Model assessment and verification - Air Pollution Modeling and Its Application - page 337

4.11 AQMEII: A New International Initiative on Air
Quality Model Evaluation

S. Trivikrama Rao 1 , Kenneth Schere 1 , Stefano Galmarini 2 , and Douw Steyn 3

1

Atmospheric Modeling and Analysis Division, National Exposure Research Laboratory,

United States Environmental Protection Agency, Research Triangle Park, NC, USA

2

Joint Research Centre, Institute for Environment and Sustainability - European Commission,
Ispra, Italy
3

The University of British Columbia, Vancouver, BC, Canada

Abstract We provide a conceptual view of the process of evaluating regional-
scale three-dimensional numerical photochemical air quality modeling systems,
based on an examination of existing approaches to the evaluation of such systems
as they are currently used in a variety of applications. A framework for model
evaluation is introduced to provide a context for the evaluation process. The
objectives of the model evaluation process include: determining the suitability of a
modeling system for a specific application; distinguishing between the performance
of different models through confidence-testing of model results; and guiding
further model development. The evaluation framework includes methods for
operational, diagnostic, dynamic, and probabilistic model evaluation. Also discussed
is a new effort, the Air Quality Model Evaluation International Initiative (AQMEII),
in which some of the new ideas in model evaluation are applied to air quality
modeling systems being used in North American and Europe, to assess the utility
of the techniques and to compare and contrast model evaluation results among
different models on both sides of the Atlantic Ocean. An AQMEII Workshop was
conducted in April 2009 in Stresa, Italy to discuss model evaluation concepts and
establish collaborative model application and evaluation projects.

1. Introduction

Regional-scale three-dimensional numerical photochemical air quality simulation
models (AQMs) are being used for air quality management decisions and for
short-term forecasting of air quality in many nations around the globe. To build
confidence in the model estimates, a model must be critically evaluated to assess
whether it is properly simulating the spatial and temporal features on the scales
resolved by the model. The evaluation also assesses whether the physical and
chemical processes are simulated correctly in the model, leading to proper model
response to changes in meteorology and emissions, the principal classes of inut
data required by AQMs. To this end, a renewed examination is needed to establish