Menu

SA028
Hummingbird foraging patterns across alpine meadows with RFID-equipped feeders in the HJ Andrews Experimental Forest, 2014-2017

CREATOR(S): Sarah J. K. Frey, Dustin Gannon, Matthew G Betts
PRINCIPAL INVESTIGATOR(S): Matthew G Betts
ORIGINATOR(S): Sarah J. K. Frey
OTHER RESEARCHER(S): Dustin Gannon, Adam Scott Hadley
DATA SET CREDIT:
B. Doyle, J. Gruden, E. Jackson, M. Paniagua, E. Sandi, J. Rousseau, S. Lundgren, A. Newell, E. Isaacs, A. Bartelt, and J. Greer contributed to field efforts. Dr. Eli Bridge assisted with RFID equipment and data collection protocol. Animal procedures were approved by the Oregon State University Animal Care and Use Committee (permit numbers 4266 and 4665). Hummingbird banding occurred under the USGS banding permit 23521. This work was supported by the U.S. National Science Foundation (grant numbers DEB-0823380 and DEB-1457837) and the Western Hummingbird Partnership.
METADATA CREATION DATE:
27 Jul 2021
MOST RECENT METADATA REVIEW DATE:
3 Aug 2021
KEYWORDS:
ecology, population dynamics, recruitment, community composition, census, forest ecosystems, birds
PURPOSE:
We aimed to test two hypotheses of how forest encroachment could reduce the functional connectivity of the landscape by influencing hummingbird foraging patterns. First, if hummingbirds fly low, foraging at flowers in sequence within meadows where flower resources are abundant and avoid flying into forested areas where floral resources are generally less abundant, forests make act as a (potentially 'leaky') barrier to movement. Second, if hummingbirds fly over the canopy to forage in disconnected meadows, flowering plants that get overgrown by woody vegetation may not be discovered by birds as they fly above the canopy. Thus, further encroachment could inhibit resource discover and lower the effective population sizes of ornithophilous plants.
METHODS:
Experimental Design - SA028:
Description: To track hummingbird visitation and movements through the landscape, we established four study sites in meadow complexes (clusters of meadows) located on high points that span the north-south extent of the H. J. Andrews Experimental Forest, Oregon, USA (see associated location information). The sites were selected such that we could establish arrays of five hummingbird feeders with one feeder in a central meadow and four satellite feeders ca. 250 m from the center, approximately along the cardinal directions. The satellite feeders were placed with at least one under closed canopy coniferous forest (total of n=5), one in a meadow separated from the center feeder by closed canopy (n=4), and one in a meadow connected to the center feeder by open habitat (n=7). We equipped each hummingbird feeder with a Radio Frequency Identification recording station that logged the date, time, and unique individual identification number for tagged hummingbirds (see Field Methods). We covered all but one port on each feeder and placed the antenna (a copper coil) around the open port such that tagged hummingbirds passed through the antenna when they drank from the feeder.
Instrumentation: We used 8 mm Passive Integrated Transponder tags to mark individual Rufous Hummingbirds (Selasphorus rufus). These tags not compatible with the recording protocol as described by Bridge and Bonter (2011), but require a minor modification to the protocol. We used readers with a reading scheme that reads 8 mm tags at 134.2 KHz following the ISO11785 protocol. The schematics for this variant can be obtained upon request from Dr. Eli Bridge of the University of Oklahoma.
Citation: Bridge, E. S., and D. N. Bonter. 2011. A low-cost radio frequency identification device for ornithological research. Journal of Field Ornithology 82:52–59.
Field Methods - SA028:
Description:

We established hummingbird feeders filled with 20% sugar water solution at the center meadow in each meadow complex (see associated site information) two weeks before trapping hummingbirds. Following the two-week habituation period, we placed Hall traps (Russel 2001) around the center feeders and trapped for a period of 5 hours (05:00 – 10:00). We collected standard measurements on each trapped individual and banded each with a unique metal leg band. The final processing step was to implant the PIT tag under the loose skin between the shoulders.

