【作者】 欧阳资文；

【导师】 关欣； 曾馥平；

【作者基本信息】 湖南农业大学 ， 生态学， 2010， 博士

【摘要】 退耕还林还草是—项长期复杂的工程,既能改善区域生态环境,也会影响退耕区的社会经济发展,切实可行的后续补偿机制直接关系着退耕还林还草工程的持续性发展。喀斯特峰丛洼地主要分布在我国西南喀斯特南端的斜坡地带,生态环境脆弱,人地矛盾突出,人为干扰强烈,石漠化现象严重,直接威胁到了我国长江、珠江流域的生态安全,生态恢复与重建迫在眉睫。有鉴于此,2002年起实施了大面积的退耕还林还草工程,客观评价其生态综合效应,并对植被土壤的耦合协调状况进行评判,是该区域研究工作中迫切需要解决的现实问题。本文以喀斯特峰丛洼地典型退耕还林还草景观单元即中国科学院亚热带农业生态研究所广西环江古周喀斯特峰丛洼地生态恢复与重建试验示范区为具体研究对象,从小尺度着手,以坡耕地为对照,选择了任豆、香椿、板栗、柑橘、任豆+桂牧1号、桂牧1号和撂荒等7个模式,通过样方设置和植被土壤的全面调查与分析,以退耕还林还草的基础理论为指导,运用经典统计学方法,比较分析了喀斯特峰丛洼地不同退耕还林还草模式的植被和土壤生态效益,用典型相关分析方法探讨了退耕还林还草模式植被与土壤相互作用机理,用主成分分析方法揭示了喀斯特峰丛洼地退耕还林还草的主要影响因子,构建了不同退耕还林还草模式的植被土壤耦合度和耦合协调度模型,筛选出了喀斯特峰丛洼地优化退耕还林还草模式。主要结论如下：(1)结合实际情况和生态意义,采用二元指示物种划分法(TWNINSPAN)在第四级水平上可将退耕区的植物划分为8个不同的群落生态类型,进一步的统计分析表明,退耕区的木本植物共有26科65种,群落多样性和结构性指标小但变异高,生活型组成复杂,退耕5年后,土壤质量比同纬度丘陵红壤高,但远低于同纬度植被发育良好的石灰岩土壤。退耕区处于自然恢复的初级阶段,植被稳定性差,土壤处于低水平的匀质状态。(2)不同退耕还林还草模式的植被效应不同。退耕还林还草模式的时限短仅为5年,生物量很小,大小顺序为任豆+桂牧1号>任豆>香椿>板栗>桂牧1号>撂荒>柑橘,生物量主要分布在目的树种和草本上,各器官分布用材林以树干为主,经济林以树干和树枝最高,所有退耕还林还草模式的物种丰富度、Shannon指数和均匀度均很低,处于一种低物种分布的较高水平状态；不同退耕还林还草模式植物各养分含量、储量及分布格局不同,有机碳和钾总储量的器官分布主要受生物量影响,氮和磷受生物量和含量的双重控制,矿质养分主要受含量的影响,主要养分总储量的分布为碳>氮>钾>磷,矿质养分含量SiO2>CaO>Al2O3>Fe2O3>MgO>MnO。矿质养分的含量远高于同纬度地区红壤上的马尾松林和油松林；从植被综合效应出发,退耕区的7种退耕模式可划分为5类,第一类为任豆+桂牧1号和桂牧1号,第二类为柑橘,第三类为撂荒,第四类为任豆,第五类为香椿和板栗。(3)喀斯特峰丛洼地存在着干旱岩溶环境下土壤含水量高的局部优势,以坡耕地为对照,不同退耕还林还草模式均不同程度地提高了土壤含水量、孔隙度、有机质、氮、磷、钾、土壤微生物生物量、微生物种群数量和BC/BN,增加了SiO2的含量,降低了其他矿质养分含量,具有良好的土壤改良效应,可分为土壤显著改良型、土壤改良型、土壤缓慢改良型和土壤改良迟滞型4种类型。土壤的主要影响因子有10个,其理化性状、矿物质特征和微生物的关系密切。(4)喀斯特峰丛洼地不同退耕还林还草模式植物养分的总吸收量、存留量、归还量及不同营养元素的吸收量、存留量、归还量、利用系数、循环系数和周转时间不同,总趋势为利用系数高,循环系数低,周转时间相似,矿质养分的利用系数和循环系数比N、P、K大,周转时间少,生物循环快,不同退耕模式、不同元素存在着相似的生物地球化学循环特征。(5)典型相关分析表明,喀斯特峰丛洼地退耕还林还草模式植被与土壤理化性状、植被与土壤微生物性状、土壤理化与土壤微生物性状、植被地上与地下矿质养分两两之间均存在着显著和极显著的典型相关,土壤速效养分对植物生长和多样性、土壤微生物种群数量对植物生长和发育、土壤微生物种群数量对土壤碱解氮和速效磷、土壤硅对植物钙的影响最大。(6)喀斯特峰丛洼地不同退耕还林还草模式的植被土壤耦合度和耦合协调度模型不同,由于退耕还林还草时间较短,均没有达到良好协调发展状态。进一步采用欧式距离、最远距离法进行综合聚类,可将喀斯特峰丛洼地7种退耕还林还草模式分为4类,各类型内的生态效应相似,不同类型的差异较大。其中第一类为香椿和板栗模式,第二类为柑橘和任豆+桂牧1号模式,第三类为任豆和桂牧1号模式,第四类为撂荒模式。(8)本文从小尺度着手,系统分析了喀斯特峰丛洼地退耕还林还草的植被土壤综合效应及相互作用机理,揭示了其主要影响因子,筛选了优化退耕还林还草模式,构建了植被土壤耦合度和耦合协调度模型,为深入研究我国西南喀斯特区域尺度范围内退耕还林还草工程生态服务功能、构建区域退耕还林还草工程的生态经济耦合度和耦合协调模型、保证该地区退耕还林还草工程的全面实施和促进该区域生态经济的健康发展提供科学的理论和实践依据。更多还原

【Abstract】 As a long-term and complicated ecological project, converting farmland into forest or grassland will locally influence the social and economic development as well as improve the ecological environment conditions. A feasible follow-up compensatory mechanism has a great importance with the project’s sustainable development. Karst peak-cluster-depression region is mainly located in the south slope of karst area in Southwest China, with the problems of fragile ecological environment, sharp conflicts between local and farmland, intensive human disturbance, and severe rocky desertification, which threatened the ecological safety of Yangtse River and Zhujiang River. Therefore, ecological recovery and restoration are urgently needed in karst peak-cluster-depression region. Given this situation, the project of converting farmland into forest or grassland was carried into execution in 2002 in the region. Now the problems that need to be crucially resolved, including evaluation the ecological effects and the coupling conditions of vegetation and soil effects.Taken a landscape unit of conversion of cropland to forest or grassland in karst peak-cluster-depression (i.e. the Guzhou Ecological Recovery and Restoration Demonstration Section of Karst Peak-cluster-depression, Institute of Subtropical Agriculture, CAS, Huanjiang, Guangxi) as the studied object and slope farmland as contrast, the paper started with the analysis of the ecological effects of vegetation and soil, and then probed into the reciprocity mechanism of vegetation and soil based on typical correlation analysis, and then demonstrated the main factors of the conversion ecosystems of cropland to forest and grassland, and then established the coupling model and harmonized coupling model of vegetation and soil, at last, by the cluster analysis, the paper screened out the optimized conversion models. The main results of the study could be concluded as follows:(1) Two-way indicator species analysis (TWINSPAN) had divided the vegetation into 8 community types in the converting areas at the fourth level, combined with the situation and ecological meaning. There were 65 woody species belonging to 26 families, with low diversity and construction index but great variation, and with complex life forms. The soil quality was better than the red soil at the same latitude, but much worse than the limestone soil covered with vegetation at the same latitude. The ecosystem of converting areas was in the primary stage of natural recovery with weak vegetation stability and soil low-level spare status.(2) Vegetation effects differed in different conversion models. The biomass was small due to only 5 years, in the order of Zenia insignis+Guimu No.1 Forage, Zenia insignis, Toona sinensis, Castanea mollissima, Guimu No.1 Forage, leaving land uncultivated, and Citrus reticulate models. The biomass mainly distributed on the aimed trees and grass, while firewood forest mainly on stocks and economic forest mainly on stocks and branches. The species abundance, Shannon index, and evenness index were low in all the converting models, which suggested the models were in the higher level with less species. The content, storage, and distribution pattern of nutrient elements had different details in the converting models. The storage distribution of organic carbon and potassium in the organs was mainly affected by the corresponding biomass, while that of nitrogen and phosphorus by both the corresponding biomass and content, and that of mineral nutrients by the corresponding content. The storage of main nutrients was in the order of carbon, nitrogen, potassium, and phosphorus. Meanwhile, the content of mineral nutrients was much higher than in Pinus massomiana plantation and Pinus elliottii plantation in the same latitude, in the order of SiO2, CaO, Al2O3, Fe2O3, MgO, and MnO. Given the vegetation integrative effects, the seven converting models could be divided into 5 classes, the models of Zenia insignis+Guimu No.1 Forage and Guimu No.1 Forage as the first class, the model of Citrus reticulate as the second class, the model of leaving land uncultivated as the third class, model of Zenia insignis as the fourth class, the models of Toona sinensis and Castanea mollissima as the fifth class.(3) There was local superiority of high content of soil water in the drought karst condition in karst peak-cluster-depression region. Taken slope farmland as contrast, all the models had some degrees to an increase soil water content, porosity, organic matter, nitrogen, phosphorus, potassium, and soil microbial biomass, the amount of soil microbial populations, the ratio of microbial carbon biomass (BC) to microbial nitrogen biomass (BN), and the content of SiO2, but a decrease in other soil mineral nutrient content. This suggested that the converting models had favorable effects on soil, and those could be divided into 4 types of soil remarkable improvement, soil improvement, soil slow improvement, and the soil lagging improvement. There were ten soil main key factors, and the relationships between soil physicochemical properties, mineral properties, and soil microbial were closely.(4) For nutrients in vegetation, the amount of absorption, retention, and return, as well as for nutrient elements, the amount of absorption, retention, and return, use coefficient, recycling coefficient, and turnover time differed in details in the seven models. The whole trend performed as great use coefficient, low recycling coefficient, and close turnover time. Mineral nutrients had greater use coefficient and recycling coefficient, fewer turn over time, and faster biological cycle than nitrogen, carbon, and potassium. This indicated that there were similar recycle characteristics of biogeography chemical cycle in the models and the elements, respectively.(5) There were remarkable and significant remarkable canonical correlations between vegetation and soil physicochemical properties, between vegetation and soil microbial properties, between soil physicochemical properties and soil microbial properties, between mineral nutrients in aboveground and belowground, based on the canonical correlation analysis. Soil available nutrients on vegetation growth and diversity, the amount of microbial population on vegetation growth, the amount of microbial population on soil available nitrogen and phosphorus, silicon in soil on calcium in vegetation, had greatest effects, respectively.(6) Both the coupling models and harmonized coupling models differed in the converting models in karst peak-cluster-depression region. The coupling situation was still not good due to the short converting time. Cluster analysis divided the converting models into 4 classes, Toona sinensis and Castanea mollissima models as first class, Citrus reticulate and Zenia insignis+Guimu No.1 Forage models as the second class, Zenia insignis and Guimu No.1 Forage models as third class, the model of leaving land uncultivated as the fourth class. There were great variations among the classes and similar ecological effects within the class.(8) On a small scale, the paper firstly analyzed the vegetation effects and soil effects, as well as the reciprocity mechanism between them, and then found the key factors of the ecosystems, and then screened out the optimized converting models, and last constructed vegetation and soil coupling models and harmonized coupling models of converting farmland to forest or grassland in karst cluster-peak-depression region, Southwest of China. This would provide theoretical and practice guide to evaluation of the ecological service function, construction of ecological and economic coupling model and harmonized coupling model, whole implementation of the converting project, and a nice ecological and economic development in the area.更多还原