Global Surface Air & Sea Temperatures: Current Conditions and Future Prospects

[Gray-Wolf]

It might help to sharpen some folks understanding of how you could increase global temps by over 2c yet not really 'notice' it getting hotter? ( like night time temp increases can do to 'daily averages'?). If we see early spring and late autumn extremes temps increase then the yearly average becomes skewed?

 

I do recall being in Chester Zoo on my Birthday ( Mar 18th) and it being around 70f!!! so the 'right airmass' does seem able to let the suns 'strength' , this time of year, be felt?

[knocker]

Temperature trend biases due to urbanization and siting quality changes

 

The temperature in urban areas can be several degrees higher than their surrounding due to the Urban Heat Island (UHI). The additional heat stress is an important medical problem and studied by bio-meteorologists. Many urban geographers study the UHI and ways to reduce the heat stress. Their work suggests that the UHI is due to a reduction in evaporation from bare soil and vegetation in city centers. The solar energy that is not used for evaporation goes into warming of the air. In case of high-rise buildings there is, in addition, more surface and thus more storage of heat in the buildings during the day, which is released during the night. High-rise buildings also reduce radiative cooling (infrared) at night because the surface sees a smaller part of the cold sky.

 

http://variable-variability.blogspot.co.uk/2015/03/urbanization-siting-quality-changes.html

[knocker]

 

Early note on March 2015 global

[BornFromTheVoid]

NCEP/NCAR reanalysis has March 2013 as the 3rd warmest on record at +0.39C (+0.80C) above the 81-10 (51-80) average. This lies behind 2010 at +0.46C (+0.87C) and 2002 at +0.40C (+0.81C).

 

 

This also means that we're 2nd warmest on record for the first 3 months of the year at +0.33C (+0.66C), just behind 2010 at +0.39C (+0.72C) and just ahead of 2007 at +0.32C (+0.65C).

 

[knocker]

Nick's been tinkering again.

 

  Dynamic trend viewer

[stewfox]

According to RSS, March 2015 averaged 0.255 degrees C. ( 0.46 deg F.) above the 1981-2010 mean for the globe (70 deg S. through 82.5 deg N. latitude). This makes March 2015 tied for 9th warmest March on record. The RSS satellite measured records go back to 1979.

 

 

http://www.accuweather.com/en/weather-blogs/climatechange/march-2015-global-satellite-me/45194068

[knocker]

Global temperature Jan/Feb 2015 from HadCRUT4

 

http://www.cru.uea.ac.uk/~timo/diag/tempdiag.htm â€¦

[knocker]

 

Global surface TempLS up slightly in March

[BornFromTheVoid]

Warmest March on record according to the JMA

 

 

1st. 2015 (+0.31°C)

2nd. 2010 (+0.28°C)

3rd. 2002 (+0.26°C)

4th. 1990 (+0.25°C)

5th. 2014 (+0.22°C)

 

 

http://ds.data.jma.go.jp/tcc/tcc/products/gwp/temp/mar_wld.html

[BornFromTheVoid]

3rd warmest March on record according to the GISS data

http://data.giss.nasa.gov/gistemp/tabledata_v3/GLB.Ts+dSST.txt

 

Top 5

2002 (+0.88C)

2010 (+0.87C)

2015 (+0.84C)

1990 (+0.71C)

2014 (+0.70C)

 

 

Warmest year to date

[Gray-Wolf]

So we are on a good start to challenge 1014's record year then? With last year seeing Feb drop the ball our warm feb may be setting us up for that 'back to back warmest' record should the rest of the year continue as is?

[BornFromTheVoid]

So we are on a good start to challenge 1014's record year then? With last year seeing Feb drop the ball our warm feb may be setting us up for that 'back to back warmest' record should the rest of the year continue as is?

 

With a strengthening NIno that most models show lasting into 2016, I'd say there's a reasonable chance of 3 record warm years in a row.

[Gray-Wolf]

With a strengthening NIno that most models show lasting into 2016, I'd say there's a reasonable chance of 3 record warm years in a row.

 

Odd that eh?

 

I wonder if they will have 'fixed' all those dodgy readings by the 3rd year or will the other place still think folk are believing a load of gish gallop?

[VillagePlank]

Here's the running HadCrut4 rank. The first value will always rank 1 (because there's only one value to compare it to)

 

 

Not quite the forever breaking records, is it? And, incidentally, there have been numerous periods where adjacent years have been the warmest in the entire series before it.

 

What would one reasonably expect on a rising trend?

 

All utterly unremarkable.

 

 

 

Edited April 15, 2015 by Sparkicle

[stewfox]

Odd that eh?

 

I wonder if they will have 'fixed' all those dodgy readings by the 3rd year or will the other place still think folk are believing a load of gish gallop?

