Battle Creek Overview | Bibliography | Home

Pool Frequency and Depth in Battle Creek

Pools are important for various life stages of salmon, steelhead and resident trout and disturbance to the landscape can cause loss of pool habitat and limit fisheries productivity. The Battle Creek Watershed Assessment (WA) (Terraqua, 2004) measured the frequency of pools and residual pool depth following Aquatic and Riparian Effectiveness Monitoring Program (AREMP) protocols (Gallo, 2002) (see note on methods). These two measures were used as an indication of suitability for salmonids according to the Ecosystem Management Decision Support (EMDS) criteria (Reynolds, 2001).

The U.S. Forest Service (Armentrout et al., 1998) studied Deer, Antelope and Mill creeks, immediately to the south of the Battle Creek watershed, and noted that low gradient reaches tended to be subject to pool filling and fines accumulation and, therefore, sensitive to sediment increases. They defined pool depths of greater than 3.3 feet as suitable for juvenile salmonid rearing.

Findings of Battle Creek Watershed Assessment

Battle Creek watershed pool frequency and depth was analyzed by Terraqua (2004): 

• "Scour pools were characterized at 50 sample sites in the watershed. Additionally, non-scour formed pools (e.g., wood- or bedrock-controlled pools) were characterized at 38 of 50 sample sites. Scour pools made up 76 percent of the total pools at the 38 sites where scour and non-scour pools were both assessed. 
• All pools were further characterized as occurring in anadromous or non-anadromous habitats. Of the 191 pools observed; sixty three (33%) were located in the 17 sample sites occurring in anadromous habitats, and 128 pools (67%) occurred in the 33 non-anadromous sample sites. Pools occurred at about the same rate for anadromous (3.7 pools/site) and non-anadromous (3.9 pools/site) sites. 
• Pool frequency (based on scour pools only) was described at 49 sample sites as the number of scour pools per 100 meters of stream. Among all sites within the Battle Creek watershed, the mean pool frequency was 1.7 scour pools per 100 meters of stream while pool frequency at individual sites ranged from 0 to 6.9 scour pools per 100 meters of stream. 
• Pool frequency decreased significantly with increasing bankfull width (F-test; p = 0.02; Figure 10) although this relationship is notable for its weakness (small coefficient of determination, r2 = 0.18; and small regression coefficient, b=-0.06) and for the fact that pool frequency was not lognormally distributed, as was expected."

This chart from the Battle Creek WA shows the distribution of scour pools versus the expected log-normal relationship with channel width, which is how the EMDS pool frequency rating curve is calculated. The green dots indicate sites where scour pool frequency is "fully favorable" for salmonids, red dots show sites where pool frequency was "fully unfavorable", and light red is "likely unfavorable". Taken from Figure 10 in the Battle Creek WA.

The Battle Creek WA shows a map of scour pool frequency at all sites measured in 2001 and 2002 with EMDS criteria displayed as color codes with green indicating "fully favorable" salmonid habitat and red "fully unfavorable" conditions. The only reaches with fully favorable pool frequency for salmonids are four within lower Battle Creek and one on the South Fork above Panther Creek. Partially favorable habitats are shown in gray and pink. Taken from Figure 11 in the Battle Creek WA.

This chart shows the raw data for scour pool frequency per 100 meters, which were collected as part of the Battle Creek WA. The highest unweighted pool frequencies were on the mainstem of Battle Creek above Coleman Hatchery and at sites on U.S. Forest Service lands in Nannie Creek and upper South Fork Digger Creek. Note that several sites had no scour pools. Chart from KRIS Battle Creek V 2.0.

Terraqua (2004) characterized pools 3 feet deep or more as suitable for anadromous fish but pools 3 feet to 10 feet deep as optimal for over-summer holding of adult salmon. Pools 1 foot deep or more were characterized as "fully favorable" for resident trout. The Battle Creek WA had the following findings regarding pool depths:

• "The mean pool depth in habitat accessible to anadromous salmonids was 0.92 meters, and the maximum observed pool depth was 2.46 meters. At sites accessible to anadromous salmonids, 60 percent of the pools were considered fully unfavorable (depths < 1.0 meter) for adult spring Chinook salmon holding. We found 25 pools fully favorable for adult spring Chinook holding within the 5,168 meters of habitat accessible to anadromous salmonids that we surveyed.
• The mean residual pool depth at sites not accessible to anadromous salmonids was 0.4 meters, and the maximum depth observed was 2.3 meters. At sites not accessible to anadromous salmonids, 40 percent of the pools were considered fully unfavorable (depth < 0.3 meters) for adult trout production."

The chart at left shows the frequency of occurrence of pools by depth at all Battle Creek WA sites. Pools in resident trout reaches were less than the 1 foot deep criteria ("fully unfavorable") at 40% of sites measured and 60% of pools at sites accessible to anadromous fish were less than three feet deep, making them "fully unfavorable". Chart from  KRIS Battle Creek V 2.0.

The Battle Creek WA (Terraqua, 2004) drew the following conclusions about pool frequency and depth in Battle Creek: 

• "Pool frequencies and depths were lower at middle and high elevation sites than predicted though pool frequencies had increased or remained the same since 1988. Approximately 150 pools with depths greater than 1 meter would be suitable as spring Chinook salmon adult holding habitat although very few of these pools are greater than 2 meters deep....
• Residual pool depth was significantly correlated with elevation (negatively), watershed area (positively), stream gradient (negatively), and the proportion of the watershed in rain-on-snow area (negatively)....
• Model exploration confirmed that road density, near-stream road density, road-stream crossings, proportion of watershed in rhyolite soils, and forest cover play no meaningful role in predicting variability in observed residual pool depth data."

The Battle Creek WA also found a relationship between decreased residual pool depth and area of the watershed at elevations susceptible to rain-on-snow effects.

References

Armentrout, S., H. Brown, S. Chappell, M. Everett-Brown, J. Fites, J. Forbes, M. McFarland, J. Riley, K. Roby, A. Villalovos, R. Walden, D. Watts, and M.R. Williams, 1998. Watershed Analysis for Mill, Deer, and Antelope Creeks. U.S. Department of Agriculture. Lassen National Forest. Almanor Ranger District. Chester, CA. 299 pp. [6.0 Mb]  

Gallo, K. 2002. Field protocols: Aquatic and Riparian Effectiveness Monitoring Program for the Northwest Forest Plan: Version 1.0. U.S. Forest Service, Corvallis, OR. 54 pp. [125 Kb]

Reynolds, Keith M. 2001. Fuzzy logic knowledge bases in integrated landscape assessment: examples and possibilities. Gen. Tech. Rep. PNW- GTR-521. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station. 24 p.

Terraqua, Inc. 2004. Battle Creek Watershed Assessment :Characterization of stream conditions and an investigation of sediment source factors in 2001 and 2002. Performed under contract to the Battle Creek Watershed Conservancy, Manton, CA. Funds from the Anadromous Fisheries Restoration Program and U.S. Fish and Wildlife Service under Agreement DCN: 11330-1-J113.