Efficacy of Filtration to Reduce Levels of Actinospores of Myxobolus cerebralis in Pond Water on the Cap K Ranch, Fryingpan River, Colorado
Principal Investigators:
R. Barry Nehring and Kevin G. Thompson
Colorado Division of Wildlife
The Fryingpan River in west central Colorado has been a destination stream trout fishery for anglers seeking a trophy trout angling experience for more than two decades. Wild rainbow trout in the Fryingpan River first tested positive for Myxobolus cerebralis (Mc) in 1995. The dynamic of infection in the stream was unusual in that wild brown and rainbow trout in the lower 9 miles of river had a much higher incidence and severity of infection than those tested from the 5 mile reach of stream immediately downstream of Ruedi Dam. In the fall of 1997, 60% of age 1+ wild rainbow trout collected from the upper 5 mile reach were infected with the Mc parasite with a mean cranial myxospore burden of 16,100. In contrast, among wild rainbow trout collected from the lower 7-mile reach of river in the fall of 1997, incidence of infection was 100% and mean cranial myxospore burden was 115,900. Abundance of age 1 wild rainbow trout in the 15-km reach upstream from its confluence with the Roaring Fork River declined 90% between 1994 and 1998 (Nehring 1999). That trend continued in 1999, 2000, and 2001. A localized area of Myxobolus cerebralis infectivity emanating from a series of off-channel ponds on the Cap K Ranch was documented (Nehring et al. 2000). The most severe reduction in abundance of age 1 wild rainbow trout has occurred downstream of this focus of infection, suggesting that whirling disease induced the decline.
The Fryingpan River is a cobble bottom stream virtually devoid of sediment and sediment accumulation that provide an 
optimum substrate for colonization by Tubifex tubifex, the aquatic oligochaete host for Mc. The lack of suitable substrate for T. tubifex coupled with no detectible decline in age 1 wild rainbow trout abundance in the 5-mile reach of river, lead to the question if it would be possible to passively filter water from the Cap K Ranch ponds to remove Mc actinospores prior to discharge into the river. If this were possible, the ambient level of infection in the river could decline to the point where survival of wild rainbow trout to age 1+ in the lower 9 mile reach of stream would occur.
Pond owners were willing cooperators on this project and allowed a number of filtration studies to be conducted onsite with funds contributed from Colorado Division of Wildlife, Trout Unlimited and the Whirling Disease Foundation. Beginning in 2000, water samples were collected and filtered simultaneously with exposures of susceptible sentinel fish to determine if sand filtration could offer a means for removal of parasites from the pond outflow. Initial tests were promising, with no fish held on sand-filtered water becoming infected. At the same time, results of laboratory studies on the efficacy of a variety of filtration methods also indicated that filtration may offer a solution for removal of parasites from the effluents of ponds (Nehring et al. 2003).
With additional funding from the Whirling Disease Initiative, a private company constructed a wetland “biofilter” on the Cap K Ranch in 2003. Monitoring of the inlet and outlet of the filter occurred during the actinospore release period over three years, ending in 2005. Actinospores were documented in the filter inlet on 14 occasions. A single actinospore was observed in the effluent on May 5, 2004. In late April through June 2005 multiple actinospores were observed in the filter effluent, indicating that the filter clearly has exceeded its useful lifespan. This was further seen in the increasing proportion of positive samples observed in the filter effluent samples.
Although the Cap-K Ranch sand filter proved to be a disappointment in the loss of water capacity experienced over a short period of use and the subsequent loss of ability to effectively filter the parasite, we have learned some important lessons. Any further efforts to construct sand filtration systems must include changes to filter design as recommended by the engineering proponent of the previous filter, namely, that the filter media be a thin layer of graded crushed glass, and that backwash air lines be laid in a much higher density than was the case with the existing filter. Other strategies for reducing infectivity from the Cap-K Ranch ponds and similar habitats appear more appropriate at this time.
Literature Cited: