Description: A colony of Humboldt penguins Spheniscus humboldti in central Chile was monitored from August 1995 to July 2000 to determine patterns of breeding and colony attendance and how these were affected by climatic (rainfall) and oceanographic (El Niño) factors. Nests were periodically checked for contents and roosting birds were counted from vantage points. Two main breeding events were observed: between August and January (spring event) and between April and June (autumn event). Whereas the spring event regularly produced offspring, the autumn event was systematically affected by rains, causing considerable nest desertion. Adults were present in the colony from August to May, abandoning the colony during winter after the nests were flooded. Juveniles occurred only between November and March. Adults moulted mainly in February, while juveniles moulted in January. During the 1997/98 El Niño episode, the number of breeding pairs was 55 to 85% lower than the mean, the onset of nesting was delayed, and abnormally heavy rainfall flooded nests. While the number of breeding pairs was significantly related to sea surface temperature anomalies (SSTA), breeding success was not. The attendance of adults and juveniles at the colony during El Niño was 25 and 73% lower, respectively, than the mean attendance. This 2-peak breeding strategy of Humboldt penguins appears to have evolved in response to the more favourable oceanographic and climatic conditions of Perú, where breeding is continuous and not interrupted by rains. Although less productive, the species probably maintains its autumnal breeding in central Chile because this provides additional offspring to supplement those regularly produced during the spring event.

Description: The spatio-temporal pattern of peak Holocene warmth (Holocene thermal maximum, HTM) is traced over 140 sites across the Western Hemisphere of the Arctic (0–180°W; north of ∼60°N). Paleoclimate inferences based on a wide variety of proxy indicators provide clear evidence for warmer-than-present conditions at 120 of these sites. At the 16 terrestrial sites where quantitative estimates have been obtained, local HTM temperatures (primarily summer estimates) were on average 1.6±0.8°C higher than present (approximate average of the 20th century), but the warming was time-transgressive across the western Arctic. As the precession-driven summer insolation anomaly peaked 12–10 ka (thousands of calendar years ago), warming was concentrated in northwest North America, while cool conditions lingered in the northeast. Alaska and northwest Canada experienced the HTM between ca 11 and 9 ka, about 4000 yr prior to the HTM in northeast Canada. The delayed warming in Quebec and Labrador was linked to the residual Laurentide Ice Sheet, which chilled the region through its impact on surface energy balance and ocean circulation. The lingering ice also attests to the inherent asymmetry of atmospheric and oceanic circulation that predisposes the region to glaciation and modulates the pattern of climatic change. The spatial asymmetry of warming during the HTM resembles the pattern of warming observed in the Arctic over the last several decades. Although the two warmings are described at different temporal scales, and the HTM was additionally affected by the residual Laurentide ice, the similarities suggest there might be a preferred mode of variability in the atmospheric circulation that generates a recurrent pattern of warming under positive radiative forcing. Unlike the HTM, however, future warming will not be counterbalanced by the cooling effect of a residual North American ice sheet.