Ecological Archives E096-262-A1
Rachael Derbyshire, Dan Strickland, and D. Ryan Norris. 2015. Experimental evidence and 43 years of monitoring data show that food limits reproduction in a food-caching passerine. Ecology 96:3005–3015. http://dx.doi.org/10.1890/15-0191.1
Appendix A. Details of feeder construction.
Feeders were made from wood and measured 30 cm high × 30 cm wide × 40 cm long. The supplemental food was placed inside the feeder at its solid rear wall, forcing the jays to enter the feeder to obtain food. We also covered the food with a layer of wire mesh to prevent small mammals such as American Martens (Martes americana) and Red Squirrels (Tamiasciurus hudsonicus) from retrieving large amounts of food if they did gain access to the inside of the feeder. Feeders were hung above the ground and a minimum of ~2 m from the trunks of trees to further minimize access by mammals. Bars around the outside of the feeder prevented larger birds, such as Common Ravens (Corvus corax), from entering.
A thin wire antenna was placed across the front of the feeder and connected to a RFID reader placed underneath the feeder (Fig. 1A). Bars along two sides of the feeder were intentionally spaced sufficiently close together that jays were forced to enter through the front and cross the antenna by walking or hopping. RFID tags were affixed to each supplemented female by gluing the tag onto existing leg bands and then wrapping the tag with clear, waterproof tape (tesa® tape, Charlotte, NC, USA; Fig. 1B). The wire antenna was able to detect tags within ~10 cm above the horizontal antenna, and the RFID reader recorded each time a female entered the feeder and thus allowed us to confirm that treatment females were indeed retrieving supplemented food. The RFID reader and antenna were connected to a battery source which was also placed underneath the feeder. In 2013, 9 V lithium-ion batteries (Energizer®, Walkerton, ON, CAN) were used and changed daily. In 2014, 12 V lead-acid batteries (Power Sonic®, Fort Worth, TX, USA) were used and changed every 3–4 days. Unfortunately, in 2013, cold weather unexpectedly reduced battery life, which meant that readers were not running continuously throughout the day. However, the readers were running for enough time each day (mean daily running time per feeder = 5 hr, 50 min) that we could estimate the number of days that the female entered the feeder. In 2014, we also estimated daily feeding rate because improved battery life allowed for continuous operation of the RFID readers throughout the day. For logistical reasons, we did not attach RFID tags to males or juveniles so we could not estimate the proportion of food that was taken by the focal female versus other jays occupying the same territory. RFID tags were also not attached to control females, although some 2014 control females had been supplemented in 2013 and retained their tags from the previous year.