干旱区生态环境知识资源中心

Reliable estimation of the mass-flux profiles of aeolian sediment is essential for predicting sediment transport rates accurately and designing measures to cope with wind-erosion. Vertical mass-flux profiles from seventeen wind-erosion events were re-evaluated using five typical models based on observed data obtained from a smooth bare field at the southern fringe of the Taklimakan Desert, China. The results showed that the exponential- function model and the logarithmic-function model exhibited the poorest fit between observed and predicted mass-flux profiles. The power-function model and the modified power-function model improved the fit to field data to an equivalent extent, while the five-parameter combined-function model with a scale constant (sigma) of 0.00001 m (different from the sigma value proposed by Fryear, which represented the height above which 50% of the total mass flux occurred) was verified as the best for describing the vertical aeolian sediment mass-flux profiles using goodness of fit (R (2)) and the Akaike Information Criterion (AIC) values to evaluate model performance. According to relationships among model parameters, the modified power model played a prominent explanatory role in describing the vertical profiles of the observed data, whereas the exponential model played a coordinating role. In addition, it was found that the vertical profiles could not be extrapolated using the five selected models or easily estimated using an efficient model without field observations by a near-surface sampler at 0 to 0.05 m.