Region I Surveys of Nutrioso and Rudd Creeks, May 16-19,2005

The Region I Fisheries Program surveyed Nutrioso and Rudd creeks in May 2005, to determine the status of Little Colorado spinedace (Lepidomeda vittata) in each stream, as well as to determine escapement of stocked rainbow trout (Oncorhynchus mykiss) from Nelson Reservoir as required by the U.S. Fish and Wildlife Service's (USFWS) effects determination from a 1995 consultation under Section 7 of the Endangered Species Act. Surveys were conducted in three areas (Figure 1): Nutrioso Creek downstream of Nelson Reservoir, Nutrioso Creek upstream of Nelson Reservoir, and Rudd Creek. Survey participants included Mike Lopez, Matt McKell, Debbie Moore, and Mike Sumner of AGFD, and Leslie Hartsell and Stewart Jacks of USFWS.

Methods
Survey stations were originally selected at random (with five stations located in each of the three areas); however, upon not detecting any spinedace in Nutrioso Creek below Nelson Reservoir and only a handful above, we began targeting areas that had maintained permanent water since surveys in 2000, from which spinedace were collected during those surveys. Random stations above Nelson were limited to four since one of the five occurred in an undefined channel immediately above the reservoir. Nonrandom stations upstream of Nelson included sites previously surveyed on EC Bar Ranch property owned by Jim Crosswhite. Surveys were conducted using a Smith-Root Model 12 POW backpack electrofisher (primarily set at J-4 and 400 volts). Randomly selected stations were approximately 100 meters in length, while nonrandom stations ranged from 100-1210 meters.

Results
Nutrioso Creek - Downstream of Nelson Reservoir
A total of 1,055 fish of five species were captured from 2,840 m of Nutrioso Creek below Nelson during these surveys (Table 1). Fish captured in order of relative abundance included fathead minnow (Pimephales promelas, PIPR) 90.8%, green sunfish (Lepomis cyanellus, LECY) 5.7%, speckled dace (Rhinichthys osculus, RHOS) 2.5%, rainbow trout (ONMY) 0.9%, and bluehead sucker (Pantosteus discobolus, PADI) 0.1%. Rainbow trout were measured for total length (TL) to the nearest millimeter and ranged from 343-413 mm. As required by the Section 7 Consultation, rainbow trout were also examined for tetracycline markings and for Little Colorado spinedace among stomach contents. Each trout contained tetracycline marks, but none of the stomachs contained spinedace, though most stomachs contained invertebrates and three contained one fathead minnow each. Nonnative crayfish were also found in the creek, abundant at three stations, common at four, uncommon at two, and absent at two.

Figure 1. Map of fish survey stations on Nutrioso and Rudd creeks, May 2005. The map of Nutrioso Creek is divided to separate stations downstream (at left, Stations 1-11) of Nelson Reservoir from those upstream (stations 12-19)

Table 1. Summary of fish collected by backpack electrofisher from stations in Nutrioso and Rudd creeks, May 2005. Station numbers correspond to those in Figure 1. See text for fish species abbreviations.

* A = abundant; C = common; NR = not recorded;  R = rare;  U = uncommon;  blank = absent.

Nutrioso Creek - Upstream of Nelson Reservoir
A total of 207 fish of four species were captured from 1,090 m of Nutrioso Creek above Nelson Reservoir (Table 1). Fish captured in order of relative abundance included fathead minnow 70.0%, speckled dace 23.2%, Little Colorado spinedace (LEVI) 3.4%, and bluehead sucker 3.4%. Spinedace ranged from 60-100 nun TL, with a mean of 80.9 mm. Nonnative crayfish were also found in the creek, abundant at one station, common at one, uncommon at one, rare at one, and absent at two. Crayfish abundance was not recorded at two stations. Rainbow trout were not captured upstream of Nelson Reservoir.

Rudd Creek
A total of six fish of two species were captured from 500 m of Rudd Creek (Table 1). A single Little Colorado spinedace was collected from one station and five speckled dace from another. No nonnative fish were captured from Rudd Creek. However, nonnative crayfish were observed in the stream and were abundant at two stations, common at one, rare at one, and absent at one. The sole spinedace was captured from the section in which crayfish were rare.

Discussion
The Little Colorado spinedace population in Nutrioso and Rudd creeks appears to have suffered seriously from recurrent drought conditions, considering that the species had been collected in relatively high numbers during previous surveys. Fifteen 50 m stations (totaling 750 m) in Nutrioso Creek upstream of Nelson Reservoir were surveyed between 1997 and 1999 and 103 spinedace were among the fish sampled. Within this same section of stream a total of 1,090 m were surveyed in 2005 and only seven spinedace were collected. Surveys downstream of Nelson Reservoir in 1994, for instance, resulted in the capture of 435 spinedace from 18 stations of 50 m each (900 m total stream length), whereas we failed to locate a single spinedace in 2,840 m of surveyed stream in 2005. Similarly, in Rudd Creek in 1994, 301 spinedace were sampled from 16 stations comprising 800 m of stream, an area that essentially overlies the five 100 m stations surveyed in 2005, from which one individual spinedace was captured.