 

Its not just dogy readings its missing readings.

 

Most of the gaps are in the colder uninhabited parts of the world. I'm also not sure what a 1/20th of a degree global temp rise will prove if it happens in the next 3 years ? One should always challenge data.

Edited April 15, 2015 by stewfox

[knocker]

Why raw temperatures show too little global warming

http://variable-variability.blogspot.co.uk/2015/04/raw-temperatures-show-too-little-global-warming.html

[Midlands Ice Age]

Why raw temperatures show too little global warming

http://variable-variability.blogspot.co.uk/2015/04/raw-temperatures-show-too-little-global-warming.html

 

 

Knocker ...

 

This looks odd to me.

 

It says  that 'even in a village' a heat island effect of a half degree was observed when the measurement was moved.

 

If this is occuring in a village how many of the guess 80% of temp recordings inside urban areas have a heat island effect?

 

I think also to claim that moving equipment to airports has a cooling effect is also dodgy. I know that airports (from plane spotting) are ALWAYS hotter than 5 miles away, unless that is there  is a breeze blowing from the other direction.

 

One only needs to look at the adjusted temps for the 1900 - 1950 period (with the severe reduction) to see that this effect already accounted for is having a much greater effect on the current warming trends.

 

As per Stewfox's post above the greatest problem today is in the areas where no readings are currently being taken. The Cowton & Way figures for these areas also look distinctly odd. I am not sure that I believe that 90% of areas under-recorded are showing larger warming trends than anywhere else. (I am not just talking about the Arctic as it is very difficult to find one area that is cooling in their homogenisations)..  

 

MIA

 

The claims of this guy  look as though his boss needs to prove something and he wants to keep his job..

Edited April 16, 2015 by Midlands Ice Age

[knocker]

MIA

 

The claims of this guy  look as though his boss needs to prove something and he wants to keep his job..

 

I shouldn't worry too much about this guy keeping his. job. He is an internationally respected scientist currently at the University of Bonn.

 

http://www2.meteo.uni-bonn.de/mitarbeiter/venema/

 

http://65.54.113.26/Author/22856449/victor-venema

[BornFromTheVoid]

The scientists have looked at the data, MIA. They theorised, experimented, analysed and came up with the corrections that, in their expert opinions, make the temperature record most consistent.

 

Cowtan and Way used Kriging and satellite data for their method. Kriging is something used in GIS and other such packages for decades and for countless areas of study. So I guess to dismiss their work, you either need to demonstrate extreme faults/weaknesses in the satelite data and/or Kriging. Also, they show a lot of cooling around the southern ocean and Antarctica for areas that aren't covered by HadCrut4

 

But yeah, fraud and conspiracy to keep jobs or something. Academia has long been known as the easy route to money...

[VillagePlank]

The scientists have looked at the data, MIA. They theorised, experimented, analysed and came up with the corrections that, in their expert opinions, make the temperature record most consistent.

 

Cowtan and Way used Kriging and satellite data for their method. Kriging is something used in GIS and other such packages for decades and for countless areas of study. So I guess to dismiss their work, you either need to demonstrate extreme faults/weaknesses in the satelite data and/or Kriging. Also, they show a lot of cooling around the southern ocean and Antarctica for areas that aren't covered by HadCrut4

 

But yeah, fraud and conspiracy to keep jobs or something. Academia has long been known as the easy route to money...

 

Kriging provides a best unbiased linear estimate for interpolation: in it's two dimensional (Cartesian) form if we know values at (x1,y1) and (x2,y2), can we determine what the value is anywhere on the line that connects the two points. In it's simplest form this is analagous to the standard linear regression (ie the slap a line on an Excel spreadsheet mathturbation etc)

 

I haven't read the paper by Cowtan etc, so I can't possibly comment on their initial assumptions which are crucially vital. Critically, one needs to select data points very carefully as Kriging, given it's linear analogue, is open to both cherry picking and is vastly affected by outliers.

 

Consider the following: if we take the temperature in London, and the temperature in Paris, can we necessarily derive the temperature at the surface in the middle of the English Channel? Anyone with even a modicum of interest in meteorology would question such an experiment as suspect (although it might well work, I don't know) The problem is, is that it is a linear solution to a non-linear problem; so all of the caveats must be considered.

Edited April 16, 2015 by Sparkicle

[BornFromTheVoid]

Kriging provides a best unbiased linear estimate for interpolation: in it's two dimensional (Cartesian) form if we know values at (x1,y1) and (x2,y2), can we determine what the value is anywhere on the line that connects the two points. In it's simplest form this is analagous to the standard linear regression (ie the slap a line on an Excel spreadsheet mathturbation etc)

 

I haven't read the paper by Cowtan etc, so I can't possibly comment on their initial assumptions which are crucially vital. Critically, one needs to select data points very carefully as Kriging, given it's linear analogue, is open to both cherry picking and is vastly affected by outliers.

