Search

We measured wood piece characteristics and particulate organic matter (POM) in stored sediments in 30 channel-spanning logjams along headwater streams in the Colorado Front Range, USA. Logjams are on streams flowing through old-growth (>200 years), disturbed (<200 years, natural disturbance), or altered (<200 years, logged) subalpine conifer forest. We examined how channel-spanning logjams influence riverine carbon storage (measured as the total volatile carbon fraction of stored sediment and instream wood). Details of carbon storage associated with logjams reflect age and disturbance history of the adjacent riparian forest. A majority of the carbon within jams is stored as wood. Wood volume is significantly larger in old-growth and disturbed reaches than in altered reaches. Carbon storage also differs in relation to forest characteristics. Sediment from old-growth streams has significantly higher carbon content than altered streams. Volume of carbon stored in jam sediment correlates with jam wood volume in old-growth and disturbed forests, but not in altered forests. Forest stand age and wood volume within a jam explain 43% of the variation of carbon stored in jam sediment. First-order estimates of the amount of carbon stored within a stream reach show an order of magnitude difference between disturbed and altered reaches. Our first-order estimates of reach-scale riverine carbon storage suggest that the carbon per hectare stored in streams is on the same order of magnitude as the carbon stored as dead biomass in terrestrial subalpine forests of the region. Of particular importance, old-growth forest correlates with more carbon storage in rivers. [ABSTRACT FROM AUTHOR]

Abstract: The channel head, defined as the upstream boundary of concentrated water flow and sediment transport between definable banks, represents the transition from hillslope processes to fluvial processes. The ability to delineate the location along a slope at which channels initiate is important for understanding hydrologic and geomorphic processes governing headwater streams. Studies demonstrating an inverse relationship between either contributing drainage area (A) and local valley slope (θ) or basin length (L) and θ for channel heads come primarily from regions with humid climates. Seventy-eight channel heads were mapped in the headwaters of rivers in the semiarid Colorado Front Range. Multiple field sites were chosen to account for variability from aspect and elevation. Surface topographic parameters were measured in the field and analyzed to test the hypothesis that surface processes control channel initiation in this region. Although simple linear regressions indicate a poor inverse relationship between A and L and no relationship between L and θ, multiple regressions indicate that surface topographic parameters explain over half the variability in the location of channel heads. This suggests that surface processes exert an influence on channel initiation, but do not explain as much of the variability as observed in previous studies from wetter regions. A threshold of erosion necessary to initiate a channel was observed at ~10,000m2 for A, although values as high as 600,000m2 were mapped for some channel heads. Variation within the study area correlated with elevation, which is a proxy for differences in volume and type of precipitation; sites at lower elevation with less precipitation, but more intense convective rainfall, tend to have smaller contributing area and basin length. Aspect did not influence surface topographic parameters. We interpret the results to reflect surface and subsurface influences on channel initiation. Nearly 70% of the mapped channel heads coincide with constrictions created by weathered bedrock knobs, suggesting that local factors strongly influence channel initiation once the threshold for contributing area is exceeded. Field-mapped channel head locations plot at or downslope from the inflection point of a regional slope–area curve generated from 10m DEMs, although some extend well downslope into the region assumed to be dissected by alluvial channels. Most actual drainage areas for channel initiation are thus an order of magnitude larger, and plot in a significantly different portion of the slope–area graph, than would result from the widespread practice of assuming channel heads are located at the gradient reversal in such curves. [Copyright &y& Elsevier]