Oecologia (2004) 141: 641–651 DOI 10.1007/s00442-004-1689-x ECOSYSTEM ECOLOGY Lara M. Kueppers . John Southon . Paul Baer . John Harte Dead wood biomass and turnover time, measured by radiocarbon, along a subalpine elevation gradient Received: 26 January 2004 / Accepted: 14 July 2004 / Published online: 26 August 2004 # Springer-Verlag 2004 Abstract Dead wood biomass can be a substantial fraction of stored carbon in forest ecosystems, and coarse woody debris (CWD) decay rates may be sensitive to climate warming. We used an elevation gradient in Colorado Rocky Mountain subalpine forest to examine climate and species effects on dead wood biomass, and on CWD decay rate. Using a new radiocarbon approach, we determined that the turnover time of lodgepole pine CWD (340±130 years) was roughly half as long in a site with 2.5–3°C warmer air temperature, as that of pine (630 ±400 years) or Engelmann spruce CWD (800±960 and 650±410 years) in cooler sites. Across all sites and both species, CWD age ranged from 2 to 600 years, and turnover time was 580±180 years. Total standing and fallen dead wood biomass ranged from 4.7±0.2 to 54 ±1 Mg ha -1 , and from 2.8 to 60% of aboveground live tree biomass. Dead wood biomass increased 75 kg ha -1 per meter gain in elevation and decreased 13 Mg ha -1 for every degree C increase in mean air temperature. Differ- ences in biomass and decay rates along the elevation gradient suggest that climate warming will lead to a loss of dead wood carbon from subalpine forest. Keywords Coarse woody debris . Decomposition . Ecosystem carbon balance . Subalpine forest . Climate Introduction Standing and fallen dead trees are ecologically significant components of forest ecosystems, contributing fuel to forest fires (Tinker and Knight 2000), sources of nutrient availability (Hart 1999), and habitat for a wide range of plant and animal species (Harmon et al. 1986). Dead wood can also account for a substantial fraction of forest biomass and stored carbon (c). Temperate forests contain highly variable amounts of standing (snags) and fallen dead trees, or coarse woody debris (CWD), depending on management history, species composition, and environ- mental conditions (Harmon et al. 1986). Forests with the highest CWD biomass are typically old-growth Douglas fir forests of the Pacific Northwest, while forests in the montane western USA are at the low end of the range for temperate forests (Fahey 1983; Harmon et al. 1986). However, where dead wood biomass is relatively low in absolute terms, it can account for a large fraction of ecosystem biomass (Arthur and Fahey 1990). Measurement of dead wood biomass is complicated by spatial variation and lack of suitable allometric equations relating snag volume or mass to girth and height, and estimates of dead wood turnover time are extremely difficult to obtain with high confidence. In addition, few studies have performed complete inventories of C stored in all living and dead biomass, and provided an estimate of the fraction of C stored in dead wood and its turnover time relative to other ecosystem C pools. As a result, this heterogeneous C pool is not well represented in ecosystem models, though it is a significant component of the forest C cycle. Electronic Supplementary Material Supplementary material is available for this article at http://dx.doi.org/10.1007/s00442-004- 1689-x L. M. Kueppers . J. Harte Department of Environmental Science, Policy and Management, Ecosystem Sciences Division, 151 Hilgard Hall, University of California, Berkeley, CA, 94720, USA J. Southon Earth System Science Department, 3200 Croul Hall, University of California, Irvine, CA, 92697, USA P. Baer . J. Harte Energy and Resources Group, 310 Barrows Hall, University of California, Berkeley, CA, 94720, USA L. M. Kueppers (*) Department of Earth Sciences, 1156 High St., University of California, Santa Cruz, CA, 95064, USA e-mail: kueppers@es.ucsc.edu Tel.: +1-831-4593504 Fax: +1-831-4593074