Stream Reach Discretization
The purpose of delineating stream reaches was to identify similar physical units for sampling. An Arc view shape-file of second-, third- and fourth-order (Strahler, 1964) stream segments in the study area was created using a 10-meter digital elevation model (DEM). In this stream layer Lookout Creek is 4th order at its mouth.
Four stream reaches from each order (12 total) were chosen at random to allow inferences to the entire basin. A computer model was used to systematically generate UTM coordinates for a population of possible stream reach locations in streams of each order. Locations were numbered and a subset of them was randomly selected for sampling. Stream reach locations were located in the field by using a hand held GPS unit to locate UTM coordinates obtained from the digital map of the study area. Selected locations were treated as the upstream end of a study reach. The downstream ends of reaches were set equal to a distance of 20 active channel widths from the head of the reach. Watershed area at the head of each reach was also calculated from the DEM. One of the randomly selected fourth-order reaches had an island that split the stream in two for the entire reach length. Both channels were surveyed and included in the analysis as separate reaches, where appropriate. This was justified because the separate channels had an overall different character, did not rejoin within twenty bank-full widths, were nearly identical in flow, and were separated in places by islands that were elevated above the active floodplain.
Stream survey methods
Stream reaches were surveyed with an auto-level and a leveling rod between June and August of 2000 and 2001. A fiberglass measuring tape was stretched between stakes driven into the streambed along the thalweg. Streambed elevations and water surface elevations, relative to an arbitrary benchmark, were surveyed at points along the measuring tape. Survey points were spaced between 0.5 and 8 meters according to stream feature sizes. This approach allowed us to efficiently capture all channel spanning slope breaks in the stream bed and stream water surface. We used definitions from Schumm (1977) as a guide for defining floodplain features and measuring floodplain and channel widths. Accordingly, the active channel was defined as the part of the floodplain that undergoes active erosion and deposition. The active width was identified by the break between bare, recently scoured alluvium and surfaces either occupied by perennial vegetation or elevated enough to have escaped recent scouring. The confined width was measured as the width of the active floodplain between the lowest terrace level, or valley wall, which ever was nearer the channel. Wetted stream channel widths, and active widths were measured at 10 meter intervals for second order streams, and at 15 meter intervals for third and fourth order streams.
Defining slope breaks and channel slope units
Longitudinal profiles of water surface elevations were systematically broken into line segments defined by each consecutive pair of survey points. Slope (gradient) was calculated for each line segment as (Dz / Dx), where Dz is a vertical distance, and Dx is a distance along the axis of stream flow. Each line segment was then assigned to one of three categories: FLAT (flat water units--slope less than 0.025), STEEP (steep water units--0.025 less than slope less than 0.13), or STEP (step units--slope greater than 0.13). Often times, two or more adjacent line segments fell in the same slope category. Collections of consecutive line segments of the same slope category define a Channel Slope Unit (CSU). These are categorized as FLAT, STEEP, or STEP according to the line segments slope criteria. Length and slope were calculated for every CSU. FLATs included what are commonly referred to as pools, runs, and glides; STEEPs included riffles and rapids; and STEPs included steps and cascades. Several values for slope categories were tested and compared, but the ones reported yielded CSUs that most closely matched the pattern of pools, riffles and steps observed in the field.
Additional study reach descriptions
As a supplement to the longitudinal profile, certain stream characteristics were documented providing a more descriptive representation of the study reaches. These were not used for analysis of the slope unit patterns, but were included in this data set for completeness. These descriptions are: the percieved cause of STEP slope units (if caused by a wood, boulders, or a bedrock step); qualitative descriptions of the channel morphologic features; and specific descriptions of stream features or materials at or near the given survey locations. The qualitative catagories used in this study are: Riffle, High Gradient Riffle, Pool, Plunge Pool, Dam Pool, Scour Pool, Trench Pool, , Rapid, Cascade, Falls, , Run, and Glide. (–Justin A., is there a reference for this naming procedure?) The specific descriptions of stream features or materials include: divergence and confluence of side channels, predominant streambed material, lateral inflows from tributaries, and types of woody debris in the channel.
Bars and secondary channels
All but one of the randomly selected stream reaches had some secondary channel development. Most of the secondary channels observed were channel splits separated from the main channel by island bars or transverse bars (Table 2). Bars in second-order streams tended to be small, poorly developed, poorly sorted and associated with large, essentially immobile boulders or wood. In third-order streams bars were larger, better developed, and had sediment that was better sorted. Bars in fourth-order streams were the largest, most well developed, and sediment was the most well-sorted. One channel split, observed in an unconfined reach (416L), was not connected to the main channel at the downstream end. An alcove was observed in the bedrock reach (403). The alcove was associated with a scoured depression between a gravel/cobble bar, and a bedrock wall confining the active floodplain.
The frequency of side channels and bar-formed secondary channels was recorded for each stream reach. Side channels are defined as extensions of the main channel that are separated by islands that are outside the active channel width (i.e. high enough to avoid periodic flooding). Bar-formed secondary channels include channel splits and alcoves. Channel splits are defined as extensions of the channel that are connected to the main channel at their head, but are separated by bars that occur within the active width of the stream. Channel splits may or may not be connected to the main channel at their tails by surface water. Alcoves are slack water channels separated from the main channel by bars; they exist in scoured bed depression connected to the main channel at the downstream end.
