Model development for estimating carbon dioxide exchange in Nordic forests and peatlands with MODIS time series data

The ongoing increase in atmospheric greenhouse gas concentrations causes an increase in the global mean air temperature. One important moderator of the atmospheric carbon dioxide (CO2) concentration is the gas exchange between the atmosphere and the biosphere; photosynthesis extracts and respiration releases CO2, resulting in a net exchange. A variety of methods are used to improve our understanding of the temporal and spatial distribution of this exchange. Since satellite sensors provide data with a continuous coverage throughout the landscape, such data are suitable for regional or global estimations of CO2 exchange across time. In this project, satellite sensor-derived data have been used in models for the gross primary productivity (GPP), ecosystem respiration (ER), and net ecosystem exchange (NEE) at sixteen Nordic forest and peatland sites. The light use efficiency (LUE) model was used as the theoretical basis for the development of empirical models where data from the Moderate Resolution Imaging Spectroradiometer (MODIS) were combined with modeled radiation data. All models were tested, calibrated, and validated with eddy covariance measurements of CO2 exchange. The results show that the GPP, ER, and NEE in forests and peatlands can be modeled with the MODIS enhanced vegetation index (EVI), MODIS two-band enhanced vegetation index (EVI2), MODIS land surface temperature (LST), and modeled incoming photosynthetic photon flux density (PPFD). It was also shown that there are nutrient-induced changes in the light use efficiency by which the vegetation converts light into biomass. In addition, near-ground sensors for spectral measurements were installed at five forest and peatland sites. This sensor network adds representative data for estimating ecosystem processes related to vegetation phenology that will aid the calibration of satellite sensor-derived data. Such knowledge will contribute to further improvements of the suggested models for the CO2 fluxes. The project has shown that the CO2 fluxes can be modeled with satellite sensor-derived data and that near-ground spectral measurements provide useful information for further model developments