Climate Change - Numerical Vs. Physical Experiments

Class 4.1 Case-study #1
Numerical vs. physical experiments

Climate change could have a large impact on precipitation (rain and snow), with consequences for drinking water, farming, etc. Clouds are obviously important to precipitation (and to climate change more broadly), and their formation is dependent on aerosols, tiny particles in the atmosphere, which occur naturally but are also products of industry and other human activities.  Cloud-formation processes are difficult to model: Boucher et al. wrote in the 2013 Intergovernmental Panel on Climate Change that, "Clouds and aerosols continue to contribute the largest uncertainty to estimates and interpretations of the Earth's changing energy budget."

In a study of the effect of temperature and aerosols on precipitation, Teller and Levin (2008) carried out a numerical experiment using a computer model.  They investigated the effect of temperature change by varying ground temperature

at 5 levels (15, 16, 17, 18, and 19°C). They also varied aerosol concentration at 4 levels (225, 600, 900, and 1530 cm-3), representing clean to polluted clouds.

Each observation in the experiment is the result of computer solution of equations representing the physical processes of cloud formation and precipitation over a period of time.

The numerical model was run under conditions typical of the eastern Mediterranean in winter.  Running the model requires the initial atmospheric conditions to be specified. It appears that these conditions were fixed for all runs in the experiment, except for the changes in temperature and aerosol concentration mentioned above. In contrast, in a real atmosphere the conditions would vary over space for any one rainfall event and would vary from day to day.

Be prepared for Class 4.1 to speculate on:

1. How a physical experiment could be run in a laboratory to investigate the same temperature and aerosol factors. What are some advantages and disadvantages of a physical experiment relative to a numerical experiment here?

Scientists can make a cloud, vary ground temperature and aerosol factors respectively and record the data in each observation. Advantages are that physical experiments can better reflect the climate change in the real world because numerical experiments are simplified, omitting many unrelated factors.

Disadvantages are that physical experiment makes it difficult for scientists to observe the effect of the factors they want to investigate, as there are so many parameters changing in the experiment.