Characteristics of daily sap flow for typical species in Jinyun Mountain of Chongqing in relation to meteorological factors
-
摘要: 运用Granier热扩散探针方法,于2012—2015年8—9月对重庆缙云山自然保护区内3个典型优势木(杉木、马尾松、四川山矾)的树干液流进行测定,并运用微型气象站同步监测太阳辐射(ES)、大气温度(T)、大气相对湿度(RH)、风速(W)、饱和水汽压差(VPD)等气象因子及土壤含水量(SWC),分析3个树种的树干液流在日尺度及典型天气条件(晴、阴、雨)下的差异和特征及其与气象因子的关系。结果表明:树种间导水能力差异表现为四川山矾>马尾松>杉木,阔叶树种蒸腾速率高于针叶树种;3个树种树干液流日变化规律均呈现“昼高夜低”的单峰走势;液流启动时间和达到峰值时间均为山矾最早,杉木最晚;典型天气条件下3个树种液流量均呈现晴天>阴天>雨天,与晴天液流量相比较,阴、雨天液流量减少幅度为41%至86%;白天树干液流贡献率表现为晴天(94.74%~98.04%)>阴天(93.63%~96.71%)>雨天(81.43%~85.43%),夜晚树干液流贡献率表现为雨天(14.57%~18.27%)>晴天(3.29%~6.37%)>阴天(1.96%~5.26%);导致雨天夜间液流贡献率最大的因子为SWC;影响3个树种树干液流的主要气象因子为ES和VPD;T、RH、W对3个树种的影响程度都很小,且略有不同。气象因子与杉木、马尾松、四川山矾的树干液流多元回归方程决定系数分别为0.873、0.873、0.903。Abstract: Using Granier-type thermal dissipation probes,we monitored sap flux density (Fd) of three species (Cunninghamia lanceolata, Pinus massoniana, and Symplocas setchuensis) in August and September during 2012--2015 in Jinyun Mountain of Chongqing, southwestern China. Solar radiation (ES), atmospheric temperature (T), atmospheric relative humidity (RH), wind speed (W), vapour pressure deficiency (VPD) and soil water content (SWC) synchronously with Fd were monitored by mini weather station. Sunny day, cloudy day and rainy day were chosen as three typical weather conditions. We aimed to analyze the differences and characteristics of sap flow among three species and among the three typical weather conditions, and to clarify the relationship between Fd and meteorological factors. The results showed that water transpiration ability of xylem for the three species was ranked as S. setchuensis>P. massoniana>C. lanceolata. Transpiration velocity of broadleaf species was higher than that of coniferous species. The variations of Fd for the three species all displayed single-peaked curves. S. setchuensis was the first one to start transpiration and reach the peak value of Fd, while C. lanceolata was the last. Sap flow for the three species under different weather conditions followed the order of sunny day>cloudy day>rainy day. Compared with sunny day sap flow, the extent of sap flow decreased on cloudy and rainy days ranged from 41% to 86%. The order of contribution rate of diurnal sap flow to whole day transpiration was sunny day (94.74%-98.04%)>cloudy day (93.63%-96.71%)>rainy day (81.43%-85.43%), and that of nighttime sap flow contribution was rainy day (14.57%-18.27%)>sunny day (3.29%-6.37%)>cloudy day(1.96%-5.26%). SWC was the major factor affecting the nighttime sap flow contribution on rainy days. ES and VPD were major meteorological factors affecting Fd. W, T and RH had little effect on Fd , and there were some differences among these three species. The regression models to relate Fd with meteorological parameters can well explain the changes of sap flow, in which coefficients of determination were 0.873, 0.873 and 0.903.
-
林分断面积和蓄积生长模型是林分生长和收获模型体系的重要组成部分[1],是研究森林生长规律的重要基础,在森林资源动态估测等方面有着重要应用价值,为森林经营提供基础的生长模型。众多研究或应用林分断面积或蓄积生长模型中,常用的模型形式有Richards和Schumacher[2−8]两大类。关于自变量的选择,应当包含年龄、密度和立地质量3个变量[9−10]。林分密度反映了林分对林地的利用程度,模型中主要使用的可变密度指标有单位面积林木株数和林分密度指数。大量实践结果表明模型中引入林分密度指数的拟合精度优于引入林木株数的拟合精度[7−9]。立地质量反映了林地的生产潜力,模型常采用立地指数或林分优势木平均高来表示立地质量[11−12]。
第8次全国森林资源清查结果显示,河南省森林面积为359.07万hm2,其中栎类(Quercus spp.)和杨树(Populus spp.)面积所占比例分别为30.97%和28.99%,在全省优势树种中位列前2名[13]。因此,开展这两个树种的断面积和蓄积生长模型研究,进而为评价立地质量提供了参考,为河南省森林资源经营决策提供了科学基础。
1. 数据与方法
1.1 研究区概况
河南省位于我国中东部、黄河中下游,地理坐标31°23′ ~ 36°22′N,112°21′ ~ 116°39′E,处于我国第2阶梯和第3阶梯的过渡地带,属暖温带−亚热带、湿润−半湿润季风气候。年均气温为12.8 ~ 15.5 ℃,年均降雨量为784.8 mm。土地总面积16.7万km2,其中林地面积504.98万hm2,森林覆盖率21.50%。森林植被类型以伏牛山主脉−淮河干流为界,南部属北亚热带常绿落叶阔叶林带,北部属南温带落叶阔叶林带。全省森林资源主要分布于太行山、伏牛山、桐柏山和大别山等山地和丘陵区,以天然阔叶林为主,主要发挥保持水土、涵养水源的生态功能;平原地区森林资源以杨树、泡桐(Paulownia spp.)等为主,主要分布于豫东黄淮海冲积平原和南阳盆地等区域[14]。
1.2 数据来源
本研究所用数据来源于河南省森林资源第6 ~ 8次(2003年、2008年和2013年)连续清查资料,样地形状为正方形,其面积为0.08 hm2。样地的调查因子主要有样地号、优势树种、起源、坡度、海拔、坡向、坡位、土壤(类型和厚度)、腐殖质层厚度、年龄、胸径、树高等以及样地的每木检尺数据。其中,乔木林的平均年龄则采用优势树种平均年龄,而平均树高的调查则是依据平均胸径大小,在主林层优势树种中选择3 ~ 5株平均样木,测定它们的树高,并利用算术平均法获取平均树高。密度指数采用Reineke密度指数,立地等级是根据立地条件划分成的5个等级。划分思路:在林分平均高生长模型中,约束与林分平均高生长过程相关性较高的立地因子(海拔、坡度、坡向、坡位、腐殖层厚度和土壤厚度)而进行树高分级[15]。详细划分方法见参考文献[16]。河南省栎类和杨树林分样地具体数据见表1。
表 1 河南省栎类和杨树样地数据统计Table 1. Summary statistics of sample plots of Quercus and Populus in Henan Province树种
Species调查年份
Survey year样地个数
Sample plot number林分年龄
Stand age林分密度指数
Stand density index林分断面积/(m2·hm− 2)
Stand basal area/(m2·ha− 1)林分蓄积/(m3·hm− 2)
Stand volume/(m3·ha− 1)栎类
Quercus spp.2003 442 5 ~ 110 42 ~ 1 597 0.82 ~ 49.05 2.68 ~ 298.34 2008 445 5 ~ 115 29 ~ 1 248 0.54 ~ 37.22 1.76 ~ 260.34 2013 484 5 ~ 120 29 ~ 1 401 0.56 ~ 39.89 1.85 ~ 279.04 杨树
Populus spp.2003 131 5 ~ 35 92 ~ 964 1.96 ~ 34.58 8.65 ~ 239.95 2008 355 5 ~ 33 45 ~ 959 0.96 ~ 28.42 4.16 ~ 189.70 2013 503 5 ~ 28 76 ~ 1 090 1.64 ~ 33.15 7.28 ~ 238.30 1.3 林分断面积和蓄积生长模型
在生长模型研究中,树木生长理论方程由于具有生物学意义而适用性广泛,主要分为Richards方程、Schumacher方程、Mitcherlich方程、Korf方程、Gompertz方程和Logistic方程等[17]。在自然环境中,林分生长极大程度上取决于林分年龄、林分拥挤程度和林地立地状况。由于林地立地状况比较复杂,因此在基础模型构建中,主要采用这些理论方程形式,建立林分断面积和蓄积与林分年龄和林分密度指数之间的关系。具体基础模型见表2。
