7227 - Laserfiche WebLink

Home Browse Search

290 ferent modifications of the curve, each made to appear more typical. Results showed that these modifications had no important effect on the out- come ofthe analyses nor on our subsequent conclu- sions. U.S. Geological Survey (USGS) annual reports were the source of temperature data for two upper- basin river reaches presently inhabited by Colora- do squawfish (Fig. 1): the Green River near Green River, Utah (data for the years 1975-1983, 1985), and the upper Colorado River near the Colorado- Utah border (1979-1986). Temperatures at the Green River and upper-Colorado locations have not been affected by upstream water-development projects (Robert Green, Regional Hydrologist, U.S. Fish and Wildlife Service, Denver, Colorado, unpublished data}. For the former range of Col- orado squawfish, the few temperature data rec- orded before modification of temperature regimes by upstream dams are from the lower Colorado River near Yuma, Arizona, for the years 1917-1924 (Dil] 1944), and from the middle Colorado River near Grand Canyon, Arizona (USGS annual re- ports for 1943-1947, 1957). Because the temperature data from the lower, middle and upper-river reaches were not concur- rent, we were concerned that possible long-term climatic changes might affect comparisons among these locations. We therefore analyzed climate da- ta provided by the U.S. National Oceanic and At- mospheric Administration for 1901-1986 for Yuma and Flagstaff, Arizona, and Grand Junction, Col- orado, monitoring stations near the sites of lower, middle and upper Colorado River data collection, respectively. Results showed that mean-annual air temperatures for the eight years for which historic lower-river temperature data were available were all cooler than the 86-yr mean for Yuma, and aver- aged 0.8° C less, whereas the six years during which the historic middle-river temperature data were collected were all warmer than the 86-yr average for Flagstaff, and averaged 0.8° C warmer. Of the ten years during which the Green River data were collected, four were cooler and six warmer than the 86-yr average for Grand Junction, and averaged 0.2° C warmer, whereas these respective data for the upper Colorado River location were two cool- er, six warmer, and averaged 0.5° C warmer. Al- though we have no means of standardizing our river-temperature data to account for the effect of these climatic differences, such adjustment would increase the average temperatures that we report for the historic lower river and reduce them for the other locations. River-temperature data were reduced to mean- monthly temperatures for each year of record and means were averaged within months to produce an average-annual temperature regime (the type of data provided by Dill [1944]) for each location. Because they consisted of once-daily measure- ments collected over a wide range of daylight hours, the Green River data had an uncorrectable bias toward the warmer temperatures that occur during daylight. However, our analyses of the con- tinuously recorded data collected at the nearby upper Colorado River gauge indicated this bias was probably no more than 1° C. Using the suitability indices that we assigned to the growth-rate versus temperature relation of Black & Bulkley (1985a), we estimated the relative suitability of each tem- perature regime by summing the indices for its average-monthly temperatures. Age-0 growth analysis Mean total lengths of age-0 Colorado squawfish captured from the Green and upper Colorado riv- ers in fall (mid September-mid October) were ob- tained from Tyus et al. (1987) and from the U.S. Fish and Wildlife Service (USFWS) and Utah Divi- sion of Wildlife Resources (unpublished data). These data were compared to the relative suitabil- ity of the annual temperature regime for their re- spective rivers for the year of capture. Population simulation We used a simple simulation technique to demon- strate how growth rate can affect survival in pop- ulations of slow- and fast-growing fish. Beginning populations consisted of 1000 female larvae 10 mm long, which we arbitrarily accepted as being the .`