Arrival and Departure Dates

https://doi.org/10.1016/S0065-2504(04)35001-4Get rights and content

Publisher Summary

This chapter reviews the evidence indicating that the arrival and departure dates of migratory birds have changed, and the relationship of these changes to climate variability. The longest continuous time series available for analysis exceed 250 years in length. All these data are observations of first arrivals of bird species that were well known to people of rural communities. In time series approaching or exceeding 100 years, it becomes readily evident that the variation of arrival times around the overall average or trend often shows periods of constant direction of change and then a shift in the direction. Correlating or regressing the arrival dates with a suitable weather (often temperature) variable often reveals a strong relationship between arrival and temperature, but a substantial part of the variation remains unexplained. A proportion of unexplained variation arises from the fact that the local temperature that is used in correlation and regression analyses is, at best, only a proxy of the weather along the migration route prior to arrival at the observation locality. Responses determining departure date probably depend on the individual species. For some species, advancement in spring may result directly in earlier departure in autumn while in others longer summer and later autumn may lead to later departure. Apparently, the direction of response depends on the importance of summer and winter seasons in the species' complete life history strategy.

Section snippets

Summary

We review the current evidence that changes in arrival and departure dates of migratory birds have taken place, and the relationship of these changes to climate variability. There is little doubt that the timing of spring migration closely follows weather variations. This is more evident and/or stronger in short-distance migrants than in long-distance migrants, but the latter have also responded to climate change. Changes of spring arrival in birds depend on climate impacts at different

A Brief History

The earliest systematic records on arrival and departure of migratory birds in Great Britain are from the early 1700s (Kington 1988, Sparks and Carey 1995) and in Northern Europe from the mid-1700s (Linné 1757, Leche 1763). These texts mark the start of phenological research, which was one of the earliest organised branches of systematic ornithological observation in many parts of Europe.

It has been well known for hundreds of years that arrivals of migrating birds are good predictors of

Variation of Arrival Dates During the Last 250 Years

The longest continuous time series available for analysis exceed 250 years in length. All these data are observations of first arrivals of bird species that were well known to people of rural communities. In time series approaching or exceeding 100 years (Sparks and Carey 1995, Ahas 1999), it becomes readily evident that the variation of arrival times around the overall average or trend often shows periods of constant direction of change and then a shift in the direction (Figure 2). Correlating

Types and Quality of Data

Different types of arrival information have been used in recent analyses. First arrivals and mean or median arrivals are the most common ones. First arrival (the day on which the first individual of the species is recorded each spring in the area) is most frequently recorded in some countries by local ornithological clubs and in others more systematically by meteorological stations or in programs run by scientific or governmental organisations (von Haartman 1956, Ahas 1999; Sparks and

How is Earlier Arrival Achieved?

For a migratory species, breeding area is by definition different from the wintering area. Therefore, birds cannot respond behaviourally to amelioration of climate in the breeding area unless the climatic conditions in the wintering area and/or in the area through which the species migrate are positively correlated. The longer the migration journey, the less likely high positive correlations in climate between areas crossed will be. Long-distance migrants, furthermore, have to pass over

Statistics

Several simple methods have been used to test whether changes in timing of migration are statistically significant. Some authors have used correlation (arrival vs. year, Pearson or Spearman) others have used linear regression (arrival vs. year) and in still other cases non-parametric smoothing methods have been used for descriptive purposes. In a few cases, multiple regression approaches have also been applied.