表 2 林分断面积和蓄积生长基础模型Table 2. Basic growth model of stand basal area and volumn模型 Model 表达式 Expression Richards (M1) BA(V)=a(1−e−b(SS0)c⋅Age)d Schumacher (M2) BA(V)=ae−bAge(SS0)c Schumacher (M3) BA(V)=ea+bAge(SS0)c+dAge Hyperbola (M4) BA(V)=a⋅(Age⋅S)2(Age⋅S+b⋅Age+cS+d)2 Linear (M5) BA(V)=ea+b⋅Age+cS+d⋅Age⋅S Mitcherlich (M6) BA(V)=a(1−e−b(SS0)c⋅Age) Korf (M7) BA(V)=ae−b⋅Age−c(SS0)d Gompertz (M8) BA(V)=ae−be−c(SS0)d⋅Age Logistic (M9) BA(V)=a1+be−c(SS0)d⋅Age 注:BA、V、Age和S分别为林分断面积、蓄积、年龄和密度指数;a,b,c,d为模型参数,S0取2 000。Notes: BA, V, Age and S are basal area, volumn, age and density index of stand, respectively, and a, b, c, d are parameters of models. And the value of S0 is 2 000. 1.4 哑变量选择
哑变量又称为虚拟变量,常用于处理定性变量,将不能够定量处理的变量量化,达到一个模型同时反映多种情况的作用,对问题描述更简明[18]。在确定林分断面积和蓄积生长基础模型后,引入树种和立地等级作为哑变量,将栎类和杨树不同立地等级上的林分建立一个统一的模型。这样不仅减少了建模工作量,而且使得不同林分的生长模型具有统一形式。因此,引入0,1变量量化定性变量树种和立地等级。
δi={1当种树为栎类或立地等级为i0否则 i=1,2,⋯,5 1.5 模型评价
候选模型的评价指标有决定系数R2、误差偏差E、均方根误差RMSE和总相对误差TRE,计算公式如下:
R2=1−∑i(Vi−ˆVi)2/∑i(Vi−ˉV)2 (1) E=1n∑i(Vi−ˆVi) (2) RMSE=√∑i(Vi−ˆVi)2/n (3) {\rm{TRE}} = {{\displaystyle\sum\limits_i {({V_i} - {{\hat V}_i})} } \mathord{\; \left/ \; {\vphantom {{\displaystyle\sum\limits_i {({V_i} - {{\hat V}_i})} } {\sum\limits_i {{{\hat V}_i}} \times 100}}} \right. \kern-\nulldelimiterspace} {\displaystyle\sum\limits_i {{{\hat V}_i}} \times 100}} (4) 式中:
n 为样本总数,Vi 为第i 个样地林分蓄积(或断面积)实测值,ˆVi 为第i 个样地林分蓄积(或断面积)估计值,ˉV 为所有样地林分蓄积(或断面积)平均值。在评价和比较哑变量模型拟合精度方面,指标有赤池信息量(AIC)、贝叶斯信息量(BIC)、决定系数R2和均方根误差RMSE
AIC=−2lnl+2p (5) BIC=−2lnl+lnn⋅p (6) 式中:p为模型中参数个数,l表示模型极大似然函数值。
针对9个林分断面积和蓄积生长模型开展参数估计和模型拟合精度评价比较,以及考虑树种和立地等级的影响而引入哑变量模型。所有计算在R3.3.2的nls函数[19]和Forstat软件[20]上实现。
2. 结果与分析
2.1 林分断面积和蓄积生长基础模型选择
根据河南省一类清查数据,选择了1 372块栎类样地和989块杨树样地拟合了林分断面积和蓄积生长模型,统计各个模型的参数估计值和评价指标。断面积生长模型结果见表3,蓄积生长模型结果见表4(只列出决定系数R2最大的前3个模型)。
表 3 栎类和杨树林分断面积生长模型参数估计及评价指标Table 3. Parameter estimation and evaluation index for growth model of basal area in Quercus and Populus stand树种 Species 模型 Model 参数 Parameter 评价指标 Evaluation index a b c d E RMSE R2 TRE 栎类
Quercus spp.M1 46.78 0.078 5.514 0.186 0 0.013 9 1.009 0.982 0.557 M2 63.66 3.606 1.042 0.044 5 1.205 0.974 0.796 M3 4.27 − 6.576 1.139 − 2.276 0 − 0.018 1 1.148 0.977 0.722 杨树
Populus spp.M1 33.15 11.843 7.815 0.134 0 − 0.005 5 0.878 0.975 0.415 M2 69.82 1.109 1.040 − 0.009 3 0.906 0.973 0.442 M3 4.24 − 1.100 1.039 0.006 5 − 0.009 3 0.906 0.973 0.442 表 4 栎类和杨树林分蓄积生长模型参数估计及评价指标Table 4. Parameter estimation and evaluation index for growth model of volume in Quercus and Populus stand树种 Species 模型 Model 参数 Parameter 评价指标 Evaluation index a b c d E RMSE R2 TRE 栎类
