (RBF) method, which as the basic data for estimating the ecological forest ecosystem services value, according to 《The Criterions of Forest Ecosystem Services Assessment》 and ((The Assessment of China’s Forest Ecosystem Services》, from the aspects of water conservation, carbon sequestration and oxygen release, soil conservation and air purification. The results showed that:The mean value of ecological forest NPP in Shunchang (g C m-2a-1) is708.43, sorted by tree species (group):broadleaved forest (764.08)> mixed wood (590.05)> C. lanceolata (518.23)> P. massoniana (474.05); sorted by age group:mature (744.73)> young (683.76); Sort by niche:suitable (711.84)> ordinary (703.22); sorted by ecological position:1st class protection (736.00)>2nd class protection (726.31)>3rd class protection (563.74).The mean value of soil erosion modulus(t-hm-2·a-1) of ecological forest in Shunchang is12.43, sorted by tree species (group):mixed wood (10.45)<broadleaved forest (10.96)<P. massoniana (12.59)<C. lanceolata (12.69); sorted by age group:mature (11.42)<young (12.44); Sort by niche: suitable (11.71)<ordinary (13.00); sorted by ecological position:1st class protection (12.11)<2nd class protection (12.64)<3rd class protection (13.86).The RMS (root mean square error) of spatial interpolation of soil nutrient is:organic matter16.42, total nitrogen0.61, total phosphorus0.16, total potassium2.68. According to soil nutrient content grading standards of2nd nationwide general soil survey, the grade of soil organic matter content mainly is rich and very rich (15～81g·kg1), the grade of total nitrogen content mainly is medium and rich (0.09～0.29g·kg-1), the grade of total phosphorus content mainly is low and middle (0.05～0.15g·kg-1), the grade of total potassium mainly is middle and low (14～24g·kg-1). The mean value of ecological forest ecosystem services in Shunchang (yuan·hm-2·a-1) is37836.4, sorted by tree species (group):mixed wood(40185.5)> broadleaved forest (39173.4)> P. massoniana (30256.2)> C. lanceolata (25428.9); sorted by age group:mature (33781.4)> young (33655.9); sorted by niche:suitable (40251.0)> ordinary (40024.1); sorted by ecological position:1st class protection (38689.5)>2nd class protection (38251.6)>3rd class protection (32509.2).Grey relational analysis showed that:net primary productivity, solar radiation and soil erosion modulus has a stronger correlation with the ecosystem services value, which are the main influence indices of ecosystem services value, while soil organic matter and total phosphorus content has relatively small impact on ecosystem services value of forest types.(4) According to (2) and (3) conclusions, the stand density and species configuration are closely related to ecosystem services value in ecological forest. Results showed that:There is a trend that ecosystem services value increases with the stand density increasing and then decreases in each forest type. Therefore, this paper used quadratic polynomial equation to fit, and its first order derivation is optimal stand density of forest type. Based on the principle of "increasing the proportion of broadleaved forest, reducing patch fragmentation and enhancing landscape heterogeneity", this paper simulated the adjustment of landscape pattern and the optimization of species configuration of ecological forest by computer. The results showed that:The optimal stand density, sorted by age groups:young> mature; sorted by niche:ordinary> suitable; sorted by tree species (group)(strain·hm-2):C. lanceolata (1395～2315)> P. massoniana (1525～2210)> mixed wood (1190～2007)> broadleaved forest (1455～2105).After optimimal adjustment of landscape pattern, the area proportion of C. lanceolata, P. massoniana reduced, while broadleaved forest, mixed wood increased (PLAND), the degree of patch fragmentation was significantly reduced (PD, FN), a slight increase of landscape heterogeneity (FDI, LSI), the connectivity and reunion between the patches were significantly enhanced (CONTAG), indicating the landscape pattern of ecological forest has significant improved after optimal adjustment and ecosystem services have been strengthened.(5) According to (4) conclusion, the stand density of patch is higher than the corresponding forest types’optimal stand density will need thinning, if less than need replanting, if similar is required to maintain. This paper, taking10plots as example, selected competition index, mingling index, aggregation index as the basis of stand structural adjustment (selection of logging tree or replanting area). This paper selected uniform angular index to compare stand structure changes before and after optimal adjustment. Based on forest road construction cost model, this paper analyzed the skidding routes; according to the principle of matching tree species with site, some recommendations about the choice of replanting trees are given. The results showed that:after thinning or replanting the uniform angular distribution shifted to left i.e. trees distribution tend to be uniform after optimal adjustment.更多还原