This paper reports on an analysis of instantaneous sediment transport in relation to wind gusts, moisture content and fetch length from a field study carried out at Skallingen, Denmark. Wind speeds were measured with cup anemometers and instantaneous sediment transport with vertical traps coupled to electronic balances. Moisture content was measured gravimetrically using samples taken along a beach profile at intervals of 1 to 2 h. Sediment transport and wind speeds were sampled at 1 Hz and a 5-s running mean was used to smooth the data from the trap because of limitations of the balance resolution. Three measures of the threshold of sediment motion—the intermittency threshold, minimum threshold and maximum threshold—were explored using data from five runs carried out on October 26, 2000. The intermittency threshold and maximum threshold were largely insensitive to wind speed but varied spatially along the fetch distance. The minimum threshold increased significantly with increasing mean wind speed as a result of reduced time for drying of the surface layer. Examination of time series for instantaneous wind speed and sediment transport for a number of runs on October 26 and November 6, 2000 showed close agreement between fluctuations in wind speed and fluctuations in sediment transport. Cross spectral analysis of instantaneous wind speed and sediment transport for one run on each day where there was continuous transport showed high coherence between the two variables over a range of frequencies from 0.2 Hz to <0.01 Hz. Plots of all non-zero instantaneous sediment transport values versus wind speed show considerable scatter. In general, transport rates are highest for traps located further from the upwind non-erodible boundary (i.e. along the fetch). Best fit power relationships show R 2 values generally ranging from 0.4 to 0.6 and exponents ranging from 4.5 to 6.94—i.e. much higher than the traditional value of 3 found in most transport equations. Traps located close to the upwind boundary had lower R 2 values and exponents <2.5.