Does this graph show that Global Warming has flattened out and perhaps even decreasing...?

Indeed it does appear to flatten out or perhaps even decrease. You can recreate this graph to look in more detail for yourself. The graph is based on the HadCRUT3 data, which is available here. http://www.cru.uea.ac.uk/cru/info/warming/gtc.csv

HadCRUT3 uses a baseline period of Jan 1961 - Dec 1990 (30 years) to define the ‘normal’ position against which ‘anomaly’ is reported. If you look at this period on the graph you see as much red as blue.

However, there are some other important things to understand about this data though before making any sweeping assumptions or statements.

The Land temperature component of HadCRUT3 is a combination of the GHCN data (worldwide stations) and USHCN data (US stations). The GHCN worldwide data is used as the Land measurement data source (excepting the US) for the HadCRUT, GISS and NOAA global datasets.

Here is a list of the stations used to compile the Global average.

ftp://ftp.ncdc.noaa.gov/pub/data/ghcn/v2/v2.temperature.inv

You will see that there are 7820 stations in the list. Of these however, only about 2,700 are currently used. See Figure 2 (page 2842) of this document.

http://www.ncdc.noaa.gov/oa/climate/research/Peter...

This shows how many stations are used to calculate the global mean in any year. The graph peaks at around 1970 when nearly 6,000 stations are used to calculate the annual mean. This falls to about 2700 by 1997 (the end of this study period).

To see which stations are dropped, do a ‘join’ of the station list linked above to this table; it contains the GHCN records that make it into the GISS data.

http://data.giss.nasa.gov/gistemp/station_data/sta...

Of the stations that are dropped in the period 1970 onwards, 79% of them are denoted “R” for Rural and 21% are denoted “U” for Urban. The dropped stations are very predominantly Rural. This has a significant warming effect on the Global mean anomaly.

The next issue is how the data is ’corrected’ to account for urban effects. This document (pages 9-11) explains how it is done.

http://hadobs.metoffice.com/crutem3/HadCRUT3_accep...

They apply a correction of “0.0055_C/decade, starting in 1900.” (you’re going to need a better thermometer!). Over a period of 100 years they adjust station measurements by just 0.055 C to account for the UHI effect.

So we have two corrections that both ‘warm’ the anomaly.

Understating the UHI raises the anomaly incrementally year by year from 1900 onwards.

Removing the Rural Stations increases the anomaly from 1970 onwards.

You can now look at that graph in armed with a little more understanding of how it’s constructed.

Dana makes the comment “Of course the CRU data excludes the Arctic, where the planet is warming the fastest. In GISTemp, which models for the Arctic using the northernmost stations, the temperature trend continues to rise as you can see in Figure 3 here:”

This is not quite true. Both HadCRUT and GISS use the same GHCN data. Of the GHCNstations currently in use 94 are above the Arctic Circle. So the statement “the CRU data excludes the Arctic” is incorrect. The important difference between GISS and HadCRUT3 is how they interpolate data into area’s that are ‘missing’ measurement stations.

HadCRUT3 divides the globe into a 5deg x 5deg grid. Where a grid cell has no measurement, one will be implied for it if there are measurements for 5 of the 8 surrounding grid cells. It leaves cells not meeting this criteria marked ‘no data’

GISS derives a value if there is any measurement within 1200 km. It also leaves cells not meeting this criterion marked ‘no data’.

The fundamental difference is that the HadCRUT grid cells are much smaller that GISS’s. This becomes particularly critical at the N pole. There are only a handful of stations close to the pole (see the station lists given above and sort by latitude). HadCRUT extrapolates anomaly over a much shorter distance than GISS, and cannot populate the grid cells close to the pole. GISS however interpolates over a much larger didtance and can span the pole (from stations in Alaska, Canada and Russia). HadCRUT say ‘we don’t know’, but GISS interpolates across the pole ignoring the fact that the pole is almost certainly colder than the stations its anomaly is derived from.

I actually love this graph. If you notice the time from 1910 to 1940, you will see that the increase in about the same rate as what we have been seeing for the past 40 years. But wait a minute, the warmers have told us for a very long time that man was creating temp increases that have not been seen in the history of the earth. That the rate of the increase is so dramatic, yet during the relatively short period of time we have records for, we see a time when the heating was every bit as "dramatic" but man had little to no influence. Also I am looking and even trying to squint at the graph, but I don't see an exponential trend. You know I only have a PhD in statistcs, perhaps I did not take the course where you model this with an exponetial trend, but I am sure that virtualguy as much more education on this matter, certainly enough to insult others.

Fourteen of the fifteen warmest years in the series have now occurred in the past fourteen years (1995-2009). The only year in the last fourteen not among the warmest fourteen is 1996 (replaced in the warm list by 1990).

http://www.cru.uea.ac.uk/cru/info/warming/

(In six months the stat will be 15 of the warmest 16 years.)

It shows that according to CRU data, the global warming trend has slowed over the past few years.

Of course the CRU data excludes the Arctic, where the planet is warming the fastest. In GISTemp, which models for the Arctic using the northernmost stations, the temperature trend continues to rise as you can see in Figure 3 here:

http://www.columbia.edu/~jeh1/mailings/2010/201001...

that is because there is no global warming, the earth warms up and cools down in long cycles back in the sixties everybody was worried about a coming ice age now global warming it was a scam for tree huggers to get money emails have been found proving this

Not to anybody who has the slightest understanding of statistics, it doesn't. Your comments about 1/100 degree thermometers show that you are not one who is included in that category. The casinos in Las Vegas are founded (and funded) on people like you.

CO2 expeller had a good point. What I think is important to keep in mind is that the red is completely arbitrary and meant to fool the gullible (i.e. the alarmists faithfull) that the warming in the last 40 years is unprecedented. The graph is biased but you pointed out in your comments why you used it. The problem for alarmists is it isn't even that alarming.

Only if you think that the recent trend is different from the other short-term ups and downs shown in the graph,

What's even more interesting is that not only has warming flattened out but it shows that the earth never reached a 1969 global warming forecast of a 7f temperature rise by the year 2000.

Is this an example of Poe's Law?