Demographic drivers of biomass carbon recovery in degraded perennial tussock grassland, with and without domestic grazing
- Landcare Research, PO Box 69040, Lincoln 7640, New Zealand
- Upland Research, 14 Chartres Lane, Arrowtown 9302, New Zealand
- Korowai Ecology, 6 Victoria Street, Springfield 7681, New Zealand
Many of New Zealand’s natural and induced tussock grasslands are in a degraded low-biomass state due to a combination of fire, overgrazing and weed invasion. The capacity of degraded grasslands to recover biomass is uncertain because legacies of degradation can strongly influence the demographic processes controlling ecosystem recovery. We develop a conceptual framework for understanding biomass carbon (C) flux in degraded perennial grassland based on demographic processes of growth, mortality and recruitment. We apply this framework to 22 years of data from Chionochloa- and Festuca-dominated perennial tussock grassland in the South Island high country, sampling two grazed sites and two sites retired from domestic grazing. Total biomass C stocks were assessed at site level using 174 temporary plots measured in 1989 and 2011. Long-term demographics were monitored using 24 permanent plots in which 7213 individual tussocks were tagged and monitored every 1–5 years from 1989 to 2011. Although biomass C stocks were generally low (2159 ± 494 kg C ha–1), there was a significant increase in total biomass C over the 22-year period at the north-facing retired site of 92 ± 47 kg C ha–1 year–1. Increases in total biomass C were generally due to recruitment and growth of tall tussock (Chionochloa spp.) and/or woody shrubs. Biomass C stock in grazed sites, and the retired south-facing site, declined by up to 65 kg C ha–1 year–1 or remained constant. Declines were due to dieback of the dominant short tussock (Festuca novae-zelandiae) across all treatments, and a lack of recruitment of tall tussock and shrub species. Our results suggest that retirement from grazing was not sufficient to ensure total biomass C gains after 22 years, and highlight the roles of recruitment limitation, shrub establishment and management history in controlling ecosystem recovery.