Bayesian estimation of input parameters of a nitrogen cycle model applied to a forested reference watershed, Hubbard Brook Watershed Six

Abstract

We present a Bayesian parameter estimation technique for improving estimates of simulation model input parameters and apply the technique to the nitrogen cycle model SINIC, which has been used to simulate the streamflow and streamflow nitrate flux at Hubbard Brook Watershed 6, New Hampshire, during the 1964–1994 period. Uncertainty in initial estimates of model input parameters was incorporated by replacing each estimate with a probability distribution of values, or “prior” distribution, usually centered at the initial estimate and having a large variance. These prior distributions were then “updated” by incorporating available data on model output variables, producing a “posterior” probability distribution of parameter values. Several key parameters used for calculating the N mineralization rate were identified as controlling the predicted nitrate export from this watershed. The level of uncertainty in these parameters was substantially reduced by incorporating the observations on streamflow and streamflow nitrate flux. The posterior distribution of predicted yearly streamflow nitrate flux shifted from year to year, with relatively large uncertainties in years with high streamflow nitrate flux.