Climate Change natural unnatural Past and Future Risks

Climate Change (natural & ‘unnatural’) … Past and Future Risks to Health A. J. Mc. Michael National Centre for Epidemiology and Population Health The Australian National University Canberra, Australia

CONVENTIONAL FRAME FOR ‘SOCIAL DETERMINANTS OF HEALTH’ Social norms, ideology, knowledge; socioeconomic equity History Marketing, social norms Production and processing systems: Human culture Built environment (cities, suburbs, roads, etc. Managed biophysical environment Mis-managed; excess demands Environmentally Sustainable Physico-chemical contaminants Nonsustainable Overload of natural environmental systems Individual health risks e. g. air pollution, contaminated food & water, agro-industrial effluent, endemic infections, social contact patterns, food, ‘substance’ items, cars, appliances, clothes Energy inputs Natural biophysical environment Human behaviours Sexual activity Food choices Alcohol & smoking Physical activity Road-use e. g. less (and nutrientpoor) food; water shortage; altered infectious disease patterns, morale & mental health disorders Social conditions and relations (Remediable) local environmental pollution Global environmental change/disruption (climate change, ozone depletion, soil loss, water depletion, biodiversity loss, etc. ) Communitylevel health risks Population health risks

Australian Govt’s Climate Commission Report on “Climate Change and Health” Public launch in Sydney, Nov 28, 2012. First of the climate change impact reports, to be prepared by the Commission.

Variations in Average Northern Hemisphere Temperature, o. C Rel. to 1960 -80; Multiple palaeo-climatic sources, averaged +4 Modelled temperature projection to 2100 – plus 3 -4 o. C (? ) Temperature variation around the smoothed graph, at half-decade scale, is approx. plus/minus 0. 5 o. C +3 Average Temp, o. C (rel. to 1960 -80) +2 Holocene Climate +1 Holocene Climatic Optimum, I & II +0. 5 0 -0. 5 Cooling event Sahara dries out -1 -2 Post-glacial warming, (after transient rapid cooling due to Younger Dryas) Accelerated warming since 1975 Roman Warm Period Mediaeval Warm Period, Europe Drought in Eastern Mediterranean region 17 th Cent Crisis, Europe Little Ice Age in Europe (similar in China) Tambora eruption 1815 536 CE Event (acute cooling) 2 -3 o. C fall -3 -411 10 9 8 7 6 5 4 3 2 1 0 2 4 6 8 10 12 14 16 18 20 22 Millennia BCE 2000 yrs ago Centuries CE

“A Safe Operating Space for Humanity” Identifying & quantifying planetary boundaries that should not be transgressed Rockstrom et al. Nature 2009 Green = estimated safe operating space Climate change Biodiversity loss Nitrogen cycle

Global Annual Temperature, 1750 -2010: The Rising Atmospheric CO 2 Concentration, plus Sporadic Volcanic Cooling, Provides the Best Modelled Fit of the Actual Observed Temperature Trend Expected temp trend – modelled – due to increasing Actual annual CO 2 concentration, temperatures punctuated by sporadic volcanic cooling Post-1950 s rise, mostly attributable to human actions (IPCC) +1. 5 o. C 1750 -2010 Systematic thermometer use begins BEST Project, UC Berkeley, 2012

Regional Temperature Change over Past ~20 Years +4 o. C +3 o. C +1. 5 o. C +2. 5 o. C -0. 5 o. C +1 o. C +2 o. C +1. 5 o. C -1. 5 o. C +1 o. C La Niña (ENSO): westward flow of cold surface water: warm water heaps up in western Pacific -1 o. C +1 o. C -1. 5 o. C Difference from 1980 -2010 average temperature -4 o. C 0 +4 o. C Adapted from: US Nat Oceanic & Atmospheric Admin (NOAA), State of Climate, 2011

Arctic Sea Ice Extent (millions km 2), Sept 17, 20 Median ice limit 1979 -2000 Nth Pole via na di an Sc d lan n ee Gr da na Ca UK ± 2 standard deviations Atlantic Ocean 1979 -2000 average 2007 2012 Source: National Snow and Ice Data Center, Boulder, Colorado

Annual Temperature Fluctuations in Western (alpine) Europe and Scandinavia, 450 -650 CE: Evidence of abrupt cooling in 536 -545 CE 2 0 -2 460 480 520 540 560 580 600 620 640 NOTE: temperature data are from seven different regional tree-ring datasets. The data are de-trended over time, and the mean is set at zero. Temperature graph from: Larsen LB et al. (2008). Geophys. Res. Lett. , 35, L 04708. doi: 10. 1029/2007 GL 032450.

