You are here:
SENSITIVITY OF THE NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION MULTILAYER MODEL TO INSTRUMENT ERROR AND PARAMETERIZATION UNCERTAINTY
Citation:
Cooter, E J. AND D B. Schwede. SENSITIVITY OF THE NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION MULTILAYER MODEL TO INSTRUMENT ERROR AND PARAMETERIZATION UNCERTAINTY. JOURNAL OF GEOPHYSICAL RESEARCH 105(D5):6695-6707, (2000).
Impact/Purpose:
This task has the following objectives:
Improve modelers' ability to focus on scientific and policy issues in modeling studies by providing software that supports composing, applying, and evaluating complex systems of models.
Improve the understanding of the interaction of the atmosphere and the underlying surface, especially the flux of mass in both directions, and EPA's ability to simulate that interaction.
Contribute to multimedia studies and assessments by applying state-of-the-art atmospheric models, estimating atmospheric contributions to multimedia issues and the sources of those contributions, and evaluating the models' strengths and weaknesses.
Description:
The response of the National Oceanic and Atmospheric Administration multilayer inferential dry deposition velocity model (NOAA-MLM) to error in meteorological inputs and model parameterization is reported. Monte Carlo simulations were performed to assess the uncertainty in NOAA-MLM deposition velocity estimates for ozone, sulfur dioxide, and nitric acid associated with measurements of meteorological variables (including temperature, humidity, radiation, wind speed, wind direction, and leaf area index). Summer daylight scenarios for grass, corn, soybean, oak, and pine were considered. Model sensitivity to uncertainty in the leaf area index (LAI), minimum stomatal resistance, and soil moisture parameterizations was explored. For sulfur dioxide and nitric acid, instrument error associated with the measurement of wind speed and direction resulted in the greatest velocity error. Depending on vegetation type, the most important source of uncertainty due to instrument error for the velocity of ozone was LAI. Of the model parameterizations studied, accurate estimation of temporal aspects of the annual LAI profile and the characterization of soil moisture supply and demand are most important to model-estimated velocity uncertainty. Considered individually, these factors can result in sulfur dioxide and nitric acid velocity estimate uncertainty of ?25% and ozone estimate uncertainty greater than 60%. For single plant species settings, reductions in estimate uncertainty should be possible with minor algorithmic modification, inclusion of more species-appropriate LAI profiles, and careful application of remote sensing technology.