Page 14 



operating the backpack electrode slighty in front of the netter. 

 The backpack operator would quickly stab the water with the 

 electrode in an effort to keep the durations of electrical 

 current to a minimum and to avoid herding or spooking the fish. 

 When fish were captured, the sites were marked with a rock 

 wrapped with yellow or fluorescent orange flagging. If mass fish 

 movement was seen, or if it was felt for any other reason that 

 the fish had moved in response to the presence of the crew, then 

 habitat measurements were not taken from that capture site. 



This sampling strategy attempted to avoid three types of 

 bias that can occur when investigating microhabitat use: 1) fish 

 tend to be displaced from focal points by electrical currents and 

 the activity of the workers; 2) certain habitat types or areas 

 tend to be disproportionately sampled relative to other habitat 

 types or areas; and 3) the efficiency of capture changes with 

 different habitat types. In most cases, this strategy was 

 probably successful in avoiding bias #1. Most sampling was in 

 clear and shallow (under 2 ft deep) water, which allowed the crew 

 to directly observe the fish and evaluate if they had been 

 displaced. Bias #2 was avoided in the sense that the entire 

 length of both sampling sections was electrof ished, and no areas 

 were left out. However, there was some bias toward riffles, 

 since these areas were investigated more thoroughly than pool 

 areas. There were only two types of situations in which bias #3 

 existed. In water depths up to about 3 feet deep it was possible 

 to capture all stunned fish, but at greater depths some fish 

 probably escaped detection or capture. Also, in areas with dense 

 mats of aquatic vegetation, the capture efficiency went down 

 because the stunned fish were concealed by the vegetation. 



Habitat measurements were taken from one day to one month 

 after electrof ishing. In six areas where large numbers of fish 

 were captured ("intensive-use areas") , thorough habitat mapping 

 was conducted. A series of transects were established 

 perpendicular to the streamflow, and typically spaced 5-10 ft 

 apart. Upstream and downstream limits of the transects were 

 designed to bracket the capture sites, so that at least one 

 transect was located upstream and one downstream from the 

 uppermost and lowermost capture sites, respectively. Between 

 these limits, other transects passed through most capture sites. 

 Water depth and velocity were measured at 1.09 ft (0.33 m) 

 intervals along each transect, although in a few backwater areas 

 an interval of 3.28 ft (1.0 m) was used. Substrate and cover 

 were determined by estimating the amount or percent in a cell, 

 which was centered on the 1.09 ft measurement sites and extended 

 halfway to the adjacent measurement sites on the same transect 

 and halfway to the neighboring transects both up and downstream 

 (Figure 2). Water depth and mean velocity measurements (at 

 6/lOths depth) were taken with a Type AA flow meter (Teledyne and 

 Price models) and rod. Substrate composition was visually 



