The structure of the atmospheric surface layer subject to local advection.

For many applications in agriculture, hydrology and meteorology simple methods are needed to determine the surface-atmosphere exchange of momentum, heat and water vapour, i.e to determine the fluxes of momentum, heat and water vapour. Most methods to calculate these fluxes are only valid for horizontal, homogeneous terrain with sufficient large dimensions. It is thus assumed that so-called advective effects can be neglected because wind speed, temperature and humidity do not change in the horizontal direction. In practice, the surface is hardly ever homogeneous. It was the objective of this study to investigate the effects of advection of heat and moisture on the fluxes. It appeared that only a few sets related to advection were available, and the available sets yielded contradictive results. Therefore an experiment was carried out in La Crau, France around a step-change from a dry and bare terrain to irrigated grass in order to measure the influence of advection on the structure of the flow and the exchange processes near the earth's surface.Before this study there was confusion about the behaviour of the flux-gradient ratios or eddy diffusivities under conditions of local advection. The flux-gradient ratios or eddy diffusivities were estimated using the calculated surface fluxes and the gradients from the profile measurements. It was found that the flux-gradient ratio for heat was smaller than that for water vapour in the lower part of the surface layer after the step-change. This was in agreement with the ratio of the observed transfer efficiencies. Higher up in the surface layer after the step-change and for weak advective conditions it was found that the flux-gradient ratio for heat was larger than that for water vapour.Also, flux determination methods were tested using a second-order closure model which was found to compare favourably with the measurements. It was found that for situations similar to that in the Crau, the so-called gradient Bowen ratio can be used at fetch-to-height ratios up to z/x = 0.02 which is high compared to what was accepted untill now. It was also found that the difference between the Bowen ratio at the surface and at some level above the surface is compensated for by the ratio of the eddy diffusivities at the height where the gradient is measured. For the Bowen ratio from standard deviations in the thermally stable surface layer the error was found to be below 10%, up to z/x = 0.02. The ratio of the transfer efficiencies was below unity which compensated the error due to the flux divergence. For the unstable surface layer z/x must be below 0.008 to achieve the same accuracy because now the ratio of the transfer efficiencies amplifies the error due to flux divergence. For the cases studied here, a number of factors may have cooperated in a favourable manner due to which the error sources cancel out. However, it is significant to note that the model can be used to forecast which conditions are favourable and which are not for the application of micro- meteorological methods to determine the surface fluxes under conditions of local advection