|
| |
| From the
1999 Southern Division of the American Fisheries Society Midyear Meeting held in
Chattanooga, Tennessee. |
Movement Rate and Barrier
Properties of Riffles: Predicting Movement of Two Cyprinids with an Individual-Based Model
Jacob F. Schaefer, University of Oklahoma, Department of
Zoology, Norman, Oklahoma 73019; E-Mail jschaefer@ou.edu
Keywords: fish, Streams, movement, barriers, model
Movement and dispersal of two Cyprinids (Notropis
boops and Campostoma anomalum) was observed in Brier Creek (Oklahoma) and
compared to predicted patterns from an individual-based random walk model. The model was
based on data from previous experiments in artificial streams that quantified movement
rate across riffles at various levels of current velocity, thalweg depth, riffle length,
and levels of predation in pools. One hundred N. boops and C. anomalum then
were marked by subcutaneous injection of acrylic paint, and released in Brier Creek at the
point of capture. Over the next 30 days, 17 snorkeling samples were conducted over a 1 km
of stream consisting of 10 pools and 9 riffles. During each survey, visual counts were
made of all species in each pool, including numbers of marked N. boops and C.
anomalum. The data from the snorkeling surveys were then compared to predicted
distributions from the model. Two sets of simulations were run to generate predicted
distributions from the model, a null model and a predictive model. The null model assumed
all riffles were not effective barriers and had no effect on dispersal. The predictive
model accounted for differences in riffle properties (thalweg depth, current velocity, and
length) that would make them less permeable. The predictive model was more accurate in
predicting marked fish distributions. The majority of marked fish remained in the same
pool as they were released, with only a few individuals moving upstream or downstream.
|
|
