The common and very widespread American Robin was the seventh most abundantly captured species in the 1992-2006 MAPS database with 12,904 adult individuals captured and 1,523 between-year recaptures at 383 stations (the largest number of stations for any species in the database) spread across 24 Bird Conservation Regions (BCRs; second only to the 25 BCRs for Downy Woodpecker and Brown-headed Cowbird).  The rather low number of between year recaptures (relative to the number of individuals captured) reflects the relatively low recapture probability for this species (its mean transient CJS model recapture probability of about 0.25 was only about half or less of the analogous recapture probabilities for the smaller thrushes).  Species like American Robin, with a body mass greater than 50 g, are more efficiently captured in 36-mm-mesh mist nets than in the 30-mm-mesh nets generally used in the MAPS program.  In addition, American Robins spend more time in relatively open areas and less time in the denser vegetation favored by the smaller thrushes, where mist nets are more effective and more often placed.  Although captures of robins were widespread all across North America, the largest numbers of robins were captured in the northwestern part of the United States (Northern Pacific Rainforest, Great Basin, and Northern Rockies BCRs), where there is a high density of MAPS stations sited at the forest edges and montane meadows favored by large populations of robins.

Temporal and spatial analyses of 1992-2006 program-wide MAPS data showed weighted mean population density indices of 2.8 and 3.8 adults per station, respectively.  These densities were slightly lower than the mean densities for all species, and much lower than analogous densities for all of the smaller thrush species except Hermit Thrush.  Annual variability in population density (5.7%) was very low, the second lowest of all species in the MAPS database, and lower than the annual variabilities of all of the smaller thrushes, including Swainson’s Thrush, which also had a very low annual variability in population density (6.4%).  In contrast, spatial variability in population density (33.8%), while still about 15% lower than the mean variability for all species, was roughly similar to the analogous variabilities for Veery and both Swainson’s and Wood thrushes, but much lower than those for both Gray-cheeked and Hermit thrushes.  The linear time model for the index of population density for American Robin showed a significant positive Beta of 0.010, suggesting, perhaps, a slightly increasing population trend.

The weighted geometric mean of the model-averaged annual lambda estimates (1.001) indicated a stable American Robin population, which agreed exactly with the population trend from the 1992-2006 program-wide North American Breeding Bird Survey (BBS; a stable lambda of 1.001).  Spatial analyses of MAPS data, however, indicated a slightly but significantly increasing population trend (the analogous mean of the BCR-specific lambda estimates was 1.007, which differed significantly from 1.0).  The annual variability of lambda for American Robin (7.9%) was much less than the mean variability for all species and less than those for each of the smaller thrushes, which might be expected for a species with a stable or slightly increasing population trend.  Spatial variability of lambda (3.1%) was also low, about half the mean for all species, but was higher than the analogous variabilities for Veery and both Swainson’s and Gray-cheeked thrushes, but lower than those for Hermit and Wood thrushes.

Temporal and spatial analyses produced estimates for adult apparent survival for American Robin (0.510 and 0.501, respectively) that were quite low for a relatively large (approximately 80-g) passerine, and perhaps reflect a relatively high emigration rate, rather than a high mortality rate.  American Robin, however, like Hermit Thrush and both Eastern and Western bluebirds, is a temperate-wintering migratory species and might, like them, be subject to harsher and more variable winter weather and, thus, be expected to show lower adult survival rates with higher annual and spatial variabilities than Neotropical-wintering migratory species.  Indeed, the annual variability in adult apparent survival for American Robin (10.5%) was higher than the analogous variabilities for Veery and both Gray-cheeked and Swainson’s thrushes, but was lower than those for both Wood and Hermit thrushes.  Spatial variability in adult apparent survival for American Robin (15.1%) was similar to the mean for all species and to the analogous variability for Gray-cheeked Thrush, but was higher than the spatial variabilities for each of the other thrush species.  The linear time model for adult apparent survival showed a significantly declining Beta of -0.045, suggesting that adult apparent survival declined over the 15 years of study.

