【作者】 王翼龙；

【导师】 刘国彬； 杜盛；

【作者基本信息】 中国科学院研究生院（教育部水土保持与生态环境研究中心） ， 生态学， 2010， 博士

【摘要】 黄土高原半干旱地区生态系统非常脆弱。严重的水土流失与干旱是该地区面临的两大主要生态问题。植被的可持续恢复是该地区治理水土流失与保障区域生态安全的根本性途径。针对该地区生态恢复中植被重建与水分不足的矛盾,本文以黄土高原半干旱区两种典型林分辽东栎天然次生林和刺槐人工林为研究对象,在延安市公路山林区,应用LI-6400便携式光合测定系统对两典型林分主要树种叶片光合、蒸腾特性及其影响因素进行了研究。利用TDP树干边材液流测定系统和自动气象站对两典型林分各树种的单木耗水特性及相应环境因子进行了定位观测;系统分析了两典型林分各主要树种树干边材液流通量密度的影响因素;估算了2008年生长季两典型林分的蒸腾耗水量。主要结果如下:1、两典型林分主要树种刺槐和辽东栎光合日进程在生长季内晴天呈双峰或单峰型曲线,两主要树种均出现明显的“光合午休”现象。造成两树种光合“午休”的原因包括气孔因素和非气孔因素。影响两树种光合“午休”的生态因子主要包括高光强、高水蒸汽压差和高温。刺槐最大光合能力高于辽东栎的最大光合能力,但是日平均光合速率却小于辽东栎生长季内日平均光合速率。两树种叶片净光合速率与气孔导度、胞间CO2浓度、光合有效辐射强度、空气温度、CO2浓度、空气水蒸汽压差和空气相对湿度极显著或显著相关。影响两树种净光合速率的主导因素是生理因子(Gs和Ci),生态因子可能通过影响两树种叶片的Gs和Ci而间接影响树木叶片光合作用。光响应曲线分析结果显示,随土壤干旱加重,两树种叶片固碳能力及利用弱光和强光的能力均降低。2、在生长季刺槐和辽东栎叶片蒸腾速率日变化呈单峰型曲线或不规则多峰型曲线。两树种叶片蒸腾速率日进程的季节差异可能与两树种不同的生物学特性有关,也可能是生长季不同时期各生态因子的差异所致。两树种叶片的蒸腾速率在生长季不同时期分别与气孔导度、光合有效辐射强度、空气温度、空气相对湿度和空气水蒸汽压差有着极显著或显著的相关关系。生长季不同时期影响两树种叶片蒸腾速率的生理生态因子不同。3、刺槐人工林和辽东栎天然次生林各树种边材液流通量密度日变化在晴天具有与太阳辐射和水蒸汽压差相似的日变化趋势。两典型林分主要树种(刺槐、辽东栎和山杏)不同径级树木间平均边材液流通量密度均存在明显差异。各样树生长季内平均边材液流通量密度与相应胸径间均无显著的相关关系,两典型林分各主要树种的边材液流通量密度可能受到多个生理生态因子的综合制约。生长季内两典型林分各树种林分平均边材液流通量密度日变化在晴天均表现为典型的单峰型曲线,在多云和阴天则总体呈双峰型曲线。各树种日平均林分边材液流通量密度均表现为晴天>多云天>阴天,可能与不同天气条件下各生态因子的显著差异有关。两典型林分各树种液流启动时间、液流持续时间、平均边材液流通量密度峰值及达到峰值的时间等方面在不同月份差异较大。日最大边材液流通量密度值总体表现为生长季初期(4月)和末期(10月)低而生长盛期高的季节动态特征。各树种边材液流通量密度在生长季大部分时间内受太阳辐射、空气温度、空气相对湿度及空气水蒸汽压差等生态因子的综合影响。生长季不同时期两典型林分各主要树种平均边材液流通量密度与各生态因子间的最优多元线性回归模型均达极显著水平,用多元线性回归模型的各生态因子能够解释大部分树木平均边材液流通量密度的变化。4、利用乘幂曲线模型可以拟合刺槐人工林和辽东栎天然次生林主要树种(刺槐、辽东栎和山杏)的边材面积与胸径间的回归方程,结合林分调查各主要树种单木的胸径可以计算相应树木边材面积。刺槐人工林和辽东栎天然次生林各树种林分平均边材液流通量密度均具有明显的季节动态变化特征。各树种林分平均边材液流通量密度的季节变化可能与生长季不同时期各树种叶片的生物学节律、气象因子以及土壤水分含量的变化有关。两典型林分各月份未测时段的林分蒸腾耗水量可以通过林分各树种蒸腾耗水量与白天日均水蒸汽压差间非线性回归模型的方法进行估算。生长季内5～7月为两典型林分蒸腾耗水量较高时期。2008年生长季内刺槐人工林和辽东栎天然次生林蒸腾耗水总量分别为73.8 mm和127.85 mm;日均林分蒸腾耗水量分别为0.41 mm day-1和0.63 mm day-1。两典型林分相对偏低的日均林分蒸腾耗水量可能与林分的叶面积指数及林分边材面积较小有关。5、除表层0～10cm外,辽东栎天然次生林地的土壤容重高于刺槐人工林地;100cm以下土层两林地土壤容重基本保持恒定。两典型林分的土壤水分动态变化与外界气候变化和林分蒸腾耗水节律紧密相关;0～300cm土层土壤水分变化均较大;300～500cm土层土壤水分保持相对稳定的低值。辽东栎天然次生林地深层(地下300～500cm)平均土壤含水量以及储水量要高于刺槐人工林相应土层的平均土壤含水量与储水量;以乡土树种辽东栎为建群种的天然次生林林下土壤水分状况要优于以外来树种刺槐为主的人工林。更多还原

【Abstract】 The semiarid region of Loess Plateau of China is characterized by its extremely fragile ecosystems. Severe soil erosion and drought are the two major ecological problems in this region. Sustainable restoration of vegetation is the fundamental approach to the control of soil erosion and ensurance of the regional ecological security. According to the contradiction between revegetation and water deficiency occurred during the process of vegetation restoration, two typical forests in the semiarid region of Loess Plateau, i.e. Robinia peseudoacacia plantation and Quercus liaotungensis forests, were selected in this study. In forests of Mt. Gonglushan of Yan’an city, the leaf scale characteristic of photosynthesis and transpiration of the dominant tree species of the two typical forest types and the corresponding influencing factors were studied using the LI-6400 portable photosynthesis system. With the thermal dissipation probe sap flow measuring system and automatic meteorological station, the stem level water use of each tree species of the two typical forests and relevant environmental factors were monitored in situ in the same forests region. The influencing factors of sap flow of each tree species in the two typical forests were analyzed systematically. The stand level water use of the two typical forests in the growing season of 2008 was assessed. The main results are as follows:1. The photosynthesis diurnal courses of the dominant tree species of