Why trust climate models if weather forecasts are often wrong? How do climate scientists test their models? We asked a climate scientist these and other questions. (More)

An Interview with a Climate Scientist

Many of us have questions about climate change. I contacted Dr. Jennifer Collins at the University of South Florida to ask her some of mine, and she was kind enough to answer every question.

Dr. Collins is an Associate Professor of Geography at the University of South Florida. Her research focuses on weather and climate. As a hurricane researcher, Dr. Collins is interested in the interaction between large scale climatic patterns such as the El Niño – Southern Oscillation (ENSO) and the Madden – Julian Oscillation and seasonal patterns of tropical cyclone activity in multiple oceanic basins. She is currently studying the environmental factors influencing the interannual variation of hurricane numbers in the Northeast Pacific and she is examining relationships between hurricane numbers in the Atlantic versus those in part of the Northeast Pacific ocean basin. In addition to her hurricane work, Dr. Collins works in other areas related to weather, climate and hazards. She works closely on projects with the National Weather Service involving tornados and fog. In addition, she collaborates with international researchers and works in the area of climate change. She is currently examining temperature changes in South America and Africa. Dr. Collins is the President of the West Central Florida Chapter of the American Meteorological Society, Director of the Climate Specialty Group of the Association of American Geographers and serves on the National Weather Association’s Specialized Operations Committee on Tropical Cyclones.

WB – We often hear complaints that short-term weather forecasts are inaccurate, so how could we possibly predict the future climate of our planet?

JC – Short term forecasts are actually quite accurate: in the 90% range for temperatures. Accuracy is more elusive with respect to timing and exact location of precipitation, although as computing resources expand forecasts increase in accuracy. Further, while there will be small local errors when forecasting the weather five days from now, the overall mean state of the atmosphere is very well represented. When predicting the future climate of the planet we are not forecasting the small scale issues such as “Will it rain next Friday?” We are forecasting the change in the overall mean state of the atmosphere – what our models do best.

WB – How do short-term weather models differ from long-term climate models?

JC – Weather and climate models come in two general categories; statistical and dynamic. Statistical models incorporate past trends to make future predictions. Dynamic atmospheric models are either: (1) point-based, where calculations are made on a grid; or, (2) spectral, where the variables are separated into wave components. Short term weather models run at a very high resolution with as many observations (weather stations, radiosondes, buoys, satellite data) input representing the initial state of the atmosphere as possible to predict very local effects over a much smaller time frame. Climate models differ from weather models in that climate models have lower resolution, longer time-steps, fewer vertical levels, and different assumptions made in parameterization of atmospheric processes, to account for factors that can change climate slowly over time and have little to no impact on a day-to-day basis. Finally, unlike some weather models that only forecast for a portion of the earth, climate models typically over the entire globe. The latest development in climate and weather modeling is coupling atmosphere models with ocean models to provide more realistic influences between the atmosphere and ocean.

WB – Do you think we understand the variables that go into long term climate change prediction, or are some significant variables still elusive?

JC – The key here is that the global temperature profile is increasing with a similar trend of increasing CO2 in the atmosphere. But we still do not understand all the variables and the physics and their total potential impact, such as stored methane in frozen areas. We still have a lot to understand about deep ocean currents as well. One of the elusive variables is volcano eruption, which can limit the solar radiation reaching the earth and cause short term global cooling. Complex variables involving cloud physics are another focus area of research, and processes that are not fully understood are handled in models (weather and climate models both) with parameterization routines.

WB – What are parametrization routines and why are they used in models?

JC – A parametrization routine is a function that replaces an atmospheric process that we cannot yet resolve with modeling. We use these for many reasons, ranging from lack of full understanding of the physics of a process (especially with cloud microphysics) to something being of too small a scale (like a cumulus cloud).

WB – How accurate are our current climate models?

JC – Climate models are constantly being improved. The models are initially run in hindcast mode for climate periods that have already occurred. We adjust parameters until the results fit those past time periods, and then assume the model will represent the future. Researchers use several models, and the IPCC predicts an increase in the range of 2 to 6 degrees Celsius by 2100 based on the various climate models currently in use.

WB – Can climate modeling reach a point where we can accurately plan for future climate change by region?

JC – While accuracy to that level is an eventually reachable goal, that level of detail will come with better models and faster computing. The time to plan is now, by taking steps to reduce dependence on fossil fuels that contribute to atmospheric pollution and CO2 increase. At this time, planning is key. The biggest threat is sea level rise that will inundate coastal cities, especially when coastal storms create storm surge. Sea level rise is caused by freshwater ice melt and through thermal expansion as ocean waters warm.

WB – Have we passed the so-called “tipping point” in climate change, with the earth now entering a positive feedback with regard to ice melt and water vapor, or do we still have time to slow or stop the warming?

JC – We’ll only know that after it has happened. It is a good idea to be good stewards of our planet.

WB – What climate research would best help us know how to slow or mitigate the effects of climate change?

JC – We already know what is leading to climate change. New research to slow or mitigate climate change will be in the realm of harnessing renewable energy sources.

WB – What sources would you recommend for BPI readers who want to keep up with developments in climate research?

JC – There are several good general sources. The National Oceanographic and Atmospheric Administration (NOAA) offer this education packet for teachers. They also offer a climate services website, ClimateWatch magazine, and responses to frequently asked questions. The Environmental Protection Agency (EPA) also have a climate change FAQ page. Your readers can help with climate change research by offering computer time at ClimatePrediction. And the Intergovernmetal Panel on Climate Change (IPCC) make all of their reports available at their website.

I and others at the University of South Florida are involved with different aspects of climate research, and you can follow my work at the USF Weather website. Also, Dr. Van Beynen in my department, Department of Geography, Environment, and Planning, does research relating to Climate and Karst environments, and our colleagues at the College of Marine Science such as Don Chambers are also involved in climate research.

I’d like to thank Dr. Collins for her time, and for helping us cut through the fog on climate change.