Categories:
Published: June 2001
Pages: 46
Author(s): Marc P. Hayes, Joseph D. Engler, Susan Van Leuven, Daniel C. Friesz, Timothy Quinn, and D. John Pierce
Abstract
Abstract
We studied overwintering behavior in the Oregon spotted frog (Rana pretiosa) at Conboy Lake National Wildlife Refuge over the 22-week interval from 21 September 2000 to 17 February 2001. Ten adult (mass > 27 g) frogs (8 females, 2 males) from two sites 3 linear km apart were tracked on a weekly basis using radio-transmitters with an external belt attachment. During the study, we obtained 162 radio-locations, and most (n = 153) were based on non-visible, submerged frogs or transmitters. During successive surveys, we recorded straight-line radio-location changes of 0 to 160 m. We interpreted most location changes > 1 m, about 67% of the data, to be frog movements. Longer and more frequent location changes before the significant decline fall temperature are thought to represent directed movements to overwintering habitat. Frogs on which we placed transmitters in lentic habitats moved to lotic ones during the temperature decline interval that preceeded freeze up, but not vice versa. During the period of ice cover, we saw significantly greater movements from sites where dissolved oxygen (DO) was lower. Frog use of less vegetated, shallower sites nearer to shorelines during this interval also agreed with the idea of DO limitation. Several frogs selected sites associated with beaver workings (i.e., beaver step dams, submerged pathways). These sites may be favorable for overwintering because they provide a better DO environment, refuge from predators, or both. Recovery of frog remains or transmitters from six frogs suggests that one slipped its transmitter, two were preyed upon, and three may have died from low oxygen stress. Regardless of cause, we estimate overwintering mortality to be high. This pattern agrees with available demographic data, which indicates that annual turnover is high and Oregon spotted frogs are relatively short-lived. Study will be needed to determine how much of the observed pattern of mortality and low oxygen stress conditions is a typically annual event or the consequence of extreme drought conditions, as observed in winter 2000-1.