An indicator of the impact of climatic

change on European bird populations

Richard D. Gregory Stephen G. WillisFrédéric Jiguet

Petr VoříšekAlena Klvaňová Arco van StrienBrian Huntley

Yvonne C. Collingham

Denis Couvet &

Rhys E. Green

Evidence is accumulating that climatic change has altered many biological phenomena

across the globe, including the geographical ranges and abundance of plants and

animals, and the timing of events in their lives such as growth, reproduction and

migration

Background

Birds laying earlier - BTO Nest Record Scheme

Leaf burst earlier in Europe

Scientists and policy makers have called for the development of indicators of the impacts of

climatic change on biodiversity based upon these

phenomena

To capture biological impacts, to describe how they are changing in an accessible way, & to raise

awareness of the consequences of climatic warming for wildlife & for people

In addition, to assist in setting targets for the reduction of impacts & help guide the

implementation of mitigation & adaptation measures

Purpose of a climatic change indicator:

1. Predictions from bioclimate envelope models (mid-end century 2070-2099)

2. Observed trends in European birds (1980-2005 derived from the PECBMS)

The indicator combines two independent strands of work:

The starting point is the EBCC’s Atlas and the climate envelope models fit to distribution data for European breeding birds

Bird distributions mapped in late 1980s -- 50-km UTM squares -- presence & absence of species

Based on the bioclimatic envelope models for each bird

species, Brian Huntley et al., have published the first

‘Climatic Atlas’ of its kind for any taxa

The ‘Climatic Atlas’ uses 3 simple bioclimate variables to

model European bird distributions:

1. ‘MTCO’ Mean temperature of the coldest month

2. ‘GDD5’ Annual temperature sum above 5 degrees C

3. ‘AET/PET’ Ratio of actual to potential evapo-transpirationThe models provided a good fit to our data (area under the curve – AUC – of a receiver

operating characteristic – ROC – plot; mean AUC of the 122 species = 0.967; lowest value = 0.907).

Serin Serinus serinus

Present simulated range

~1961-1990

Future ‘potential’ range under a modelled climatic change scenario: HadCM3 B2 for ~2070-2099

We have the PECBMS population trends e.g. European Wild Bird

Indicator 2008

40

60

80

100

120

1980 1985 1990 1995 2000 2005Year

Pop

ulat

ion

inde

x (1

980=

100)

-43% Common Farmland (33 species)

-15% All common (124 species)

-14% CommonForest (28 species)

European trends for 124 common bird species were available from the PECBMS

1

10

100

1000

1980 1985 1990 1995 2000 2005Year

Pop

ulat

ion

inde

x (1

980=

100)

Jynx torquilla

Picus canus

Picus viridis

Dryocopus martius

Dendrocopos major

Dendrocopos minor

1

10

100

1000

1980 1985 1990 1995 2000 2005Year

Pop

ulat

ion

inde

x (1

980=

100)

Troglodytes troglodytes

Prunella modularis

Erithacus rubecula

Luscinia megarhynchos

Phoenicurus phoenicurus

Turdus merula

Turdus philomelos

Turdus viscivorus

1

10

100

1000

1980 1985 1990 1995 2000 2005Year

Pop

ulat

ion

inde

x (1

980=

100)

Hippolais icterina

Sylvia borin

Sylvia atricapilla

Phylloscopus sibilatrix

Phylloscopus collybita

Phylloscopus trochilus

Regulus regulus

Muscicapa striata

Ficedula albicollis

Ficedula hypoleuca

We developed the indicator in two steps:

First, we tested the performance of projections of change in the extent of species’ geographical range

(termed ‘CLIM’, based upon climatic envelope models) as predictors of observed interspecific variation in

population trends of European birds

Testing the performance of envelope models is necessary to address concerns about their accuracy in

predicting species’ responses to climatic change

We expect a positive correlation between observed change in abundance and

‘CLIM’

Having found a robust relationship of this kind, our second step was to construct an indicator based upon the divergence in population trends between species expected to be positively and negatively affected by

climatic change

The ‘CLIM’ value for a species is the loge of the ratio of the extent of the future potential range

to that of the recent simulated range(CLIM >0 predicts range expansion, CLIM <0 predicts range

contraction)

We also looked at the influence of habitat choice, migratory behaviour & body mass (as a proxy for life

history characteristics) in predicting bird trends

Step One

To test for sensitivity of the scenario projections we considered results from:

