Keywords城市环境 热消散技术 树木水分利用 环境因素 冠层导度 影响机制 臭氧吸收
Thermal dissipation probe method
tree water use
Full recordShow full item record
Abstract城市植物是城市生态系统的重要组成部分，能提供降温增湿、吸收大气污染物等多种生态服务功能。然而，城市化深刻地改变了城市植物赖以生存的环境条件，进而影响植物的水分代谢。因此，研究城市环境下植物水分利用特征及影响机制对于发挥城市植物生态服务功能、选择适合的绿化树种、建立科学合理的绿地水分管理制度、提高水资源的利用效率、促进城市森林的健康经营具有重要意义。<br />本研究利用热消散技术和气象站对北京典型绿化树种（银杏、七叶树、紫玉兰、刺槐、油松、雪松）的树干液流及其影响因子进行为期三年的连续同步监测，明确了城市环境下典型绿化树种水分利用时间变化规律，详细分析了环境因素（包括水、热和空气污染物）对研究树种的独立影响和交互影响，探讨了研究树种冠层气孔导度对环境因子的响应规律，以便更好地理解环境因子在冠层水平上对其水分利用的影响机理。同时，基于研究树种树干液流测定值计算的冠层气孔导度，结合外界大气中的臭氧浓度，揭示了其臭氧吸收特征及限制因子。最后综合各研究结果对绿化树种的选择和绿地管理提出建议。主要研究结果如下：<br />（1）在昼夜尺度上，银杏、七叶树和紫玉兰树干液流密度多呈窄单峰，峰值出现在9:00－14:00<br />，春夏季节其蒸腾午休现象明显；刺槐、油松和雪松树干液流密度呈宽单峰，峰值出现在9:00－<br />16:00。在年内尺度上，玉兰是春季耗水性树种，银杏、七叶树、雪松、油松、刺槐为夏季耗水型树种；冬季树木液流显著下降。在年际尺度上，同一树种液流年际差异不显著。<br />（2） 把研究树种树干液流点的测定值推广到树干液流平均密度考虑了树干液流的方位变异和径向变异。不同方位每小时液流密度之间高度相关(p &lt; 0.0001)。因此，可以基于这种关系准确地计算其它方位的液流(R2 &gt; 0.91, p &lt; 0.000 1)。油松和雪松液流密度与方位之间的关系较为固定，而刺槐液流密度与方位之间的关系表现出随机性。不同深度树干液流之间密切相关, 因此可以利用较浅处的液流外推其他深度的液流(R2 &gt; 0.89, p &lt; 0.000 1)。结合误差分析，采取北向15 mm和75 mm 深处的液流密度均值来估算整树耗水较为准确。基于研究树种边材面积和冠层投影面积，把树干液流平均密度推广到冠层尺度。冠层蒸腾由小到大依次是刺槐(32.44mm)&lt;油松(171.95 mm)&lt;雪松(166.80 mm)&lt;银杏(235.49 mm)&lt;七叶树(242.93 mm)&lt;紫玉兰(492.74mm)。落叶树种的冠层蒸腾显著高于常绿树种。幼树的冠层蒸腾显著高于老树。<br />（3）研究树种存在夜间液流，最小值出现在凌晨03:00 到06:00。午夜前的夜间液流比午夜后的夜间液流高得多，且波动更大。夜间液流比日间液流值小得多，两者相关关系显著(p&lt;0.001)，表明研究树种的夜间液流用于补充树干白天的水分亏缺。常绿树种的夜间液流与水汽压亏缺等关系显著，解释量高，说明可能存在夜间蒸腾。与常绿树种相比，落叶树种夜间液流对总蒸腾的贡献较高，且波动较大。<br />（4）日尺度上，影响城市环境下树木水分利用的关键因子有大气温度、土壤温度、总辐射、水汽压亏缺和臭氧。小时尺度上，空气相对湿度取代土壤温度成为城市环境下树木水分利用的关键影响因子。研究树种的日液流响应总辐射(Rs)的增加呈直线变化；日液流响应水汽压亏缺(D)的增加呈幂指数变化，拐点不同；油松和刺槐日液流明显受土壤湿度影响。研究树种树干液流密度(Js)的变化均滞后于Rs，提前于D；除了油松，其它树种Js 与Rs 之间时滞范围(10～70 min)比Js 与D 之间时滞范围(47～130 min)短，其中刺槐、油松和雪松种间差异显著；城市树种Js 与蒸腾驱动因子之间的时滞主要受树形（胸径、树高、冠层投影面积、边材面积）以及夜间水分补充量的影响，而与树种无关。为解决环境因子之间共线性的问题，采用方差分解研究水、热和空气污染物三类因子对城市环境下树木水分利用变化的独立作用和交互作用，发现其水分利用大多是由环境因子的交互作用决定，而各因子独立作用很小。因此，根据研究树种树干液流综合城市环境因子的响应规律，建立了简易的树干液流模型。<br />（5）在各个季节研究树种的冠层导度日变化基本均呈单峰曲线，日变化曲线最大峰值为7 月份最高，10 月份最低。研究树种的冠层蒸腾速率呈现由早晨日出后迅速增加，至中午前后达到一天中的最大值、保持到下午15:00 左右，而后再降低的日变化规律。而冠层气孔导度日出后迅速增加，在上午9、10 点左右达到最大值，其后再降低的规律。由于研究树种的冠层导度是控制整树蒸腾的主要因素，我们分析了冠层导度与环境因子之间的关系。太阳总辐射、水汽压亏缺和气温是影响冠层导度的三个重要因素。2009 年生长季节，使用包括太阳总辐射、水汽压亏缺和气温三个变量的Jarvis 模型可以较好地模拟研究树种冠层导度对环境因子的响应特征，不同树种的模型参数有较大差别。<br />（6）生长季节落叶树种（银杏和七叶树）成年树木单位冠层投影面积的平均臭氧吸收速率显著高于常绿树种（油松和雪松）(p=0.002 &lt; 0.05; 平均值: 2.43vs 1.45 nmol m-2 s-1)。但是，两者一年的累积冠层臭氧吸收通量没有显著差异(p=0.065 &gt; 0.05; 平均值: 14.26 vs 10.81 mmol m-2)。研究树种的臭氧吸收速率依赖于外界大气中的臭氧浓度和冠层导度。我们的结果表明不同树种对臭氧的冠层导度不同导致其臭氧吸收的区别。研究树种冠层对臭氧的导度对水汽压亏缺和总辐射的响应相似，这种响应方式导致进入城市树木叶片的臭氧通量更大。根据臭氧对树木造成危害的新临界浓度，城市环境下的落叶树种比常绿树种更容易受到臭氧的危害。除此以外，我们发现基于树干液流测定的树木的年臭氧吸收通量大大低于基于模型的结果，意味着在臭氧风险评估中考虑特定树种的臭氧吸收通量是很必要的。<br />总之，城市环境下植物的水分利用在时间变化规律、环境影响机制、冠层导度的调控等方面都发生了重大的改变以适应城市环境。根据研究树种水分利用的时间变化规律，建议花期适量灌溉、蒸腾高峰期和停滞期勿灌溉，能够减缓耗水。住宅区或者商业区，可考虑种植紫玉兰和七叶树等吸收空气污染物较多的树种。为实现节水和提高空气质量的双重目标，建议刺槐、油松、雪松与银杏等配植。城市环境下植物的水分利用特征与绿化树种的选择、绿地管理有重大关系，对城市环境的影响已不容忽视，更多的研究亟待开展。提出了今后城市环境下植物水分利用研究的重点将主要集中在植物适应城市环境的机理、提高植物的水分利用效率、减缓城市热岛效应等城市环境效应等方面。
Urban trees can provide many ecological services such as temperature and microclimatic modifications, removal of air pollutants, and so forth. However, urbanization has profoundly changed the urban environmental conditions, thereby<br />affecting plant water physiology. Studies on characteristics of water use by typical<br />landscape tree species and its impacting mechanism under urban environment would greatly enhance our ability to improve the urban tree ecosystem service, select proper landscape species, maintain green spaces, and ultimately develope the health management of urban forest.