Looking at the Impacts of Climate Change: Uncertainty, Precaution and Regulation Tim Swanson

Science of Climate Change

Science of Climate Change

Radiation Forcing – Past Present Future

Science of Climate Change

Carbon stocks and commitment to warming 400 ppm CO2e 450 ppm CO2e 550 ppm CO2e 650ppm CO2e 750ppm CO2e 5% 95% Eventual temperature change (relative to pre-industrial) 0°C A summary of recent evidence of the commitment to warming for different stabilisation levels. This new evidence allows us to apply probabilities to projections of future climate change Ends of range – lowest probability; near centre of range – highest probability The red bars indicate range (90% confidence) based on two studies – IPCC Third Assessment Report (2001) and Hadley Centre (2004). IPCC because internationally agreed. Hadley because more recent, more sophisticated on probabilities and central of recent studies. The grey bars cover the range of eleven recent studies. Detail and sources: The lower bound is the 5th percentile from a study based on the IPCC Third Assessment Report (Wigley and Raper 2001). The upper bound is the 95th percentile is from a recent study by the Hadley Centre that explores the uncertainty in climate models (Murphy et al. 2004). The grey lines indicate the results over 11 recent studies (from Meinshausen (2006)) – 5% point from lowest, and 95% from highest. [newer studies are still emerging and give results within this range] Now at 430ppm and rising 2.5ppm per year [rate varies significantly year to year, 2004 it was 2.2ppm/yr, average 2.3ppm per year 1980-2004]

The impacts of climate change Global temperature change (relative to pre-industrial) 0°C 1°C 2°C 3°C 4°C 5°C Food Falling crop yields in many areas, particularly developing regions Possible rising yields in some high latitude regions Falling yields in many developed regions Water Significant decreases in water availability in many areas, including Mediterranean and Southern Africa Small mountain glaciers disappear – water supplies threatened in several areas Sea level rise threatens major cities Ecosystems Extensive Damage to Coral Reefs Rising number of species face extinction The projected impacts of climate change over multiple dimensions, aggregated across several studies. The arrows indicate the approximate temperature at which we might start to see these impacts – where the darkening shade shows the increasing risk and intensity of impacts as temperatures rise. Already committed to 0.5 – 1degC of temperature change above pre-industrial. BAU gives serious risks of exceeding 5degC. Developing countries are especially vulnerable – increased water stress, falling farm incomes, malnutrition and disease increase pressure for migration and conflict. Developed countries not immune – water stress (Europe California), extreme weather events leading to increased insurance costs. Specific Examples: Food: (1) from 1degC expect severe impacts in marginal Sahel, (2) 2 – 3degC may see 25 – 60% more people at risk from hunger, with half the increase in Africa and W. Asia (assuming weak carbon fertilisation). (3) 4 – 5degC, crop yields decline by up to a third in Africa Water: (1) 2 – 3degC, more than 30% degree in water flows in Mediterranean and Southern Africa. (2) 3degC, more than one billion people suffer water shortages, many in Africa (year:2080s). Ecosystems: (1) 3degC, 20 – 50% of species face extinction (2) 1 – 2degC, irreversible damage or even loss of large fraction of coral reefs. Extreme Weather Events: (1) 3degC, small increases in hurricane intensity lead to doubling of damage costs in the US. (2) 3degC serious droughts in Europe occur once every 10 years. Abrupt and Major CC: (1) 2 – 3degC onset of irreversible melting of the Greenland ice sheet. (2) 3 – 5degC dangerous ((W. Atntarctic ice sheet and ocean currents) Extreme Weather Events Rising intensity of storms, forest fires, droughts, flooding and heat waves Risk of Abrupt and Major Irreversible Changes Increasing risk of dangerous feedbacks and abrupt, large-scale shifts in the climate system

Regional Variation in Impacts

Uncertainty in Climate Change Impacts

Ways to Think about Uncertainty “Mean” Risk replace distribution of outcomes with a single point 2) “Margin for Error” Risk Management look at some band around the mean 3) “Worst Case” Risk follow the bottom of the distribution 4) “Waiting” for Information postpone regulation until information arrives

Using the mean to estimate risk

Comparing mean and “worst case”

“Waiting” to discover the cost of impacts – the cost of regulation vs the cost of non-regulation

Dealing with Risk 3 Types of “Uncertainty” Known determinate risk (e.g. 3% chance of light rain – event is understood and probability is understood – impacts are bounded and determinate)  predicted outcome is a single point

Dealing with risk 2) Known risk with known distribution of potential outcomes (e.g. global warming example given previously) “Risk” derives from: “spread” or variance of distribution “fatness” in tails Likelihood of impact Amount of damage from impact

Dealing with risk – third type of uncertainty (unknown distribution) Distribution A – feedback effect is positive Distribution B – feedback effect is negative

Precautionary Principles – weak versions Uncertainty should not preclude regulation (need not wait until all uncertainties are resolved) 2) Regulation should consider the range of possible outcomes not just the mean – must incorporate a margin of safety to account for the range of possibilities

Weak PP Range of outcomes should not disallow regulation Regulation must take into account the spread of risk as well as the mean risk How? Margin for error? (confidence intervals) Fatness of Tails? (length of tail) Role of Learning? (value of time)

Precautionary Principles – strong versions If uncertainty includes a significant possibility of large negative impacts, then regulation must impose a BAT rather than a balanced approach (unless the proponent shows “no harm” from the activity) 4) If uncertainty includes a significant possibility of large negative impacts, then regulation must prohibit the activity (unless the proponent shows “no harm” from the activity)

Strong PP Significant Uncertainty (of large negative impact) Requires Regulatory Action Significant Uncertainty (of large negative impact) Requires Shift of Burden of Proof Why? Burden of proof (why only shift here?) Significant Uncertainty (importance of learning?)

Regulation of Risk PP about: Ability to bound the range of outcomes Size of tails (poss of hugely neg outcomes) Shifting of burden of proof Importance of regulation and learning