Limits...
Change in spring arrival of migratory birds under an era of climate change, Swedish data from the last 140 years.

Kullberg C, Fransson T, Hedlund J, Jonzén N, Langvall O, Nilsson J, Bolmgren K - Ambio (2015)

Bottom Line: There was a larger change in spring phenology in short-distance migrants than in long-distance migrants.Interestingly, the results further suggest that climate change has affected the phenology of short-distance migrants more in southern than in central Sweden.The results suggest that the much earlier calculated arrival to southern Sweden among short-distance migrants mirrors a change in location of wintering areas, hence, connecting migration phenology and wintering range shifts.

View Article: PubMed Central - PubMed

Affiliation: Department of Zoology, Stockholm University, 106 91, Stockholm, Sweden, cecilia.kullberg@zoologi.su.se.

ABSTRACT
Many migratory bird species have advanced their spring arrival during the latest decades, most probably due to climate change. However, studies on migratory phenology in the period before recent global warming are scarce. We have analyzed a historical dataset (1873-1917) of spring arrival to southern and central Sweden of 14 migratory bird species. In addition, we have used relative differences between historical and present-day observations (1984-2013) to evaluate the effect of latitude and migratory strategy on day of arrival over time. There was a larger change in spring phenology in short-distance migrants than in long-distance migrants. Interestingly, the results further suggest that climate change has affected the phenology of short-distance migrants more in southern than in central Sweden. The results suggest that the much earlier calculated arrival to southern Sweden among short-distance migrants mirrors a change in location of wintering areas, hence, connecting migration phenology and wintering range shifts.

No MeSH data available.


Related in: MedlinePlus

Regression lines for estimated first arrival day of year (DOY) in relation to latitude for the historical (a, c) and present day (b, d) datasets in short-distance (a, b) and long-distance (c, d) migrants, respectively. Southern Sweden = 55.5°N and central Sweden = 61°N. DOY 50 represents February 19, while DOY 100 represents April 10
© Copyright Policy - OpenAccess
Related In: Results  -  Collection


getmorefigures.php?uid=PMC4289004&req=5

Fig1: Regression lines for estimated first arrival day of year (DOY) in relation to latitude for the historical (a, c) and present day (b, d) datasets in short-distance (a, b) and long-distance (c, d) migrants, respectively. Southern Sweden = 55.5°N and central Sweden = 61°N. DOY 50 represents February 19, while DOY 100 represents April 10

Mentions: For the historical data, we used a linear mixed model analysis (R Core Team 2014; Bates et al. 2014) examining the relation between arrival day and the following factors: latitude, altitude, and distance to larger water bodies. Factors were nested within year. Data on altitude and distance to large water bodies had a highly skewed distribution and were thus square-root transformed. Since present-day data were not based on instruction for first sightings, we used quantile regression (Koenker 2013) to analyze patterns of early reports in relation to latitude. With the linear quantile model, we obtained a regression for a given percentile of the dataset giving us an estimate of arrival date in relation to latitude for each species. We chose to use the 5% percentile of the data for this estimate. In accordance with the historical dataset, we used latitude, altitude (square-root transformed), and distance to larger water bodies (square-root transformed) as factors also for the present-day dataset. After obtaining intercept and regression coefficients for the effect of latitude on arrival date for each species and year, we calculated mean intercept and coefficient for each species in the present-day data. The obtained intercept and coefficient for each species from the historical and present-day datasets were then used to calculate arrival date at specific latitudes and analyzed to reveal patterns of change depending on migratory strategy of the species under study (Table 1; Fig. 1). Since the two datasets contain data that were gathered differently using different methods, we cannot compare the calculated arrival dates of the two models directly. We therefore present modeled arrival dates and perform statistical analyses only on relative differences between the historical and present day model in the results. When analyzing effects of migratory strategy, we compared trans-Saharan migrants (long-distance migrants; 6 species) with migrants that do not cross the desert (short-distance migrants; 8 species).Fig. 1


Change in spring arrival of migratory birds under an era of climate change, Swedish data from the last 140 years.

