Uncertainty in wind speed and direction measurements from the ARM Doppler lidars

Poster PDF

Description

A key component of the on-going reconfiguration of the ARM Southern Great Plains (SGP) site involves the establishment of a network of four boundary facilities around the existing central facility that would provide continuous height-resolved measurements of temperature, humidity, and wind. Characterization of measurement uncertainty is particularly important for current plans to use these observations to support model data assimilation and development of a continuous, long-term model forcing data set. The wind measurements, in particular, will be obtained using scanning coherent Doppler lidars, and efforts are currently underway to procure several systems for deployment to the new boundary facilities. In this study, we focus specifically on quantifying the accuracy of horizontal wind measurements from one of the existing ARM Doppler lidars using observations from a recent off-site field campaign at the Boulder Atmospheric Observatory (BAO). In March and April of 2015 the ARM Doppler lidar that was formerly at operated at the Tropical Western Pacific site in Darwin Australia was deployed to the BAO for the eXperimental Planetary boundary-layer Instrument Assessment (XPIA) field campaign. During XPIA the 300-m tower at the BAO site was instrumented with sonic anemometers at six levels. These sonic anemometers provided highly accurate reference measurements against which the lidar was compared. This provided a rare opportunity to characterize the measurement accuracy of the ARM Doppler lidar because ARM does not currently operate any towers taller than the lidar’s minimum measurement height of approximately 100m. The results of the lidar-tower comparison are presented. Tests were conducted using three different uncertainty estimation methods, and three different PPI scanning strategies. We examine the uncertainty estimates produced by the lidar’s wind-retrieval algorithm and their reliability as measures of data quality. The method that produced the best overall results was a 3D retrieval using a multi-pass PPI scan with 8 evenly spaced beams around the compass. Results from that trial show a negligible bias of ~1cm s-1 in the lidar wind speed, and a standard deviation of approximately 50 cm s-1. The wind direction measurements also show a negligible bias of ~1o and a standard deviation of about 6o. Additionally, the uncertainty estimates produced by this technique were in better agreement with the actual observed lidar-tower differences.