Variable Variability: Malcolm Turnbull, how should Australia adapt to climate change without science?

Adapting to climate change needs information on local changes in the mean, weather variability and extremes. Observed changes in the means are not enough.

If you don't like what #climate science is telling you, just fire all the climate scientists

Miles Grant

The "conservative" government of Australia plans to gut its climate research and kill the groups doing climate research at Australia's main research institute, CSIRO. Australia's opposition leader rightly said the Prime Minister Malcolm Turnbull should "hang his head in shame".

The destruction is not for lack of quality of the research. CSIRO's new chief Larry Marshall send an email to its employees stating:

"CSIRO pioneered climate research ... Our climate models are among the best in the world and our measurements honed those models to prove global climate change.

From this the strange conclusion is drawn:

That question has been answered, and the new question is what do we do about it, and how can we find solutions for the climate we will be living with?"

[UPDATE. Judith Curry agrees with this strange sentiment: "Now that the UN’s community of nations has accepted consensus climate science to drive international energy and carbon policy, what is the point of heavy government funding of climate research, particularly global climate modelling?"]

Just because we know climate change is real, does not mean that we understand everything. Projecting increases in the global mean temperature is easy. Saying something about the changes in the hydrological cycle is much harder. We know how much the global mean precipitation will increase because we can estimate the additional evaporation and what goes up must go down, but say where and how it goes down is hard. These assessments naturally have their uncertainties and it certainly pays to reduce them to make better political decisions.

Much more important than the uncertainties in the changes in the global means, for "solutions for the climate we will be living with" (adapting to climate change) we will need local predictions. That is a lot harder and very uncertain. Locally the changes can be very different from the global change. As Roger Pielke Jr. writes about storms on the US East Coast: "So those who argue for a simple relationship between increasing water content of the atmosphere and storm strength, data do not support such a claim over this multi-decadal period, in this region." (my emphasis)

open flames and smoke in a rural Texas landscape

open flames and smoke in a rural Texas landscape

Much more important than the uncertainties in the changes in the global means, for adaptation we need information on changes in weather variability and extremes. Especially for a country like Australia that knows very large variations due, for example, to El Nino.

One of the strategies of the mitigation skeptics is to pretend that adaptation is straightforward and cheap. When the sea level goes up 1 mm, just make the dikes 1 mm higher. However, the sea dikes will break during spring tide and a strong storm. Thus we also need to understand the storms to know how much stronger the dikes need to be. They will break during a once in a century storm. Or at least during what used to be a once in a century storm. Try to estimate from observations during a changing climate whether the 100-year storms are getting worse.

"With climate change, we can’t drive by looking in the rear view mirror. We’re in a new normal."
Climate scientists Berrien Moore and Katharine Hayhoe

Was the flooding of New Orleans due to [[Hurricane Katrina]] a unique event or the "new normal"? During the flooding last year in South Carolina in some locations the rain amounted to a 1,000-year event (in a given year there is a 1 in 1,000 chance of observing rainfall of that magnitude of more). Does South Carolina have to adapt to this because this will happen more often or will this remain an outlier? Parts of the United Kingdom were hit three times by 100 year rain events the last few years. How often will they have to suffer this before we know from waiting and seeing that the weather has changed, people will have to move and the infrastructure needs to be more more robust?

"There's no point putting in flood defenses that respond to mean climate change if you haven't thought of what a one-in-a-hundred-year event will look like in a warmer world... They don't want to know what the climate will be like, they want to know what the weather will be like in 20, 30, 50 years time."

Chief Scientist UK MetOffice Dame Julia Slingo on adaptation

The same goes less visibly for droughts. When your farm takes a hit due to a drought, do you build it up again when the rain comes back or is your land no longer profitable. Do you want to do this blindly? Or do you prefer some scientific guidance? For planning crops and managing reservoirs during droughts, seasonal and decadal climate predictions reduce costs and hardship. For planning new reservoirs and desalination plants long-term climate projects give guidance.

Meteorologists and climatologists are building seamless prediction systems. Going from short term weather predictions and nowcasting using observations during severe weather, to long-term weather predictions to prepare for bad weather, to seasonal and decadal predictions for planning and climate projections for adaptation. In many wealthy countries governments are setting up national climate service centers to help their societies adapt. The World Meteorological Organization is building a Global Framework for Climate Services (GFCS) to coordinate such efforts and help poorer countries understand the changes their region will see. While Australia sticks its head in the sand.

We will need very good science, a very good understanding of the coming climatic changes to adapt. The Australian government destroys climatology at a moment people, communities and companies need it most to adapt to the climatic changes that we have set in motion. This is about as stupid as the US states where the civil servants are no longer allowed to talk about climate change, which will mean that these communities will suffer the consequences without being prepared for the changes.

The same is true for (nearly?) every impact of climate change. In the past we could use long-term observations to determine what kind of extremes we could expect. Now, after all the delays to solve the problem, humanity is becoming more and more dependent on climate science and climate models, the models the mitigation skeptics who campaign for more global warming claim not to trust.

If you do not know which climatic changes you need to adapt to, you need to adapt to everything. Preparing for the worst case scenario in every direction is very expensive.

Never attribute to maladaptation that which can be adequately explained by stupidity.

When Australia notices what a blunder they are making it will easily take over a decade until Australia's climate research is again where it started. It takes years until you understand a climate model or a data set well and start to be productive. Science is a social profession and once you are proficient you can start building your network. Then you notice the kinds of expertise still missing in your freshly build up institute. Unfortunately, like trust losing scientific expertise goes much faster than building it up.

Variable Variability

Pages

Friday, 5 February 2016

Malcolm Turnbull, how should Australia adapt to climate change without science?

Related reading

2 comments: