A Comparison of Sensible and Latent Heat Flux Calculations Using the Bowen Ratio and Aerodynamic Methods

Abstract
An analysis technique is outlined that calculates the sensible and latent heat fluxes by the Bowen ratio and aerodynamic methods, using profile measurements at any number of heights. Field measurements at two sites near Churchill, Manitoba, comprising over 4000 hours of data from June through September, are used to assess the relative accuracy of the two methods. If hourly data are eliminated when the Bowen ratio method gives the wrong sign for the fluxes, or when the Bowen ratio is close to −1, the two methods agree very well. The results also indicate that the aerodynamic method can provide acceptable results with only a single measurement of wind speed and a good estimate of the surface roughness length. The relative error associated with the use of only a single anemometer is small for unstable conditions, and only becomes significant under very stable conditions when the fluxes themselves are small. The overall comparison of the two methods yields a mean bias error of less than 10 W m−2, and... Abstract An analysis technique is outlined that calculates the sensible and latent heat fluxes by the Bowen ratio and aerodynamic methods, using profile measurements at any number of heights. Field measurements at two sites near Churchill, Manitoba, comprising over 4000 hours of data from June through September, are used to assess the relative accuracy of the two methods. If hourly data are eliminated when the Bowen ratio method gives the wrong sign for the fluxes, or when the Bowen ratio is close to −1, the two methods agree very well. The results also indicate that the aerodynamic method can provide acceptable results with only a single measurement of wind speed and a good estimate of the surface roughness length. The relative error associated with the use of only a single anemometer is small for unstable conditions, and only becomes significant under very stable conditions when the fluxes themselves are small. The overall comparison of the two methods yields a mean bias error of less than 10 W m−2, and...

This publication has 0 references indexed in Scilit: