|
<br />01191
<br />
<br />HU~IPB.-\Cr;: CHUB HABITAT MOD::LLlNG
<br />
<br />381
<br />
<br />preferentially select shorelines composed of vegetation, talus. or ~ebri:, fans and arc found in much Im...'er densities
<br />over cobble, bedrock, and sandy substrates (Con\'crse el al., 1998). II is hypothesized that a decrease in the fre-
<br />quency of low.flow periods due to dJOl operations has reduced suitable habitat availability and, coupled with lower
<br />water tempcra:ures, hJve probably inflicted Lln extreme negative effecl onjuvenile humpback chuh (Converse et aI.,
<br />1998), This hypothesis is a critical component of the logic used to d<\elop the SASF regime, Valdez and Ryel
<br />(1995) also hypothesized that 'recruitment of young [chub) may be d'r~ndent on their ability to remain and mature
<br />in habitats required by adults'. that is. in habitat within relatively clo" proximity to the LCR, Thus. differences in
<br />discharge and water temperatures in the mains[em during the dispers.:.! paiod could potentially affect !he ability of
<br />YOY and juvenile chub to find and rem,in in their preferred habitat; in the m,instem downstream of the LCR.
<br />Young fish dispersed further downstream are assumed to have much lower survival ra(es because of reduced habitat
<br />availability and large pred,tor populations and a reduced likelihood of being able to migrate back to the LCR to
<br />spawn if they do survive as juveniles.
<br />Numerical h.i::lbit.i::ll models have been widely used to predict the efi~':IS of flow regulation on fish habitat (Reiser
<br />el al.. 1989). but such models have not been used to dote to e"luate hypotheses related to GCO effects on
<br />humpback chub in Gr<l11d Canyon, The most commonly applied tool is the physical habitat simul,tion model
<br />(PHABSIM). which combines predictions of water depth and velocit) from one. or two-dimensional hydraulic mod-
<br />els with data or professional judgment on fish habitat prete-rence CO pr;:-jjctlhe porenrial effect of flow regulation on
<br />fish h,bitat qu,lity, which is indexed by wh,t is termed . weighted use,ble area' (WUA; Bovee. 1982), The v,lidity of
<br />modelling approaches like PHABSIM to assess instre,m flow requirements has been repeatedly questioned (Mathur
<br />el aI., 1985; Studley el ai" 1996; Williams er a{" 1999), Fish preferen.:e for depth. velocity. and substrate changes
<br />with a variety of factors including time of d,y. season. physical conditions (turbidity. temperature, discharge).
<br />and biological factors (food availability, pred,tion risk), More imporuntly, the link between WUA and population
<br />parameters such as abundance, growth, survival, or recruitment has never been well documented. These
<br />weaknesses apply to all numerical habiw.t mod~l:'i. including 'ho~ based on two-dimensional flow fields (e.g.
<br />Guay e/ al,. 2000) or that include more detoils about biology (e,>- individu,lly-based models; see Van Winkle
<br />el al,. 1997),
<br />Managers and scientists involved in the decision-making process on the future operation of GCD are faced with
<br />managing releases to meet existing \.....ater delivery agreements, hydropower generation, and [he needs of environ-
<br />mental resources in Grand Canyon, including threatened and endangered species. Although a conceptual model for
<br />the effects of flow on humpback chub has been articubted (Walters '-1ld Korman. 1999; Walters el at,. 2000), there
<br />is not sufficient infonnalion to deyelop a numerical model to help design experimental flow regimes. The Glen
<br />Canyon Dam Adaptive Management Program has a mandate to evaluate flow regimes and other management
<br />actions by monitoring changes in a wide range of resources, including the abundance and distribution of native
<br />and exotic fish species, The well-publicized experimental flood in 1996 and the LSSF experiment conducted in
<br />2000 are the most notable experimenls conducted to date; however. there was also considerable experimentation
<br />with steady flow regimes in Ihe early 1990s, A ne\\' experimental flow regime. which includes increases in daily
<br />flow fluctuations over the wimer to disadvantage non-native rainbo',I, and brown trouf, was implemented in 2003.
<br />Managers are well aware that it will take yenrs, if not decades, before the native fish responses to alternative flow.
<br />regimes are potentially known. In the short tenn, the ESA essentially requires that Ihe Bureau of Reclamation
<br />design and implement the experimental flow regime identified in the Biological Opinion, In the face of this man:
<br />agement dilemma, the use of a numerical habitat model seems warranted, at least as a first-cut tool to examine
<br />some of the assumption.s regarding juvenile chub habitat availabilit, and dispersal that are currently being used
<br />in a qualitative way to design these experimental flow regimes (e,g, Valdez er al.. 1999).
<br />Prediction of changes in suitable habitat as a function of discharge from GCO is also relevant to Ihe interpreta-
<br />tion of results from the sampling programmes being used to monitor fish populations in Grand Canyon. A signifi-
<br />cant component of current and previous monitoring consists of measuring catch per unit effort (ePE) at index sites
<br />throughout the canyon, In Grand Canyon, CPE for boat electrofishing is computed as the catch divided by the
<br />number of seconds of electofishing effort, The use of CPE to index abundance relies on the assumption that catch
<br />rales are proportional to stock size (N);
<br />
<br />CPE = qN
<br />
<br />(I)
<br />
<br />Copyright i!d 2004 John Wiley & Sons, Ltd.
<br />
<br />Rj~'~r Res. AppJic. 20: 379-400 (2004)
<br />
|