摘要

It is shown that the sensible heat flux, H, can be estimated without measurements of wind speed. Over grass, H was estimated deploying three fine-wire thermocouples in the inertial sub-layer. Two thermocouples are required to measure the mean gradient and the other is required to measure the air temperature at high frequency to determine the mean ramp dimensions. Regardless of the stability conditions, excluding those cases where the gradients fall within the measurement error, the H estimates were close to the measured using the Eddy Covariance method, H-EC. The cases where the method is not applicable (i.e., when the gradient is small), are near-neutral and under free convection conditions. In near-neutral cases, the air temperature trace does not show clear ramps and, therefore, H is close to zero. Under convection, H can be estimated using the free convection limit approach. It is shown that the new method can be used to calibrate the earlier surface renewal method which requires only one thermocouple to estimate H. Regardless of the stability cases, the earlier SR method calibrated against the EC method or against the new SR method performed close (the root mean square error, RMSE, were RMSE %26lt; 1 W m(-2)). %26lt;br%26gt;The new SR method was consistently best correlated (0.78 %26lt;= R-2 %26lt;= 0.90, where the highest and lowest R-2 correspond to stable and unstable cases with positive net radiation, respectively) than the earlier SR method (0.71 %26lt;= R-2 %26lt;= 0.74) to H-EC. The RMSE, however, for the new and the earlier SR method were comparable, in general, RMSE %26lt;= 25 W m(-2). It is concluded that for field applications the implementation of the new and the earlier SR method is an affordable and reliable alternative to the EC method to obtain long series of half-hourly H.

  • 出版日期2013-10-15