ABSTRACT

The remote sensing approach of Mapping Evapotranspiration with the Internalized Calibration (METRIC) model is one of the accurate evapotranspiration models but the its complexity motivated us to seek for a simplified version of the original model. This study was designed to propose and evaluate the performance of a new methodology called Simplified Mapping Evapotranspiration with the Internalized Calibration (S-METRIC) model to estimate Actual Evapotranspiration (AET). S-METRIC is a simpler technique to implement while it requires relatively limited input data without losing tangible precision. Thanks to the simpler equations, it is faster computationally. The election of the mathematical form of this formula is mainly based on two purpose: first, it was upon the simplification of the formulas and then, their precision to approximate the formulas of METRIC. Polyfitn function and the synthetic meteorological data were used to fit corresponding curves for the four formulas of S-METRIC ($R_{\mathrm{s}\downarrow }$, ${\tau }_{\mathrm{s}\mathrm{w}}$, $\mathrm{C}\mathrm{o}\mathrm{s}{\theta }_{\mathrm{h}\mathrm{o}\mathrm{r}}$ and $\mathrm{C}\mathrm{o}\mathrm{s}{\theta }_{\mathrm{r}\mathrm{e}\mathrm{l}}$). The monthly AET calculated by S-METRIC was compared using a number of statistics indexes, including mean absolute error (MAE), root-mean-square error (RMSE), BIAS, relative error (RE), coefficient of determination (R²) and Nash-Sutcliffe efficiency coefficient (NSE). The results of the monthly AET comparison suggested that both models had similar performance to estimate monthly AET in the Vanak Basin with R² = 0.97–0.99, NS = 0.90–0.99, MAE = 0.49–8.7 mm month⁻¹ and RMSE = 0.8–13.1 mm month⁻¹. Also, the results of the monthly estimated $R_{\mathrm{s}\downarrow }$METRIC and $R_{\mathrm{s}\downarrow }$S-METRIC values comparison suggested that both models had showed a similar performance to estimate monthly $R_{\mathrm{s}\downarrow }$with R² = 0.97–0.99, MAE = 5.9–11.38 W m² and RMSE = 7.17–13.12 W m².

摘要

利用内标化（METRIC）模型映射蒸散量的遥感方法是精确的蒸散量模型之一，但其复杂性促使我们寻求原始模型的简化版本。本研究旨在提出并评估一种称为“简化映射蒸散发”的新方法的性能，该方法使用内部校准（S-METRIC）模型来估计实际蒸散发（AET）。S-METRIC是一种较简单的实施方法，它需要相对有限的输入数据而又不会损失有形的精度。由于方程式更简单，因此计算速度更快。选择该公式的数学形式主要基于两个目的：首先，是根据公式的简化，然后，它们的精度近似于METRIC的公式。使用Polyfitn函数和合成气象数据拟合S-METRIC（${\mathrm{[R}}_{\mathrm{s}\downarrow }$， ${\tau }_{\mathrm{s}\mathrm{w}}$， $\mathrm{C}\mathrm{Ø}\mathrm{s}{\theta }_{\mathrm{H}\mathrm{Ø}\mathrm{[R}}$ 和 $\mathrm{C}\mathrm{Ø}\mathrm{s}{\theta }_{\mathrm{[R}\mathrm{Ë}\mathrm{升}}$）。使用许多统计指标对S-METRIC计算的每月AET进行比较，包括平均绝对误差（MAE），均方根误差（RMSE），BIAS，相对误差（RE），测定系数（R ²）和Nash-Sutcliffe效率系数（NSE）。每月AET比较的结果表明，两种模型在Vanak盆地中估计每月AET的性能相似，R ²  = 0.97-0.99，NS = 0.90-0.99，MAE = 0.49-8.7 mm月^-1，RMSE = 0.8- 13.1毫米月^-1。另外，每月估算的结果${\mathrm{[R}}_{\mathrm{s}\downarrow }$METRIC和 ${\mathrm{[R}}_{\mathrm{s}\downarrow }$S-METRIC值比较表明，两个模型都表现出相似的性能以按月估算 ${\mathrm{[R}}_{\mathrm{s}\downarrow }$其中R ²  = 0.97–0.99，MAE = 5.9–11.38 W m ²和RMSE = 7.17–13.12 W m ²。