Kullberg C, Fransson T, Hedlund J, Jonzén N, Langvall O, Nilsson J, Bolmgren K - Ambio (2015)

Regression lines for estimated first arrival day of year (DOY) in relation to latitude for the historical (a, c) and present day (b, d) datasets in short-distance (a, b) and long-distance (c, d) migrants, respectively. Southern Sweden = 55.5°N and central Sweden = 61°N. DOY 50 represents February 19, while DOY 100 represents April 10
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC4289004&req=5

Fig1: Regression lines for estimated first arrival day of year (DOY) in relation to latitude for the historical (a, c) and present day (b, d) datasets in short-distance (a, b) and long-distance (c, d) migrants, respectively. Southern Sweden = 55.5°N and central Sweden = 61°N. DOY 50 represents February 19, while DOY 100 represents April 10
Mentions: For the historical data, we used a linear mixed model analysis (R Core Team 2014; Bates et al. 2014) examining the relation between arrival day and the following factors: latitude, altitude, and distance to larger water bodies. Factors were nested within year. Data on altitude and distance to large water bodies had a highly skewed distribution and were thus square-root transformed. Since present-day data were not based on instruction for first sightings, we used quantile regression (Koenker 2013) to analyze patterns of early reports in relation to latitude. With the linear quantile model, we obtained a regression for a given percentile of the dataset giving us an estimate of arrival date in relation to latitude for each species. We chose to use the 5% percentile of the data for this estimate. In accordance with the historical dataset, we used latitude, altitude (square-root transformed), and distance to larger water bodies (square-root transformed) as factors also for the present-day dataset. After obtaining intercept and regression coefficients for the effect of latitude on arrival date for each species and year, we calculated mean intercept and coefficient for each species in the present-day data. The obtained intercept and coefficient for each species from the historical and present-day datasets were then used to calculate arrival date at specific latitudes and analyzed to reveal patterns of change depending on migratory strategy of the species under study (Table 1; Fig. 1). Since the two datasets contain data that were gathered differently using different methods, we cannot compare the calculated arrival dates of the two models directly. We therefore present modeled arrival dates and perform statistical analyses only on relative differences between the historical and present day model in the results. When analyzing effects of migratory strategy, we compared trans-Saharan migrants (long-distance migrants; 6 species) with migrants that do not cross the desert (short-distance migrants; 8 species).Fig. 1

Bottom Line: There was a larger change in spring phenology in short-distance migrants than in long-distance migrants.Interestingly, the results further suggest that climate change has affected the phenology of short-distance migrants more in southern than in central Sweden.The results suggest that the much earlier calculated arrival to southern Sweden among short-distance migrants mirrors a change in location of wintering areas, hence, connecting migration phenology and wintering range shifts.

View Article: PubMed Central - PubMed

Affiliation: Department of Zoology, Stockholm University, 106 91, Stockholm, Sweden, cecilia.kullberg@zoologi.su.se.

ABSTRACT
Many migratory bird species have advanced their spring arrival during the latest decades, most probably due to climate change. However, studies on migratory phenology in the period before recent global warming are scarce. We have analyzed a historical dataset (1873-1917) of spring arrival to southern and central Sweden of 14 migratory bird species. In addition, we have used relative differences between historical and present-day observations (1984-2013) to evaluate the effect of latitude and migratory strategy on day of arrival over time. There was a larger change in spring phenology in short-distance migrants than in long-distance migrants. Interestingly, the results further suggest that climate change has affected the phenology of short-distance migrants more in southern than in central Sweden. The results suggest that the much earlier calculated arrival to southern Sweden among short-distance migrants mirrors a change in location of wintering areas, hence, connecting migration phenology and wintering range shifts.

No MeSH data available.


Related in: MedlinePlus