<br />Sap flux density (Js) of selected urban tree species (Ginkgo biloba, Aesculus<br />chinensis, Magnolia denudata, Robinia pseudoacacia, Pinus tabulaeformis, Cedrus<br />deodara) was investigated by use of thermal dissipation method at Beijing Teaching Botanical Garden in Beijing from April, 2008 to April, 2011. The environmental factors were monitored at automatic weather station at the same frequency with sap flow measurements. The objective of this study was to clarify characteristics of water use in urban tree species, to investigate environmental effects on tree water use, to calculate and simulate of the canopy conductance and its regulating mechanism, to indicate ozone uptake by urban tree species and its limiting factors.The hints to tree species selection, configuration and maintenance of urban green space were also disscussed. The main conclusions are as followed:<br />(1) On the diurnal scale, diurnal courses of the sap flux density in G.biloba,<br />A.Chinensis, and M.denudata followed a narrow single peak pattern with the<br />maximum occurring between 900 to 1400 h, adding that in sunny days of summer and spring, diurnal courses of the sap flux density in these species followed a double single peak pattern with obvious depression at noon. While the diurnal pattern in R.pseudoacacia, P.tabulaeformis and C. deodara exhibited the broad single peak pattern with the maximum occurring between 900 to 1600 h. On the annual scale, the sap flux density was much higher during the months of spring season than during the other seasons for M.denudata, while the sap flux density was much higher during the months of summer season than during the other seasons for G.biloba, A.Chinensis,R.pseudoacacia, P. tabulaeformis and C. deodara. On the inter-annual scale, the daily sums of sap flux density (&Sigma;Js) of the study urban tree species in 2008 did not vary significantly from that in 2009, at p &gt; 0.05.<br />(2) Sap flux density of the study species was scaled from single point measurement to averaged sap flux density, based on the knowledge of spatial sap flow profiles in the trunk. Hourly Js over 24 hours at different aspects were highly<br />correlated. Therefore, mean Js may be accurately estimated based on the measurement obtained on one aspect. There was apparent relationship between sap flux density and aspect in P. tabulaeformis and C. deodara), but no apparent relationship in R.pseudoacacia. Hourly Js over 24 hours at different depths were highly correlated, so Js at a particular depth could be extrapolated as a multiple of Js at the depth of 15 mm. In conclusion, sap flux density on the north side at the depth of 15 mm and 70 mm could give a more exact estimation of the whole-tree transpiration. Sap flux density was scaled from averaged sap flux density to the whole tree, based on the knowledge of tree features (sapwood area, canopy area). The canopy transpiration of the selected species was as follows: 32.44 mm in Robinia pseudoacacia, 166.80 mm in Cedrus deodara, 171.95 mm in Pinus tabulaeformis, 235.49 mm in Ginkgo biloba, 242.93 mm in Aesculus chinensis, 492.74 mm in Magnolia denudata). The canopy transpiration in decidous species was significantly higher than that in evergreen species. The canopy transpiration in young trees was significantly higher than that in old trees.