The Use of Game Camera and Scent Attractants as a Mammalian Species Index on Clark's River Refuge

The use of game cameras has been used widely in monitoring mammals.  We looked at using different scents as attractants that have been rumored to work (apple and grease) as well as a commercial attractant (Caven’s “Hiawatha Valley”).  Three upland and bottomland sites were chosen on Clark’s River Wildlife Refuge near Benton Kentucky.  Scents were placed along game trails to maximize productivity.  A different scent was used for each site for a week at a time and then rotated till each scent was used at each site.  Cameras were placed near the scent to capture animal activity around the scent.  Cameras were also used to count different species present and compared to what GAP program showed to be present to test the cameras use as in index of species in an area.  There was no significant difference in the scents or the areas used.  The cameras only captured 46.7% of possible species on the bottomland sites and 33.3% of the upland sites.  More trap nights could improve both scent attractant and number of species seen.

Searching for mammal traps at night can be difficult. Time spent searching may mean more time in traps for animals. Reflective tape can assist in finding traps at night which would eliminate stress to the animal as well as money spent to replace lost traps. We used 120 traps in three locations to test the effect of reflective tape on trapping success. Sixty traps were set in an over-grown field and sixty were set in two wooded areas. Of  these, sixty were marked with reflective tape in one of three strategic locations. These small Sherman Live Traps were baited with a mixture of peanut butter and oats. Traps were left out for four days and checked three times daily. A Chi-square homogenaity tests was run and it was found that reflective tape has no significant effect on trapping success. The same test was used to determine if, amongst the tape types there was a significant difference. Tape near the back door had a significantly lower capture rate than traps with tape in the other two locations.

Animal Species Prone to Automobile Fatality Along the Trace

BJ Riden and Matt Dennis

Road kill sampling of large geographic areas has been widely used and accepted for biologists. Road kill surveys are used as indicators of local population densities over time.  A road kill survey was conducted on the Trace in Land Between the Lakes between 13:00 and 15:00 every Wednesday and Sunday from October 14 to November 14, 2007.  We predicted that Didelphis virginiana would be the most observed road kill due to its local abundance, personal observations, and more easily spotted body size    Thirteen specimens of mammals were collected, with Sciurus carolinensis being the most abundant with six specimens. A road kill index was calculated by dividing the number of road mortalities for each species by the distance of the sample route (Gehrt 2002).    

---

Success rates between two live traps, the Tomahawk and the Havahart

In this research project, we tested the success rates between two types of live traps, the Tomahawk and the Havahart.  We did a random placement of 10 traps at the first two sites and a random placement of five at each of the last two, a total of four sites.  At each site the traps were set in a line.  All of the traps used were relatively the same size.  Each trap was baited with pre-mixed peanut butter and tuna.  All traps were set and checked within 12 hours, although they were normally checked before that mark.  Two specimens were trapped in Tomahawk traps (a female Eastern gray squirrel, Sciurus carolinensis and a female Virginia opossum, Didelphis virginiana).  They were then sexed and released.  These traps were then reset.  We hypothesized that there would be a difference in the success rates between the two types of trap.  We used a two-proportional z-test to determine whether the alternative hypothesis was supported or not.  We used our derived z-score of 1.45 to determine our P-value, which was 0.9265, meaning our data was statistically insignificant.  Even though both of our successful traps were in the same type of trap, due to our insufficient amounts of data, the results of the experiment were insignificant. 

---

Bait Efficiency in Trapping for Gray Squirrel (Sciurus carolinensis) under Fruit Failure Conditions

James Mayes and Robert Schaefer

 

The gray squirrel (Sciurus carolinensis) is the subject of many studies and most of them involving trapping squirrels. For our experiment we wish to see if there is a preference in bait for the squirrels under fruit failure conditions, being that this year’s oak crop has been drastically reduced as well as other primary forage crops. We tested three baits in total, walnut (Juglans nigra), pecan (Carya illinoinensis), and peanut butter (Jif^® Extra Crunchy), by trapping with the baits.  After 80.08 hours of trapping, we caught 17 gray squirrels.  The peanut butter traps showed to be the least efficient, not catching any squirrels, while the pecan and walnut traps showed to be the best.  A chi-squared test was performed between the data obtained from the pecan and walnut sets to find there was not a significant difference (P-value = 0.467) between the two. Meaning that there is not a preference between walnut and pecan, but the squirrels clearly do not like Jif^® peanut butter.

---

Exploratory Behavior of Mus musculus and Peromyscus leucopus

Katie Marks and Jessica Purcell

We expect there to a difference in the exploratory patterns of Peromyscus leucopus and Mus musculus.  Katie and I defined exploratory patterns as the time each mouse took to reach the reward, number of turns, backtracks, pauses, and attempts to take shortcuts to the reward.  We selected these two species based on the differences in their habitat.  P. leucopus primarily lives in wooded areas or along edges.  M. musculus inhabits mostly inhabited or uninhabited buildings, fence rows, and abandoned grain fields.  In order to measure exploratory behavior, we created a four-level, 68.6 by 68.6 centimeter maze.  The walls were interchangeable, allowing us to create three different mazes.  Six mice of each species were tested in the various mazes.  A one-tailed T-test indicated a significant difference between the two species in number of pauses {.04}, number of turns {.02}, and number of shortcuts taken {.01}.  We can attribute these results to the differing habitats each species occupies.

---

Tree cutting preferences in Castor canadensis

Justin Moorman

The diet preferences of the beaver (Castor canadensis) have been moderately studied in the past.  However, these studies were done in different regions of the country which have dissimilar dominant species of vegetation.  It was my goal to determine the local vegetative preferences of beavers in fall, and my hypothesis was that depending on what trees are available the preferences of the species will stay on the same gradient (i.e. Betula spp. over Alnus spp.).  A beaver lodge was identified at Ferguson Spring; the primary feeding sites were determined and roped off in a 30 x 4 meter area.  The number of individual tree cuttings were tallied and analyzed to find the species most commonly utilized.  Trees were identified to species and loosely categorized by diameter.  Results were analyzed using standard percentage calculations.  River Birch (Betula nigra) was found to be utilized over all other trees when in abundance with Common Alder (Alnus serrulata) being an alternate choice.  Castor canadensis’s primary and secondary preferences stayed consistent with previous findings.