A study of wind flow around a refinery
Assessment of diffuse emissions using remote sensing needs appropriate information on vertical profiles of wind speed and wind direction. This information has then to be combined with concentration, or path-integrated concentration, in order to estimate the mass flux of hydrocarbons through the measurement plane.
The presence of plant units, buildings, tanks, process areas, etc. modifies the wind as it passes across the refinery. It is not practical to make wind measurements at alongside concentration measurements in order to determine flux explicitly. Assumptions about the wind spatial and temporal resolution are therefore necessary and contribute to overall uncertainty in flux determinations.
To provide information on variability in wind parameters, Concawe carried out a programme of wind measurements on a refinery site using a combination of traditional meteorological instruments mounted on fixed masts and a wind LIDAR which was made mobile by mounting it on a pick-up truck. The study was carried out during a 9-day period. The purpose of the exercise is to investigate whether there are significant differences in wind data gathered at various locations.
This report is confined to an inter-comparison of the fixed mast data. Review of the LIDAR data revealed obvious errors in some of the LIDAR measurements (see Appendix B) and so these were not used in the analysis.
It was found that the wind vector (speed, direction) is modified as the wind interacts with the refinery. Measurements at one reference station are not always representative of measurements elsewhere. The difference in wind vector between stations varies with time denying the development of a correction factor. The difference is greater for measurements made at 3.7 m height compared to those made at 10 m height.
Recommendations have been made for the placement of wind masts to provide data to support remote sensing campaigns.
Recommendations are also made for both the evaluation of wind data and exclusion of time periods for meaningful interpretation of data. The need for sensitivity calculations to account for uncertainty in wind parameters in the derivation of emission flux is identified.
Although there were problems in this wind campaign with the use of a truck-mounted wind LIDAR, pre-campaign experience with a static mounted LIDAR was positive and the technique has advantages for measuring vertical profiles.