• 3 General Circulation Models (GCM): HadCM3, Echam4 & GFDL

• 2 Scenarios from the Special Report on Emissions Scenario (SRES): A2 & B2

• = 6 variants termed ‘CLIMHaA2’, ‘CLIMHaB2’, ‘CLIMEcA2’, ‘CLIMEcB2’, ‘CLIMGfA2’ and ‘CLIMGfB2’

• We also calculated the average of these 6 to create an ‘ensemble forecast’, termed ‘CLIMEns’

Step One

Observed increase

Observed decrease

Retrodicted range decrease Retrodicted range increase

Population trends of 108 bird species in 20 European countries 1980 – 2005 correlated significantly with projected trend in climate suitability from the climate envelope models

-0.2

-0.1

0

0.1

-0.03 -0.02 -0.01 0 0.01 0.02 0.03

CLIM value

Obs

erve

d tre

nd

We found a highly significant +ve correlations between interspecific variation in recent population trends & the CLIM projections

-0.1

0

0.1

0.2

0.3

0.4

0.5

CLIM

HaA

2

CLIM

HaB

2

CLIM

EcA

2

CLIM

EcB

2

CLIM

GfA

2

CLIM

GfB

2

CLIM

Ens

CLIM

Ens -

CLIM

Ens+

CS

T

Climatic Response Predictor

Sta

ndar

dise

d re

gres

sion

coe

ffici

ent

?

Lots of assumptions. One is that the bioclimate variables have changed since 1980 in the direction of the GCMs for the

longer-term predictions

• We tested this by examining the relationship of CLIM & the recent trend in climate suitability based upon observed climate change 1980-2005

• We used the climate envelope models and the annual values of the bioclimate variables to calculate probability of occurrence in each year for each species

• We then regressed these against year for each species and the slope of this line is what we call the ‘Climate Suitability Trend’ (CST)

Encouragingly, we found:1. A highly significant relationship between

interspecific variation in CLIM and CST

- So climate suitability for species is changing just as we’d predict

2. A marginally significant relationship between observed population trend and CST when controlling for confounding variables

- So bird numbers are changing just as we’d predict over this period, but the link is quite weak

Our second step was therefore to construct an indicator from the observed population

trajectories of 122 bird species with data available for any part of the period 1980 – 2005

We divided these species into those for which the climatic envelope model projection indicated

an increase in potential geographical range (CLIMEns+) and those with projected decreases

in geographical range (CLIMEns-).

Step Two

For each of the two groups of species, we calculated a multi-species population index from population

indices for individual species, with the weight of the contribution of each species to the index being being

its absolute value of CLIMEns

Extreme CLIM values for species (+ve or –ve) have greater influence on the line

So birds predicted to be strongly affected by climate in

our models strongly influence the direction of the index

Step Two

60

65

70

75

80

85

90

95

100

105

110

1980 1985 1990 1995 2000 2005

Year

Wei

ghte

d po

pula

tion

inde

x

(A) Weighted population trend of species predicted to gain range in response to climatic change (30 species)

60

65

70

75

80

85

90

95

100

105

110

1980 1985 1990 1995 2000 2005

Year

Wei

ghte

d po

pula

tion

inde

x(B) Weighted population trend of species predicted to lose

range in response to climatic change (92 species)

Multi-species population indices for both species

groups declined in the early 1980s,

but from the latter part of that

decade onwards, CLIMEns+ (30

species) increased, whilst CLIMEns- index

(92 species) continued to

decline

The impact of climatic changes (both +ve and -ve) on bird populations can then be summarised in a

single indicator, the ‘Climatic Impact Indicator’ (CII)

This is calculated in a given year as the ratio of the index for CLIMEns+ species to that for CLIMEns- species, and has 95% confidence limits obtained

using a bootstrap method

Step Two

The Climatic Impact Indicator (CII), reflecting the divergence of the indices for the two groups, declined

slightly in the early 1980s, but has shown a roughly linear increase from then onwards

1980 1985 1990 1995 2000 2005

Inde

x of

clim

atic

cha

nge

impa

cts

on b

ird p

opul

atio

ns

50

60

70

80

90

100

Index valuePiecewise regression

160B

120

140

We can present the CII in a more accessible fashion for a wider general audience

60

70

80

90

100

110

120

130

140

1980 1985 1990 1995 2000 2005Year

Inde

x of

clim

atic

impa

cts

on b

ird p

opul

atio

ns

Increasing climatic impact on bird populations

Decreasing climatic impact on bird populations

Note that the pattern in the CII closely resembles that of observed climatic change in Europe

1980 1985 1990 1995 2000 2005

Inde

x of

clim

atic

cha

nge

impa

cts

on b

ird p

opul

atio

ns

50

60

70

80

90

100

Index valuePiecewise regression

160 B

120

140

Year

1980 1985 1990 1995 2000 2005

Stan

dard

ised

clim

atic

indi

ces

-3

-2

-1

0

1

2

GDD5MTCOMTEMPPiecewise regression

C

But what does the CII show?

