More crooked Warmist statistics: Butterflies are NOT emerging earlier

Comment on Kearney et al., 2010 by Marc Hendrickx

I have obtained the same data used in this study as Kearney et al. and am unable to confirm the results for the historical observation data. I count 239 observations made in Oct-Dec from 1942 to 2009. The annual data show a wide range of earliest observation dates (Figure 1), and at face value the use of 5 year or 10 year averages appears to be a convenient statistical method that hides the very wide spread of observation dates.

Applying a linear regression to a graph of the earliest observation date for each year indicates a trend of -0.7 days per decade. However, with an R2 of 0.0091 the trend has no statistical significance. Based on a 10 year average of earliest observance dates, Kearney et al., 2010 claim -1.5 days per decade with R2 of 0.766. This is an artifact of averaging the dataset, and misrepresents the wide spread of observation dates and resulting uncertainty in trends.

Regardless of any trend noted, there remains a major problem using this "opportunistic" data as a proxy for emergence. This has been poorly discussed in the paper and requires further comment. Indeed the caption for Figure 1a is incorrect and misleading. The graph is in fact a measure of earliest "observance" times, not emergence. This should be amended here and elsewhere in the paper (eg Abstract).

Using this "opportunistic" data to establish emergence is like dating a volcanic eruption based on collection dates of samples housed in a museum. The historical trends identified simply reflect variation in the time collectors have ventured out to observe and collect butterflies. The databases in question do not record a single observation of natural emergence of H.Merope. Indeed no work has been published that records natural emergence times for the butterfly concerned.

In order to establish a change in emergence, the authors should actually be observing emergence. The proxy used is simply not close enough. I understand this is difficult because the "bugs" are small and difficult to observe under natural conditions.

There remains considerable temporal bias in the data, with over 50% of total observations post dating 1990. There is also a considerable bias in observation locations, with the vast majority collected in Melbourne's east and none in the vicinity of Laverton, the weather station that was used to characterise temperature change over the whole of the study area (Figure 2).

The other issue relates to the use of this Laverton weather station to characterise temperature over the very large and geographically diverse study area, amounting to approximately 12,000km2 (37.60-38.54 S, 144.17-145.48 E). The paper does not mention well documented Urban Heat Island effects over Melbourne that encompasses Laverton that have clearly affected temperature at this station over the period of study (see Morri and Simmonds, 2000 and Torok et al., 2001).

Close examination of other stations in the study area shows a wide variety of temperature trends (Figure 2). It seems the authors have chosen one station that favours their theory without adequately explaining why others should be rejected. The choice of Laverton with its inherent problems of Urban Heat Island effects are not sufficiently explained.

Trends for other stations (eg Durdidwarrah) fall well within the limits of natural temperature change indicated by Kearney's Figure 1d and provide an indication that observed temperature trends over parts of the study area can be adequately explained by natural factors without recourse to warming through increased green house gases.

Based on these points, I believe that the authors' conclusions remain unsupported by the data presented.

In addition, there is apparently an error in the discussion section where the trend from the previous version (-1.6) is used.


Kearney, Michael R., Natalie J. Briscoe, David J. Karoly, Warren P. Porter, Melanie Norgate, and Paul Sunnucks. "Early emergence in a butterfly causally linked to anthropogenic warming" Biol. Lett. published online before print March 17, 2010

Morris C.J.G and Simmonds I., 2000. Associations between varying magnitudes of the urban heat island and the synoptic climatology in Melbourne, Australia. International Journal of Climatology 20: 1931-1954.

Torok S.J, Morris C.J.G., Skinner C. and Plummer N., 2001. Urban Heat Island features of southeast Australian towns. Australian Meteorological Journal 50:1-13.


Posted by John J. Ray (M.A.; Ph.D.). For a daily critique of Leftist activities, see DISSECTING LEFTISM. To keep up with attacks on free speech see TONGUE-TIED. Also, don't forget your daily roundup of pro-environment but anti-Greenie news and commentary at GREENIE WATCH . Email me here

No comments:

Post a Comment

All comments containing Chinese characters will not be published as I do not understand them