I am working with an environmental dispersion model, that I am trying to verify with measured data. We have a six receptors (measurement stations) in the field that measures gas concentrations that could be attributed to release from a plant. For each receptor I have 28 (14 years of semi-annual measurements) univariate samples, with the distribution of data from each station generally following a log-normal distribution (to be expected for environmental data). Using the dispersion model, I can change a number of input parameters, and model the resulting gas concentrations at each receptor. At the end of the day, for each receptor, I have a measured distribution of data, and a single model value for each set of different input parameters into the dispersion model.

My question is, how do I best determine which set of input parameters into our model best matches my measured data across all receptors, and how do I determine how closely our model matches the measured data.

Thanks in advance!