Keywords: Uncertainty analysis | Monte Carlo analysis| sensitivity analysis
What data needs were addressed? Identifying the key sources of uncertainty in the national inventory.
Why was the data needed? Finland began reporting cattle emissions using a Tier 2 approach in the 1990s, but Tier 1 was used for other livestock. Uncertainty analysis was used to identify emission sources and parameters in the Tier 2 model for which improved estimation methods could reduce overall uncertainty of the inventory.
Methods used: Monte Carlo analysis, sensitivity analysis
How was the data gap addressed? In the early 2000’s, Finnish researchers applied uncertainty analysis to the national inventory in order to identify emission sources to target for improved estimation. The analysis, reported in Monni et al. (2007), used Monte Carlo analysis. The uncertainty of activity data was estimated by examining the data for representativeness and possible bias, informed by interviews with relevant experts. For example, cattle have individual ear marks that enable very accurate assessment of animal numbers (uncertainty of ±3%), but uncertainty in animal numbers for other species on farms is higher (±5%). For animal weight, the researchers divided the standard deviation of the total population by the square root of the number of animals in each category to obtain a standard deviation of the mean value. Additional uncertainty was added, based on expert judgement, to reflect the effects of estimating animal weights using heart girth measurements. The distribution of data for each parameter was established following IPCC guidelines, i.e. assume normal distribution for empirical data unless other distributions fit the data better. Monte Carlo simulation was used to combine uncertainties, and sensitivity analysis was used to identify the most important factors affecting uncertainty. The analysis identified higher uncertainty of emission factors for bulls, heifers and calves than for dairy cattle, mostly affected by digestibility and net energy for maintenance. It concluded that using a Tier 2 approach for all animal types would reduce uncertainty in the agriculture inventory by 3%.
Monni S, Perälä P, Regina K. 2007. Uncertainty in agricultural CH4 and N2O emissions from Finland–possibilities to increase accuracy in emission estimates. Mitigation and adaptation strategies for global change.
Author: Andreas Wilkes, Values for development Ltd (2019)