Atmospheric Pollution Dispersion Simulation and Assessment: Technical Principles, Methods, and Appli

Atmospheric pollution dispersion simulation and assessment is a computational modeling process that predicts how pollutants emitted from industrial stacks, vehicles, and area sources are transported, diffused, and transformed in the atmosphere. By integrating meteorological data, emission source parameters, and validated numerical models such as AERMOD, CALPUFF, and CFD, the assessment determines whether predicted ground-level concentrations comply with ambient air quality standards such as GB 3095 and international benchmarks. This article provides a comprehensive overview of the technical principles, assessment inputs, evaluation items, modeling methodologies, industrial applications, and frequently asked questions related to dispersion simulation services.

Technical Overview

Atmospheric pollution dispersion simulation and assessment applies boundary-layer meteorology, atmospheric physics, and numerical modeling to quantify the transport, diffusion, chemical transformation, and deposition of pollutants released into the atmosphere. The core objective is to predict short-term (1-hour and 24-hour average) and long-term (annual average) ground-level concentrations at designated receptors, including environmentally sensitive targets such as residential zones, schools, hospitals, and nature reserves, and to compare these predictions against ambient air quality criteria specified in GB 3095 and relevant international guidelines.

Modern dispersion assessment combines multiple data streams, including surface and upper-air meteorological observations, prognostic meteorological fields generated by mesoscale models such as WRF, digital terrain elevation data, land use classification, building geometry for downwash analysis, and detailed emission inventories. Depending on the spatial scale, terrain complexity, and averaging period of concern, assessors select appropriate modeling systems ranging from steady-state Gaussian plume models to non-steady-state Lagrangian puff models and full computational fluid dynamics (CFD) simulations.

The results support environmental impact assessment (EIA) approvals, emission permit applications, stack design optimization, industrial site selection, chemical accident risk analysis, and environmental management verification under frameworks such as ISO 14001. All modeling work is typically performed with regulatory-grade models and validated against local monitoring data to ensure scientific defensibility and regulatory acceptance.

Testing Samples

In dispersion simulation projects, the "samples" are the input datasets and source characterizations that define the modeling scenario. The following categories of input data are collected, reviewed, and processed before simulation.

Point Source Parameters

Industrial stacks and vents are characterized by stack height, inner diameter, exit gas velocity, exit gas temperature, and pollutant emission rates. These parameters determine plume rise, effective release height, and downstream concentration distribution. Data are obtained from emission tests, engineering design documents, or material balance calculations.

Line and Area Sources

Vehicle emissions from roadways, fugitive dust from material storage yards, evaporative losses from wastewater treatment units, and diffuse VOC leaks from equipment are modeled as line, area, or volume sources. Source dimensions, release heights, and activity-based emission factors are compiled into the emission inventory.

Meteorological Data

At least one full year, and preferably three to five consecutive years, of surface meteorological data (wind speed, wind direction, temperature, cloud cover, pressure) and upper-air sounding data are required to capture the local dispersion climatology. For data-sparse regions, prognostic meteorological fields generated by the WRF model are used to reconstruct site-specific conditions.

Terrain, Land Use, and Building Data

Digital elevation models, land use and land cover datasets, and building dimension surveys are processed to account for terrain-induced flow channeling, surface roughness effects, and building downwash, all of which significantly influence near-field concentrations.

Receptor and Background Data

Discrete sensitive receptors and structured receptor grids are defined around the emission sources. Regional background concentrations of relevant pollutants, obtained from government monitoring stations, are incorporated to enable cumulative impact evaluation.

Testing Items

The assessment evaluates the following core items, tailored to project characteristics and regulatory requirements:

• Prediction of maximum 1-hour, 24-hour, and annual average ground-level concentrations of criteria pollutants, including SO2, NO2, PM10, PM2.5, CO, and TSP.
• Simulation of industry-specific pollutants such as NH3, H2S, HCl, HF, chlorine, formaldehyde, benzene, VOCs, heavy metals (lead, mercury, cadmium, arsenic), and dioxins and furans.
• Determination of maximum incremental concentration and the downwind distance at which it occurs.
• Cumulative impact assessment combining predicted increments with measured background concentrations against GB 3095 limit values.
• Stack height, location, and emission rate optimization studies for engineering design.
• Odor impact assessment based on odor threshold concentrations at sensitive boundaries.
• Accident scenario simulation, including toxic gas release, dense gas cloud dispersion, jet release, and evaporating pool scenarios for risk zoning.
• Dry and wet deposition analysis for particulate-bound and gaseous pollutants affecting soil, vegetation, and water bodies.
• Secondary pollutant formation assessment, including sulfate and nitrate particulate formation from SO2 and NOx.

