You are here:
Evaluation and optimization of the Surface Tiled Aerosol and Gaseous Exchange (STAGE) resistance model in CMAQv5.4 with long term ozone flux observations at multiple sites
Citation:
Bash, J., C. Hogrefe, B. Cheng, Kiran Alapaty, D. Schwede, Z. Wu, M. Coyle, E. Fredj, I. Goded, G. Manca, O. Gazetas, L. Horváth, Q. Li, E. Tas, I. Mammarella, T. Vesala, J. Munger, R. Staebler, L. Zhang, T. Weidinger, AND O. Clifton. Evaluation and optimization of the Surface Tiled Aerosol and Gaseous Exchange (STAGE) resistance model in CMAQv5.4 with long term ozone flux observations at multiple sites. International Technical Meeting on Air Pollution Modelling and its Application, Chapel Hill, NC, May 22 - 26, 2023.
Impact/Purpose:
Low signals can often be found in satellite observations of minor atmospheric species with weak spectral signals (e.g. ammonia (NH3)). Omitting these non-detects can high-bias averaged measurements in locations that exhibit conditions below the detection limit of the sensor. Here we utilize the information content from the satellite signal to explicitly identify non-detects and then account for them with a consistent approach. This methodology is applied to the CrIS Fast Physical Retrieval (CFPR) ammonia product and results in a more realistic averaged dataset under conditions where there are a significant number of non-detects when compared to surface observations. In areas with large emission sources these non-detects occur in less than 5% of the observations and in areas with ammonia concentrations less than 1 ppb the non-detects can account for more than 70% of the observations. If the non-detects are discarded, this results in a positive bias in the satellite retrieval as only the higher concentrations are retained. The methodology here utilizes available measurements to determine if the observation is valid but below the detection limit and then applies a consistent approach to estimate the concentration resulting in a better comparison with in-situ observations when the concentrations are lower than 1 ppb.
Description:
A box model of the Surface Tiled Aerosol and Gaseous Exchange (STAGE) option in CMAQ was applied to meteorological, observed canopy, and flux data at seven long term O3 micrometeorological flux sites. These sites represent evergreen needleleaf, deciduous broadleaf, grass land and shrub biomes/plant functional types. 2,000 random hourly samples from each site were selected, representing approximately 2.9% to 65.1% of the observed data at those sites, for an aggregate evaluation to prevent the overrepresentation of sites with longer data records. The modeled deposition velocities and fluxes were typically underestimated in the winter and nighttime conditions and overestimated in the afternoon and summer. Modeled biases at individual sites using the full record of observations ranged from 43% to -87% and 34% to -85% for fluxes and deposition velocities respectively. The randomly selected data was equally divided into eleven subsets for a ten-fold cross-validation and a validation dataset for the optimization of non-stomatal resistances applied to both deposition velocities and fluxes. Each training sub-set was used to optimize the soil, cuticular, and snow resistance using quasi-Newton and conjugate-gradient algorithms. The cuticular resistance was assumed to be a function of plant functional type and relative humidity. Soil and snow resistances were assumed to be constant due to a lack of correlation in the observed fluxes and deposition velocities with observed data. Optimized model results reduced model biases and errors and increased correlation with observations for both the validation data set and for the range of biases and errors in the full observations data sets. These revised parameterizations were incorporated in the CMAQ v5.4 and preliminary model simulations result in improvements in the evaluation of ambient O3 concentrations against network observations.