Effects of land use and management on ecosystem respiration in alpine meadow on the Tibetan plateau

Land use change and management practices have greatly affected global carbon dynamics, especially the critical process of ecosystem respiration (Re). Our aim was to assess how Re and Re intensity (Re per unit aboveground biomass) are affected by land use (i.e. native alpine meadow with winter grazing (NAM), abandoned cropland/pastureland (APL), perennial Elymus nutans (PEN) and annual Avena sativa L. pasture (AO)) and management practices (i.e. nitrogen (N) inorganic fertilizer or sheep manure for APL, PEN and AO, tillage and no-tillage for AO) in a Gelic Cambisol soil underlying alpine meadow on the Tibetan plateau from 2008 to 2010. In the present study we hypothesized that (1) the conversion of abandoned cropland/pastureland to pasturelands would increase Re and Re intensity; (2) natural restoration following the abandonment of cropland/pastureland would decrease Re and Re intensity; and (3) management practices mentioned above would reduce the magnitude of the effects of land use change on Re and Re intensity in the alpine meadow ecosystem. Generally, our results did not support our hypotheses. There were no significant differences between annual average Re in NAM and APL, while the annual average Re for native meadows were 2–3 times higher than that of pasturelands in 2008. Re intensities for PEN and AO were 1.7 times lower than NAM and APL, and Re intensity was higher for APL than for NAM in 2010. Management practices did not significantly affect Re and Re intensity for PEN and AO treatments, except for the interaction between tillage and year on the Re of AO in 2010. The results suggest that conversion of abandoned cropland/pastureland to pastureland decreases Re and Re intensity, while natural restoration following the abandonment of cropland/pastureland increased Re intensity. Management practices did not significantly modify the patterns of the effects of land use on Re and Re intensity.

► Land use change did not significantly affect annual average Re except first year.
► Cropland significantly decreased Re intensity compared with abandoned cropland.
► Annual average Re and Re intensity was not significantly influenced by N fertilizer and sheep manure.