Effects of large woody debris . stream channels and benthic

Large woody debris (LWD) was added as an experimental stream restoration technique in two streams in southwest Virginia. Additions were designed to compare human judgement in log placements against a randomized design and an unmanipulated reach, &d also to compare effectiveness in a lowand a high-gradient stream. Pool area increased 146% in the systematic placement and 32% in the random placement sections of the low-gradient stream, lending support to the notion that human judgement can be more effective than placing logs at random in low-gradient streams. Conversely, the high-gradient stream changed very little after LWD additions, suggesting that other hydraulic controls such as boulders and bedrock override LWD influences in high-gradient streams. Logs oriented as dams were responsible for all pools created by additions regardless of stream or method of placement. Multiple log combinations created only two pools, while the other seven pools were created by single LWD pieces. Total benthic macroinvertebrate abundance did not change as a result of LWD additions in either stream, but net abundances of Plecoptera, Coleoptera, Trichoptera, and Oligochaeta decreased, while Ephemeroptera increased significantly with the proportional increase in pool area in the low-gradient stream. R&urn6 : De gros debris ligneux (GDL) ont ete ajoutes darts deux cours d’eau dans le sud-ouest de la Virginie pour &udier une technique exp&imentale de remise en &at des cours d’eau. Ces debris ont ete ajout& pour comparer le jugement humain dam le placement des billes de bois par rapport a un placement aleatoire et a un troncon non manipule, et egalement pour comparer l’efficacite de ces m&odes dam un ruisseau a forte declivite et dans un ruisseau a faible declivit& La superticie de la nappe d’eau a augmente de 146% dam le cas du placement systematique comparativement a 32% dam les troncons a placement aleatoire du ruisseau 9 faible declivit.& ce qui sous-tend l’idee que le jugement humain peut &re plus efficace que le hasard dans le placement des billes dans les misseaux a faible declivite. lnversement, le ruisseau a forte d&&it6 a t&s peu change apres l’addition de GDL, ce qui donne a entendre que d’autres facteurs hydrauliques tels les rochers et le substratum rocheux, l’emportent sur les effets des GDL dans les ruisseaux a forte declivite. Les billes orient&es de maniere a former des barrages ont ete a l’origine de tous les bassins c&s par addition quel que soit le ruisseau ou la methode de placement. Les combinaisons de plusieurs billes n’ont cre6 que deux bassins, tar&s que les sept autres bassins ont ete le fait de GDL uniques. L’abondance totale des macroinvert&es benthiques n’a pas change par suite de l’addition de GDL dans l’un ou l’autre type de ruisseau, mais l’abondance nette de plecopteres, de col&opt&es, de trichopteres et d’oligochetes a diminue alors que celle des ephemeropt&es a augment6 de man&e statistiquement signiflcative en fonction de l’augmentation de la superticie de bassins darts le ruisseau a faible declivite. [Traduit par la Redaction] Introduction *: The importance of large woody debris (LWD) is well documented in temperate stream ecosystems (Harmon et al. 1986). LWD influences geomorphic processes (Keller and Swanson k 1979; Swanson et al. 1982), transport and storage of organic Received November 6,1995. Accepted September 27, 1996. 513150 RH. Hilderbrand,’ A.D. Lemly,’ C.A. Dolloff: and K.L. Harpster. Department of Fisheries and Wildlife Sciences, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061-0321, U.S.A. 1 Author to whom all correspondence should be sent at the following address: Department of Fisheries and Wildlife, Utah State University, Logan, UT 84322-5210, U.S.A. e-mail: [email protected] * Affiliation: U.S. Forest Service, Southern Research Station Coldwater Fisheries Research Unit. Can. J. Fish. Aquat. Sci. 54: 931-939 (1997) materials (Bilby and Likens 1980; Speaker et al. 1984, Smock et al. 1989; Trotter 1990), fish habitats (Bryant 1983; Dolloff 1986; Bisson et al. 1987), and aquatic invertebrate habitats (Anderson et al. 1978; Benke et al. 1984; O’Connor 1991). It also serves as an interface linking terrestrial and aquatic systems (Triska and Cromack 1980). Most pools in old-growth streams are associated with LWD (Bilby 1984; Amlrus et al. 1988; Carlson et al. 1990; Robison and Beschta 1990). Its presence is integral in maintaining stream channel complexity by creating and maintaining these pools (Swanson et al. 1976, 1984; Keller and Swanson 1979; Bisson et al. 1987; Maser et al. 1988; Fausch and Northcote 1992). An&us et al. (1988) found that 70% of the pools in their streams were formed by LWD. Similarly, Carlson et al. (1990) reported that 50% of the debris within the bank full channel was contributing to pool formation. In contrast, streams in disturbed forests often have lower LWD loadings (Silsbee and Larson 1983; Sedell et al. 1988; Richmond 1994) and lower pool number and area (Ralph et al. 1994).