MICS-Asia II: Model intercomparison and evaluation of ozone and relevant species

https://doi.org/10.1016/j.atmosenv.2007.07.031Get rights and content

Abstract

Eight regional Eulerian chemical transport models (CTMs) are compared with each other and with an extensive set of observations including ground-level concentrations from EANET, ozone soundings from JMA and vertical profiles from the TRACE-P experiment to evaluate the models’ abilities in simulating O3 and relevant species (SO2, NO, NO2, HNO3 and PAN) in the troposphere of East Asia and to look for similarities and differences among model performances. Statistical analysis is conducted to help estimate the consistency and discrepancy between model simulation and observation in terms of various species, seasons, locations, as well as altitude ranges. In general, all models show a good skill of simulating SO2 for both ground level and the lower troposphere, although two of the eight models systematically overpredict SO2 concentration. The model skills for O3 vary largely with region and season. For ground-level O3, model results are best correlated with observations in July 2001. Comparing with O3 soundings measured in the afternoon reveals the best consistency among models in March 2001 and the largest disparity in O3 magnitude in July 2001, although most models produce the best correlation in July as well. In terms of the statistics for the four flights of TRACE-P experiment, most models appear to be able to accurately capture the variability in the lower troposphere. The model performances for NOx are relatively poor, with lower correlation and with almost all models tending to underpredict NOx levels, due to larger uncertainties in either emission estimates or complex chemical mechanism represented. All models exhibit larger RMSE at altitudes <2 km than 2–5.5 km, mainly due to a consistent tendency of these models towards underprediction of the magnitude of intense plumes that often originate from near surface. Relatively lower correlation at altitudes 2–5.5 km may be attributed to the models’ limitation in representing convection or potential chemical processes. Most of the key features in species distribution have been consistently reproduced by the participating models, such as the O3 enhancement in the western Pacific Ocean in March and in northeast Asia in July, respectively, although the absolute model values may differ considerably from each other. Large differences are found among models in the southern parts of the domain for all the four periods, including southern China and northern parts of some Southeast Asia countries where the behaviors of chemical components and the ability of these models are still not clearly known because of a lack of observational databases.

Introduction

East Asia has been experiencing continuous economic growth in the past two decades (Streets et al., 2003), but brought about noticeable degradation of air quality and ecosystem at the same time. Significant increases in emissions and changes in environment in East Asia also have important implications for atmospheric chemical cycling and climate change at both regional and global scales. A large number of projects have been launched by Asian countries to promote our understanding of air pollution problems and to conserve our living environment in a practical and effective way.

Chemical transport models (CTMs) serve as useful tools in scientific research and policy making. These models are normally composed of a series of modules, which represent complex physical, chemical and biological processes. Due to limitations in our understanding and computational resources, many processes are usually handled by parameterization. Various treatments and numerical approaches at differing levels of complexity have been incorporated into models of different structure, and modelers apply what they think the most appropriate parameters and inputs for specific model simulation. As a result, models may provide a variety of simulation results, and such disparity among models might be intensified due to nonlinear interactions among many kinds of processes. To better understand model performance and accompanying uncertainty, model evaluation is needed and model intercomparison is a good way for promoting model's ability through identifying potential uncertainties.

A number of air quality models or CTMs have been developed and widely used to explore a series of concerned problems regarding the transport, transformation and deposition mechanism of chemical components and source-receptor relationship in East Asia (An et al., 2002; Carmichael et al., 2003; Chang and Park, 2004; Engardt, 2000; Han et al., 2005; Hayami and Ichikawa, 2001; Holloway et al., 2002; Ichikawa et al., 2001; Kajino et al., 2004; Ueda et al., 2000; Wang et al., 2002). Several intercomparisons of global scale models have been conducted with focuses on radon, carbon monoxide, sulfur dioxide and sulfate, as well as photochemistry (Jacob et al., 1997; Kanakidou et al., 1999; Barrie et al., 2001; Olson et al., 1997), some of which are supported by IPCC and IGAC. However, model intercomparison study specifically for Asia is very limited (Phadnis and Carmichael, 1998; Kiley et al., 2003) and require more efforts. The MICS-Asia project was initiated in 1998. Carmichael et al. (2002) introduced the findings and outcome of MICS-Asia phase I, which focused on long-range transport and deposition of sulfur in East Asia. MICS-Asia phase II was initiated in 2003, as an extension of phase I, it includes more species of concern, such as ozone, nitrogen compounds, aerosols in additional to sulfur. Detailed information on this project is referred to an overview paper (Carmichael et al., 2008).

