by David Wojick
Our goal here is to begin to articulate a research program into the role of recent long-term natural variability in climate change.
Long-term natural variability has implications for the modeling of future climate changes, on the scale of decades to centuries. It is called dec-cen variability. Dec-cen variability also relates to explaining the climate changes that have occurred over the last century or so. This is what is called the attribution problem; that is, how much of these historical changes are attributable to human activity, versus natural variability?
Our investigations indicate that the $2.5 billion a year US Global Change Research Program (USGCRP) is doing very little research of this sort [link]. There is a great deal of research on short-term variability, on the scale of a season to a decade or so. There is also some paleoclimate research looking at long periods of past climate, which may be useful. But there is very little research specifically on near term dec-cen variability, especially in relation to observed climate change over the last century or so.
This lack of research is unfortunate because the attribution problem is clearly the central policy-relevant question in climate change science. The National Academy of Sciences put the importance of attribution very succinctly way back in 1998, in their report titled “Natural Climate Variability on Decade-to-Century Timescales (NAP, 1995). The Preface of the 1998 Report provides a clear statement of the attribution problem:
“The climate change and variability that we experience will be a commingling of the ever changing natural climate state with any anthropogenic change. While we are ultimately interested in understanding and predicting how climate will change, regardless of the cause, an ability to differentiate anthropogenic change from natural variability is fundamental to help guide policy decisions, treaty negotiations, and adaptation versus mitigation strategies. Without a clear understanding of how climate has changed naturally in the past, and the mechanisms involved, our ability to interpret any future change will be significantly confounded and our ability to predict future change severely curtailed.”
In stark contrast, the USGCRP seems to assume that human activity is all that matters and this is a great mistake. For example, semantic analysis of USGCRP annual reports indicates that their attention is heavily weighted to what is called “anthropogenic global warming” or AGW. Then too, analysis of NSF research awards under the program that arguably anchors the USGCRP indicates that the vast majority of awards are directed at short-term variability, typically on a scale from a season to a decade. Modeling makes up a great deal of climate research and it too looks to be biased toward AGW. It might even be argued that AGW-based modeling dominates climate change science.
In contrast to the above, it is entirely possible that much, perhaps most, of the climate change observed over the last century or so is natural. We simply do not know because the crucial research is not being done. This central question is the attribution problem.
Our Proposal:
The USGCRP needs to be expanded or redirected to look deeply into the attribution problem. Here is our candidate list of research topics for a research program on recent long-term natural variability.
1) Low climate sensitivity to CO2 increases. Recent research suggests that climate sensitivity is much lower than most models assume.
2) Sun-climate mechanisms, especially indirect effects. Several indirect solar effects have been proposed.
3) Natural oscillations (ENSO, AMO, PDO, etc.). The role of these natural oscillations in recent long-term climate variation should be a major USGCRP research area.
4) Ocean circulation (upwelling, Gulf Stream, conveyor belt, etc.). Changes in ocean circulation are thought to be able to produce large rapid temperature changes. What role they play in recent long-term changes needs to be determined.
5) Long-term natural variations (Little Ice Age, Medieval Warm Period, etc.). We need to know if climate oscillates naturally on the decade to millennial scale.
6) Negative feedbacks (Lindzen’s Iris, convection, etc.). The climate models generally do not include strong negative feedbacks, but these have been proposed.
7) Chaotic oscillations. Climate is known to be chaotic on relatively small time scales. Whether it is on larger scales needs to be investigated. It might explain the long-term natural variations.
8) Alternative model parameterizations and assumptions.
9) Other hypotheses and new approaches.
10) Modeling the above. (It will be important to do new modeling, to explore these various processes and hypotheses, and their potential role in recent long-term climate change.)
Congress and the USGCRP should work together to develop and fund this Dec-cen Research Program.
Moderation note: As with all guest posts, keep your comments civil and relevant.Filed under: Attribution