9697 - Laserfiche WebLink

Home Browse Search

YOUNG COLORADO PIKEMINNOW RECRUITMENT I 30 September ~ I ~gust Q I ^ 12 July c:::: <ll ::l D"" <ll .... '"' I ~ 14 June I ~ ~June . 30 40 50 60 70 80 Red shiner TL (mm) FIGURE 3.-Red shiner length frequency distributions used in the individual-based model. Distributions were based on field collections in the middle Green River, 1994. Daily distributions were estimated by linear interpolation of distribution means and standard deviations for simulation modeling. selected at random. This daily attack rate was then multiplied by predator density (predators!m2) to estimate mean number of total attacks. The actual number of attacks on the larva for that day was determined by drawing a value from a Poisson distribution with a mean equal to mean total attacks. Red shiner density and size structure.-Estimates of red shiner densities and size distributions used in the mM were based on 1994 field collections from the middle Green River, Utah (Haines and Tyus 1990; K. Bestgen, unpublished data). The size distribution and relative abundance of red shiners changed over time (Figure 3). In June, the population was predominated by relatively large age-l and age-2 adults. From late June to late July, the size of fish in the red shiner population declined as old and reproductively spent fish died. Then, fish size increased as age-l fish grew until water temperatures declined and red shiner growth rates slowed in autunml. Those size distribu- tions and patterns of change over the summer were assumed representative for other reaches and years, based on summer size distributions of red shiners from Oklahoma and New Mexico, where large adults were common in early summer but uncommon in later summer (Farringer et al. 1979; Gido and Propst 1995; Gido et al. 1997). Limited sampling in Green River backwaters in July 1996 also showed a size shift 1729 similar to that observed in 1994, where large red shiner were relatively common in early July, but were rare later. During a simulation, red shiner size distributions were updated each day by linear interpolation between the mean and standard deviations of the data-based distributions. Each time a larva encountered a red shiner, predator total length was randomly drawn from the size distribution for that day. Adult red shiners regularly attain densities greater than five individuals! m2 in Green River backwaters (Haines and Tyus 1990). Unless otherwise noted, we used a relatively conser- vative and seasonally constant density of three red shiner/m2 in simulations. Temperature-dependent growth.-To evaluate growth rate effects on predation, simulations were conducted using different Colorado pikemirmow base- line growth rates (GRbaseline)' Those rates were adjusted for water temperature using the equation, described by Bestgen (1996). Daily growth rate (GRdaily) was calculated using the equation GRdaily = GRbaseline[( -0.279 + 0.0387T - 0.OOO637T2)jO.283], (4) where T was mean daily water temperature (oC). Each fish was assigned a baseline growth rate (mm/d) by randomly drawing a value from a normal distribution with a mean that was specified for each simulation and a coefficient of variation of 14.1% (Bestgen 1996). Mean baseline growth rates included in the analysis ranged from 0.2 to 0.6 mm/d, a range that represented growth rates of wild fish that survived to autumn. Baseline growth rate was adjusted as a function of water temperature via a proportion that increases or decreases growth rates when mean daily water temperature is higher or lower than 240C, respectively. The proportion was based on a temperature-dependent and ration-dependent growth equation (equation 5 in Bestgen 19%). The value 0.283 in equation (4) is the solution to the temperature-dependent and ration- dependent equation at 240C and high ration. In most simulations, we used a moderate mean baseline growth rate of 0.3 mm/d. The mM simulations were conducted using two different summer temperature regimes (1 June to 30 September) from the Green River near Jensen, Utah (U. S. Geological Survey gauge 09261000), a relatively warm thermal regime during 1994 and a cooler regime in 1983 (Figure 4). We used main-channel water temperatures in simulations be- cause they probably represent the average temperature present in backwaters on a daily basis. This is true because backwater temperatures are higher during the day but colder at night than the main channel (personal observations, KRB). Use of main-channel water