494 
THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 3. September 2012 
TABLE 3. Model selection for conditional logistic regressions describing the effect of percent substrate cover on 
probability of microsite use by breeding Magellanic Plovers. Sample size (n), number of parameters (k), Akaike's 
Information Criterion (AIC), AIC corrected for small sample size (A1C,). difference in AIC, (AAIC,.). and Akaike's weight 
(h 7) are reported for the top five models (AAIC, < 4) and constant model of lb candidate models describing the effect of 
substrate cover on microsite occupancy of the Magellanic Plover. 
Model 
n 
k 
AIC 
AIC. 
AAIC, 
wi 
Clay, cobble 
50 
3 
14.186 
20.708 
0.000 
0.37 
Clay 
50 
2 
17.055 
21.310 
0.602 
0.27 
Clay, vegetation 
50 
3 
17.040 
23.562 
2.854 
0.09 
Clay, gravel, cobble 
50 
4 
15.492 
24.381 
3.673 
0.06 
Clay, cobble 
50 
3 
18.065 
24.587 
3.879 
0.05 
Constant 
50 
1 
34.657 
36.741 
16.033 
0.00 
lakes throughout the accessible portion of this 
area. The inaccessible interior of the province is 
likely to have similar proportions of occupied 
lakes. Lakes with plovers were large (with large 
perimeters), and contained large aeolian lunettes. 
All nests were on the windswept shorelines of 
lakes with lunettes, and on the side of the lake 
with the lunette. The substrate created by strain 
winds was predominantly clay and cobble and. oi 
these shorelines, both adults and their nests wen 
extremely cryptic. 
Plovers occupied sites further from vegetatioi 
than unused sites at the territory scale allowing 
for good visibility by breeding pairs. Thus, lik« 
Charadrius plovers (Nguyen et al. 2003), thi 
species probably nests in open habitats to maxi 
mize predator detection, while minimizing visibil 
ity of adults, nests, and eggs (Graul 1973. Solis an< 
De Lope 1995. Nguyen et al. 2007). Most occupies 
lakes were saline, although we did find nests ant 
successfully Hedged young on two Ireshwate 
lakes, suggesting the species can nest successfully 
using food derived from freshwater sources. 
We did not lmd any pairs nesting on rive 
banks. Previous reports of their occurrence then 
(Ferrari et al. 2003) suggest a larger sample o; 
nvers, as well as a thorough survey of potentiall\ 
suitable habitats on Tierra del Fuego needs to be 
conducted to modify current estimates of popula 
tion size. Our lack of ability to find a relationship 
between number of territories (e.g.. abundance 
and any lake measure that we used suggests 
predicting which lakes might be occupied will be 
aided by oar research, but predicting abundance 
close n™'" d ' ffi f Ul1 ' For exam P lc - 'wo lakes in 
close proxtmuy had 14 and I pair, respectively 
and we were enable to visually assess any 
'“Tnib t CharaCterisli ‘ :s 0<£ lunette 
' Sal,mty - sha P e ° r major ecological differ¬ 
ences). One variable we did not quantify and was 
unavailable on the satellite image was the number 
of small streams entering occupied lakes. Plovers 
at the territory scale nested near freshwater sources, 
and we often observed breeding plovers feeding 
where streams entered the lakes; this may be an 
important variable for future study. Ideally,finding 
a characteristic that can be measured via remote 
sensing that correlates with abundance, will allow 
more accurate estimates of population size. Cur¬ 
rently. we are unable to reline the global population 
estimates of between 1.000 (Jehl 1975) and 10.000 
(Birdl.ife International 2012) individuals. 
Magellanic Plovers within occupied territories 
avoided placing nests on clay substrates. Avoid¬ 
ance of vegetation in nest-site selection and place¬ 
ment of the one or two eggs directly on gravel or 
cobble rather than clay are probably additional 
anti-predator adaptations (Solis and De Lobe 
1995; Nguyen et al. 2003, 2007). Clay retains 
moisture better than gravel or cobble, and nests on 
this substrate could cool unnecessarily when 
adults are not in attendance, 
Poor detection due to a highly cryptic plumage 
has been a critical issue for research on this 
species (Ferrari et al. 2003). Surveys using our 
technique ot parallel walking, repeated monthly, 
detected 90% of pairs, in part, due to vocaliza¬ 
tions given by breeding individuals. 
We documented reproductive success at 20 
nests. Hatching success was higher than those 
reported for other under-studied South American 
shorebirds (11.6%. Charadrius wilsonia. Ruiz- 
Gucrra et al. 2008; 46%. C. Jalklandicus. St. Clair 
et al. 2010), and may be a result of the highly 
cryptic nests ot the Magellanic Plover. None cl 
the habitat variables that we measured were linked 
lo nest success, but our sample size was small, 
ieffecting low densities of this rare species across 
