Guest post by Sam I. Outcalt
Abstract: The historic record of the mean annual air temperature record at Boulder does not display the major “global” climate transitions detectable in the total annual precipitation record. Hurst ReScaling is demonstrated to be a powerful tool to examine climate change.
Background: The citizens of Boulder, Colorado are extremely active in efforts to combat, research, and attenuate the impact of “Global Warming”. It is therefore instructive to look at the temperature record of this city as a means of understating the high level of concern that has led to municipal and county carbon taxes and other mandates by the local green machine. The Modern Warming Regime began in 1976 with the “hockey stick” upturn in “global” temperature data. There is a large body of evidence indicating that the warming ended with the 20th Century. The footprint of that event on station thermal records is spatially irregular. Strong 1976 integral signals are found in the mean annual temperature (MAT) data of some stations in Colorado. These stations are concentrated in the Northwest quadrant of the state. Fort Collins is in the group but Boulder is not.
The Data: Let us begin with an analysis of the air temperature history. The temperature record at Boulder beginning in the late 19th Century extending just past the the end of the first decade of the 21st Century is presented as Figure 1.
Figure 1. The mean annual air temperature station record at Boulder, Colorado.
Several years ago Outcalt et.al.(1997) discovered that inflections in the integral traces of serial climate data indicated regime transitions. In Figure 1 air temperature history integral inflections in 1932, 1967 and 1986 break the record into five regimes. The Boulder station record is missing strong integral inflections that are present around 1936 and 1976 in “global” temperature records (see Figure 1 and Appendix GLOB).
The 1976 inflection became famous as the base of the “hockey stick”, which signaled a steep temperature rise in the “global” air temperature data. There is a steep regime warming trend between1986 and 2011 in the Boulder station air temperature record but it’s end may be signaled by the steep downturn in the 7 year moving average in 2004, near the record end. It is clear that while a large area of the globe suffered the impact of Global Warming, Boulder escaped ! Is the record at Boulder representative of a larger area ? As a check the averaged GHCN and HadCRUT3v grid air temperatures were examined in a 10×10 degree parch centered on Boulder. This pooled grid data is presented as figure 2.
Figure 2. The integral of the pooled gridded data for the period from 1940 to 2008 displays the well known 1976 transition and a transition in 1998 followed by a steep downward trend which has a much steeper cooling slope than the preceding warming trend.
The gridded regional data displayed integral defined weak warming trend from 1976 to 1998. This indicated that there was no pronounce warming during period of modern warming from 1976 to 1998.
To check if the regional and station records are synchronous they are they were plotted together for the short time span of the gridded data as Figure 3.
Figure 3. The records appear synchronous indicating that the long term station record must crudely reflect the climate transitions in a wider region.
Having established the synchronous character of the overlapping records the difference between the station and grid record is displayed as figure 4.
Figure 4. The range of deviations and clustering of deviations with the same sign may indicate the suppression of real regime transitions and the introduction of false positives.
The stark differences in regime trend slopes in the station and grid records at the end of the 20th Century may be an artifact of record length although other sources of this contrast are possible.
It should be noted that the Hurst ReScaling method used to detect climate regime transitions is also sensitive to station moves to new sites (Runnalls and Oke (2006)). This extreme sensitivity suggests that station site move and/or raw data adjustment may introduce false positive regime transitions and/or mask real regime transitions. The clusters of deviations of the same sign in Figure 4 suggest some distortion may have been introduced into one or both data sets.
As precipitation is not as sensitive air temperature to urban effects as air temperature an exploration of the long term Total Annual Precipitation [ TAP ] record might prove interesting as earlier investigations showed that the signature of both the 1936 and 1976 regime transitions in Arizona TAP data. TAP records are also not as visible in the climate debate as MAT records so they may be free of the semi infinite adjustments suffered by the MAT records. The Boulder TAP record from the same source as the MAT record is displayed as Figure 5.
Figure 5. The Boulder station TAP record and it’s integral trace. Note the 1936 and 1976 “global regime transitions” are visible as major inflections in the integral trace.
In Figure 5 that the major “global” regime transitions that occurred near 1936 and 1976 are prominent. This clearly illustrates the power of the rescaling technique in climate analysis.
Conclusion: This brief examination indicates that the effects of “global warming” were minimal at Boulder in the shorter gridded data set and insignificant in the context of the long term station record. In short the city that has a hyper active response to “Global Warming” never experienced it ! However, Boulder did experience the impact of the regime transition in its precipitation history. It is evident that the Hurst Rescaling technique is a robust but underutilized tool in the analysis of climate change and other serial data.
Outcalt,S.I.,Hinkel,K.M.,Meyer,E . and Brazel,A.J.(1997) The application of Hurst rescaling to serial geophysical data. Geographical Analysis 29, 72-87.
Runnalls,K.E. and Oke,T.R.(2006) A technique to detect micro-climatic inhomogeneities in historical records of screen-level air temperature. Journal of Climate 19: 959-978.
Major inflections near 1936 and 1976 form a broad minimum on the integral traces of global temperature departures from the record mean from 1880-2008.
Hurst ReScaling, more info: http://wattsupwiththat.files.wordpress.com/2012/07/sio_hurstrescale-1.pdf
Sam Outcalt is a retired Professor of Physical Geography, University of Michigan, Ann Arbor