The Northwest Golden-winged Warbler Working Group

Blood Sample

The Northwest Golden-winged Warbler Working Group formed in 2008 to address research and conservation needs for GWWA in Manitoba and north-western Ontario. The group includes representatives from Bird Studies Canada (BSC), Environment Canada (EC), Parks Canada Agency- Riding Mountain National Park (RMNP), Louisiana-Pacific (LP), Manitoba Conservation (MC), and the Ontario Ministry of Natural Resources (OMNR). Our group is working collaboratively to gain a greater understanding of the distribution, abundance and demography of GWWA in our region. In so doing, we hope to monitor threats and examine habitat management options to preserve and/or enhance this population.

On-going research suggests that the Manitoba population may be one of the last genetically pure populations of Golden-winged Warbler. Due to the apparent uniqueness of the north-western population, our group is attempting to document the demography of GWWA at the north-western periphery of their range. To support these efforts, EC is also collaborating with the Golden-winged Warbler Genetic Atlas Project by providing blood and feather samples for genetic screening, and by conducting stable-isotope analyses to help examine long-distance dispersal and it's implications for gene-flow. EC also recently began an international collaboration, attempting to link wintering ground populations to their breeding ground origins using stable-isotope analysis. Thus far, EC has collected samples from 29 birds in south-eastern Manitoba, 16 birds in western Ontario (Rainy River, adjacent to the Manitoba and Minnesota borders), 183 birds from Riding Mountain National Park and vicinity, and 14 birds from the Duck Mountain area.

Surveys for GWWA in our region are being conducted by all members of our group. BSC has taken a lead role in conducting broad-scale surveys for GWWA in Manitoba, conducting over 3000 point counts with GWWA playback in 2009 alone. The start point of all transects was randomly selected and points spaced 400m apart. After locating several new sites for this species in Southeast Manitoba in 2008, we were excited to confirm breeding presence in Manitoba’s Interlake and several areas of western Manitoba away from known strongholds. In western Manitoba, we found GWWA up to 60km south of Riding Mountain National Park but none were found in the so-called southwest Manitoba upland areas including the Turtle Mountains, Brandon Hills and Pembina Hills.

BSC (in collaboration with OMNR) also conducted a short survey in the Rainy River area of northwest Ontario and found very high densities of Golden-winged Warbler there in several habitat types. In Rainy River, GWWA were detected on 15.7% of the 428 point counts. This area appears to have a denser population than the core area of south-eastern Manitoba, where we detected GWWA on 8.5% of 809 point counts in 2008 (6.1% of 1383 point counts throughout south-eastern Manitoba). Volunteers also completed over 500 repeat measure point counts in 2009 in south-eastern Manitoba, with a very similar detection rate to 2008. Western Manitoba has several small areas where GWWA density appears very high; however, the overall detection rates (2.7% of 2001 point counts in 2009 throughout western Manitoba and 2.7% of 582 points in Riding Mountain National Park in 2008) are lower than in the east.

The broad-scale efforts of BSC are in addition to existing avian monitoring programs lead by RMNP and LP. Both RMNP and LP have also been augmenting their regular avian monitoring with playback surveys. In future, they hope to develop spatially explicit occupancy models to predict the occurrence of GWWA on their landbases, and examine habitat management options for GWWA.

In 2008, the combined number of GWWA territorial males located on survey transects from all members of the Northwest Golden-winged Warbler Working Group was over 330 and we located over 100 additional territories in new areas of the province in 2009. The Golden-winged Warbler population in Manitoba is clearly much higher than the previous COSEWIC estimate of 105 to 270 pairs.

In addition to the existing survey efforts, Rob Rempel is exploring an automated detection technique for assessing the presence of GWWA from digital recordings. The approach uses computer algorithms (similar in concept to speech identification used in cell phones) to determine at which point within a recording GWWA may have been singing. Trained interpreters then listen to that short segment of the recording to confirm whether GWWA was indeed singing. Initial testing of the approach using Riding Mountain Park recordings was very successful. This could be a big step forward in detecting the presence of species at risk, as it would allow biologists to evaluate many hours of recordings in a short time using the automated technique.

Photo Credits: Rachel Vallender