Comparing dune migration measured from remote sensing with sand flux prediction based on weather data and model, a test case in Qatar

Abstract

This study explores validating and calibrating the wind regime predicted by Global Circulation Models (GCM) on Earth and other planets using optical remote sensing of dune dynamics. We use Spot-5 images to track the migration of 64 Barchan dunes in Qatar using the COSI-Corr technique. We estimate the volume of the dunes using a scaling law calibrated from one particular dune, which was surveyed in the field. Using volume and migration rate, we determine the sand flux from a single dune, Q_(Dunes), and scale this estimate to the whole dune field. We compare the measured sand flux with those derived from wind velocity measurements at a local meteorological station as well as with those predicted from ERA-Interim (a Global Circulation Model). The comparison revealed that the wind velocity predicted by ERA-Interim is inappropriate to calculate the sand flux. This is due to the 6-h sampling rate and to systematic bias revealed by a comparison with the local wind data. We describe a simple procedure to correct for these effects. With the proposed correction, similar sand flux are predicted using the local and ERA-Interim data, independently of the value of the value of the shear velocity threshold, u_(*t). The predicted sand flux is about 65% of Q_(Dunes). The agreement is best assuming the value u_(*t)=0.244 m/s, which is only slightly larger than the value of u_(*t)=0.2612 m/s estimated based in the sand granulometry measured from field samples. The influence of the dune topography on the wind velocity field could explain the underestimation. In any case, the study demonstrates the possibility of validating GCM model and calibrating aeolian sand transport laws using remote sensing measurements of dune dynamics and highlights the caveats associated to such an approach.