A drained boreal bog with active drainage ditches and the introduction of forage grass was compared with a natural bog site to assess the effect of agricultural drainage on the photosynthetic capacity of boreal peatlands. In this study, the net carbon dioxide uptake and the ecosystem respiration were measured using a pair of clear and dark chambers, connected to a greenhouse gas analyzer in order to estimate the gross primary production (GPP). Environmental factors, including photosynthetically active radiation (PAR), peat soil temperature, soil moisture, electrical conductivity and water table depth (WTD), and dissolved organic carbon (DOC) in the soil pore water were also investigated at each site. Our results confirmed a correlation between water table depth, peat soil temperature, electrical conductivity and peatland productivity. Changes in electrical conductivity, water table depth, and peat soil temperature together explained 61% of gross primary productivity variance at our drained peatland pasture site. In general, plant functional types/growth forms did not have any significant difference in their GPP at the drained site except shrubs which had a significantly higher GPP at PAR level greater than 1000 μmol/m2/s (F(3,30.84) = 11.64, p < 0.001). At the natural site, GPP was significantly higher at the hollow (0.168 ± 0.02 mg/m2/s) compared to the hummock (0.098 ± 0.01 mg/m2/s) where relatively the higher WTD was recorded. We therefore concluded that water shortage to the shallow rooting depth of shrubs on our hummock subplots may be the cause of the low productivity. At PAR greater than 1000 μmol/m2/s, productivity at the hollow was still significantly higher (0.17 ± 0.002 mg/m2/s) compared to the hummock (0.10 ± 0.00 mg/m2/s). This suggests that, for ecosystem carbon modelling, the minimum PAR level at which the GPP is estimated is very crucial. All the growth forms at the drained site recorded a significantly higher GPP than the microforms at the natural site. We speculate that not only hydrological changes via agricultural drainage can affect the carbon balance of boreal peatland ecosystem through photosynthesis, but also the hydrological changes in the peat due to the differences in peat soil makeup, structure and density. We, therefore, conclude that differences in microtopography and the corresponding changes in vegetation, land-use practice and its effects on peat soil properties and hydrology should be taken into account when modelling the gross primary production in a peatland ecosystem.