The main objectives of this paper are to evaluate the abilities of eight regional scale CTMs in simulating ozone and relevant chemical species, to identify the consistency and discrepancy among models, and to explore the potential factors responsible for the difference among models and deviation of model results from observations, as well as to reveal the characteristics of key chemical components in the troposphere of East Asia. A wide variety of observations are collected from EANET (Acid Deposition Monitoring Network in East Asia), JMA (Japan meteorological agency) and TRACE-P (transport and chemical evolution over Pacific) experiment that allows for a rigorous and comprehensive evaluation of model performances. Several key trace species are investigated, namely SO2, O3, NO, NO2, HNO3 and PAN, with focus on O3, NOx and SO2 due to relatively abundant observations and their significance in atmospheric chemistry.

This paper appears to be the first attempt to provide a perspective of model intercomparison and evaluation for ozone and relevant species in East Asia. It is expected to provide valuable insights into the abilities and limitations of current regional CTMs and to indicate directions for future model development.

Section snippets

Model description and relevant parameters

In total, eight model groups participated in MICS-Asia phase II. All of the models are regional 3-D Eulerian models. These models are from Seoul National University (Chang and Park, 2004), PATH (pollutants in the atmosphere and their transport over Hongkong) from Hongkong Environmental Protection Department, RAQM (regional air quality model) from Acid Deposition and Oxidant Research Center, Japan (Han et al., 2004), MSSP (model system for soluble particles) from Disaster Prevention Research

Comparison with ground-level observations of SO2, NO2 and O3

Table 2 shows the statistics for monthly mean concentrations of SO2, NO2 and O3 over all available sites for the combination of four periods. Results of seven models and ensemble means are compared with observations. M5 is not included because it only reports one period. A common domain of 90°E–145°E, 20°N–50°N (shown in Fig. 1) is chosen to ensure an identical statistical analysis. Each of the models generally contains 69, 53 and 42 pairs of samples (modeled and observed monthly means) for SO2

Conclusions

Evaluation and intercomparison of eight regional scale CTMs were conducted with a wide variety of observations from EANET, JMA and the TRACE-P experiment. The ability of CTMs in simulating O3 and relevant species in the troposphere of East Asia were rigorously evaluated in terms of different seasons, locations and altitude ranges. Potential factors responsible for deviation of models from observations and disparity among models were investigated. Important features in spatial distribution were

Acknowledgements

This research is sponsored by Acid Deposition and Oxidant Research Center (ADORC) and partly supported by National 973 project (2005CB422205), in collaboration with International Institute for Applied System Analysis (IIASA), Austria. The first author was also supported by the 100-Talent Project of Chinese Academy of Sciences. We would like to express our sincere thanks to all the model groups and participants for their great efforts and contributions to this project.

References (37)

  • H. Ueda et al.

    Behaviors of volatile inorganic components in urban aerosols

    Atmospheric Environment

    (2000)
  • C.J. Walcek et al.

    A simple but accurate mass conservative peak-preserving, mixing ratio bounded advection algorithm with Fortran code

    Atmospheric Environment

    (1998)
  • Wang

    MICS-Asia II: Model Inter-comparison and evaluation of acid deposition

    Atmospheric Environment

    (2008)
  • M.L. Wesely

    Parameterization of surface resistances to gaseous dry deposition in regional-scale numerical models

    Atmospheric Environment

    (1989)
  • L.A. Barrie

    A comparison of large-scale atmospheric sulphate aerosol models (COSAM): overview and highlights

    Tellus

    (2001)
  • E. Berge

    Coupling of wet scavenging of sulfur to clouds in a numerical weather prediction model

    Tellus

    (1993)
  • A. Bott

    A positive definite advection schemes obtained by non-linear renormalization of the advective flux

    Monthly Weather Review

    (1989)
  • J. Boutahar et al.

    Development and validation of a fully modular platform for numerical modelling of air pollution: Polair3D

    International Journal of Environmental Pollution

    (2004)
  • Cited by (0)

    View full text