Wild weather

Volume 7 Number 3 March 14 - April 10 2011

The immense power of the Earth’s climate has long been felt in Australia. But with the recent monster floods, destructive cyclones and intense bushfires wreaking havoc right across the continent, many are wondering why such events are happening and what we should expect in the future. Sally Sherwen talks to the experts.

Cycles of droughts, floods, bushfires and cyclones are ingrained in the Australian experience and over time we have become relatively effective at predicting the likelihood of impending wild weather events.

Astonishing developments in atmospheric knowledge and computer power now allow us to forecast weather up to seven days in advance and even predict seasonal conditions months ahead. But long before such technology was developed, the Indigenous people of Australia recognised the seasonal fluctuations in climate using an entirely different method.

Dr Marie Keatley from the University of Melbourne’s School of Land and Environment explains that the Aboriginal people have an intimate knowledge of the environment and have used the timing of species life stages to classify seasons.

“The timing of natural events such as the first flowering of a plant species or the first clutch of eggs laid by a bird are useful indicators of environmental conditions because such stages in a species’ life are significantly influenced by the weather,” Dr Keatley says.

“Aboriginal knowledge of climate is fine-tuned to conditions in specific regions, and this means there are differences in the Aboriginal classification of seasons according to location, and can range from three seasons to thirteen in a year.”

The Aboriginal seasonal calendars recognise the diversity of weather in Australia across different regions, whereas the European four-season calendar adopted in Australia during colonial settlement classifies climate more generally based on the geometrical positioning of the sun.

Over time, weather prediction has developed into an invaluable and widely used tool.

Huge amounts of work go into providing the seven-day weather forecasts we see on the news every night, and researchers such as Dr Todd Lane from the School of Earth Sciences are continuously working to refine weather forecasting techniques.

According to Dr Lane, weather forecasting has undergone significant developments over time.

“Interestingly, weather forecasting used to be based solely on human knowledge,” Dr Lane says.

“Forecasters who had extensive knowledge of the atmosphere, would map and analyse current conditions and, based on their knowledge, would predict what weather conditions were coming,” he says.

“Now we use complex computer models to do the forecasts for us. But the models aren’t trusted exclusively; humans still play an important role.”

Dr Lane explains that the computer models are essentially an extensive series of equations that estimate how current conditions will change according to air flow.

“They are a mathematical way of predicting future events based on what is happening now and on physical equations of air movement,” he says.

“We have a thorough scientific understanding on the dynamics of air movement, and because short-term weather patterns are influenced by air movement around the atmosphere, we can use this to predict how conditions will change with time.

“So before we can make a weather forecast, we need to collect data on current atmospheric conditions using weather stations, weather balloons and satellite sensors. We then incorporate the current conditions into the computer model and let it run to solve the equations.”

Such computer models have been used since about the 1950s and are continually being improved, Dr Lane explains.

“This model-based forecasting is reliant on computer power as it takes a huge number of computations to solve the equations. So as computers get more powerful, we expect weather forecasting to get more advanced – we have already moved from one-day forecasts to seven-day forecasts.”

But Dr Lane stresses that the models are still far from perfect.

“We’re at the point where large weather systems like highs and lows are accurately predicted but small severe events are still very difficult to predict and rely on human knowledge.”

This is where Dr Lane’s research comes in.

“I am working to test and refine new models to make them more accurate so we can predict the smaller events that current models miss,” he says.

But the researchers aren’t stopping at seven-day forecasts. The models can be altered to predict average seasonal conditions months ahead of time.

Professor David Karoly from the School of Earth Sciences works on long-term seasonal and climate predictions.

“With seasonal predictions of climate, we’re estimating the average conditions we expect to see in the seasons to come. For example we’re looking at the probability that summer will be hotter than normal or winter colder than normal,” Professor Karoly explains.

“We look at averages over seasons because individual weather events are unpredictable any longer than seven days ahead.

“To do this, we use similar numerical models to those used for short-term weather forecasting, but we run them forward in time for a longer period and therefore different processes come into play,” he says.

“So short-term weather forecasts generally focus on air movement, but when we start looking at longer-term predictions of seasonal conditions months into the future, ocean temperature changes also have to be considered because they strongly influence the climate over time.”

There are two main phenomena that describe ocean temperature changes in the Pacific that affect Australia, El Niño and La Niña, Professor Karoly explains.

“El Niño leads to extensive warming of ocean temperatures in the Pacific, and La Niña leads to extensive cooling, and both strongly influence climatic conditions – El Niño usually brings droughts and La Niña usually brings floods.

