Global Data Hound

AUTHOR: W. Fred Zimmerman, Research Scientist, ISCIENCES, L.L.C.

Executive Summary

In the 22nd century, if GHG emissions are consistent with the predictions of the IPCC’s high-growth “A2″ scenario, humans, and all other species, will experience novel climates that will cover between 12 and 39 percent of Earth’s land surface ^[1]. In these areas, the seasonal mean precipitation and temperature will be like nothing we have experienced in the 20th century. Indeed, with regard to temperature in particular, Williams et al. observe that “because the pre-Industrial Revolution climate system was already in a warm state, further increases … are likely to be novel not just relative to the 20th century but also to climates for at least the last million years” ^[2].

Under the same scenario, humans, and all other species, will find that familiar climates that now cover between 10 and 48 percent of the Earth’s land surface will have disappeared; in other words, few places on Earth will offer the same combination of climate parameters as the ones that exist today in those locations.

How Do They Know That?

Using a gridded global map, a current “climate vector” was created for each cell. The climate vector used four (4) variables: mean surface air temperature and precipitation for June, July, and August (JJA) and for December, January, and February (DJF). These variables were chosen for a combination of substance and convenience. These variables were considered sufficient because

1. seasonal temperature and moisture availability are known controls on species distribution and abundance (Prentice et al. 1992)
2. they correlate well with other bioclimatic controls on species distribution (Stephenson 1998; Sitch et al. 2003).

These variables were also considered convenient because

1. GCMs are good at predicting seasonal means
2. previous work on “no-analog” climates used the same set of variables

The next step was to create future climate vectors for each cell. The future values (for the same variables) were drawn from the results of nine GCMs used by the IPCC Fourth Assessment report. (list of models used in the Williams study)

Then, the current and future values of each cell were compared using a measurement known as the Standardized Euclidean Distance (SED). (For a simple example, remember high-school analytic math: the Euclidean Distance between a point with x, y coordinates (3,3) and another point with coordinates (2,2) is the square root of ( (3-2)^2 + (3-2)^2 ), or the square root of 1^2 + 1^2, i.e. the square root of 2. This is the same thing, except in four dimensions.)

High SED values (high distances between the 20th and 21st-C values for a cell) scorrespond to high degrees of predicted local climate change. To understand how Williams et al. arrived at their estimates of “novel” and “disappearing” climates, we must take a brief trip into hyperspace, aka “climate space”.

An Excursion into Climate Space

Just as a two-dimensional scatter plot looks like a “spray” of points, and a three-dimensional scatter plot looks like a “cloud” of points, a four-dimensional scatter plot would look like a four-dimensional cloud … if we could see it.

The 20th century climate cloud is “home.” Some points in the 21st century climate cloud are “close” to home, i.e. they have four-dimensional locations that are within the boundaries of the current cloud. The variable that the authors call SEDmin is the minimum distance from the 21st century cloud to the closest point in “home.”

The 21st century points that are outside the (interpolated) boundaries of the current cloud, i.e. they have an SEDmin > 0, are those that are “novel”: they correspond to combinations of seasonal precipitation and surface temperature that are not found in today’s climate anywhere on the world. The higher the distance from “home,” the more novel the conditions. By definition, today’s species, including ours, are not accustomed to such conditions.

The 20th century points that are not found within the boundaries of the 21st century cloud are those that are “disappearing.” What this means is that those species–including ours–that are accustomed to taking these current conditions for granted are in for a rude shock.

Impacts and Implications

The most obvious implications of this study pertain to strategies for conservation and preservation of biodiversity in the face of disappearing climates. Indeed, as of May 2009, ISI Web of Science reported 37 citing references since 2007 and Google Scholar reported 60.

“There is a close correspondence between regions with disappearing climates and previously identified biodiversity hotspots; for these regions, standard conservation solutions … such as assisted migration and networked reserves … may be insufficient to preserve biodiversity” (Williams et al., 2007, at 5738) If there is no suitable climate available anywhere on Earth, let alone in an effectively preserved conservation area, the prospects for survival of an endangered species must be bleak indeed.

The implications for biodiversity are heightened by the significance of contiguity for habitat preservation. In a nutshell, if the current climate in the Andes disappears at the end of the 21st century, but there is still an analog habitat in the Himalayas, that does nothing to help save the llama. (Or, in the reverse scenario, the yeti.) To deal with this issue, Williams et al. built a second set of maps that for each cell, limit the pool of analog climates to locations within 500 km. As the map below illustrates, this makes things look a lot worse.

The study also predicts the appearance of novel climates. As noted in the executive summary, these novel climates will mostly be climates that are warmer than today’s. Other studies have found that CO2 concentrations exceed any recorded for the last 650,000 years ^[3]. In other words, substantial portions of the Earth’s surface will have conditions that are entirely outside the “fitness landscape” in which the human species evolved.

References

1. ↑ Williams, J.; Jackson, S. & Kutzbach, J. (2007), ‘Projected distributions of novel and disappearing climates by 2100 AD’, Proceedings of the National Academy of Sciences 104(14), 5738.
2. ↑ Overpeck, J.; Whitlock, C. & Huntley, B. (2003), ‘Terrestrial biosphere dynamics in the climate system: past and future’, Paleoclimate, Global Change and the Future. Springer-Verlag, Berlin-Heidelberg-New York, 81–103
3. ↑ Siegenthaler, U.; Stocker, T.; Monnin, E.; Luthi, D.; Schwander, J.; Stauffer, B.; Raynaud, D.; Barnola, J.; Fischer, H.; Masson-Delmotte, V. & others (2005), ‘Stable carbon cycle-climate relationship during the late Pleistocene’, Science 310(5752), American Association for the Advancement of Science, 1313–1317.