Quercus spp.M1 396.58 0.015 2.513 0.426 3 0.234 4 11.821 0.925 3.092 M4 2 067 1 879 38.97 − 6 494 0.346 3 12.514 0.916 4.028 M7 396.45 4.168 0.663 0.730 4 − 0.136 7 12.149 0.921 3.761 杨树
Populus spp.M1 244.99 0.667 4.152 0.264 9 − 0.050 2 9.686 0.926 1.447 M2 495.11 2.238 1.078 − 0.077 1 9.940 0.922 1.542 M3 6.21 − 2.289 1.083 − 0.039 4 − 0.079 8 9.940 0.922 1.544 由表3可以看出:栎类和杨树林分断面积生长模型拟合效果整体上良好,决定系数R2都在0.973以上,最高达到0.982;Richards方程、Schumacher方程的两种形式拟合效果在9个模型中位列前3名;无论是栎类还是杨树,Richards方程拟合效果均最佳。由表4可以看出:栎类和杨树林分蓄积生长模型拟合效果良好,决定系数R2都在0.916以上,最高达到0.926;栎类林分蓄积生长模型中,Richards方程、Korf方程和Hyperbola方程拟合效果位列前3名,杨树林分蓄积生长模型拟合效果跟断面积一致,但拟合效果最好的都是Richards方程。从表3和表4结果得出:(1)林分断面积和蓄积生长模型以Richards方程形式最佳,选择模型M1作为基础模型;(2)林分断面积生长模型与蓄积生长模型可以使用相同模型形式。
2.2 哑变量模型构建
考虑树种对林分断面积和蓄积生长模型的影响,将树种作为哑变量,以模型M1为基础模型构建统一模型,分别将哑变量加入到参数a、b、c、d及其组合上。利用ForStat软件计算的结果见表5。
表 5 带树种哑变量模型的评价指标Table 5. Evaluation indices for the different alternatives of models with dummy in species哑变量 Dummy variable 断面积生长模型 Growth model of basal area 蓄积生长模型 Growth model of volume AIC BIC RMSE R2 AIC BIC RMSE R2 a, b, c, d 9 048 9 103 0.949 0.980 14 804 14 858 10.9 0.930 a, b, c 9 059 9 107 0.954 0.980 14 825 14 873 11.0 0.929 a, b 9 058 9 099 0.954 0.980 14 826 14 867 11.0 0.929 a 9 062 9 096 0.955 0.980 14 842 14 876 11.1 0.928 b 9 061 9 095 0.955 0.980 14 841 14 875 11.1 0.928 c 9 308 9 342 1.060 0.975 15 110 15 145 12.4 0.909 d 9 145 9 179 0.989 0.978 14 885 14 919 11.3 0.925 从表5可以得出:不同参数及其组合上哑变量模型拟合效果相当,断面积生长模型与蓄积生长模型表现效果一致;仅参数c或d上考虑哑变量效果较差,而全部参数引入哑变量效果最佳,但这时模型最复杂,并且R2并未有提升。从简化模型角度考虑,最终选择参数a、b上带哑变量的模型。断面积和蓄积生长模型分别为:
BA=(32.87B0+46.83B1)×(1−e−(3.759B0+0.0845B1)(S2000)5.898⋅Age)0.1751 (7) V=(212.4B0+461.6B1)×(1−e−(0.5780B0+0.0974B1)(S2000)2.847⋅Age)0.3835 (8) 式中:
B0=1 ,B1=0 ,表示优势树种为杨树;B0=0 ,B1=1 ,表示优势树种为栎类。在参数a、b上进一步引入立地等级作为哑变量,断面积和蓄积生长模型评价结果见表6。
表 6 带树种和立地等级哑变量模型的评价指标Table 6. Evaluation indices for the different alternatives of models with dummy in species and site class哑变量 Dummy variable 断面积生长模型 Growth model of basal area 蓄积生长模型 Growth model of volume 树种 Species 立地等级 Site class AIC BIC RMSE R2 AIC BIC RMSE R2 a, b a, b 8 869 9 018 0.874 0.983 14 670 14 819 10.2 0.938 a a, b 8 891 8 986 0.885 0.983 14 681 14 775 10.3 0.937 a, b a 8 899 8 994 0.889 0.983 14 687 14 781 10.3 0.937 a b 8 893 8 954 0.888 0.983 b a 8 900 8 961 0.891 0.983 14 693 14 754 10.4 0.937 从表6得出:断面积和蓄积生长模型拟合效果较仅带树种的哑变量模型均有所提升;立地等级加到参数b上效果要好于参数a;参数a、b上同时带树种和立地等级效果最佳。但考虑到BIC以及模型的简化性,最终选择参数a上带树种、参数b上带树种和立地等级作为哑变量的模型(图1)。具体表达式如下:
