|publications > paper > a field test of attractant traps for invasive Burmese pythons in southern Florida > methods
A field test of attractant traps for invasive Burmese pythons (Python molurus bivittatus) in southern Florida
We conducted the experiment east of Everglades National Park in an area known as the Frog Pond (Fig. 1). Prior to our experiment, this area had been intensively managed for crop production for >30 years. We selected this site because in previous years 165 pythons had been found dead in fields (505 ha total area) following mowing or disc-harrowing operations in the Frog Pond (2005: 22 pythons, 2006: 44 pythons, 2007: 55 pythons, 2008: 44 pythons). Many of these pythons were found by personnel from the National Park Service or South Florida Water Management District in the days following agricultural treatment by looking for vultures feeding on dead pythons in the treated fields.
The study site was disc-harrowed and planted in okra in April 2009. Harvesting continued until 01 July 2009; after this the fields were allowed to go fallow in preparation for ecological restoration of the area. Grasses, forbs, and woody vegetation quickly spread throughout the site; by the time that trapping started in August, vegetation was >1 m high throughout the site, and in some places was approaching 3 m high. Vegetation continued to grow profusely during trapping.
We established four trapping sites in the study area, totaling 60 traps. One site consisted of a block of 40 traps placed at 33 m intervals along each of 10 transects cut through the vegetation, with 10 m spacing between transects, while three other sites (4, 7, and 9 traps respectively) were placed around small tree islands.
Traps measured 1830 mm long x 690 mm wide x 510 mm tall, with a frame constructed of untreated 51 x 51 mm lumber (standard 2 inch x 2 inch pine). The sides and bottom were sheathed in metal hardware cloth (wires spaced at 12.7 mm (=0.5 inch) intervals), while the top was made of 12.7 mm plywood with a large access door (Fig. 2A). All traps had a one-way entrance at each end, precluding captured animals from escaping through the ingress point once they had entered completely. Half of the traps were equipped with entrance funnels (430 mm long at a 30 to 40˚ angle) made of plastic mesh with 12.7 mm square holes. The funnels led to round one-way entrance flaps constructed of hardware cloth suspended vertically within a PVC plastic sleeve (152 mm diameter, 76.2 mm long; Fig. 2B). The other half of the traps were equipped with ramps set at a 30˚ entrance angle and made of hardware cloth, leading to rectangular entrances. Entrance flaps for the rectangular entrances consisted of four separate one-way flaps made of hardware cloth (each 102 mm tall x 152 mm wide) attached by metal door hinges; these combined to form an entrance 610 cm wide (Fig. 2C).
All traps were baited with a live laboratory rat (Rattus norvegicus) contained in a separate wire cage (400 mm long x 300 mm wide x 150 mm tall, with hinged entrance at one end for servicing) affixed to the underside of the trap access door. Rats were provided with a food block (commercial parrot mix in a paraffin matrix) and a potato for water (Rodda et al. 1999b); potatoes and food were replenished as needed.
Traps were baited and opened on 05 August 2009 and checked daily until 16 November 2009.
To assess the relative capture rates of traps and visual surveys, we conducted standardised visual surveys for pythons in the study area roughly once per fortnight between 03 August and 13 November 2009 (N = 16 surveys). Surveys were conducted between 0800 and 1000 h, and totaled 87.9 person-hours. We also conducted opportunistic visual searching while conducting daily trap checks. To avoid inducing behaviors that might cause pythons to flee the trapping site, we did not disturb or capture pythons observed during standardised or opportunistic visual searches. We identified all species sighted, and recorded locations using a hand-held GPS unit.
Captured pythons were measured and marked in the field and then released within 5 m of the trap. Individuals were sexed and measured [snout-vent length (SVL) and tail length], weighed, implanted with a subcutaneous passive integrated transponder (PIT tag), and scale-clipped with a unique numeric code as backup in case the PIT tag was lost. Non-target species were noted but not disturbed during visual surveys, and non-target animals captured in traps were identified to species and immediately released within 10 m of the trap.
Post-trapping site treatment
In previous years, a large area
of the Frog Pond including our study site had been treated with a disc harrow
at the end of each crop production season; numerous dead pythons were observed
in the fields after these treatments (see above). After conclusion of the trap trial, 81 ha in and around the
study site were disc-harrowed as part of vegetation control activities, so as
to expose pythons concealed in the extremely dense vegetation. Disc-harrowing occurred during 17 - 24
November 2009, and 2 - 4 observers followed immediately behind the harrow,
searching for pythons and other organisms killed or otherwise revealed by the
U.S. Department of the Interior, U.S. Geological Survey
This page is: http://sflwww.er.usgs.gov/publications/papers/python_traps/methods.html
Comments and suggestions? Contact: Heather Henkel - Webmaster
Last updated: 15 January, 2013 @ 12:43 PM (KP)