Spinedace in Nutrioso Creek on the EC Bar Ranch were captured in the only site (station 17) of the three sites (17-19) on that property that had retained water throughout the protracted drought period (Jim Crosswhite, personal communication).

Rudd Creek has been affected by drought conditions to the extent that fish were absent from three stations and crayfish were absent or rare at two stations. Incidentally, the stream banks at the station in which the single spinedace was captured appeared to have been severely impacted by heavy ungulate use. Spinedace are probably still present in Rudd Creek, though greatly reduced in abundance.

A comparison of results of previous fisheries surveys with precipitation patterns (and associated streamflow conditions) indicates that fluctuating water levels in the stream, especially downstream of Nelson Reservoir, have likely been a major factor in the success or failure of the spinedace population. The comparison shows evidence of high spinedace survivorship following periods of average or higher-than-average annual precipitation, indicated by the collection of relatively high numbers of individuals, as well as successful natural recruitment into the population during these same periods as evidenced by the low mean TL of captured spinedace.

The above correlation is further supported by a substantial decrease in the number of spinedace collected, coupled with an increase in mean TL of spinedace sampled following a lower-thanaverage water year. Results of surveys in the years following the water year of 1996 demonstrate this effect, in that spinedace numbers decreased considerably from 176 individuals in fall 1996 to just three in spring 1997, while exhibiting an increase in average TL following 1996 surveys to nearly 95 mm in 1997-1998 from 60 mm in the previous four years (1993-1996). The trend of reduced representation among the total catch continued until surveys in spring 1999, which followed an average rainfall year in 1998. It is noteworthy that, as might be expected from this sort of pattern, there is a lag between the onset of drought conditions (or simply occurrence of a poor water year) and a decrease in spinedace catch, and vice versa when rainfall is at or above normal.

Although Nelson Reservoir did not spill at any time between spring runoff in 2000 and early 2005 (not even during the above average 2001 water year), and surveys were not conducted during this sustained period of low precipitation, the precise response of the species during this time is unknown. However, given the above pattern, it is expected that spinedace probably responded in a manner similar to the response that followed the 1996 water year. To corroborate this supposition, if water did not spill from Nelson Reservoir into Nutrioso Creek, then the environmental impetus provided by seasonal runoff or a high-flow event thought to induce spawning in spinedace was lacking, resulting in fewer individuals, especially of the young of year size class. And, if environmental conditions in a stream preclude spawning by a relatively short-lived fish species such as spinedace (probably 4-6 year lifespan), reason suggests that the species may not persist through a prolonged period of such conditions.

Spinedace are likely still present in lower Nutrioso Creek, though that portion of the Nutrioso/Rudd population is apparently severely reduced and distribution highly localized. Rainbow trout do not appear to be a factor in the decimation of the spinedace population, though the other three nonnative aquatic species (fathead minnow, green sunfish, and crayfish) likely play at least an incidental role in spinedace decline. Water, or lack thereof, seems to be the primary cause of spinedace population deterioration. Additionally, much of lower Nutrioso Creek was overrun with dense mats of aquatic vegetation as well as extensive siltation, which also likely negatively impact the species by reducing habitat available for spawning.

Recommendations
Future survey efforts should be conducted again in spring 2006, and concentrated downstream of Nelson Reservoir once reservoir spilling has ceased, to determine whether the remaining isolated portion of this population was able to successfully reproduce during periods of high flows in spring 2005, which would be evidenced by a rebound in spinedace numbers and low mean TL. It may be helpful to begin surveys in sections not surveyed in 2005. Alternatively, a full, single electrofishing pass, as conducted in 1999-2000, could be completed from the Apache-Sitgreaves National Forest Boundary upstream to Nelson, all of which is designated Critical Habitat for Little Colorado spinedace. Nonnative fish should be manually removed at every encounter during surveys. Use of an herbicide in Critical Habitat may not be a viable option, but some means of reducing pervasive aquatic vegetation may be necessary.

Surveys should again be coordinated with Jim Crosswhite to monitor spinedace on the EC Bar Ranch, as well as again gaining access through his property to disjointed portions of the stream on Forest Service land to survey at or near site 16, from which spinedace were also collected in 2005.

Consideration should be given to construction of a livestock/elk exclosure on portions of Rudd Creek that sustain repeated ungulate traffic, and would likely benefit spinedace by protecting both instream and riparian habitats.