Abstract:Forested wetlands are recognized as long-term terrestrial carbon (C) sinks, so they are very important in global C accounting. Harvest disturbance plays an important role in forested wetland ecosystems, since harvest can turn wetlands into a C source if soil respiration is stimulated more than photosynthesis. Daxing'anling supports the main area of Chinese forested wetlands; however, the effects of harvesting on vegetative C storage and C sequestration in these wetlands are still poorly understood. The aim of this study was to evaluate changes in biomass, C concentration, C storage, net primary productivity and C sequestration under four harvesting intensities (0.0%-an unharvested control, and 25%-low, 35%-moderate and 50%-high intensity selective cutting) 5 years after harvest conducted in Larix gmelinii-Carex schmidtii forested wetlands in the Daxing'anling, northeast China. The study used experimental cutting and sample tree analysis methods to reveal the influence of selective cutting on wetland C sinks and provide a theoretical basis for sustainable wetland management. The main results follow: (1) Total vegetation biomass varied from (135.03±7.72) to (204.71±1.71) t/hm2 at the various selective cutting sites and decreased significantly by 8.7%-34.0% as harvest intensity increased when comparing the three intensities of selective cutting (P<0.05). The decrease in total vegetation biomass mainly resulted from significant reductions of tree biomass, which comprise the dominant component (91.5%-96.0%) of this forested wetland type, although selective cutting could increase the understory biomass and reduce the litter biomass, both of which had only limited impacts on total vegetation biomass. (2) The C concentration of trees such as Larix gmelinii and Betula platyphylla ranged from (439.05±9.70) to (508.41±27.09) g/kg; selective cutting resulted in a significant decrease in the root and leaf C concentrations by 4.1%-11.7% (P<0.05). The C concentration of shrubs, (444.87±5.40)-(472.52±9.44) g/kg, and litter, (433.64±16.23)-(468.82±21.27) g/kg, declined by 3.8%-5.9% and 6.0%-7.5% (P<0.05) at the low-and high-intensity selective cutting sites, respectively. However, there was no significant difference in the herb C concentration (399.34±83.65)-(419.20±23.75) g/kg among the four treatments. (3) Total vegetation C storage varied from (61.16±0.67) to (99.61±1.47) t·C/hm2 at the different selective cutting sites, following a similar pattern as total vegetation biomass. All selective cutting intensities caused significant decreases, by 15.5%-38.6% (P<0.05), in total vegetation C storage which was mainly caused by the reduction of tree C storage which had previously accounted for 91.8%-96.2% of the total vegetation C storage. (4) Total vegetation net primary productivity and C sequestration varied from (6.48±0.28) to (11.87±0.92) t·C·hm-2·a-1 and from (3.52±0.21) to (6.29±0.92) t·C·hm-2·a-1 for the various selective cutting sites, respectively. Both low-and moderate-intensity selective cutting led to significant increases in both net primary productivity and C sequestration by 69.1%-83.2% and 52.0%-78.7% (P<0.05), respectively. Also, the increase in the total vegetation, net primary productivity and C sequestration of the forested wetlands mainly came from an increase in tree growth in the low-intensity selective cutting area and in understory growth with moderate-intensity selective cutting, respectively. Therefore, it appears both low-and moderate-intensity selective cutting may be capable of enhancing the efficiency of C sequestration as well as increasing the C sequestration capacity of this type of forested wetland.