2010 Hottest Year Ever?

2010 is coming to a close at the end of this month, at least in a meteorological sense. You see for meteorological purposes the year ends at the end of November instead of December because December is the starting month for the winter season. So how is 2010 shaping up? Well according to most of the analysis 2010 will not be the Hottest Year on Record.

Figure 1

Here in Figure 1 we see the monthly anomalies for each of the analysis datasets. The reason it only goes back to 1979 is for two reasons, the first being 1979 is as far back as you can go for the satellite datasets and that for all of the datasets they have the Hottest Year on Record after 1979. As can be clearly seen the only dataset that has 2010 as the possible record year is the GISS analysis.

So lets zoom the focus in on the record:

Figure 2

The reason to go from 1998 here in Figure 2 is that 1998 is the earliest Hottest Year of Record for any of the datasets. Also take note that even though the datasets use different baselines the analysis are remarkably close in 1998, however they diverge the next year. Now from there you can see how in the Satellite records the Hottest Year has been 1998, in Hadcrut it looks the same and possibly also NCDC. GISS is the only record I know that has 2005 as the Hottest Year on Record and that is due to the fact they have the monthly anomalies for 2005 stay high for the entire year, unlike 1998, 2002 and 2007 where they have a spike followed by a sharp drop in temperatures.  So that brings us to 2010,  will it be like 1998, 2002 and 2007 where you get the sharp increase followed by a decrease or would it be like 2005 where the temperatures anomalies increase and basically stay elevated all year?

Figure 3

Here in Figure 3 we have 2010 just by itself and there is a couple of things of note.

1. NCDC and HadCrut have not updated since September yet, where as RSS, UAH and GISS have.

2. GISS is the only dataset that shows an increase in the anomaly since July 2010, all the other ones show decreases with both RSS and UAH both showing a sharp drop in October.

Funny how NCDC shows a drop of about .15°C, HadCrut -.125°C, RSS  -.3°C and UAH has a drop of about -.025°C since July and with RSS and UAH both showing large drop offs in the last month all the while GISS shows an increase of about .2°C. Of course GISS is the only one to pass off that they “know” what the temperatures are in the Arctic.

Here is the links to each Dataset:

GISS: http://data.giss.nasa.gov/gistemp/tabledata/GLB.Ts+dSST.txt

NCDC: ftp://ftp.ncdc.noaa.gov/pub/data/anomalies/monthly.land_ocean.90S.90N.df_1901-2000mean.dat

HadCRUT3: http://www.cru.uea.ac.uk/cru/data/temperature/hadcrut3gl.txt

UAH: http://www.nsstc.uah.edu/public/msu/t2lt/uahncdc.lt

RSS: ftp://ftp.ssmi.com/msu/monthly_time_series/rss_monthly_msu_amsu_channel_tlt_anomalies_land_and_ocean_v03_2.txt


6 responses to “2010 Hottest Year Ever?

  1. bob November 14, 2010 at 9:20 pm

    The difference between HadCrut and GISTEMP since 1998 is down to the different handling of the arctic. GISTEMP shows 2005 beating 1998 for top spot because it estimates the arctic as warming much faster than the global average. HadCrut, by omitting it, is effectively saying the arctic has warmed at the same rate in recent years as the global average. I think HadCrut is an underestimate. The satellite records of course are not measuring the surface so are not readily comparable in the arctic where the warming trend may be larger at the surface.

  2. boballab November 15, 2010 at 4:28 am

    Actually none of the datasets used by CRU, NCDC or GISS measure the surface temperature, they are measuring the temperature of the air above the surface.

    Now as to GISS and GISTEMP, that temperature trend across the Arctic is completely make believe on the Grid Cell level. When you start looking at when a Grid Cell only has 1 station in it, that station’s trend becomes the Grid Cell’s trend take that away and use the 1200km or 250km infill options of GISTEMP and you will get two different trends. You can see this by looking at the Alert Canada station. GISS from GHCN has useful data for Alert running from 1951 to 1990. During those years the Grid Cell Temperature for that station sits in matches exactly the station data. After 1990 you get one trend for 1200 KM and another for 250 KM infilling, however you don’t need to do infilling if you go to Environment Canada (EC) and check the station data: They have data past 1990 and all the way up to 2005. The EC data for 1951-1990 matches the GISS data but the actual data that EC collected from 1991-2005 for Alert does NOT match the GISS infilled data trend. It is not even close the actual data for Alert shows a .4°C rise between 1951-2005 where the GISS infilled data has a trend of 1.16°C almost 3 times as much.
    See the graphs and full post here: https://boballab.wordpress.com/2010/03/09/giss-infilling-the-true-hypothetical-cow/

    So I would say that no HadCrut is not an underestimation, GISS is seriously over estimating the Arctic temperatures and trends. Matter of fact in documents released under FOIA GISS admits that HadCrut is what people should look at not GISS for Global Trends since they are best at it:

    My recommendation to you is to continue using NCDC’s data for the US means and Phil Jones’ data for the global means. Our method is geared to getting the global mean and large regional means correctly enough to assess our model results.

