Decomposition of roots in a seasonally flooded swamp ecosystem

Abstract

Decomposition rates of roots to a depth of 40 cm were estimated by two methods in four plant communities in the periodically flooded Great Dismal Swamp. The community dominants were Chamaecyparis thyoides (L.) BSP., Taxodium distichum (L.) Richard, mixed Quercus spp. and Acer rubrum L.- Nyssa spp. respectively. Modified litter bags and a core method were simultaneously employed on three flooded sites and an unflooded site. Long vertical litter bags were developed to measure root decay over a vertical soil profile with minimal disturbance to the soil. Reciprocal samples (litter from each of the other sites) were placed on each site to examine the effects of litter quality. Roots in the cores exhibited higher decay rates than in the litter bags; rates in the bags ranged from 0.48 to 1.00 mg g −1 day −1 and the range for the cores was 1.15–2.74 mg g −1 day −1. The core method was ineffective in estimating decay rates for the >5 mm diameter roots because of high sample variability. Reciprocal samples revealed statistically significant differences between root types, with roots from the mixed Quercus site being most resistant to decay. Just as leaf litter quality has been shown to regulate aboveground decomposition, root quality appears to play a major role in belowground decay. The present study also demonstrated the importance of environmental influences since root decay was slowest on the sites ( Chamaecyparis and Acer-Nyssa) with the longest duration of soil saturation. Both techniques exhibited slow decay rates with increasing depth. The litter bag technique is the recommended approach; several problems make the core technique a less than satisfactory method.