Using a clean cotton applicator, we applied lidocaine gel to the exposed skin on the upper back of the hummingbird and left it to sit for a minimum of two minutes for the numbing to take effect. While waiting for the lidocaine gel to take effect, we placed the PIT tag and insertion needle in 70% ethyl alcohol to clean and disinfect the surface before implantation. Once the lidocaine numbed the skin where the needle will be inserted, we applied a betadine antiseptic solution with a new/clean cotton applicator to disinfect the area before implanting the tag. With clean forceps, we gently lifted the skin between the shoulders and then insert the needle, the bevel facing up, just under the lifted skin. We then carefully attached the clean plunger to the needle and inserted the 8 mm tag. Once inserted, we carefully closed the hole with vetbond and forceps.

Over the course of four summers (mid-June through July, 2014-2017), we captured and implanted 163 Rufous Hummingbirds with PIT tags. Feeders were maintained with 20% sucrose solution for 1.5-12 weeks during the summer season (June-August) per year from 2014 to 2017. Relocation data were downloaded from data loggers and backed up weekly.

Instrumentation: Tags: 8mm Passive Integrated Transponder tag (weight 0.034 g) from Biomark, MiniHPT8 8.4 mm X 1.4 mm, 134.2 kHz ISO FDX-B, N165 Needles (https://www.biomark.com/n165), MK165 Implanter (https://www.biomark.com/mk165), Akorn Lidocaine hydrochloride jelly USP 2% (https://www.akorn.com/prod-details-list.php?group=Lidocaine+Hydrochloride+Jelly%2C+USP&topicals=true), Betadine antiseptic solution (povidone-iodine, 10%) (https://www.3m.com/3M/en_US/company-us/all-3m-products/~/3M-Vetbond-Tissue-Adhesive/?N=5002385+3294397973&preselect=8710646&rt=rud), Foam pad on which to perform the procedure, Hand sanitizer, Forceps - flat, no ridges, with blunted ends, 70\% ethyl alcohol, Cotton swabs, RFID readers and antennae.
Citation: Russell, S. M. 2001. The North American Banders’ Manual for Banding Hummingbirds. North American Banding Council.
Processing Procedures - SA028:
Description: To create the forest.tif raster of forested and non-forested areas along Frizzel Ridge, we downloaded the National Land Cover dataset (NLCD; Homer et al. 2012) and clipped it to the bounding box defined for this study. We then reclassified the raster, assigning coniferous, broadleaf, and mixed forests a value of 1 and non-forest cells a value of 0. To improve the resolution of cell assignment, we additionally digitized meadows and forest openings. We rasterized the shape file with all digitized meadows and forest openings and multiplied the resultant raster with the binary forest raster from the NLCD. This ensured that cells that fell within a digitized meadow or forest opening were classified as non-forest.
Citation: Homer, C. G., J. A. Fry, and C. A. Barnes. 2012. The National Land Cover Database. Report, Reston, VA.
Quality Assurance - SA028:
Description: The RFID readers and data loggers occasionally glitched and produced nonsensical records. To ensure these were removed from the data, we used a custom R script to cross-reference each record against the list of tags implanted in hummingbirds, removing any records with tag numbers that were not included in the list. We additionally checked that dates and times were recorded correctly by checking that they fell into the time interval during which we conducted field research.
SITE DESCRIPTION:
We established four study sites in meadow complexes (clusters of meadows) located on summits that span the north-south extent of the H. J. Andrews Experimental Forest (AND), Oregon, USA (44.212 N, -122.256 E). The sites were selected such that we could establish arrays of five hummingbird feeders with one feeder in a central meadow and four satellite feeders ca. 250 m from the center. The satellite feeders were placed with at least one under closed canopy coniferous forest (n=5), one in a meadow separated from the center feeder by closed canopy (n=4), and one in a meadow connected to the center feeder by open habitat (n=7).
TAXONOMIC SYSTEM:
The Sibley Guide to Birds
GEOGRAPHIC EXTENT:
Frizzel Ridge, H. J. Andrews Experimental Forest, Western Cascades, Oregon
MEASUREMENT FREQUENCY:
Irregular
PROGRESS DESCRIPTION:
Complete
UPDATE FREQUENCY DESCRIPTION:
asNeeded
CURRENTNESS REFERENCE:
Ground condition
RELATED MATERIAL:
R code used to analyze these data to test the hypotheses described in the Purpose section of the metadata can be found on the public GitHub repository https://github.com/Dusty-Gannon/RUHU-movements.