Scott Saleska

University of California, Berkeley, Energy Resources Group

At a research site in Gothic, Colorado, Saleska and colleagues are heating a meadow. They are conducting a field experiment that applies suspended overhead heaters to a series of plots to discern effects of global warming on complex natural ecosystems. The site - a high-altitude, alpine meadow on the western slope of the Rocky Mountains - is expected to be especially responsive to global warming due to snow-melt response, albedo feedbacks, and soil that has 20 kilograms of carbon per square meter (about twice the US average).

Electric resistance heaters apply twenty watts per square meter of infrared heat to ten heated plots alternated with unheated plots. Probes in the soil at depths of 5, 10, and 25 centimeters log temperature and moisture every two hours. The researchers also measure whole system carbon dioxide flux.

Among the not-surprising micro climate results were 2° C higher daily average soil temperature in the heated plots relative to the control plots.

Surprises include:

1. Micro climate results The soil temperature difference between heated and control plots contains a mid-day spike of almost 6° - more than expected - probably because the heaters dried out the soil, so more of the sun's energy striking the heated plots went into warming the soil rather than evaporating water from it.
2. Ecosystem carbon storage The researchers used a special chamber to measure carbon flux in all plots every four hours, six times a day. Results revealed that from June to September, there is consistently less carbon storage in the heated plots than in the control plots. Though it is early to draw clear conclusions, extrapolating these results to the global scale would yield a significant impact. Globally, there are 10 million square kilometers of meadow; if the effect on carbon storage measured here were global, (everything else being equal) there would be an extra gigaton of carbon per year in the atmosphere of a greenhouse-warmed world. Some sort of carbon-flux feedback from warming should not be surprising but is currently not included in the general circulation models. There is a high correlation between carbon dioxide concentration and temperature for the past 160,000 years as evidenced in the ice-core record. Could this be due to release of carbon from tundra during warming periods?
3. Shifts in albedo In the heated plots of the upper zone, which is more representative of the area as a whole, sagebrush (which is drought tolerant) is doing better than forb and grasses (there is a higher rate of sagebrush recruitment and growth). The albedo (reflectivity) of sagebrush is less than forb by ~3% (~12% for sagebrush compared to ~15% for forb), resulting in a net increase in absorption to the system on the order of 10W/m². So if the system converts from forb and grasses to sagebrush, there will be more solar absorption. The affects of warming on species composition go beyond changes in albedo, and eventually may result in a less diverse ecosystem.

• encourage innovation and unconventionality in scientific research
• encourage interdisciplinary approaches to research

Saleska would like to see the establishment of programs explicitly designed to fund and otherwise support innovative, unconventional and interdisciplinary global change research, i.e., NASA global change fellowships, Pew Scholars Programs, and Aspen Global Change Institute programs.

To know one's ignorance is the best part of knowledge. To work to reduce ignorance, we should make public policy that is flexible and takes ignorance (uncertainty and surprise) into account. Since surprise is inevitable, we should strive to make policy that accounts for it.