BA=a(1−e−b(S2000)5.688⋅Age)0.1788a=34.85B0+42.25B1b=2.764B01+2.280B02+2.180B03+1.848B04+1.258B05+0.210B11+0.161B12+0.139B13+0.121B14+0.0870B15 (9) V=a(1−e−b(S2000)2.717⋅Age)0.3935a=218.1B0+309.7B1b=0.5965B01+0.4872B02+0.4725B03+0.4024B04+0.2899B05+0.0417B11+0.0306B12+0.0276B13+0.0238B14+0.0168B15 (10) B0j={1树种为杨树,立地等级为j0否则 B1j={1树种为栎类,立地等级为j0否则 j=1,2,⋯,5 3. 结论与讨论
本文利用河南省森林一类清查数据,建立了栎类和杨树两个树种的林分断面积和蓄积生长模型。首先,在9个适用性强且具有明确生物学意义的备选模型中,利用决定系数R2、平均误差E和均方根误差RMSE指标综合选择拟合效果最佳的具Richards形式的M1模型作为基础模型,栎类林分断面积和蓄积生长模型的决定系数R2分别为0.982和0.925,杨树林分断面积和蓄积生长模型的决定系数R2分别为0.975和0.926;然后,引入树种作为哑变量,用统一的模型来表达断面积和蓄积生长规律,这样不仅减少了建模工作量,而且使不同林分的生长模型具有统一形式;最后,进一步考虑立地质量对模型的影响,同时引入树种和立地等级作为哑变量,断面积和蓄积生长模型精度进一步提高,决定系数R2分别为0.983和0.937。公式(9)和公式(10)可以用来描述河南省栎类和杨树林分断面积和蓄积生长过程,丰富了河南省森林经营的基础生长数据。
采用生长理论方程的优势是参数具有生物学意义,这里Richards方程中参数a表示渐进值,反映林分断面积或蓄积生长能够达到的上限值。从基础模型到带哑变量模型,通过参数a的估计值(断面积:杨树34.85,栎类42.25;蓄积:杨树218.1,栎类309.7)可知杨树林分每公顷断面积或蓄积极限值低于栎类林分。参数b反映林分断面积或蓄积的生长速率,同样通过这些模型参数b的估计值的比较,可知在相同林分密度情况下,早期杨树林分断面积或蓄积生长速率高于栎类林分(图1)。生长速率与立地质量关系密切,通过公式(9)和公式(10)可得出,随着立地等级下降,反映栎类和杨树林分生长速率的参数b也依次降低。事实上,根据样地实际情况,杨树林分平均年龄9年、林分平均密度指数434株/hm2,栎类林分平均年龄23年、林分平均密度指数444株/hm2,计算出杨树和栎类5个立地等级上断面积和蓄积的生长量(表7),发现这些生长量也随着立地等级下降而下降,与参数b表现的规律一致。这些规律与人们的直观经验是吻合的,说明所构建的林分断面积和蓄积生长模型客观上描述了河南省栎类和杨树的生长规律,并具有可靠性。
表 7 栎类和杨树林分断面积和蓄积生长量Table 7. Growth increment of stand basal area and volume in Quercus and Populus stand树种
Species项目
Item林分年龄/a
Stand age/year林分密度指数
Stand density index立地等级 Site class 1 2 3 4 5 杨树 Populus spp. 断面积 Basal area 9 434 0.259 0 0.251 0 0.249 0 0.242 0 0.226 0 蓄积 Volume 3.400 0 3.170 0 3.140 0 2.970 0 2.640 0 栎类 Quercus spp. 断面积 Basal area 23 444 0.094 1 0.089 8 0.087 5 0.085 3 0.080 4 蓄积 Volume 1.030 0 0.916 0 0.880 0 0.831 0 0.726 0 注:断面积和蓄积的单位分别是m2/hm2和m3/hm2。Note: units of basal area and volume are m2/ha and m3/ha, respectively. -
[1] CHENG J, OUYANG X, HUANG D W, et al. Sap flow characteristics of four dominant tree species in a mixed conifer-broadleaf forest in Dinghushan[J] .Acta Ecologica Sinica,2015,35(12):4097-4104.
[1] 程静,欧阳旭,黄德卫,等.鼎湖山针阔叶混交林4种优势树种树干液流特征[J].生态学报,2015,35(12):4097-4104. [2] ZHAO C Y, SI J H, FENG Q, et al. Stem sap flow research: progress and prospect[J]. Journal of Northwest Forestry University,2015,30(5):98-105.
[2] NADEZHDINA N. Sap flow index as an indicator of plant water status[J]. Tree Physiology,1999,19(13):885-891.
[3] COHEN M, GOLDHAMER D A, FERERES E, et al. Assessment of peach tree responses to irrigation water deficits by continuous monitoring of trunk diameter changes[J]. Journal of Horticultural Science and Biotechnology,2001,76(1):55-60.
[3] CHEN B,CHEN L X,LIU Q Q,et al. Transpiration of Pinus sylvestris var. mongolica and its response to urban environmental factors in semi-arid area[J]. Acta Ecological Sinica,2015,35(15):5076-5084.
[4] CHI B, CAI T J, MAN X L, et al. Effects of influencing factors on stem sap flow in Larix gmelinii in northern Da Hinggan Mountains, northeastern China[J]. Journal of Beijing Forestry University,2013,35(4):21-26.
[4] 赵春彦,司建华,冯起,等. 树干液流研究进展与展望[J].西北林学院学报,2015,30(5):98-105. [5] LIU X Z, KANG S Z, LI F S. Simulation of artificial neural network model for trunk sap flow of Pyrus pyrifolia and its comparison with multiple-linear regression[J]. Agric Water Manage,2009,96(6):939-945.
[5] WANG W J, SUN W, QIU L, et al. Relations between stem sap flow density of Larix gmelinii and environmental factors under different temporal scale[J]. Scientia Silvae Sinicae,2012,48(1):77-85.
[6] YU M M, ZHANG X J, YUAN F H, et al. Characteristics of sap flow velocities for three tree species in a broad-leaved Korean pine forest of Changbai Mountain,in relation to environmental factors[J]. Chinese Journal of Ecology,2014,33(7):1707-1714.
[6] ORTUNNO M F, CONEJERO W, MORENO F, et al. Could trunk diameter sensors be used in woody crops for irrigation scheduling? A review of current knowledge and future perspectives[J]. Agricultural Water Management,2010,97(1):1-11.
[7] MA L Y, WANG H T, LIN P. Comparison of water consumption of some afforestation species in Beijing area[J]. Journal of Beijing Forestry University,2003,25(2):1-7.
[7] TYREE M T. A dynamic model for water flow in a single tree: evidence that models must account for hydraulic architecture[J]. Tree Physiology,1988,4(3):195-217.
[8] KÖHLER M, SCHWENDENMANN L, HÖLSCHER D. Throughfall reduction in a cacao agroforest: tree water use and soil water budgeting[J]. Agricultural and Forest Meteorology,2010,150(7):1079-1089.
[8] WANG H T, ZHAO W F, MA L Y. Spatial variation of sap flow of Platycladus orientalis and its affecting factors[J]. Scientia Silvae Sinicae,2006,42(7):21-27.
[9] XU J L, MA L Y, YAN H P. Relationship between process of sap flow of P. tabulaeformis and solar radiation[J]. Science of Soil Water Conservation,2006,4(2):103-107.
[9] SPERRY J S, HACKE U G, OREN R, et al. Water deficits and hydraulic limits to leaf water supply[J]. Plant, Cell Environment,2002,25(2):251-263.
[10] SCHAEFFER S M, WILLIAMS D G, GOODRICH D C. Transpiration of cottonwood/willow forest estimated from sap flux[J]. Agricultural and Forest Meteorology,2000,105(1):257-270.
[10] WANG R H, MA L Y, XI R C, et al. Fluctuation of Acer truncatum sap flow in rapid growth season and relevant variables[J]. Chinese Journal of Ecology,2006,25(3):231-237.
[11] GRANIER A, HUC R, BARIGAH S T. Transpiration of natural rain forest and its dependence on climatic factors[J]. Agricultural and Forest Meteorology,1996,78(1):19-29.
[11] ZHANG H D, WEI W, CHEN L D, et al. Analysis of sap flow characteristics of the Chinese pine in typical Loess Plateau region of China[J]. Environmental Science,2015,36(1):349-356.
[12] ZHANG H, SIMMONDS L P, MORISON J I, et al. Estimation of transpiration by single trees: comparison of sap flow measurements with a combination equation[J]. Agricultural and Forest Meteorology,1997,87(2):155-169.
[12] GUO B N, ZHANG J J, WANG Z, et al. Characteristics of water consumption by Robinia pseudoacacia forest on Loess Plateau of western Shanxi Province, China[J]. Chinese Journal of Ecology,2012,31(11):2736-2741.
[13] GREEN S R. Radiation balance, transpiration and photosynthesis of an isolated tree[J]. Agricultural and Forest Meteorology,1993,64(3):201-221.
[13] MA Q Y, MA J F, KANG F F, et al. Functions of water conduction in sapwood and heartwood of Pinus tabulaeformis in Taiyue Mountain of Shanxi Province, north China[J]. Journal of Beijing Forestry University,2005,27(Suppl.2):156-159.
[14] 陈彪,陈立欣,刘清泉,等. 半干旱地区城市环境下樟子松蒸腾特征及其对环境因子的响应[J].生态学报,2015,35(15):5076-5084. [14] YIN L H, HUANG J T, WANG X Y, et al. Characteristics of night time sap flow of Salix matsudana and Populus simonii in Yulin, Shaanxi[J]. Journal of Northwest A F University(Natural Science),2013,41(8):85-90.
[15] FAN C, QIU Y P, LI Z Q, et al. Relationships between stem sap flow rate of litchi trees and meteorological parameters[J]. Acta Ecologica Sinica,2014,34(9):2401-2410.
[15] 池波,蔡体久,满秀玲,等. 大兴安岭北部兴安落叶松树干液流规律及影响因子分析[J].北京林业大学学报,2013,35(4):21-26. [16] 王文杰,孙伟,邱岭,等. 不同时间尺度下兴安落叶松树干液流密度与环境因子的关系[J].林业科学,2012,48(1):77-85. [16] SUN Z W, ZHAO P, NIU J F, et al. Seasonal variations of sap flow and transpiration water consumption of introduced tree species Acacia auriculaeformis and Eucalyptus citriodora[J]. Atca Ecologica Sinica,2014,33(10):2588-2595.
[17] 于萌萌,张新建,袁凤辉,等. 长白山阔叶红松林三种树种树干液流特征及其与环境因子的关系[J].生态学杂志,2014,33(7):1707-1714. [17] WANG S F, WANG X K, OUYANG Z Y. Path analysis on environmental factors controlling soil organic carbon and total nitrogen contents in the upstream watershed of Miyun Reservoir, North China [J]. Ecology and Environmental Sciences,2014,8(8):1378-1383.
[18] ZHAO Y X, CHEN J D. The path analysis model and application in research of decision degree of ecological factors[J]. Journal of Sichuan Normal University(Natural Science),2007,30(1):120-123.
[18] 马履一,王华田,林平. 北京地区几个造林树种耗水性比较研究[J].北京林业大学学报,2003,25(2):1-7. [19] 王华田,赵文飞,马履一. 侧柏树干边材液流的空间变化规律及其相关因子[J].林业科学,2006,42(7):21-27. [19] WANG Y B, DE Y J, XIONG W, et al. The characteristics of nocturnal sap flow and stem water recharge pattern in growing season for a Larix principis-rupprechtii plantation[J]. Acta Ecologica Sinica,2013,32(5):1375-1385.
[20] 徐军亮,马履一,阎海平. 油松树干液流进程与太阳辐射的关系[J].中国水土保持科学,2006,4(2):103-107. [20] ZHANG Y Y, ZHOU Z F, DANG H Z, et al. A study on the sap flow of Pinus sylvestris var. mongolica with thermal dissipation probe[J]. Research of Soil Water Conservation,2006,13(4):78-80.
[21] WANG X W, MAO Z J. Effects of conducting tissue structure on sap flow density and stem CO2 efflux[J]. Journal of Beijing Forestry University,2013,35(4):9-15.