Temporal and spatial analyses produced somewhat different estimates for the productivity index for American Robin (0.266 and 0.308, respectively).  These estimates were intermediate between the higher productivity indices for Gray-cheeked and Hermit thrushes, and the lower productivity indices for Veery and both Swainson’s and Wood thrushes.  Annul variability in productivity (15.9%) for American Robin was only about a third of the mean annual variability for all species, and was lower than analogous variabilities for all five other thrushes.  In contrast, spatial variability in productivity (54.5%) was higher than the mean spatial variability for all species and about as high or higher than the analogous variabilities for each of the five other thrush species.  The linear time model for productivity, like that for adult apparent survival, showed a significantly declining Beta of -0.011, suggesting a 15-year decline in productivity.

Temporal analyses among American Robin vital rates showed that lambda was moderately but non-significantly positively correlated with post-breeding effects and rather weakly positively correlated with productivity, but weakly negatively correlated with adult apparent survival.  This suggests that annual variation in post-breeding effects was the strongest driver of annual variation in lambda, which also provides a possible explanation for why estimates of lambda indicated a stable or increasing population, while linear time models indicated decreasing trends for both adult apparent survival and productivity.  Temporal analyses also showed strong and highly significant negative correlations between post-breeding effects and both adult apparent survival and productivity, as well as a rather weak and non-significant positive correlation between productivity and adult apparent survival.  These results suggest that at least some degree of density dependence likely existed on both the breeding and non-breeding grounds.

The suggestion of density dependence on the breeding grounds was supported by the moderate but not significant negative correlation between the index of adult population density and lambda, and by the strong and significant negative correlation between adult population density and productivity.  Temporal analyses also showed a weak negative correlation between adult population density and adult apparent survival, and a moderate positive correlation between adult population density and post-breeding effects.  Overall, these results suggest that the relatively stable populations of American Robin may have been regulated to a large extent through density-dependent mechanisms on the breeding grounds, especially those effecting productivity, but that the slight increase in overall populations during the 15 years of study could have been driven by factors affecting the first-year survival of young birds on their non-breeding grounds.  The weak positive temporal correlation between adult apparent survival and productivity suggests that there was no cost of reproduction for this abundant, multi-brooded species.

Spatial analyses of American Robin MAPS data found different patterns of correlations among vital rates than were found with temporal analyses.  Lambda, for example, tended to be spatially positively correlated with all three vital rates, although none of the three correlations was strong or significant.  This suggests that spatial variation in all three vital rates contributed positively toward spatial variation in lambda.  In addition, the order of strength of the spatial correlations was exactly opposite to the order of strength of the temporal correlations, with adult apparent survival showing the strongest, but still rather weak, spatial correlation with lambda.  Moreover, the spatial correlations between adult apparent survival and both productivity and post-breeding effects were opposite to the analogous two temporal correlations.  These differences are difficult to explain, but probably derive from different mechanisms driving population change in different parts of the species’ wide range, complicated by different mechanisms being operative during different portions of the 15 years of study.  For both spatial and temporal analyses, however, post-breeding effects were highly significantly negatively correlated with productivity, perhaps suggesting density-dependent regulation of first-year survival of young birds during the non-breeding season or of recruitment of the surviving young during the breeding season.  These suggestions, however, seem difficult to reconcile with the weak and non-significant positive temporal correlation between post-breeding effects and the index of adult population density.

Summary of research and management hypotheses – Because of the American Robin’s widespread abundance and stable or slightly increasing population trend, research and management efforts for this species are at present not required.  Considering, however, that climate change will likely modify the population dynamics of most species, we provide suggestions for research and management of robin populations should their populations begin to decline.  We suggest that research and management efforts should be directed first toward determining and addressing causes of low survival for both young and adult birds, particularly during years and within regions when and where low survival tends to drive annual or spatial declines in lambda.  Similar research and management should then be directed toward determining and addressing causes of low productivity.  Because little is known regarding migratory connectivity in American Robin, or even whether strong migratory connections exist, research on migratory connectivity will need to be initiated.

Please cite this narrative as:  DeSante, D. F., D. R. Kaschube, and J. F. Saracco.  2015.  Vital Rates of North American Landbirds.  www.VitalRatesOfNorthAmericanLandbirds.org: The Institute for Bird Populations.