“These weather changes due to El Niño and La Niña are not really noticed in short-term forecasts because they are slow-cycling, but they are evident in long-term averages of seasonal weather conditions, and therefore our seasonal prediction models include their effects.”

Professor Karoly says long-term predictions that tell us if we are likely to have a hotter than normal summer or a drier than normal winter, are useful for climate-dependent industries such as the agricultural industry, as predictions may help with planning for the season.

And the recent heavy rain and flooding in Queensland didn’t come as a surprise to Professor Karoly, as he said the models predicted such heavy rains associated with a strong La Niña.

Climate modelling can even go a huge step further and be used to make predictions for long-term climatic trends decades ahead of time.

“Once again, we use extensive models to predict future events based on physical processes and mathematical equations. But for climate projection modelling, we look at trends over 20 or 30 year blocks,” Professor Karoly says.

“We have moved from forecast models focusing on air movement, to seasonal models focusing on ocean temperature changes as well as air movement, and now climate projection models that incorporate all of these interactions plus other factors that influence the heat being received and retained in the atmosphere, such as orbit changes of the Earth, volcanic eruptions and trapped greenhouse gases.”

Over decades, any changes in heat being received and retained in the atmosphere will be of importance, so climate models analyse the potential impact of factors that influence this heat retention, Professor Karoly explains.

“We run the models under different scenarios and levels of heat retention to predict what future climate would be like under different management plans,” he says.

“For example, we can run the models for a situation where all human-related greenhouse gas emissions stop completely or for a situation where greenhouse gas emissions continue to increase, and we can see what the climate would be like under each scenario decades ahead.”

The models are validated by running them under past climatic conditions and comparing the predicted results with the actual observed results, Professor Karoly said.

So what do the models predict?

“Climate models predict that the climate will warm if greenhouse gas levels keep rising as they are now. The models project a long-term warming of global air temperature by around 3°C if the concentration of carbon dioxide in the atmosphere doubles,” Professor Karoly says.

“Changes of this magnitude can have major long-term ramifications for the environment and for human livelihood.

“Climate change and warmer sea surface temperatures will mean the effects of El Niño and La Niña will be exacerbated, leading to more extreme weather events. Wet places will get wetter, and dry places will get drier.

“Individual events like bushfires, floods and cyclones cannot be attributed to climate change but the recent extremes of such events are in line with predictions.”

However these large-scale dramatic impacts may not even be our biggest problem, an impact of a changing climate that is less well-recognised, at least in non-Indigenous cultures, is the timing of natural events, as discussed earlier.

Dr Keatley explains that because temperature and rainfall influence life cycle stages, any change in climate will have effects on timing of such natural events, which in turn can have significant consequences for human health, biodiversity and agriculture.

“Research has shown that there are changes in the life cycle stages of certain Australian plants and animals,” she says.

“For example, Dr Michael Kearney from the Zoology department here at Melbourne University discovered that Melbourne’s Common Brown Butterflies are emerging at least 10 days earlier in spring than they did in 1945.”

“Such changes impact entire ecosystems because species interact with each other in a complex system, so other species that rely on this butterfly will face problems.”

Dr Keatley explains that humans rely on ecosystems for services such as clean air and water and carbon cycling and disruptions to the system will affect these services.

“There may also be direct impacts on human health,” Dr Keatley says. “Asthma, hay fever, allergic conjunctivitis and eczema can all be influenced by pollen – part of the flowering cycle. So changes in the timing of plant life cycles could lead to extended and more extreme hayfever seasons.

“For example in Europe, the pollen season has lengthened with species flowering earlier in spring and later into summer.”

“Understanding the life cycle stages of plants such as asthma weed and the influence of climate on it will assist in the management of these diseases.”

However, Dr Keatley explains that the full impact of changes in life cycle timing is not completely understood as Australia lacks long-term records of the timing of life cycle events in the past.

To solve this problem, Dr Keatley and her colleagues from the EarthWatch Institute and the Bureau of Meteorology are working to develop an Australian-wide picture of the impacts of climate change on the timing of natural events. Together they developed a website called Climate Watch, where members of the public observe and record information about what’s happening in their own backyards, streets and local parks.

“This information will assist in the understanding, modelling and reporting of climate change impacts in both natural and management systems and can therefore be used by planners and policy-makers to assist in outlining adaptation strategies to climate change,” Dr Keatley says.