• It shows conformity between observed population trends & projections of how each species’ population should respond to climatic warming

• The CII increases when population trends go in the direction predicted by the models

• The CII decreases when population trends go in the opposite direction predicted by the models

We can also create the CII adjusting for the confounding effects of habitat, migratory

behaviour & body mass on the trends – but it is basically unchanged

Year

1980 1985 1990 1995 2000 2005Inde

x of

clim

ate

chan

ge im

pact

s on

bird

pop

ulat

ions

50

60

70

80

90

100

Adjusted index valueUnadjusted index value

120

140

160

Key messages1. Climate change is having a detectable effect on common

bird populations at a European scale, including evidence of negative as well as positive effects

2. The number of bird species whose populations are observed to be negatively impacted by climatic change is 3 times that of those positively affected in our sample

3. The Climatic Impact Indicator (CII) has increased strongly in the past 20 years, coinciding with a period of rapid warming

4. Potential links between changes in bird populations and ecosystem functioning are not well understood. It is suggested that increasing climatic effects might alter ecosystem functioning & resilience

The novelty of the findings:• Shows a strong link between observed population

change and forecast change in range extent in a large species assemblage (widespread/common European birds)

• New observation that this link is apparently equally strong for species predicted to be negatively & positively impacted by climatic change

• Application of these results into an index of biotic impact of climatic change, provides first time a robust, accessible indicator of a phenomenon of global concern

So what does this mean for the birds?

Potentially, at least, wide-scale changes in bird communities across Europe with:

1. Sardinian Warbler2. Subalpine Warbler3. Bee-eater4. Cirl Bunting5. Cetti’s Warbler6. Hoopoe7. Golden Oriole8. Goldfinch9. Great Reed Warbler10.Collared Dove

A few winners (?) ‘Top 10’ - Increasing birds projected

to increase

And many losers (?) ‘Bottom 10’ - Declining birds

projected to decline

1. Snipe2. Meadow Pipit3. Brambling4. Willow Tit5. Lapwing6. Thrush Nightingale7. Wood Warbler8. Nutcracker9. Northern Wheatear10.Lesser Spotted Woodpecker

Special thanks to the PECBMS network

Special thanks to the data providers and organisations responsible for national data collection and analysis: Adriaan Gmelig Meyling (Statistics Netherlands). Norbert Teufelbauer, Michael Dvorak, Christian Vansteenwegen, Anne Weiserbs, Jean-Paul Jacob, Anny Anselin, Karel Šťastný, Vladimír Bejček, Jiří Reif, Henning Heldbjerg, Michael Grell, Andres Kuresoo, Frederic Jiguet, Risto Väisänen, Martin Flade, Johannes Schwarz, Tibor Szép, Olivia Crowe, Lorenzo Fornasari, Ainars Aunins, Ruud P. B. Foppen, Magne Husby, Przemek Chylarecki, Geoff Hilton, Juan Carlos del Moral, Virginia Escandell, Ramón Martí, Åke Lindström, Hans Schmid, David G. Noble, Juha Tiainen, Romain Julliard, Ward Hagemeijer, David G. Noble, Norbert Schäffer, Nicola Crockford, Zoltan Waliczky, David Gibbons, Simon Wotton, Adrian Oates, Gregoire Loïs, Dominique Richard, Anne Teller, Jeremy Greenwood, Lucie Hošková, Václav Zámečník, Lukáš Viktora, Tomáš Telenský, & Zdeněk Vermouzek.

Gregory R.D., Willis, S.G., Jiguet, F., Voříšek, P., Klvaňová, A., van Strien, A., Huntley, B Collingham, Y.C., Couvet, D. &

Green, R.E. (2009). An indicator of the impact of climatic change on European bird

populations. PLoS ONE 4(3): e4678. doi:10.1371/journal.pone.0004678

FREELY AVAILABLE AT:http://www.plosone.org/article/info:doi/10.1371/jour

nal.pone.0004678

Next steps

• Update CII with new trend data• Repeat at national and regional scales• Build non-breeding ranges for migrants• Explore CII trend pattern and trends• Explore new modelling approaches and

climate/data• Correlate projected range change with observed

range change

• We are looking for funding