You are here:
Evaluating Seasonality and Trends in Modeled PM2.5 Concentrations Using Empirical Mode Decomposition (2019 ITM)
Citation:
Luo, H., M. Astitha, AND C. Hogrefe. Evaluating Seasonality and Trends in Modeled PM2.5 Concentrations Using Empirical Mode Decomposition (2019 ITM). 37th International Technical Meeting on Air Pollution Modeling and Its Application, Hamburg, GERMANY, September 23 - 27, 2019.
Impact/Purpose:
By comparing observed and CMAQ-simulated long term changes in aerosols, the work in this study addresses the question of how well CMAQ performs when applied to simulate the effects of changing emissions and meteorological variability on ambient particulate matter. To this end the study utilizes different statistical techniques to determine the time-scale dependency of model performance. Results of this analysis can inform model development efforts by focusing such efforts on time scales most directly connected to the models ability in reproducing observed variability and trends.
Description:
Empirical Mode Decomposition (EMD) of the time series of daily average fine particulate matter (PM2.5) and its species enables us to assess how well regional-scale air quality models simulate the time-dependent long-term trend and seasonality seen in the observations. In this study, we applied EMD to extract the intrinsic long-term trend annual cycles in the PM2.5 concentrations observed and simulated by the fully coupled Weather Research and Forecasting (WRF) - Community Multiscale Air Quality (CMAQ) modeling system separately for the 2000-2010 period. Results for total PM2.5 and its chemical components (SO4, NO3, NH4, OC, EC, and Cl) for three sites in Atlanta, GA; Reno, NV and Quabbin Summit, MA are presented. In general, the changing rate and the magnitude of the long-term trend components are well represented in total PM2.5 and SO4. Also, the model reproduces the magnitude of the annual variation of total PM2.5, SO4 and OC. However, the phase difference in the annual cycles for total PM2.5, OC and EC reveal a shift of up to 183 days (half year), indicating a potential challenge in the allocation of emissions during this 11-year period and the urgent need for the recent completed model updates in the treatment of organic aerosols compared to the version employed for this set of simulation. Here, we present why the EMD approach is selected and how it is applied to the evaluation of PM2.5 and discuss the findings in relation to other previous studies that evaluated the same set of model simulations.