ARC Seibersdorf research GmbH Treatment of uncertainty in greenhouse gas emission inventories: General approach and specific experience for Austria Wilfried Winiwarter

ARC Seibersdorf research GmbH IPCC Expert Group on Good Practice in Inventory Preparation CHAPTERS 1. Introduction 2. Energy 3. Industrial Processes 4. Agriculture 5. Waste 6. Quantifying Uncertainties in Practice 7. Methodological Choice and Recalculation 8. Quality Assurance and Quality Control ANNEXES Annex 1. Conceptual Basis for Uncertainty Analysis Annex 2. Verification Annex 3. Glossary Annex 4. List of Participants

ARC Seibersdorf research GmbH Issues relevant to uncertainty calculation  Quality Assurance and Quality Control (procedural recommendations)  Methodological Choice and Recalculation (selection criteria for key sources to focus resources)  Verification (quantitative checks and comparison with ambient air)  Conceptual Basis for Uncertainty Analysis (mathematical background for uncertainty calculations)  Quantifying Uncertainties in Practice (from uncertain inputs to resulting uncertainties)

ARC Seibersdorf research GmbH Uncertainty calculations for Austria  Input uncertainties:  Statistical differences  Reported variation  Expert judgement  Estimation  Treatment of systematic errors  Magnitude of unknown systematic errors assumed to be of the same size as errors identified  Monte-Carlo approach  impressive  easy to operate  allows handling of co-variances  supports sensitivity analysis

ARC Seibersdorf research GmbH Sensitivity studies  Input uncertainties total uncertainty influenced by few inputs requirement to deep investigation limited  Uncertainty by gas similar variances for CO 2, CH 4, N 2 O when decreasing variance of any gas  little change to overall uncertainty when increasing variance of any gas  increase of overall uncertainty  Robustness to assumptions on probability density function

ARC Seibersdorf research GmbH International comparisons CO 2 CH 4 N 2 O HFCs PFCs SF 6 Total Total trend ( ) Austria Norway Netherlands UK USA all uncertainties in % (2s), trend in %-points

ARC Seibersdorf research GmbH Conclusions  It is possible to assess the uncertainty of national GHG emission inventories  Personal judgement (of experts...) does have influence on uncertainty estimates  Scenario analysis and sensitivity runs allow to assess this influence and to understand / evaluate it  Intuitive aspect gains weight when uncertainties are larger

ARC Seibersdorf research GmbH Acknowledgements: Kristin Rypdal for setting the concept into the perspective of IPCC; Environmental Federal Agency (UBA) Vienna, specifically Klaus Radunsky; Rudi Orthofer for important comments and ideas; and the experts willing to contribute to our survey....