Invasive Reed Canary Grass (Phalaris arundinacea) and Carbon Sequestration in a Wetland Complex

Abstract

Terrestrial carbon sequestration is one of several proposed strategies to reduce the rate of carbon dioxide (CO2) accumulation in the atmosphere, but the impact of plant invasion on soil organic carbon (SOC) storage is unclear. The results of past studies are often confounded by differences in vegetation and environmental conditions. Reed canary grass (Phalaris arundinacea) is an herbaceous species that invades riparian fringes and wetlands throughout North America, including Beanblossom Bottoms – a wetland complex in south-central Indiana. Because of the prolific growth of P. arundinacea, it was hypothesized that significant alterations in SOC pools and dynamics would occur at invaded sites within the wetland complex. To test this hypothesis, study plots were established in areas colonized either by native herbaceous species or by P. arundinacea. Above and below-ground biomass were collected at the middle and end of the growing season and were analyzed for cellulose, lignin, acid detergent fiber, total phenolics, and organic carbon and nitrogen concentration. Soil samples were analyzed for SOC and nitrogen, bulk density, pH, and texture. The biomass of Scirpus cyperinus – a native wetland species was found to contain significantly (P < 0.05) more lignin (168 g kg-1 versus 98 g kg-1) and phenolics (19 g kg-1 versus 3 g kg-1), and had a higher C to N ratio (28 versus 20) than P. arundinacea biomass, suggesting greater recalcitrance of S. cyperinus tissues compared to P. arundinacea biomass. Results of a laboratory incubation study were consistent with the residue biochemistry data and showed that S. cyperinus biomass degraded at much slower rates than the biomass of P. arundinacea. However, measurements of SOC pools (0-30 cm) showed larger pools under P. arundinacea (25.5 Mg C ha-1) than under stands of S. cyperinus (21.8 Mg C ha-1). Likewise, SOC stocks under stands of mixed native vegetation were significantly (P < 0.05) smaller (18.8 Mg C ha-1) than in areas invaded by P. arundinacea. Biomass of the mixed native vegetation was also considered more recalcitrant than that of P. arundinacea based on residue biochemistry. Therefore, contrary to the study hypothesis, residue quality was not a good predictor of SOC stocks in the wetland soils. Thus, it appears that traditional laboratory assessments of biomass recalcitrance and decomposition do not accurately simulate the various biological interactions occurring in the field.