1206 Ecological Applications, 11(4), 2001, pp. 1206–1223  2001 by the Ecological Society of America CONTROL OF LITTER DECOMPOSITION IN A SUBALPINE MEADOW–SAGEBRUSH STEPPE ECOTONE UNDER CLIMATE CHANGE M. REBECCA SHAW 1,2,3 AND JOHN HARTE 2 1 Carnegie Institution of Washington, Department of Plant Biology, 260 Panama Street, Stanford, California 94305 USA 2 Energy and Resources Group, 310 Barrows Hall, University of California, Berkeley, California 94720 USA Abstract. Litter decomposition is an important component of the global carbon and nitrogen cycles. Because climate exerts strong controls over rates of litter decomposition, climate change may alter both cycles. Climate change can influence litter decomposition rates directly, or indirectly through changes in litter quality. The relative importance of climate and litter quality in controlling rates of decomposition varies across ecosystem types. Thus, ecosystem responses to climate change are not readily predictable. This study examines in situ litter decomposition rates of native plant litter of different growth forms (grass, forb, and shrub) in two microclimates (xeric and mesic). In a Colorado subalpine meadow–sagebrush steppe ecotone, a climate-warming treatment was used to determine the influence of litter quality, microclimate, and warming on the rates of decomposition. Three one-year litter bag experiments were performed using senescent leaf litter of the three growth forms from a xeric microclimate (shrub, Artemisia tridentata; grass, Festuca thurberi; forbs, Delphinium nuttallianum, Erigeron speciosus) and a mesic microclimate (shrub, Pentaphylloides floribunda; grass, Festuca thurberi; forbs, Erythronium grandiflo- rum, Ligusticum porteri). A reciprocal transplant litter experiment was performed in the third year to determine the direct effect of warming on litter quality and subsequent litter mass loss rates. Evidence suggests that decomposition was limited by moisture in the xeric zone and by temperature in the mesic zone. Decomposition rates were strongly correlated with the initial lignin:N ratio of the litter. The forbs had a much lower initial lignin:N ratio and, therefore, decomposed at a much higher rate than did the grasses or shrubs. In a changing climate, initial microclimate and changes in litter quality of the bulk litter produced as a result of shifts in species composition may be more important in determining rates of litter decomposition than the direct effect of warming. Key words: carbon cycling; climate change; growth form diversity; litter decomposition; litter quality; nitrogen cycling; sagebrush steppe; soil microclimate; species composition change; subalpine meadow. INTRODUCTION The atmospheric concentration of CO 2 is increasing by 1.8 L/L each year because of anthropogenic emis- sions. General circulation models (GCMs) predict that this rise in CO 2 will increase global mean surface tem- peratures by 1.5°–3.5°C (Intergovernmental Panel on Climate Change [IPCC] 1995). High-altitude and high- latitude regions are expected to warm disproportion- ately with rising atmospheric temperatures (IPCC 1995). The temperature increase is expected to directly alter ecosystem processes such as carbon (C) and ni- trogen (N) cycling (Bonan and Van Cleve 1992, Jon- asson et al. 1993). Impacts may include the modifi- cation of litter quality, litter decomposition rates, and nutrient mineralization rates that control the availabil- ity and cycling of nutrients (Nadelhoffer et al. 1992, Hobbie 1996). Under increasing atmospheric concen- Manuscript received 12 October 1999; revised 31 July 2000; accepted 10 August 2000. 3 Address correspondence to Carnegie Institution of Wash- ington. E-mail: shaw@jasper.stanford.edu trations of CO 2 , litter quality may improve (Couteaux et al. 1996), decline (Ko ¨rner and Miglietta 1994), or remain unchanged, depending on the system (Chu et al. 1996, Franck et al. 1997, Norby et al. 1999). In- creasing atmospheric temperatures may alter plant spe- cies composition (Chapin and Shaver 1985, Wookey et al. 1993, 1994, Chapin et al. 1995, Harte and Shaw 1995) and therefore alter litter production and quality indirectly through species-specific responses (Hobbie 1992, Cornelissen 1996). Other investigators have dis- cussed these changes for tundra (Shaver and Chapin 1991, Hobbie 1996), grassland (Kemp et al. 1994, Vin- ton and Burke 1995, Chu et al. 1996), and forest eco- systems (Alban and Pastor 1993, Zou et al. 1995), as the total bulk litter produced varies in litter quality and decomposability with shifts in species composition. In- creasing temperatures could increase rates of litter de- composition directly by increasing microbial activity, releasing C and N from the soil organic matter (SOM; Nadelhoffer et al. 1992). Decomposition could also be hindered if litter moisture substantially declines with