    We are basically a modeling group and were forced into rudimentary analysis of global observed data in the 70’s and early 80’s since nobody else was doing that job at the time. Now we happily combine NCDC’s and Hadley Center’s data to get what we need to evaluate our model results. For that purpose, what we do is more than accurate enough. But we have no intention to compete with either of the other two organizations in what they do best.


  3. bob November 15, 2010 at 4:14 pm

    I don’t think looking at one station is useful. GISTEMP seems to be working well overall in the arctic, I wouldn’t expect the method to get individual stations correct, but I would expect the error to be evenly distributed so that over sufficient time the error will largely cancel out. The DMI trend compared to the GISTEMP one is very close.

    The hadley center acknowledge that HadCrut3 is likely an underestimation of warming:
    “New analysis released today has shown the global temperature rise calculated by the Met Office’s HadCRUT record is at the lower end of likely warming….The new analysis estimates the warming to be higher than that shown from HadCRUT’s more limited direct observations. This is because HadCRUT is sampling regions that have exhibited less change, on average, than the entire globe over this particular period.”

    Given the arctic has warmed so much faster than the global average as shown by datasets other than GISTEMP, HadCrut’s omission of large parts of the arctic produces a cooling bias. That’s just the way it rolls because HadCrut’s method is to only use stations and not to extrapolate in the arctic, so they have to caveat their record especially in recent years when the arctic temperature has jumped up (coincidentally around 2005 which is when GISTEMP broke past 1998). GISTEMP does extrapolate the arctic, which is fine. It’s not clear that produces an overestimate, which means the values are estimates and will probably be wildly wrong in some months (sometimes overestimating warming in an extrapolated area, sometimes underestimating it), but over the longterm I expect extrapolation will be a lot closer to the actual trend than omission.

  4. boballab November 15, 2010 at 6:53 pm

    That is where you are making your mistake, that one station is what GISS is basing multiple Grid Cell trends on: Get that one station wrong you got all the other Grid Cells it is based on wrong. While they were using the actual data from Alert the Grid Cell it was in matched the Alert Station trend exactly. The Grid Cells around that one were off that trend by just .03° C +/- depending if it was east or west of that grid cell. Once they stopped using the actual data from the Alert station the trend for the Grid Cell is inflated by 3 times the amount and the adjoining Grid Cell trends also are inflated by 3 times the amount. That means that the “greater warming” of the Arctic is an artifact of the interpolation done from lack of data and not represented by actual data.

    Also the Satellite datasets only reach to 82.5° North, the NCDC dataset is based on GHCN just like HadCrut and GISS so they also do not have actual data from the Arctic. So which one of these global datasets show actual measured data that is statistically matching the GISS Interpolated Polar trend?

    Hell the UAH Arctic trend (60° North to 82.5° North) is only .47° C over the time period of 79 to Now, while GISS annual trends for the same Polar regions from 1979-2009 are .75°C to over 2°C depending on Grid Cell. You can see that clearly by going to the GISS map maker program and making a 1200KM Polar projection trend map for Annual Nov-Oct and years 1979-2009. From there just look at the Zonal mean plot for between 60° and 90°.
    Matter of fact here is the image of that Zonal Mean plot from GISS:
    Image and video hosting by TinyPic

    Take note that only the Arctic has the amplified warming, even though BOTH Polar regions are supposed to have it by theory. The reason the Antarctic doesn’t have it is that they are using actual temperature measurements from there. Again more evidence that the amplified warming of the Arctic is an artifact of GISS interpolation from lack of having enough data points. That is the rub: GISS would be a lot more accurate if they had more datapoints to work from, instead they are trying to make an extrapolation off of an extrapolation off of 1 to 2 datapoints to get the temps at the pole.

  5. Pingback: Stating the obvious concerning global warming projections - Page 4 - VolNation

  6. MostlyHarmless February 7, 2012 at 3:55 pm

    A dummy knows it gets colder the further north you go – only GISS says otherwise by projecting temperatures from the few stations around or inside the Arctic circle north. Other dummies throw up their hands in horror at the red Arctic areas on GISS global maps. Then you have GISS subtracting anything up to 2°C from temperatures prior to around 1960, and adding to those later – “swinging” the plot anti-clockwise and upwards. Global warming isn’t man-made, it’s GISS-made.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: