Do not get me wrong - I am a great supporter of global climate models. Twenty years spent researching statistical downscaling methods would not have been possible without them. However, louder calls for even greater investment in high-resolution climate modelling - justified partly by more cost-effective adaptation strategies - require pause for thought.
The idea of scenario-driven planning is not new. It emerged in the early 1970s and is widely attributed to Royal Dutch Shell. In 1977 the US Academy of Sciences published a panel study on Climate, Climatic Change and Water Supply. The earliest climate sensitivity studies for water resource systems began to appear shortly afterwards.
Some of the first work to feed climate model scenarios directly into impact models was published in a special issue of the journal Climatic Change on “The sensitivity of natural ecosystems and agriculture to climate change”. [Readers might like to identify even earlier research of this kind.]
The “top down”, climate-impact framework is now firmly enshrined in the psyche of the climate science community. It is implicit in the order of the Working Groups of the Intergovernmental Panel on Climate Change (IPCC). It frequently defines the session lists of conferences. It is reinforced by undergraduate courses and texts: greenhouse gas emissions scenarios first, global change and regional climate modelling next, impact assessment, and (maybe) adaptation response last.
It could be argued that the use of climate models in this way represents a departure from their original purpose: to quantify the links between rising concentrations of greenhouse gases and global climate change. After 25 years of scenario-led impacts research it is hard to point to any examples where an adaptation decision has been based exclusively on climate predictions. So what could have been achieved without climate models?
We would still have sensitivity studies showing the complex, often non-linear, responses of natural and human systems to prescribed perturbations in climate. We would know that for every unit reduction in rainfall, there is typically an even greater reduction in river flow. We would surmise the poleward and upward rates of migration of species and biomes for given amounts of warming. The humble sensitivity study might be experiencing an even greater renaissance in sectors such as flood risk management.
There would be no regional climate downscaling inter-comparison studies [and at a sweep a large part of my publication list is wiped out]. Practitioners would not be bogged down with computationally demanding, pseudo probabilistic analysis, generated from ever larger ensembles of climate and impact model permutations. We would still know that the future regional climate is errr … highly uncertain …
With no climate models to drive our adaptation thinking we would have to approach risk and vulnerability from the “bottom up”. Measures that are low regret, reversible, incorporate safety margins, employ ‘soft’ solutions, are flexible, and deliver multiple co-benefits would be our starting point. Individual options would still have to be tested for social, economic, technical and political feasibility using plausible climatic and non-climatic narratives.
Early warning systems (based on remotely sensed data), land-use zoning, and contingency planning for extreme events would be just as effective.
Without the large opportunity costs of climate modelling we might choose to invest more in surface and upper-air global observing networks. Even a cursory glance at the World Meteorological Organisation’s map of silent meteorological stations reminds us that there are still many gaps to be filled. We would recognise that real-time measurements are our most important assets for adaptively managing dynamic systems. Such data would be a common good, freely shared between and within nations.
Sites with unbroken, long-term, homogeneous records would be protected. We would then be better placed to detect and attribute any changes. We would also have clearer insight to physical processes and variability operating over many decades. Agencies could benchmark adaptation interventions and conduct more field trials. Governments would be less fearful of the associated recurrent costs.
The IPCC would almost certainly not exist in its present form. The Working Groups and Fifth Assessment Reports would be re-structured around cross-cutting themes. These are better suited to Earth systems thinking and for evaluating multiple anthropogenic pressures. We would be developing a more integrated set of strategies for tackling poverty, biodiversity loss, resource exploitation, pollution, and so forth. Climate change would not be singled out.
Of course, this is just day-dreaming. Climate models are maturing into Earth system models with even greater potential for what if analyses. But this should not distract us from doing more of the above.
Discuss in our Forums
See what other readers are saying about this article!
Click here to read & post comments.
20 posts so far.