Estimation of plant biomass and plant water mass through dimensional measurements of plant volume in the Dund-Govi Province, Mongolia
Estimation of plant biomass and plant water mass through dimensional measurements of plant volume in the Dund-Govi Province, Mongolia
复制标题
DOI：
10.1111/j.1744-697x.2007.00096.x
复制
发表时间：
2007-12-01
影响因子：
1.3
通讯作者：
Hongo, Akio
中科院分区：
文献类型：
--
作者：
Hirata, Masahiro;Oishi, Kazato;Hongo, Akio
研究了三种类型的函数（线性、二次和对数）和自变量（高度、树冠面积/覆盖度和体积），以建立估算植物生物量（灌木和高大草本植物：g干物质[DM][植物](-1)）的最佳回归方程；矮灌木和矮草本植物：g DM [0.25 m（2)](-1)）和植物水分质量(灌木和高草本植物：g H2O[植物](-1)；蒙古敦煌-戈维省干旱区矮灌木和矮草本植物：g H2O [0.25 m（2)](-1)）。研究了5种优势灌木，包括1种矮灌木、1种高草和9种矮草本植物。根据确定系数（r(2)）、计算简便、避免估算误差较大等特点，表明具有量纲体积的线性回归方程最适合估算优势灌木和草本植物的生物量和植物水分。计算残差值（估计值与测量值之间的误差绝对值）和残差比（残差比），验证各植物线性回归方程估计值的准确性。虽然残差值随单位面积植物总生物量（g DM m(-2)）和植物总含水量（g H2O m(-2)）的增加而增加，但残差率随总生物量和总含水量的增加而减小。82.6%的估算植物总生物量和67.4%的估算植物总水量分别落在+/- 40%的残余误差范围内。因此，利用各线性回归方程可以估计出单位面积植物总生物量和植物总水量。
Three types of functions (linear, quadratic and logarithmic) and independent variables (height, crown area/ coverage and volume) were examined to develop the best regression equation for the estimation of plant biomass (shrub and tall herbaceous plant: g dry matter [DM] [ plant](-1); dwarf shrub and short herbaceous plant: g DM [0.25 m(2)](-1)) and plant water mass (shrub and tall herbaceous plant: g H2O [plant](-1); dwarf shrub and short herbaceous plant: g H2O [0.25 m(2)](-1)) by dimensional measurements in the arid area of Dund- Govi Province, Mongolia. Five dominant shrubs including one dwarf shrub, one tall grass and nine short herbaceous plants were investigated in this study. According to coefficients of determination (r(2)), simple and easy calculation, and avoidance of large error occurrence in the estimation, it was indicated that the linear regression equation with the dimensional volume was best applicable to estimate the plant biomass and the plant water mass of the dominant shrubs and herbaceous plants. Residual error values (absolute values of errors between the estimation and the measurement) and residual error ratios (ratios of residual errors) were calculated to validate the accuracy of the estimated values obtained from linear regression equations of each plant. Although the residual error values increased with the increase in total plant biomass (g DM m(-2)) and total plant water mass (g H2O m(-2)) per unit area, the residual error ratios tended to decrease with the increase in those total amounts. 82.6% of the estimated total plant biomass and 67.4% of the estimated total plant water mass fell within +/- 40% of the residual error ratios, respectively. Hence, it was indicated that the total plant biomass and the total plant water mass per unit area could possibly be estimated by the use of each linear regression equation.