Aksum’s main trade routes, land sea during 300700 CE. ( red lines/arrows)

Europe’s coldest period, 1570 -1660, during Little Ice Age Food yields down, grain prices tripled … conflict , war, displacement Cold Period 1570 -1660 Nth Hemisphere Temperature Variation o. C 1644 European Temperature Variation standardised units Famine-yrs/decade x 2 Epidemic rate x 3 Adult height ↓ 1. 5 cm Rate of Migrations, Europe War Fatality Index, Europe 1500 1550 1600 1650 1700 1750 1800 Zhang et al. , PNAS, 2011

Climate Change and Human Health: Types of Health Risk Assessment (past, present, future) Temperature Scenario-based modelling: Statistical study: Estimating the range of future Estimating the plausible risks attributable burden of disease +2. 0 o. C +1. 5 o. C Empirical Studies: Risk per o. C Empirical Observations: Attributable to climate change? 1950 Natural temperature variation 2050 Now Modelled future temperature(s)

Other global/systemic environmental changes – coexisting with climate change Glacier Altered loss, sea- surface level rise water Reduced food yields Climate Change ~ temperature Mean climatic conditions and Variability and rainfall Ecosystem Property damage loss heatwaves, extreme weather events Infra. Tourism structure and damage recreation Physical Direct Loss of jobs, River flows, hazards livelihoods economic impacts Direct impacts of Altered microbial heat: physiological, ecology Fresh. Food organ damage, (host-animals, behavioural water prices, vectors, pathogen availability choices multiplication) Community & Risk of injury Household hygiene family morale: and death Nutrition: Local food yield mental health Infectious child devt; problems Post-event disease risks depression, etc. adult health shortages, prices: competition People Conflicts, Relocation displacement

Deaths Attributable to Climate Change: Year 2000 Estimated annual deaths due to climate change from: malnutrition (~80 K), diarrhoea (~50 K), malaria (~20 K), flooding (~3 K) 14 WHO statistical regions scaled by estimated annual mortality (in 2000) due to change in climate since c. 1970. Selected major causes of death. (Patz, Gibbs et al, 2007: based on Mc. Michael, Campbell-Lendrum, et al, 2004)

Climate Change Influences on Health in Australia Already apparent: preexisting risks amplified by climate change Uptrend in av annual no. of heat-days deaths, hospitalisations Increase in no. /severity bushfires injury/death, resp. hazard, mental hlth Probable ongoing health impacts: but not yet clearly identified Rise in some food-borne diarrhoeal diseases Altered air quality: ozone formation, aeroallergens Mental health impacts, esp. in some (drying) rural regions: e. g. MD Basin Predicted future health impacts Extreme weather events: trauma, infectious disease, depression Water shortages: affecting food yields, domestic hygiene Shifts in mosquito-borne infection patterns: Dengue, Ross River virus (& chikungunya? ), Barmah Forest virus, Japanese encephalitis, etc. Thermal stress in outdoor workers: accidents/injuries, organ damage

Typical Relationship of Daily Temperature to Daily Death Rate in Temperate-Zone Countries High Daily Death Rate Comfort zone Low Source: Mc. Michael, 2012 Relative Risk of (daily) Death Source: Huang, Barnett et al, 2012 Daily Temperature (o. C) Brisbane Melbourne 24 o. C 22 o. C Daily Temperature (o. C)

Evidence of a Warming Australia Bureau of Meteorology: '2009 will be remembered for extreme bushfires, dust-storms, lingering rainfall deficiencies, areas of flooding and record-breaking heatwaves. ' Australia’s 2009 -2010 mean temperature was ~1°C above the 1961 -90 average. 2010 Decadal mean temperature 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010

Heatwaves, Illness Events, Mortality Melbourne, Australia, 2009 Temperatures 27 -31 January 2009: max temperatures 12 -15°C above summer norm. 28 -30 Jan: > 43°C Number of ambulance call-outs Ambulance call-outs for heat-related illnesses in Metropolitan Melbourne, 19 Jan - 1 Feb, 2009 50 250 45 40 200 Maximum Temperature 25 Heat Stress 100 20 0 Jan: 19 Temp o. C 15 Dehydration 10 Heat Stroke Call-outs, deaths 126 out-of-hospital deaths (vs. 44 expected deaths) 30 150 50 January 29 -30: Great increase in ambulance call-outs 35 5 0 20 21 22 23 24 25 26 27 28 29 30 31 1 Feb Date SOURCE: January 2009 Heatwave in Victoria: an Assessment of Health Impacts. State of Victoria 2009

Relationship of Salmonella gastro-enteritis occurrence to daily temperature, Sydney, 1990 -2010 Relative Risk As maximum daily temperature increases, the risk increase with min. daily temperature steepens [26 o. C] [22 o. C] [19 o. C] Minimum Daily Temperature o. C Source: Ainslie Butler, NCEPH/ANU, unpub

Dengue Fever: Estimated ‘receptive’ region for Ae. aegypti mosquito vector, under alternative climate-change scenarios for 2050 Darwin . . Katherine . Darwin . . Broome Port Hedland . . Katherine . Townsville Port Hedland . . Current risk region for dengue transmission . Mackay Carnarvon Townsville Rockhampton Darwin . Brisbane . . Katherine . Broome . Port Hedland . . . Rockhampton Cairns Townsville . . Mackay Risk region for high emissions scenario, 2050 Carnarvon NCEPH/CSIRO/Bo. M/Univ. Otago, 2003 . Risk region for medium emissions scenario, 2050 . Cairns Broome Rockhampton

Climate Change and Malaria Potential transmission in Zimbabwe Baseline 2000 2025 2050 Climate suitability: * red = high; blue/green = low Harare Low probability Highlands Medium probability High probability Bulawayo * Temperature + minimum seasonal rainfall Ebi et al. , 2005

Climate Change and Malaria Potential transmission in Zimbabwe Baseline 2000 2025 2050 Climate suitability: red = high; blue/green = low Harare Bulawayo Ebi et al. , 2005

CLIMATE CHANGE: Poor Countries Projected to Fare Worst MODELLED CHANGES IN CEREAL GRAIN YIELDS, TO 2050 20 Plus climate-related: • Flood/storm/fire damage • Droughts – range, severity • Pests (climate-sensitive) • Infectious diseases (ditto) 36 80 64 Percentage change in yields to 2050 -20 0 +20 +50 +100 UN Devt Program, 2009

Now Skipjack tuna, Pacific Ocean 2050 A 2 1, 727, 000 m tonnes/yr 2100 432, 000 mt Loukos H, Monfray P, Bopp L, Lehodey P. (2003) Potential changes in skipjack tuna habitat from a global warming scenario: modeling approach and preliminary results. Fisheries Oceanography, 12(4): 474 -482 2100 A 2 Source: P. Lehodey

Adaptation and Mitigation Adaptation: Risk management Time-frame Options: - Immediate protection (heatwave warnings, dams/sea-walls) - Long-term protection (urban design/greening, resilient rural communities, vaccines) Priority/equity (who is vulnerable? ) Evaluate (cost-effectiveness? ) Mitigation … and its Health Co-Benefits to local population - Cleaner urban air: ↓cardio-resp disease - More walking/cycling: fitness, social contact - Healthier climate-proof insulated housing - Diet: local fresh foods less (ruminant*) red meat ↓obesity, cardiovasc disease, some cancers * cattle, sheep, goats, camels, buffaloes CH 4 Brrrp CH 4

Three Key Messages 1. Human-induced ‘climate change’ is real, increasing … and quite distinct from ‘local envtl pollutants’ in the type, range and significance of health impacts 2. Health professionals will, increasingly, encounter changing rates of differential diagnoses, emergency hopital admissions, enquiries from concerned public 3. Health sector will need to adapt (continuously) by: -- Understanding climate-related exposures and risks -- Curbing the sector’s greenhouse emissions -- Upgrading/changing community surveillance practices -- Working more closely with other sectors -- Assisting with public education, and political advocacy

That’s All

40 Total Arctic Sea-Ice Volume (1000 km 2) “… between 5 -30% of the decrease in sea ice is due to the Atlantic Multi-decadal Oscillation …[which] implies that 70 -95% of the changes are caused by human-induced global change. " JJ Day, Univ. of Reading, July 2012 30 1979 -2011 average 20 (av. 1979 -2011) 10 (2012) 0 Adapted from: Polar Science Center (Univ. Washington), 2012 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Estimated age-standardised incidence rate per 100, 000 Melanoma of skin, both sexes, all ages ≤ 0. 4 30 o. N ≤ 40 < 0. 4 < 0. 8 < 1. 6 < 3. 9 30 o. S < 36. 7 Source: IARC, 2008. Globocan http: //globocan. iarc. fr, accessed on 20/08/2012