ballester j 20150708_1730_upmc_jussieu_-_room_105
TRANSCRIPT
Will Climate Warming Decrease
Winter Mortality in Europe?
Joan Ballester
Institut Català de Ciències del Clima (IC3)
with the collaboration of
Jean-Marie Robine, François Richard Herrmann, Xavier Rodó
Our Common Future Under Climate Change
Health and Climate Change: the Need for a Diversity of Approaches
UPMC Jussieu, Paris, July 8th 2015
The research leading to these results has been funded by
the Marie Curie Actions of the European Commission’s
Seventh Framework Research Programme.
Sensitivity to cold temperatures
Large differences in the slopes of the relationship
Higher sensitivity in the Mediterranean and the British Isles
Lower sensitivity in central, eastern and northern Europe
Ballester et al. (2011) Nature Communications
Healy (2003)
Year-to-year relation in the UK Staddon et al. (2014) Nature Climate Change
Declining year-to-year relation between
– Number of winter days
– Excess winter deaths (EWDs)
Conclusion: climate warming will not decrease winter mortality
Vivid scientific debate Hajat and Kovats (2014) and Staddon et al. (2014)
Comment by
Hajat and Kovats (2014)
Nature Climate Change
Reply by
Staddon et al. (2014)
Nature Climate Change
Adaptation to cold winter days
Regression coefficients between: – daily winter temperature
– daily winter mortality
Countries with mild winters are more sensitive to cold temperatures
Mild climate
More sensitive
Cold climate
Less sensitive
Coherent with Healy (2003)
Adaptation to winter mean temperatures
Year-to-year correlations between: – winter mean temperature
– winter mean mortality
Large values everywhere, except the UK, the Netherlands and Belgium
Coherent with Staddon et al. (2014) Coherent with Healy (2003)
Comparison at the country level
No spatial relation between regressions and correlations,
suggesting two types of acclimatization processes.
Country Year-to-year correlation
Portugal -0.97
Croatia -0.97
Italy -0.96
Austria -0.96
Slovenia -0.93
Switzerland -0.88
Spain -0.88
Czech Republic -0.88
Germany -0.85
France -0.83
Luxembourg -0.81
Poland -0.70
Denmark -0.60
United Kingdom -0.12
Netherlands 0.08
Belgium 0.38
Country Day-to-day
regression coeff.
Portugal -1.13
Spain -0.67
Italy -0.52
United Kingdom -0.38
France -0.26
Czech Republic -0.25
Austria -0.24
Croatia -0.23
Belgium -0.22
Denmark -0.21
Slovenia -0.20
Poland -0.18
Luxembourg -0.17
Germany -0.16
Switzerland -0.15
Netherlands -0.14
Spatial
correlation:
r = 0.16 (not signif.)
Low
vulnerability
Large
adaptation
High
vulnerability
Low
adaptation
Daily vs. seasonal acclimatization
Temperature PDF
T/M relationship
Interannual anomalies:
∆T: temperature
β small
β large
Daily vs. seasonal acclimatization
β small
β large α ≈ 1
α ≈ 0
Temperature PDF
T/M relationship
Interannual anomalies:
∆T: temperature
∆R: response
Long-term adaptation to mean temperatures
The spatial variability of the comfort temperatures suggests
an (incomplete) long-term adaptation to environmental temperatures
spatial correlation: 0.69
spatial regression: α’ = 0.5
∆R’ = α’ · ∆T’ + ε
M
T
T’ R’
Implications: projections of mortality
Ballester et al. (2011) Nature Communications
Very few studies have considered the “seasonal acclimatization”
Mortality projections are highly sensitive to the adaptation coefficient
x106
Take-home messages
Results confirm (and generalize) the seasonal acclimatization to
winter temperatures from Staddon et al. (2014).
Potential for inter-country comparison of adaptation policies:
– daily: northeastern vs. southern Europe,
– seasonal: UK+Benelux vs. others.
Mortality projections should include adaptation coefficients.
The contribution of temperature rise to the decline of winter
mortality has been overestimated in some countries.
Thanks!
Poster session:
“Integrating climate information into decision support tools for public health”
UPMC Jussieu – Posters (Block 24)
––––– ––––– ––––– ––––– –––––
The research leading to these results has been funded by
the Marie Curie Actions of the European Commission’s
Seventh Framework Research Programme.