[21] 王瑞辉,马履一,奚如春,等. 元宝枫生长旺季树干液流动态及影响因素[J].生态学杂志,2006,25(3):231-237. [22] 张涵丹,卫伟,陈利顶,等. 典型黄土区油松树干液流变化特征分析[J].环境科学,2015,36(1):349-356. [22] SUN H Z, LI Y P, WANG C, et al. Comparative study on stem sap flow of non-and ring-porous tree species[J]. Chinese Journal of Ecology,2005,24(12):1434-1439.
[23] CHEN L X, ZHANG Z Q, LI Z D, et al. Nocturnal sap flow of four urban greening tree species in Dalian, Liaoning Province, China[J]. Journal of Plant Ecology,2010,34(5):535-546.
[23] 郭宝妮,张建军,王震,等. 晋西黄土区刺槐林耗水特征[J].生态学杂志,2012,31(11):2736-2741. [24] WANG H, ZHAO P, CAI X A, et al. Partitioning of night sap flow of Acacia mangium and its implication for estimating whole-tree transpiration[J]. Journal of Plant Ecology,2007,31(5):777-786.
[24] 马钦彦,马剑芳,康峰峰,等. 山西太岳山油松林木边材心材导水功能研究[J].北京林业大学学报,2005,27(增刊2):156-159. [25] 尹立河,黄金廷,王晓勇,等. 陕西榆林地区旱柳和小叶杨夜间树干液流变化特征分析[J].西北农林科技大学学报(自然科学版) ,2013,41(8):85-90. [25] HU X B, HAN L, ZHANG D, et al. Analysis on the dynamics of nighttime sap flow of Ulmus pumila and Platycladus orientalis in semi-arid area of Loess Plateau[J]. Science of Soil Water Conservation,2010,8(4):51-56.
[26] YANG Z G, SHI Y, YU X X, et al. Characteristic of stem sap flux velocity of individual trees and its response to environmental factors in the Beijing mountain area[J]. Research of Soil Water Conservation,2012,19(2):195-200.
[26] 凡超,邱燕萍,李志强,等. 荔枝树干液流速率与气象因子的关系[J].生态学报,2014,34(9):2401-2410. [27] 孙振伟,赵平,牛俊峰,等. 外来引种树种大叶相思和柠檬桉树干液流和蒸腾耗水的季节变异[J].生态学杂志,2014,33(10):2588-2595. [28] WULLSCHLEGER S D, MEINZER F C, VERTESSY R A. A review of whole-plant water use studies in trees[J].Tree Physiology,1998,18(8):499-512.
[29] 王淑芳,王效科,欧阳志云. 环境因素对密云水库上游流域土壤有机碳和全氮含量影响的通径分析[J].生态环境学报,2014,8(8):1378-1383. [30] 赵益新,陈巨东.通径分析模型及其在生态因子决定程度研究中的应用[J]. 四川师范大学学报(自然科学版) ,2007,30(1):120-123. [31] 王艳兵,德永军,熊伟,等. 华北落叶松夜间树干液流特征及生长季补水格局[J].生态学报,2013,32(5):1375-1385. [32] 张友焱,周泽福,党宏忠,等. 利用TDP茎流计研究沙地樟子松的树干液流[J].水土保持研究,2006,13(4):78-80. [33] 王秀伟,毛子军. 输导组织结构对液流速度和树干CO2释放通量的影响[J]. 北京林业大学学报,2013,35(4):9-15. [34] 孙慧珍,李夷平,王翠,等. 不同木材结构树干液流对比研究[J].生态学杂志,2005,24(12):1434-1439. [35] 陈立欣,张志强,李湛东,等. 大连4种城市绿化乔木树种夜间液流活动特征[J].植物生态学报,2010,34(5):535-546. [36] 王华,赵平,蔡锡安,等. 马占相思夜间树干液流的分配及其对整树蒸腾估算的影响[J]. 植物生态学报,2007,31(5):777-786. [37] 胡兴波,韩磊,张东,等. 黄土半干旱区白榆和侧柏夜间液流动态分析[J]. 中国水土保持科学,2010,8(4):51-56. [38] 杨芝歌,史宇,余新晓,等. 北京山区典型树种树干液流特征及其对环境因子的响应研究[J].水土保持研究,2012,19(2):195-200. -
期刊类型引用(1)
1. 伊丽茜,岳永杰,张军,舒洋,王雅倩. 额尔古纳市兴安落叶松中龄林植被碳储量研究. 现代园艺. 2021(23): 55-56+74 . 百度学术
其他类型引用(6)
计量
- 文章访问数: 2013
- HTML全文浏览量: 320
- PDF下载量: 28
- 被引次数: 7