The statistical complications, of which one must be aware of, depend partially on the

Confounding Factors

Simple analyses of temperature effect and trend are inevitably coarse. Arrival and departure dates depend on other factors as well. Some of these have been identified and incorporated in the models. We divide the confounding factors in two groups: (1) those depending on the observers, and (2) those depending on birds. An example of the former is the possible trend of observer activity and skill (Sparks et al. 2001, Butler 2003, Vähätalo et al. 2004, but cf.Mason 1995, Tryjanowski et al. 2002),

Overview of Changes of Timing in the Latter Part of the 20th Century

Burton 1995, Zalakevicius and Zalakeviciute 2001, Bairlein and Winkel 2001, Fiedler 2001 have earlier summarised evidence for changes in arrival and departure dates of migratory birds. Impacts on birds have been included also in more general reviews and meta-analyses (Hughes 2000, Root et al. 2003). We give below an overview on the present state of affairs knowing that much work is currently under way and new reviews will soon be needed. We collected published time series of migration times

Consequence of Changing Timing

Large changes in arrival and departure dates may have important consequences for other parts of a species' annual cycle. The final step—a species or an appreciable part of its population becoming sedentary—has also been documented with different example species in different areas (Zalakevicius et al. 1995, Berthold 1998; Fiedler et al., 2004, this volume). Partial migrants are continuously “monitoring” the weather and food conditions and they can be especially fast in taking whatever advantage

Future Work

The quality of data still needs further study. There are several points of concern as pointed out in this review, but despite this, the sign of change and the signal are most likely correct. What needs to be done is to identify and quantify the other factors that cause arrival and departure dates to shift in the same direction as climate and weather. At least in some areas, detectability of first arrivals may have changed and there is no careful analysis on how important a factor this might be.

References (80)

  • R. Harrington et al.

    Tree

    (1999)
  • L. Hughes

    Tree

    (2000)
  • R. Ahas

    Int. J. Biometeorol

    (1999)
  • M. Ahola et al.

    Global Change Biol

    (2004)
  • S. Åkesson et al.

    Behav. Ecol. Sociobiol

    (2000)
  • T. Alerstam

    Bird Migration

    (1990)
  • A.A. Ananin
  • F. Bairlein et al.
  • R.T. Barrett
    (2002)
  • P. Berthold

    Acta Congr. Int. Ornithol

    (1991)
  • P. Berthold

    Control of Bird Migration

    (1996)
  • P. Berthold

    Naturwissenschaftliche Rundschau

    (1998)
  • C. Both et al.

    Nature

    (2001)
  • N.L. Bradley et al.

    Proc. Natl Acad. Sci. USA

    (1999)
  • J. Burton

    Birds and Climate Change

    (1995)
  • C.J. Butler

    Ibis

    (2003)
  • W.S. Cleveland

    Am. Stat

    (1981)
  • T. Coppack et al.

    Ardea

    (2002)
  • T. Eeva et al.

    Ecography

    (2002)
  • N. Elkins

    Weather and Bird Behaviour

    (1995)
  • W. Fiedler
  • M.C. Forchammer et al.

    J. Anim. Ecol

    (2002)
  • Frederick II of Hohenstaufen (1244–48) De Arte venandi cum avibus (translated by C.A. Wood and F.M. Fyfe (1943). The...
  • W. Gatter

    J. Orn

    (1992)
  • A. Gilyazov et al.

    Avian Ecol. Behav

    (2002)
  • E. Gwinner

    Circannual Rhythms: Endogenous Annual Clocks in the Organization of Seasonal Processes

    (1986)
  • E. Gwinner

    J. Exp. Biol

    (1996)
  • M. Hario et al.

    Finnish Game Res

    (1988)
  • M. Hario et al.

    Suomen Riista

    (1993)
  • R. Heino

    Finn. Meteorol. Inst. Contrib

    (1994)
  • A. Helbig
  • N. Huin et al.

    Bird Study

    (1998)
  • N. Huin et al.

    Bird Study

    (2000)
  • O. Hüppop et al.

    Proc. R. Soc. Lond., Ser. B Biol. Sci

    (2003)
  • D.W. Inouye et al.

    Proc. Natl Acad. Sci. USA

    (2000)
  • D. Jenkins et al.

    Bird Study

    (2000)
  • L. Jenni et al.

    Proc. R. Soc. Lond. B

    (2003)
  • J.G. Justander

    Specimen calendarii florae et faunae aboensis

    Aboæe

    (1786)
  • A.M.G. Klein Tank et al.

    J. Clim

    (2003)
  • Cited by (320)

    View all citing articles on Scopus
    View full text