Modelling photochemical air pollutant formation in Hungary using an adaptive grid technique

Abstract

A regional air quality model has been developed that describes the transport and chemical transformation of photochemical oxidants across Central Europe using an adaptive gridding method to achieve high spatial resolution. High-resolution emission inventories for Budapest and Hungary were utilised. The air pollution episode in August 1998 was modelled using a fixed coarse grid (mesh size 70 km) a fixed fine grid (17.5 km) and an adaptive, variable sized (from 17.5 to 70 km) grid. The fine and the adaptive grid models provided similar results, but the latter required 50% longer computing time. High ozone concentrations appeared downwind of Budapest and the plume extended up to about 150 km from the city at 17.00 on the simulated day. The simulation results were compared with ozone concentrations measured at the K-puszta and Hortobágy monitoring stations.

Adaptive Grid Use in Air Quality Modeling

“…The adaptive gridding method of Tomlin et al [22] has been subsequently applied to model the dispersion of nuclear contamination [26] and air pollution formation across central Europe [27]. These simulations extended into the regional scale and incorporated vertical layering and mixing, space-and-time-varying three-dimensional meteorological fields, dry deposition, and pollutant transformations (e.g., nuclear decay or chemical reactivity).…”

“…This differs from most adaptive grid plume simulations previously reported, centered on ozone photochemistry. Additionally, results were compared to measurements at air quality monitoring sites, an assessment approach that had only been previously reported for adaptive grids by Lagzi et al [27]. Three-dimensional meteorological and emissions data were prepared by a numerical weather prediction model and emissions processors.…”

“…Ghorai et al [28] found that performing grid adaptation every 20 min is sufficient to produce accurate results and reduces the computational requirements of a three-dimensional elevated plume simulation. A 5 min adaptation interval is used in the regional air quality case presented in Lagzi et al [27]. In their simulation of air quality over East Asia, Constantinescu et al [24] tested adaptation intervals from 1 to 6 h and determined that 3 hours provided an adequate balance between solution accuracy and computational cost.…”

“…The plume simulation carried out by Tomlin et al [25] addressed the issue by placing the modeled stack within a subdomain and using the subdomain concentration as an internal boundary condition, an approach similar to grid nesting. Subsequent applications of this adaptive grid method have commenced from grids initially refined around major pollutant sources (e.g., point sources [28], cities [26,27]). A grid preadaptation step was proposed by Srivastava et al [23].…”

Dispersion modeling of air pollutants in the atmosphere: a review

“…H-refinement dynamically changes the total number of grid elements within a computational domain for which the original structure remains fixed. This technique is also known as mesh enrichment or local refinement, and it is carried out by subdividing grid elements into smaller self-similar components [16,[77][78][79][80]. An example ofrefinement is shown in Figure 6.…”