Often when analysing and comparing climate data we have to choose a reference period (or baseline) to calculate anomalies. But it is not often discussed why a particular baseline is chosen. Our new paper (open access in BAMS) considers this issue and asks: does the choice of reference period matter?
Consider simulations and reanalyses of global mean temperature, without using anomalies (Fig. 1). Each climate model has a different value for global mean temperature (with a min-max range of about 3K), but all show similar changes over the historical period (grey). The range amongst the different reanalyses (based on assimilating observations) is about 0.5K, highlighting that is difficult to estimate the real world global mean temperature precisely (colours). To compare the simulated changes with those from reanalyses, a common reference period is therefore needed.
Two immediate questions arise: do these differences in mean temperature matter? And, does the choice of reference period matter?
Do these differences in global mean temperature matter?
Fig. 2 shows that future changes in global mean temperature are not strongly related to the simulated global mean temperature in the past – there is no robust correlation between the two, and there are good physical arguments for this (see Appendix A in Hawkins & Sutton, or this RealClimate post).
Does the choice of reference period matter?
Short answer: yes. The lower panels in Fig. 1 show the same data as anomalies, using two different but equally valid choices of reference period. Note that the impression of where the reanalyses fall within the model range is markedly different when making the two choices. You might draw different conclusions based on the left or right panel only.
The choice of reference period also matters when assessing expected changes in global mean temperature over the next couple of decades, and for assessing when the 2°C above ‘pre-industrial’ global temperature level might be reached. These examples are discussed in detail in Hawkins & Sutton, but here we use an analogy instead.
Different time series can be thought of as stiff (and “wiggly”) wires that are required to pass through a fixed length of tube. Different length reference periods correspond to tubes of different lengths, with longer tubes required to have wider diameters. There is little constraint on how the wires spread outside the tube, and for longer tubes, less constraint on how they vary within the tube, thanks to a larger diameter. In the extreme, a tube that is one time point long would have zero diameter because all of the wires can be forced to pass through the same point. The constraint on where the wires are positioned vertically, relative to each other and relative to the tube, varies as the tube is slid horizontally along the loose bundle of wires.
Interpreting the wires as time-series of annual mean global mean temperature illustrates the effect of choosing a reference period. Which is the warmest timeseries at later times depends on the choice of reference period (or tube position). The black dashed lines show the range of possible futures for a larger set of time-series demonstrating that the uncertainty shrinks for later reference periods.
Overall, there is no perfect choice of reference period. A strong recommendation is that any studies that seek to draw quantitative conclusions from analyses that involve use of a reference period should explicitly examine the robustness of those conclusions to alternative choices of reference period.
Paper: Hawkins & Sutton, Connecting climate model projections of global temperature with the real world, BAMS, in press (open access)
[Thanks to Francis Zwiers for suggesting the wire analogy.]