Sample-Based Estimation of Greenhouse Gas Emissions From Forests—A New Approach to Account for Both Sampling and Model Errors

• DOI: 10.5849/forsci.13-005
• Publication Date: 2014-01-26T01:34:41Z
• Authors: Göran Ståhl; Juha Heikkinen; Hans Petersson; Jaakko Repola; Sören Holm
• Keywords: Sample (material)

The Good Practice Guidance (GPG) for reporting emissions and removals of greenhouse gases from the land use, land-use change, forestry (LULUCF) sector United Nation's Framework Convention on Climate Change states that uncertainty estimates should always accompany net emissions. Two basic procedures are suggested: simple error propagation Monte-Carlo simulation. In this article, we argue these methods not very well-suited assessments in connection with sample-based surveys such as national forest inventories (NFIs), which provide a majority data LULUCF several countries. We suggest more straightforward approach would be to use standard sampling theory assessing errors; however, it may important also include contribution biomass other models applied requires new variance estimation. method assessment, including both model errors, is developed using NFIs Finland Sweden. study revealed combined sampling-model mean square ratio estimators aboveground forestland amounted about 10% estimating 5-year change corresponding stocks, permanent units, was reduced less than 1%. smaller impact case estimation due fact any tendency either over- or underestimate random parameter errors will same at beginning end period. fairly small contributions our large number sample trees used fitting Sweden; could expected substantial. proposed framework applies only gas but traditional NFI of, e.g., growing stock uncertainties typically neglected applications.