Earth & Sky

More extreme weather expected as climate warms

Tim Palmer talks about predicting floods, droughts, even disease.

Tim Palmer is Head of the Probability and Seasonal Forecasting Division at the European Centre for Medium–Range Weather Forecasts, based in the United Kingdom. Palmer and his colleagues have developed a system that relies on climate forecasts to predict malaria outbreaks in Africa several months in advance.

In March 2006, Dr. Palmer spoke with Earth & Sky’s Marc Airhart on the role of weather prediction in our changing climate.

Airhart: Climate scientists agree that Earth’s climate is changing. What should we be doing to adapt?

Palmer: As we go into a new climate, climate extremes are going to become more commonplace. Flooding is going to become more commonplace, heat waves are certainly going to become more commonplace, droughts for other regions are going to become more commonplace.

In order for us to be able to at least try to adapt to this new environment, it will be even more important to know months ahead of time whether this particular year, for example, the rains will fail, or whether there will be excessive flooding, or whether this will be a period of an intense heat wave. So you have to remember on top of the gradual process of climate change, we’re going to get these extremes. And we can use the same climate models.

Airhart: The same climate models that are used to predict, say, our daily weather?

Palmer: Well, we use climate models to make predictions for the end of this century, 100 years ahead. We do that by specifying how we think CO2 is going to increase over the coming decades. So that’s a long–term prediction.

In addition to that, the same models can be used to make much shorter range forecasts – maybe over a few months into the future. So this is a little bit like a weather forecast but maybe going out a little bit further than a weather forecast, going out for a few months ahead.

As we get into the future it’ll become much more important to make much shorter range forecasts for, let’s say, a season ahead as to whether in that particular year or for that particular period of time in the future, we will expect to experience a climatic extreme, either an intense heat wave or an intense flooding event or an intense drought or whatever, because then people can adjust for it.

Airhart: Can you give me some examples of how we could use those kinds of predictions?

Palmer: Farmers, for example, can plan to grow drought–resistant crops rather than their normal crops. Health authorities can prepare for a lot of heat stress amongst the elderly population. In areas of flooding, authorities can make sure that fresh water is available.

In terms of health, certain diseases might become more prevalent under certain climatic conditions. Some parts of Africa have epidemics of malaria which are known to be brought on by heavy rains. So if we know about a particular occurrence a few months ahead of time, then the authorities can start to give out bed nets, or start spraying insecticides in relevant parts.

Airhart: You have just recently come out with a paper in the journal Nature on a new way to forecast malaria using climate models. Can you tell me about that work?

Palmer: The malaria work is a specific example of how climate models can be used as an aid to society, an increasingly important aid as we go into this new era of climate change when extremes of climate are going become more prevalent and more commonplace.

There is a group of us here in Europe who have been doing work collaboratively with scientists at the Earth Institute, looking at climate models and their ability to forecast malaria incidence. The country we chose to look at was Botswana, in southern Africa.

We chose Botswana for two reasons. Firstly, it’s a region affected by what’s called epidemic malaria rather than endemic malaria. There are parts of Africa in which malaria is endemic – it occurs year on year.

However there are other parts in the more semi–arid areas of Africa where malaria is epidemic. It may not occur for many years, and then suddenly there’s a major outbreak. And these major outbreaks are known to be caused by climatic conditions, in particular, when the seasonal rainfall during the rainy season is well above average. These tend to be the years with outbreaks of epidemic malaria. Basically, the parasite and mosquito thrive under such conditions. And the community doesn’t have the immunity that it would have perhaps in the endemic regions.

Secondly, Botswana is a good country because it’s actually kept quantitative records of malaria incidence over the last 20 years.

So, what we’ve done is developed a prediction system based on climate models, and linked these climate models to malaria models. The malaria model takes the climatic information as input and as output, it predicts malaria incidence.

We tested this system on the epidemiological data from Botswana for the last 20 years and found that our prediction system is remarkably skillful in predicting malaria incidence well before the onset of the rainy season – so that’s several months, maybe four or five months, in advance of the actual malaria epidemic outbreak.

What we’re doing is buying time for the aid agencies and health workers in the area to target resources to help alleviate this effect ahead of time. They can provide bed nets, they can provide insecticides, spraying stagnant waters, and spray people’s houses.

Airhart: Dr. Palmer, thank you for speaking with me today.