the two typical forests, i.e. R. peseudoacacia and Q. liaotungensis, were bimodal curve or single-peak curve during clear days of the growing season. Both the two dominant tree species had a significant photosynthetic midday depression phenomenon; and the main reason for this depression of the two tree species included stomatal factor and non-stomatal factor. High solar radiation, high VPD and high temperature are the major ecological factors that induced the photosynthetic midday depression of the two tree species. The maximum net photosynthetic rate of R. peseudoacacia was higher than it of Q. liaotungensis while the daily mean photosynthetic rate of R. peseudoacacia was less than it of Q. liaotungensis. The net photosynthetic rate of R. peseudoacacia and Q. liaotungensis had extremely significant correlation or significant correlation with stomatal conductance, intercellular CO2 concentration, available photosynthetic radiation, air temperature, air CO2 concentration, air vapour pressure deficit (VPD) and air relative humidity during growing season. The physiological factors, including stomatal conductance and intercellular CO2 concentration, were the major factors influencing the net photosynthetic rate of the two tree species. The ecological factors had indirect effect on the net photosynthetic rate of the two tree species by influencing stomatal conductance and intercellular CO2 concentration. The analysis of light response curve of R. peseudoacacia and Q. liaotungensis showed that the ability of carbon fixation and ability of using low light and high light of the two tree species were decreased obviously with the enhancing of soil water stress.2. The diurnal courses of transpiration of leaves of R. peseudoacacia and Q. liaotungensis were single-peak curve or irregular multi-peak curve during clear days of growing season. The seasonal differences of the diurnal courses of transpiration of leaves of the two tree species during growing season may be related to the different biologic characteristics of the two tree species or the differences of ecological factors of the different stages of growing season. The transpiration rate of R. peseudoacacia and Q. liaotungensis leaves had extremely significant correlation or significant correlation with stomatal conductance, available photosynthetic radiation, air temperature, air relative humidity and VPD during growing season. The ecophysiological factors influencing the transpiration rate of the two tree species varied with the different months during growing season.3. The diurnal courses of sap flux density of each tree species of R. peseudoacacia plantation and Q. liaotungensis forests was similar to the diurnal courses of solar radiation and VPD in clear days. The average sap flux density of each diameter at breast height (DBH) class of dominant tree species (R. peseudoacacia, Q. liaotungensis and Arrmeniaca sibirica) of the two typical forests differed among DBH classes significantly. However, we did not find a correlation between DBH and average sap flux density of R. peseudoacacia, Q. liaotungensis and A. sibirica in this study. This implies that sap flux density of each dominant tree species of the two typical forests may be controlled by multiple ecophysiological factors. The diurnal courses of the stand average sap flux density of each tree species of the two typical forests were single-peak curve in clear days and bimodal curve in cloudy days and overcast days during growing season. The daily mean stand average sap flux density of each tree species in different weather conditions