<br />(3) The nocturnal sap flow of the study species was substantial, with the minimum sap flux density occurred from 03:00 to 06:00 h each day. During the evening low sap flux density period, the sap flux density was relatively higher and more fluctuating before midnight and afterwards approaching a steady state. The<br />nocturnal sap flow was much lower than daytime sap flow and showed seasonal<br />variation similar to the patterns observed for daytime sap flow(p&lt;0.05), from which it can be concluded the nocturnal sap flow was mainly used for refilling water in the trunk. The vapor pressure deficit was significantly correlated with nocturnal sap flow of evergreen species, adding that they could explain a large fraction of the variance in nocturnal sap flow, suggesting that nocturnal leaf transpiration exsited in evergreen species. Compared with evergreen species, the contribution of nocturnal water recharge to the total daily transpiration was larger and fluctuated more in decidous species.<br />(4) Daily transpiration rate was most affected by air temperature, soil<br />temperature, total radiation, vapor pressure deficit, and ozone. Relative humidity<br />would replace soil temperature when factors influencing hourly transpiration rate was considered. Increases in the average daily total radiation (Rs) led to linear increases in daily sap flux density (&Sigma;Js). There was indication that a plateau-style relationship existed between &Sigma;Js and the average daily vapor pressure deficit (D). Soil water content was a very important factor influencing &Sigma;Js of Robinia pseudoacacia and Pinus tabulaeformis. Js lagged behind Rs and was ahead of D. With the exception of P. tabuleformis, time lags of Js at diameter at the beast height (DBH) with Rs was shorter (10-70 min) than with D (47-130 min). Moreover, the R. pseudoacacia, P.tabulaeformis and C. deodara individuals differed significantly among these species. It was suggested that the time lag was mainly correlated with tree features (DBH, tree height, canopy area, sapwood area) and nocturnal water recharge, regardless of species. To sort out colinearity of the explanatory variables, their individual and joint contributions to variance of tree transpiration were determined by the variation and hierarchical partitioning methods. Majority of the variance in transpiration rates was associated with joint effects of variables in heat and water groups and variance due to individual effects of explanatory group were in comparison small. Atmospheric pollutants exerted only minor effects on tree transpiration. Therefore, a simple model was developed to predict tree sap flow responses to a synthetic environmental variable generated by a principle components analysis.<br />(5) In order to clarify the environmental factors affecting the water use in typical<br />urban tree species Ginkgo biloba, Aesculus chinensis, Magnolia denudata, whole-tree sap flow responses to air temperature, air relative humidity, radiation, wind speed, soil temperature, soil water content, and precipitation were investigated. 8 environmental factors affecting whole-tree water use can be divided into 3 categories including the evaporative demand index, the soil index, and the rain index. The evaporative demand index（air temperature, air relative humidity, total radiation, wind speed, vapour pressure deficit）could explain 60% of the variation in sap flow of individual trees.Sap flow reached an asymptote where higher light and evaporative demand could not cause sap flow to increase further. Soil index（soil temperature, soil water content）and rain index（rainfall）had little influence on sap flow. These results can provide references for selecting reasonable parameters for sap flow models and compare the water use characteristics among urban tree species.