Ebola and Climate-Change Models

Ebola and Climate-Change Models

That title should upset somebody. Perhaps get a few hits for those looking to blame the Ebola outbreaks on climate change. I’m not doing that here. I do have some history of using Ebola to talk about climate change.

What got me thinking about Ebola again was a report on News Hour called Using the power of prediction to halt Ebola in its tracks. I will get to Ebola and Climate-Change Models after a few words for the sake of background. I did a modeling tutorial back in 2012. Long ago, I wrote an entry where I described types of models and a shorter series on models and use of models. In another entry, I wrote about uncertainty and the use of models. Last year, I did an entry on using models in El Niño prediction. Finally, with luck in the next couple of weeks, Andrew Gettelman and I will be completing a book on demystifying climate models.

Models are Everywhere in our modern world and used for all sorts of planning and decision making. Recently, I have been working with scenario modeling, which is used, for example, to conceive possible futures and contingency planning. This is the sort of planning that I like to think airline pilots do whenever they are not actually flying. One function of models is to provide plausible representations of events to come, and then to place people into those plausible futures. It is a way to anticipate and manage complexity. Though most times these models do not give an exact story of the future, the planning and decision making that comes from these modeling exercises improves our ability to anticipate the unexpected and to manage risk.

In Using the power of prediction to halt Ebola in its tracks, one of the first exercises was to determine how bad Ebola would get without intervention. This alerted and engaged nations, and helped to mobilize attention to the problem. More detailed uses of the models were to identify places of particular vulnerability and to identify the impacts of different interventions. The models helped make it clear that the only effective intervention was to change human behavior and to disrupt the transmission of the disease. They identified tipping points – goals that, if achieved, they could gain the upper hand on the disease. (Webb and others, 2015, describe a particular Ebola model)

There are two points I want to make here. First, there is a certain element of common sense in the Ebola model that, perhaps, you don’t need a model to tell you. Namely, if you don’t do anything, it might get very bad. Climate models share this element of the obvious – if you keep pumping the atmosphere full of stuff that holds energy close to the surface of the Earth, the Earth will warm up. Second, when you look at the details of Ebola models, infectious disease models in general, there are significant errors and uncertainties (King and others, 2015). The same is true of climate models. People choosing to identify and amplify these errors and uncertainties is a tried and true technique of creating doubt and disrupting societal responses to climate change.

To be sure, there are those in the epidemiological community who do not embrace a model-based approach to response and to management of outbreaks and epidemics. Aside from those experts who are skeptical, there is also the fact that, in the African Ebola outbreak, there are people “who believe in witchcraft and home remedies. Some are convinced Ebola is a plot by white people, if it exists at all.” That is, as with climate change, there exists skepticism, some anchored in science-based arguments, some not.

It is in the messy middle, between the two extremes of the models representing the obvious and the inability of the models to represent the entirety of the world’s complexity that the models find their usefulness. Not only do they allow the practitioner and planner to think about complexity, through the physical and statistical principles and rules expressed in the models, they constrain the range of behaviors to one that is, nominally, plausible. Models, sometimes, reveal modes of behavior that are previously unobserved, perhaps low probability events with high risk. The models, also, suffer from the fundamental limitation of their simplicity, which might preclude them from representing important behavior realizable in nature, but not in the model.

In a 2015 review of infectious disease modeling, Heesterbeek and others state that the behavior of “pathogens play out on a wide range of interconnected temporal, organizational, and spatial scales that span hours to months, cells to ecosystems, and local to global spread.” They came to the conclusion that, “Faced with this complexity, mathematical models offer valuable tools for understanding epidemiological patterns and for developing and evaluating evidence for decision-making in global health.” With very little change in words, the same can be said for climate models, and models used in virtually all practice of modern science.

I make my living thinking about that messy middle and how to use models in planning and management. The common-sense part of climate modeling and societal behavior assure us that the changes that are being observed are part of a trend of changing climate. The observed changes and the predicted changes are aligned well enough, their cause and effect understood adequately, to let us know that the projections of the future are plausible. The common-sense part of the models, also, lets us know that if we do not work to limit our greenhouse gas emissions, then we can anticipate weather and impacts far outside of any previous experience. Therefore, planning, expenditure, and preparation are not only justified, but prudent.

What the Ebola modeling example suggests is that the motivation and practice of modeling of complex systems is shared across many fields of scientific investigation. If a politician or organization takes on climate science as in someway deficient, it is difficult, then, not to reject all modern science. That would be a philosophical belief, rather than a knowledge-based conclusion. It is a belief that, if pursued as public policy, increases societal risk and reduces societal competiveness.

r



Figure 1: “Modeling for public health. Policy questions define the model’s purpose. Initial model design is based on current scientific understanding and the available relevant data. Model validation and fit to disease data may require further adaptation; sensitivity and uncertainty analysis can point to requirements for collection of additional specific data. Cycles of model testing and analysis thus lead to policy advice and improved scientific understanding.” Heesterbeek and others, 2015. The same description can be applied to climate modeling.

Series links:

Models, Water, and Temperature

Models are Not All Wet: Series Introduction

Models are Everywhere

Ledgers, Graphics, and Carvings

Balancing the Budget

Point of View

Looking Under the Cloak of Complexity

The Free Market and the Climate Model