Comparing field inventory with mechanistic modelling and light-use efficiency modelling based approaches for estimating forest net primary productivity at a regional level

Monitoring of forest carbon fluxes for the purpose of national greenhouse-gas inventorying and reporting are commonly based on repeated large-scale field measurements. Alternate approaches based on modelling of forest growth offers potential benefits such as cost savings and detailed assessments of involved carbon fluxes. We calculated the net primary productivity (NPP) of Swedish forests using two methods based on mechanistic and light use efficiency (LUE) modelling. The results were evaluated using data from traditional field inventories, and showed large variations in calculated NPP for the two methods. The national mean NPP for each method ranged between 0.35 and 0.59 kg C m–2 year–1, with an average regional difference of ±50%. Despite the large differences in calculated NPP, mechanistic modelling was promising for estimating the spatial distribution with an r2 value of 0.92 for predicting NPP of mainland Sweden.