Record Details
Potential effects of ungulate exclusion fencing on displaying Hawaiian Petrels (Pterodroma sandwichensis) at Hawai'i Volcanoes National Park
ScholarsArchive at Oregon State University
| Field | Value |
|---|---|
| Title | Potential effects of ungulate exclusion fencing on displaying Hawaiian Petrels (Pterodroma sandwichensis) at Hawai'i Volcanoes National Park |
| Names |
Swift, Roberta
(creator) Coblentz, Bruce E. (advisor) |
| Date Issued | 2004-12-07 (iso8601) |
| Note | Graduation date: 2005 |
| Abstract | Recently the National Park Service has proposed raising fence heights to exclude mouflon sheep (Ovis musimon) from conservation areas at Hawai'i Volcanoes National Park. Concern over previous mortality of Hawaiian Petrels (Pterodronia sandwichensis) due to collision with fences prompted this study to quantify the potential effects of raising fence heights on this critically endangered seabird. Avian perception and navigation capabilities were researched, and vision was judged to be the dominant sense that Hawaiian Petrels may use to detect and avoid fences. Previous studies and the techniques they used to assess the risk of bird collision with obstructions were also reviewed. In the current study, we used behavior to quantify the ability of Hawaiian Petrels to detect and avoid fences, and the relative collision risk of three different fence designs. We observed night-time flight behavior of Hawaiian Petrels using night vision goggles in a breeding colony display area for seven weeks during the summer of 2003. We recorded petrel behavior around three simulated fence designs: (1) a 1.2-m hogwire fence, (2) a 1.8-m hogwire fence, and (3) a 1.8-m hogwire fence with white flagging added for visibility. We also recorded behaviors during a control observation period, when no fence was present, to represent the natural flight behavior of the birds. Fences used during the trials were made of surrogate materials to mimic hogwire fencing, including 13-cm square fabric netting and padded bamboo poles that would not harm the birds, should they collide with them. Because collisions with fences were rare, we quantified the risk of petrels colliding with each simulated fence by counting the number of passes attempted below fence height, as evidenced by late avoidance behaviors and collisions. We compared the proportions of late avoidance behaviors and collisions among fence types to investigate the effects of fence type on fence strike risk. Counts during the control period were a measure of the proportion of passes when birds were at risk of colliding with fences of different heights during unobstructed flight. Hawaiian Petrels were significantly more likely to attempt to pass at heights above ground level below fence height when no fence was present (during the control period) than when the 1.2-m fence or 1.8-m simulated fences were present. This result indicates that although petrels flew below fence height when no fence was present, they were able to detect and avoid 1.2-m and 1.8-m fences in their flight path. However, one petrel did collide with the simulated 1.2-m fence on a foggy night, suggesting that Hawaiian Petrels may have more difficulty avoiding fences during poor visibility conditions. There was no significant difference in the likelihood of Hawaiian Petrels to exhibit late avoidance behaviors and collisions between the two fence heights. Therefore, the 1.2-m and 1.8-m simulated fences apparently posed similar fence strike risk to birds. There was a suggestive difference in the likelihood of petrels to exhibit avoidance behaviors and collisions between flagged and unflagged fences. This suggests that Hawaiian Petrels were able to detect and avoid flagged fences at greater distances, possibly reducing fence strike risk. We also investigated the effect of fence orientation relative to slope on petrel avoidance behavior. Fence orientation had no detectable effect on the proportion of late avoidance behaviors and collisions exhibited by displaying Hawaiian Petrels. However, investigations into the effect of fence orientation may be more appropriate when studying behavior of petrels commuting to or from breeding colonies rather than that of displaying birds following circuitous flight paths. In conclusion, fences help to protect essential habitats of native species and pose little risk to displaying Hawaiian Petrels. Fence strike risk for this critically endangered species may be further minimized by adding visible materials, such as white flagging, during construction. The methods developed in this study could be used to test fence designs proposed for future construction, particularly designs that would exclude both introduced predators and feral ungulates from Hawaiian Petrel breeding habitat. |
| Genre | Thesis/Dissertation |
| Topic | Fences -- Environmental aspects -- Hawaii -- Hawaii Volcanoes National Park |
| Identifier | http://hdl.handle.net/1957/22758 |
