Published research emphasizes rapid downstream export of terrestrial carbon from mountainous headwater rivers, but little work focuses on mechanisms that create carbon storage along these rivers, or on the volume of carbon storage. Here we estimate organic carbon stored in diverse valley types of headwater rivers in Rocky Mountain National Park, CO, USA. We show that low-gradient, broad valley bottoms with old-growth forest or active beaver colonies store the great majority of above-and below-ground carbon. These laterally unconfined valley segments constitute o25% of total river length, but store B75% of the carbon. Floodplain sediment and coarse wood dominate carbon storage. Our estimates of riverine carbon storage represent a previously undocumented but important carbon sink. Our results indicate that: not all mountainous rivers rapidly export carbon; not all valley segments are equally important in carbon storage; and historical changes in riverine complexity have likely reduced carbon storage.

We evaluate the validity of the beaver‐meadow complex hypothesis, used to explain the deposition of extensive fine sediment in broad, low‐gradient valleys. Previous work establishes that beaver damming forms wet meadows with multi‐thread channels and enhanced sediment storage, but the long‐term geomorphic effects of beaver are unclear. We focus on two low‐gradient broad valleys, Beaver Meadows and Moraine Park, in Rocky Mountain National Park (Colorado, USA). Both valleys experienced a dramatic decrease in beaver population in the past century and provide an ideal setting for determining whether contemporary geomorphic conditions and sedimentation are within the historical range of variability of valley bottom processes. We examine the geomorphic significance of beaver‐pond sediment by determining the rates and types of sedimentation since the middle Holocene and the role of beaver in driving floodplain evolution through increased channel complexity and fine sediment deposition. Sediment analyses from cores and cutbanks indicate that 33–50% of the alluvial sediment in Beaver Meadows is ponded and 28–40% was deposited in‐channel; in Moraine Park 32–41% is ponded sediment and 40–52% was deposited in‐channel. Radiocarbon ages spanning 4300 years indicate long‐term aggradation rates of ~0.05 cm yr‐1. The observed highly variable short‐term rates indicate temporal heterogeneity in aggradation, which in turn reflects spatial heterogeneity in processes at any point in time. Channel complexity increases directly downstream of beaver dams. The increased complexity forms a positive feedback for beaver‐induced sedimentation; the multi‐thread channel increases potential channel length for further damming, which increases the potential area occupied by beaver ponds and the volume of fine sediment trapped. Channel complexity decreased significantly as surveyed beaver population decreased. Beaver Meadows and Moraine Park represent settings where beaver substantially influence post‐glacial floodplain aggradation. These findings underscore the importance of understanding the historical range of variability of valley bottom processes, and implications for environmental restoration. Copyright © 2011 John Wiley & Sons, Ltd.

No abstract

Summary We assessed the influence of agricultural land use on aquatic–terrestrial linkages along streams arising from changes in the emergence of aquatic insects. We expected that terrestrial predators would respond to a change in the abundance and/or the size structure of the emerging aquatic insects by an increase or decrease in population size. We measured the flux of emergent aquatic insects and the abundance of terrestrial invertebrate predators and birds along 10 streams across a forest‐to‐agriculture land‐use gradient. We also performed stable isotope analyses (hydrogen, carbon and nitrogen) of terrestrial invertebrate predators. Small aquatic insects (Nematocera) were most abundant under agricultural land use, whereas larger bodied aquatic insects (Plecoptera and Trichoptera) were more associated with forest land use. Carabid beetles and linyphiid spiders were associated with agricultural streams (where there was a high abundance of small aquatic insects), whereas lycosid spiders and birds were associated with forest streams and a high abundance of large‐sized aquatic insects. The contribution of aquatic insects to the diets of riparian Lycosidae, Linyphiidae and Carabidae was estimated to be 44%, 60% and 43%, respectively, indicating the importance of aquatic subsidies to the terrestrial system. Our results show that agricultural land use in an overall forested landscape can have significant effects on the abundance and diet of terrestrial consumers through its impact on the size structure of the assemblage of emerging insects, rather than the overall magnitude (numbers) of the aquatic subsidy. Hence, our results suggest that the composition, not just quantity, of a cross‐habitat resource may influence the recipient system.

We reviewed follow-up studies from Finnish and Swedish streams that have been restored after timber floating to assess the abiotic and biotic responses to restoration. More specifically, from a review of 18 case studies (16 published and 2 unpublished), we determined whether different taxonomic groups react differently or require different periods of time to respond to the same type of restoration. Restoration entailed returning coarse sediment (cobbles and boulders) and sometimes large wood to previously channelized turbulent reaches, primarily with the objective of meeting habitat requirements of naturally reproducing salmonid fish. The restored streams showed a consistent increase in channel complexity and retention capacity, but the biotic responses were weak or absent in most species groups. Aquatic mosses growing on boulders were drastically reduced shortly after restoration, but in most studies, they recovered after a few years. Riparian plants, macroinvertebrates and fish did not show any consistent trends in response. We discuss seven alternative explanations to these inconsistent results and conclude that two decades is probably too short a time for most organisms to recover. We recommend long-term monitoring using standardized methods, a landscape-scale perspective and a wider range of organisms to improve the basis for judging to what extent restoration in boreal streams has achieved its goal of reducing the impacts from timber floating.