Testing Methods

The assessment follows a staged modeling methodology consistent with national technical guidelines for atmospheric environmental impact assessment and internationally accepted regulatory modeling practice.

• Screening modeling: AERSCREEN or equivalent screening tools provide conservative first-cut estimates of maximum ground-level concentrations to identify whether refined modeling is required.
• Refined steady-state modeling: AERMOD with the PRIME algorithm is applied for near-field transport distances up to approximately 50 kilometers, handling point, area, volume, and line sources, building downwash, and elevated or complex terrain.
• Non-steady-state modeling: CALPUFF is used for long-range transport, stagnant wind conditions, coastal and complex wind circulations, chemical transformation, and deposition studies extending over hundreds of kilometers.
• Computational fluid dynamics: CFD simulation resolves three-dimensional flow around buildings, in street canyons, and across steep terrain where Gaussian assumptions break down, providing detailed concentration fields for ventilation and near-field exposure analysis.
• Emergency release modeling: Dense gas and jet release models quantify concentration footprints and isolation zones for accidental releases of hazardous substances.
• Model validation and performance evaluation: Predicted concentrations are statistically compared with field monitoring data using normalized mean square error, fractional bias, and fraction within a factor of two metrics, supported by sensitivity and uncertainty analyses.
• Scenario and sensitivity analysis: Worst-case meteorological combinations, alternative operating loads, and mitigation measures are simulated to bracket the range of potential outcomes.

Applications

Atmospheric dispersion simulation and assessment serves a wide range of regulatory, engineering, and management purposes:

• Environmental impact assessment and emission permit applications for new, relocated, and expanded industrial projects in power generation, petrochemicals, metallurgy, cement, chemicals, and waste incineration.
• Stack height optimization and rational layout of emission sources within plant boundaries.
• Site selection and environmental feasibility studies for industrial parks and land use planning authorities.
• Chemical process hazard analysis, safety zoning, and emergency response planning for facilities handling toxic or flammable gases.
• Odor nuisance investigation and mitigation design for wastewater treatment plants, landfills, rendering facilities, and food processing operations.
• Regional air quality planning, emission reduction pathway evaluation, and policy decision support for municipal and provincial authorities.
• Environmental management system verification and continuous improvement evidence under ISO 14001 programs.
• Litigation support and technical due diligence in environmental disputes and property transactions.

FAQ

What is the difference between AERMOD and CALPUFF?

AERMOD is a steady-state Gaussian plume model best suited for near-field assessments within about 50 kilometers under hourly steady meteorological conditions. CALPUFF is a non-steady-state Lagrangian puff model that handles calm winds, recirculation, long-range transport, chemical transformation, and deposition, making it appropriate for regional-scale and complex meteorological applications.

What data must a client provide for the assessment?

Clients typically provide emission source parameters and emission rates, plant layout drawings, building dimensions, project location coordinates, production process information, and any existing monitoring reports. The assessment team supplements these with meteorological, terrain, and land use datasets.

Which standards govern the assessment?

Ambient air quality evaluation is performed against GB 3095 for criteria pollutants and applicable industry emission standards, while environmental management context is provided by ISO 14001. Model selection and procedures follow national technical guidelines for atmospheric environmental impact assessment and internationally recognized regulatory modeling practice.

How long does a typical dispersion assessment take?

Screening-level assessments can be completed within one to two weeks. Refined AERMOD studies generally require two to four weeks depending on source count and receptor density, while CALPUFF or CFD studies involving mesoscale meteorological preprocessing may require four to eight weeks.

When is CFD modeling necessary?

CFD is recommended when emissions occur in dense building clusters, street canyons, or steep terrain where flow separation and recirculation invalidate simpler models, or when detailed near-field exposure, ventilation design, or intake location optimization is required.

How reliable are the simulation results?

Regulatory-grade models incorporate decades of field-experiment validation. When high-quality input data are used and model performance is verified against monitoring data, predicted concentrations typically fall within a factor of two of observed values, which is considered acceptable accuracy for regulatory decision-making.