Data from the Swedish NFI (NFI; Ranneby et al., 1987 and Axelsson et al., 2010) were used for greenhouse gas predictions. These data were suitable for two reasons: (i) they comprise individual tree data from about 30,000 permanent sample

plots first inventoried before 1990 (base year of the KP) and re-inventoried every 5–10 years thereafter, (ii) national representative BiEqs and volume equations are available for all three species (Näslund, 1947, Marklund, 1987, Marklund, 1988 and Petersson and Ståhl, 2006). The data are learn more summarized in Table 1. The Swedish NFI (Axelsson et al., 2010) is a systematic cluster sample inventory that includes annual data for all land and fresh water areas (ca. 45 mill. ha), except for the high mountains in the northwest

(ca. 2.3 mill. ha), which are not covered by trees, and urban areas (ca. 1.1 mill. ha). The clusters are square-shaped with sample plots along each side and are distributed throughout the country but have a higher density in southern than northern Sweden. Each year, about 6000 permanent PD0325901 sample plots are inventoried. For each circular sample plot (radius 10 m), extensive information is collected about the trees, stand and site. The main purpose of the Swedish NFI is to monitor forests for timber production and environmental factors. In the present study, the FAO definition (FAO, Adenosine triphosphate 2004) of forest land was used, i.e., land areas spanning more than 0.5 ha with a tree crown cover of at least 10% and a minimum height of trees of 5 m. The values for crown cover and minimum height refers to trees maturing in situ, and the predominant land use must be forestry. Marklund,

1987 and Marklund, 1988 pioneered the use of single-tree BiEqs for predicting the biomass of tree components, such as needles (not leaves), branches, bark, stem, stump and roots, of Scots pine (Pinus sylvestris), Norway spruce (Picea abies) and birch (Betula pendula and Betula pubescens, not stump and roots for birch). In deriving the BiEqs, the total fresh weight of each component per tree, and the fresh weight of samples from different components were measured in the field. The dry weight of each sample, defined as the constant weight at 105 °C, was measured in the laboratory and used for developing biomass equations per component. Trees were selected from 123 stands from different parts of Sweden, covering a wide variety of stand and site conditions. The resulting data were representative of Swedish forests at a national scale with the selected species constituting about 92% of the standing stem volume ( SLU, 2010). Broad-leaved species constitute most of the remaining 8% and equations based on birch were applied for all broad-leaved species.