4 
THE WILSON JOURNAL OF ORNITHOLOGY • Vol 124, No. 1. March 2012 
TABLE 1. Phenology of migration stages of three North¬ 
ern Black Swifts". 
Departure from Colorado 
Arrival at wintering area 
Time spent at wintering area 
Departure from wintering area 
Arrival in Colorado 
Duration of southbound 
migration 
Duration of northbound 
migration 
14 Sep (10-19 Sep) 
5 Oct (28 Sep-12 Oct) 
220 days (209 to 227 days) 
13 May (9-20 May) 
1 Jun (23 May-18 Jun) 
21 days (18 to 23 days) 
20 days (14 to 29 days) 
“ Arrival and departure dates are presented as means with range in 
parenthesis. 
RESULTS 
Geolocators recovered from two females at 
Fulton Resurgence Cave and one male at Box 
Canyon Falls represent a 15% recovery rate. The 
Black Swifts initiated fall migration from Colo¬ 
rado beginning on 10 September and continued 
through 19 September 2009. We used longitudinal 
information around the time of the autumnal 
equinox and documented the swifts arrived at 
their wintering location in South America be¬ 
tween 28 September and 12 October 2009. 
Approximate dates of migration initiation north 
from wintering areas began on 9 May and 
continued through 20 May 2010. Dates of arrival 
at breeding sites began 23 May and continued 
through 18 June 2010 (Table L Fig. I). 
Kernel density estimates indicate all three birds 
over-wintered primarily in the lowland rainforest 
of western Brazil with some kernels extending 
into Bolivia, Colombia. Peru, and Venezuela 
(Fig. 2). Average nearest-neighbor distance anal¬ 
ysis for all geolocators exhibited clustering with 
nearest-neighbor ratios = 0.89 (P = 0.02), 0.74 
(P < 0.001), and 0.77 (P < 0.001) for geolocator 
#553, #554. and #556, respectively. The distance 
between the Ouray Box Canyon Falls breeding site 
and the center of the wintering range in Brazil 
(#554) is 6,901 km and the average distance 
between the Fulton Resurgence Cave breeding site 
and the center of the wintering range in Brazil 
(#553 and #556) is 7.025 km. The swifts traveled at 
an average speed of 341 km/day during the 2009 
fall migration and an average speed of 393 km/day 
duiing the 2010 spring migration. The inaccuracy 
ol geolocators precludes precise calculation of an 
average daily distance covered by each bird. 
The land cover overlay maps for 50% kernel 
density areas for all three birds indicate a 
FIG. 1. Spring migration routes for three (#'s 553. 554. 
5561 individual Northern Black Swifts marked in Colorado. 
dominant land cover (>86%) of closed to open 
broadleaved evergreen or semi-deciduous forest 
and a small percentage (2-10%) of the kernel 
density areas are classified as closed to open 