<br />(6) Diurnal course of canopy stomatal conductance (Gc) in the selected species<br />exhibited consistently the mono-peak patterns during the year. The maximum Gc<br />appeared in July and the minimums in the late October. The canopy transpiration (Ec)showed rapid increase after sunrise, reached the maximum around the noon and kept steady, then reduced. While Gc rapid increase after sunrise, reached the maximum around 0900h, and then reduced rapidly. As Ec of the selected species was strongly controlled by Gc, we analyzed the relationships between Gc and environmental parameters. The total radiation, vapor pressure deficit and air temperature were the most important factors affecting Gc. A Jarvis-type model was then set up to simulate the response of the Gc to the environmental factors. It performed well to express the variation observed in Gc. The model parameters of different species are very different.<br />(7) After being scaled to the ground area, the average O3 uptake rate by adult<br />trees of two deciduous species (Ginkgo biloba, Aesculus chinensis) was significantly greater than that by adult trees of two evergreen species (Pinus tabulaeformis, and Cedrus deodara) (p=0.002 &lt; 0.05; mean: 2.43 vs 1.45 nmol m-2 s-1), during the growing season. However, the difference of the accumulated stomatal ozone flux for the entire year between deciduous and evergreen species was not significant (p=0.065 &gt; 0.05; mean: 14.26 vs 10.81 mmol m-2). The O3 uptake rate depends on both ambient O3 concentrations and canopy conductance. Our results indicated that the difference in O3 uptake rate was due to the different canopy conductance to ozone (GO3) between the contrasting species. The responses of GO3 to vapour pressure deficit (D) and total radiation (RS) were similar among the tree species, which may lead to the relative high flux of O3 into the leaves of urban trees. Our results showed that urban trees of deciduous species are more susceptible to O3 damage than trees of evergreen species,<br />according to the new critical levels for O3 effects on tree. In addition, we found the<br />annual O3 uptake by trees estimated by our sap flow measurements is greatly less than that from modeling approaches. This difference may suggest the necessity to consider the species- specific flux in O3 risk assessment. In conclusion, water use by landscape tree species under urban environment has<br />been dramatically changed in the temporal variation, environmental impacting<br />mechanism, canopy conductance regulation to adapt the environment change. Based on the temporal variation of water use in study species, flowering period with proper amount of irrigation, transpiration peak and stagnation period without irrigation, are suggested to relieve water shortage. In residential or commercial area, we may consider planting such as Ginkgo biloba, Aesculus chinensis trees which can absorb more air pollutants than other species. To save water and improve air quality, we propose to plant Robinia pseudoacacia, Pinus tabulaeformis, Cedrus deodara with Ginkgo biloba.Water use by landscap tree species under urban environment connected to tree species selection and green space management should not be ignored, and additional research should be conducted. A few important fields for the future researches on plant water use under urban environment have been pointed out, such as the adaption mechanism of tree water use to changing environment, the improvement on water use efficiency in landscape sepecies, the urban heat island effect modification, and so forth.