Northern hemisphere warming rates: More than you may have heard

Guest post by muoncounter

Winter is the time of year when denials of global warming seem to come from all sides.  It is therefore a useful time to determine just what our current rates of warming are.

The very thorough SkS post Assessing global surface temperature reconstructions put the average global warming trend at 0.14-0.16 deg C/decade.

This rate is typical of the linear trend of modern temperature records which in some cases go back to 1880.  And some skeptics take comfort in this rate, claiming that it is nothing to worry about.  But is this rate of change an accurate description of what we are currently seeing?

A more in-depth look at surface temperature trends reveals obvious differences in hemispheric warming rates.

 --- NASA GISS, 7 Jan 2011

The northern latitudes (upper graph) have warmed more than the rest of the globe.  This figure also makes it clear that the rate of warming since approximately 1970 is much greater than the linear trend over the entire record.  

In his 2007 testimony to the US Senate, Dr. Kevin Trenberth described such an observable increase in warming rates in recent decades:

For the global average, warming in the last century occurred in two phases, from the 1910s to the 1940s (0.35°C or 0.63ºF), and more strongly from the 1970s to the present (0.55°C or 1.0ºF) at a rate of about 0.16ºC (0.3ºF) per decade. An increasing rate of warming has taken place over the last 25 years, and 12 of the 13 warmest years on record have occurred in the past 13 years.

A recent SkS article on Canada showed a warming trend of 0.5 degC/decade during the last 30 years.  Closely related are the 30 year warming trends (0.4 degC/decade typical) calculated for individual stations in northwestern US and southwestern Canada.

This begs the question:  What is the recent (30 or more year) warming rate in other parts of the northern hemisphere?

A surprisingly straightforward way to analyze temperature trends over large areas is by comparison of the long term (100 year) trend to the more recent (50 year) trend.  This Mapview interface to the HadCrut/GHCN data makes this task very easy.  Individual 5x5 degree grids may be selected and a linear fit applied (as shown in Figure 1a).  The time window may then be altered from this graph screen, giving a recalculated linear fit (Figure 1b) upon redraw. The slope of each linear fit in degrees C per decade, along with an estimate of statistical significance, is given in the right-hand panel of each graph.

Figure 1a.  100 year trend for 5x5 degree grid in central Europe(50-55N, 10-15E, roughly centered on Berlin)

Figure 1b, 50 year trend, same 5x5 degree grid

Table 1a shows the results of applying this process to 5x5 grids across a large part of Europe, with trends (all in deg C/decade) calculated for the interval 1910-2009.

The arithmetic average of all these 100 year trends is 0.09 degC/decade.

Table 1b shows the trends for the same grids with linear fits for the interval 1960-2009:

The average of all these 50 year trends is 0.28 degC/decade.  Red indicates greater than 0.35, yellow 0.3-0.35, green 0.25-0.3 and purple 0.2-0.25 degC/decade.

Figure 2 is a map showing the 5x5 grids color-coded according to their 50 year trends from Table 1b.

It is clear from this analysis that during the last 50 years, large parts of Europe warmed at 2-3 times the ‘average’ global rate.  This is at the high end of the 'land' warming rates shown in the bar graph above.  Some will say that this is a cherry-pick, but 50 years over an entire continent is one very large, long-lived cherry. 

We know that early 20th century warming, thought to be largely driven by increasing solar irradiance, was followed by a mid-century cooling episode thought to be due to increasing aerosols.  When we quote the hundred year linear trend, we are averaging all of those independent events together with the recent, more rapid CO2-driven warming.  We are understating the magnitude of what has actually taken place over the last 50 years. 

As we watch the snow fall this winter, it is time to stop softening our own message.  Begin with 'there is no evidence whatsoever for cooling.'  Continue by talking about what is actually happening: Large areas of the northern hemisphere are warming at twice the rate we've been quoting.

Posted by muoncounter on Friday, 14 January, 2011

The Skeptical Science website by Skeptical Science is licensed under a Creative Commons Attribution 3.0 Unported License.