was: Clear days > cloudy days > overcast days. This may be resulted from the significant difference of ecological factors among different weather conditions. The daily sap flow starting time, duration, peak value and the time of peak value of each tree species of the two typical forests differed evidently among the different months during growing season. In general, the daily maximum stand average sap flux density was lower at the beginning and the end of growing season and higher in the peak of growing season respectively. The stand average sap flux density of each tree species was influenced by the combined factors of solar radiation, air temperature, air relative humidity and VPD during most of growing season. The stepwise multiple regression model of ecological factors and stand average sap flux density of each tree species of the two typical forests in different months during growing season were all extremely significant. And, ecological factors of the stepwise multiple regression models are able to explain the most variation of the stand average sap flux density of each tree species in different months during growing season.4. The power function model can be used to fit the relationship between sapwood area and DBH of the dominant tree species of R. peseudoacacia plantation and Q. liaotungensis forests. Sapwood area of each tree of the dominant tree species in experimental plot can be calculated by DBH of tree and the corresponding fitted model. The stand average sap flux density of each tree species of R. peseudoacacia plantation and Q. liaotungensis forests were differed obviously among different months of growing season. The seasonal dynamics of stand average sap flux density of each tree species may be related to the differences in leaf phenology, meteorological factors and availability of soil water. The missing stand transpiration data of the two typical forests can be estimated on the basis of the nonlinear regression model between stand level transpiration of each tree species and the mean daily daytime VPD (VPDm) of each month. Stand transpiration of the two typical forests were higher from May to July. Total stand transpiration of R. peseudoacacia plantation and Q. liaotungensis forests during the growing season of 2008 were 73.8 mm and 127.85 mm, respectively. Daily mean stand transpiration for R. peseudoacacia plantation and Q. liaotungensis forests in the region were 0.41 mm day-1 and 0.63 mm day-1, respectively. The relatively lower daily mean stand transpiration for the two typical forests may be related to the relatively low leaf area index and a low ratio of sapwood area to ground area.5. Except for the 0-10cm layer, the soil bulk density of Q. liaotungensis forests was higher than it of R. peseudoacacia plantation. Soil bulk density of the two typical forests remained constant below underground 100cm. The variations of soil moisture of the two typical forests were related with the climate factors changes and stand transpiration dynamics. The variations of soil moisture in 0-300cm soil layer of the two typical forests were higher while the soil moisture of 300-500cm soil layer of the two typical forests kept a relatively stable low value. The mean soil moisture and cumulative soil water storage of underground 300-500cm of Q. liaotungensis forest was higher than the corresponding value of R. peseudoacacia plantation. Soil moisture of Q. liaotungensis forests dominanted by the native tree species was better than it of the plantation which was mainly composed by the exotic tree species (R. peseudoacacia).更多还原