On the Effect of a Low-level Jet on Atmospheric Pollutant Dispersion: A Case Study Over a Coastal Complex Domain, Employing High-Resolution Modelling

Abstract

A low-level-jet (LLJ) event that occurred over a coastal area in complex terrain is analyzed to investigate its effect on the dispersion of potential air pollutants released in the area. The atmospheric model RAMS (Regional Atmospheric Modeling System) is employed with a high vertical resolution close to the surface, adopting a rarely used nesting approach, in order to allow a detailed analysis of the flow and to characterize the specific features of the LLJ. After a comparison with meteorological variables measured by radiosondes, numerical experiments are performed adding a scalar tracer in the simulation. As a first test, the tracer is distributed uniformly throughout the domain to follow the dynamics of the LLJ and its effect on the tracer dispersion. Then, continuous releases from virtual point sources are simulated to address their possible impact in the area under LLJ conditions. This allows the identification of “hotspots” of pollutant accumulation due to very local circulations and convective cells that develop from the combined effect of terrain-induced flow and the interaction of the LLJ flow with complex topography. An original mass analysis is applied on the dispersion results for an advanced exploration of the LLJ impact on the tracer. The RAMS model provides reliable results demonstrating that with which atmospheric numerical models are useful tools with which to study LLJ dynamics and their effect on local circulation and pollutant dispersion.

Appendix: Bulk Jet Richardson Number

The data presented in Fig. 6 are calculated based on the bulk jet Richardson number Ri_J defined in Banta et al. (2003) as

$${Ri}_{J}=\frac{g}{\varTheta}\frac{ \Delta \varTheta/\Delta Z}{{\left(W{S}_{max}/{Z}_{max}\right)}^{2}} .$$

(1)

Here, we describe the calculation for each model level, at each site and at each hour of the LLJ occurrence, as follows. For each site and time the LLJ peak was identified, with the corresponding maximum wind speed WS_max and its elevation Z_max giving a Richardson number as

$${Ri}_{J}{|}_{i}=\frac{g}{{\varTheta}_{i}}\frac{ {\Delta\varTheta }_{i}/{\Delta Z}_{i}}{{\left(W{S}_{max}/{Z}_{max}\right)}^{2}} ,$$

(2)

where ${\Delta \varTheta}_{i}/{\Delta Z}_{i}= \left({\varTheta}_{i}-{\varTheta}_{i-1}\right)/\left({Z}_{i}-{Z}_{i-1}\right),$ g is the acceleration due to gravity, WS_max, Z_max are the wind speed and elevation of the LLJ peak, respectively, Θ, Z are the potential temperature and elevation respectively, Θ_i, Z_i are the potential temperature and model elevation of ith level. The ith Ri_J|_i value for each model level was calculated and plotted against the corresponding ith TKE (TKE|_i) value. Figure 6 includes two sets of five half-hourly points calculated as explained above, from 0600 to 0800 UTC—one set from the ground level to Z_max and a second set from Z_max to 600 m.