Izvorni znanstveni članak
DISPERSION MODEL INPUTPARAMETERS FROM NUMERICAL WEATHER PREDICTION OR SYNOPTIC OBSERVATIONS
Douglas R. Middleton
; Met Office FitzRoy Road Exeter Devon EX1 3PB U.K.
Sažetak
In the UK, local and regulatory dispersion modelling has usually been based upon input of hourly synoptic observations (OBS) from a not too distant station. Key variables arc wind speed and direction, cloud amount, and precipitation. Developments in weather forecasting mean that copious amounts of output data from numerical weather prediction (NWP) are produced at intervals, e.g. hourly or 3-hourly. These NWP data are available for use in air quality management. They can include friction velocity, heat flux and boundary layer depth, which arc not routinely measured. Operational air quality forecasts by the NAME III model rely on full 3-D NWP fields of data. Regulatory models need such data reduced to suit model input formats. Initial comparisons of NWP (at two grid scales, ~60 km and ~12 km) and OBS data as used for dispersion models have been presented at the Harmo 11 Conference (Middleton, 2007). Weibull distributions of wind speed were fitted and the parameters compared. The present study describes the results from using parallel NWP and OBS data as input data to the meteorological pre-processor of the Met Office NAME III dispersion model. This processor is very similar in form and function to that of the ADMS model, in having similar boundary layer algorithms. Pre-processor results for dispersion parameters have been compared, namely friction velocity, heat flux, Monin Obukhov length and boundary layer depth. These parameters play a major role (along with source emissions and other factors) in influencing the dispersion calculations. The study compares values from NWP with pre-processor results using OBS data. It discusses potential advantages and disadvantages of NWP and OBS data in the pre-processors of regulatory dispersion modelling. The talk ends with a discussion of which NWP output variables might be best suited to managing air quality.
Ključne riječi
numerical weather prediction data; dispersion model; pre-processor; friction velocity; heat flux; stability
Hrčak ID:
64083
URI
Datum izdavanja:
12.12.2008.
Posjeta: 1.399 *