The effect of wind speed averaging time on the calculation of sand drift potential: New scaling laws

Abstract

Drift potential (DP) is a key factor used to quantify aeolian sand transport, and it is often used for predicting wind erosion and dune mobility under climate change scenarios. It is proportional to the mean of the cube of the wind speed at a 10-m height when that wind speed is above 6.2 m/s (12 knots), a speed threshold above which sand can be transported by the wind. DP depends on the temporal resolution and the averaging time of the wind velocity measurements. Usually, either a 10-min or 1-h mean wind speed is used to calculate the DP, although different studies have used different averaging periods for calculating the mean wind speed and, consequently, the DP. Here, we perform a comprehensive analysis of an extensive, long-term wind speed database from different locations around the world and from reanalysis data sources, and show that DP decreases with the averaging time of the wind speed. The decrease of DP versus the averaging time interval can be approximated by a stretched exponential function. In addition, the relative decrease of DP, as a function of the maximum DP, is smaller for high wind speed sites. This result is a consequence of the fact that high DP sites are characterized by longer clusters of successive wind speed events above the threshold velocity; they are thus less affected by the time averaging process. Interestingly, the clusters' lengths decay exponentially with wind speed and faster for low DP sites. Our new results have important implications for calculating the effective DP and for better estimating the biases that result from using wind data with different averaging times, which is a common problem in aeolian research. We also found an approximate linear relation between DP and the wind power that is close to the analytic approximation for the relation between the two. This new finding has many important applications since it will allow one to calculate DP from wind power data and vice versa. In addition, it implies that the same scaling laws found for DP also exist for the wind power.