**Estimation of Terrestrial Carbon Cycle Parameters Using a Particle Filter**
*
Kemp, Sarah ^{1}; Terwisscha van Scheltinga, Arjen^{1}; Scholze, Marko^{2}
^{1}University of Bristol, UNITED KINGDOM; ^{2}Lund University, SWEDEN
*

More accurate estimates of the net exchange of CO2 between the terrestrial biosphere and the atmosphere are crucial in improving predictions of future atmospheric CO2 levels. Better modelling of the carbon cycle strongly depends on improving the parameterisation of the underlying processes. Parameters can be estimated via data assimilation methods; however, often these rely on a linearisation of the model and assume Gaussian statistics. We present a novel method, to this field, of estimating parameters using an ensemble data assimilation approach - the particle filter, which is non-linear and non-Gaussian. Using the particle filter method, an initial sample of parameter vectors is drawn from a distribution based on the mean and standard deviation of each parameter. Each parameter vector is then assigned a weight depending on its cost function value. A probability distribution is constructed for each of the parameters based on this weighting. From which, a new sample of parameter vectors is drawn. The parameter space is resampled in such a manner for a number of iterations. We apply this method to estimate the posterior probability density function (PDF) for 19 parameters of the Biosphere Energy Transfer and Hydrology (BETHY) scheme. Our results show that the particle filter is a good solution to this problem.