 

Consider the following: if we take the temperature in London, and the temperature in Paris, can we necessarily derive the temperature at the surface in the middle of the English Channel? Anyone with even a modicum of interest in meteorology would question such an experiment as suspect (although it might well work, I don't know) The problem is, is that it is a linear solution to a non-linear problem; so all of the caveats must be considered.

 

Kriging involves a more than simple linear interpolation that one can use in excel with 2 points in excel. It involves local, regional and global spatial trends (not just linear), with covarience based weightings, etc. The accuracy was mainly assessed by removing known temperature data, and filling in with kriging and hybrid data and comparing to the real observations.

 

With your example, multidirectional trends would be taken into account, as using kriging to fill in a simple line of data points would be no better than other estimates. So rather than just the NW-SE line, other regions of data would be taken into account and the effect of the land/sea boundary could be easily accounted for. Different methods could be tested and assessed, then the most accurate used.

 

Anyway, the method they found that was the most accurate involved using the HadCrut4/UAH hybrid and kriging, where their respective errors were lowest.

 

Personally, I think trying to fill in gaps caused by missing data does provide an area for a researchers bias to get in the way, but it's better than pretending that these areas don't exist. Also, I'd hope the peer review process and the open nature of their work (all code and data are provided on their university web pages) should help with spotting any obvious errors or bias.

[Midlands Ice Age]

I shouldn't worry too much about this guy keeping his. job. He is an internationally respected scientist currently at the University of Bonn.

 

http://www2.meteo.uni-bonn.de/mitarbeiter/venema/

 

http://65.54.113.26/Author/22856449/victor-venema

 

OK  Knocker....

 

Are you suggesting that Victor actually did this fieldwork?. Or was it given to one of his undergrads to perform the work?

 

I think you will find your data is more likely to prove my contention than your own interpretation of it.

 

In no way is the bosses job at risk.  I was not talking about the boss losing his job. (only the undergrad judging by the inferences coming out of the report.) 

 

Can you really see how a 0.5C increase in a village compared to its surrounding areas, means that all sites moved have this effect, or that  the village increase in heat island effect should not be applied to all suburban sites across the world?

 

Also why is it that if you look over on the other blogsite (which publishes up to date temperature data to within 30 mins), the airports up and down the country are always in at least 5  positions of the top 10, and hardly ever feature in any low temperatures?.

 

It just doesn't feel correct to me -  No matter how many papers Victor V has allowed his name to be put on!!

. 

 

BFTV, to talk in terms that the theory for the Cowtan and Wey homogenisation data being correct, when you actually examine the reports above by Stewfox it all looks beyond mere coincidence that about 90% of all missing areas have increases in temperatures.

Natural climate variability will ensure that a reasonable balance of above and below temperatures would be achieved (as in the rest of the world).. It simply doesn't happen  with the C & W dataset that Stew has posted and I commented on last year for the NOAA and NCDC data.

 

Note I am not a denier, but I beiieve we need to keep our assumptions within reasonable bounds if you expect to win the battle. 

The above results and conclusions just do not pass the reasonability test!

 

MIA                       

[knocker]

Jan, Feb & Mar of 2015 were the warmest start to any year on record according to NASA

 

http://data.giss.nasa.gov/gistemp/tabledata_v3/GLB.Ts+dSST.txt?ns_mchannel=social&ns_campaign=bbc_weather&ns_source=twitter&ns_linkname=news_central

[mikeworst]

What I

would like to know is how anyone including yourself Knocker can take Nasa Giss seriously after all that has been exposed. I proceed to duck.

[VillagePlank]

Kriging involves a more than simple linear interpolation that one can use in excel with 2 points in excel. It involves local, regional and global spatial trends (not just linear), with covarience based weightings, etc. The accuracy was mainly assessed by removing known temperature data, and filling in with kriging and hybrid data and comparing to the real observations.

 

With your example, multidirectional trends would be taken into account, as using kriging to fill in a simple line of data points would be no better than other estimates. So rather than just the NW-SE line, other regions of data would be taken into account and the effect of the land/sea boundary could be easily accounted for. Different methods could be tested and assessed, then the most accurate used.

 

Anyway, the method they found that was the most accurate involved using the HadCrut4/UAH hybrid and kriging, where their respective errors were lowest.

 

Personally, I think trying to fill in gaps caused by missing data does provide an area for a researchers bias to get in the way, but it's better than pretending that these areas don't exist. Also, I'd hope the peer review process and the open nature of their work (all code and data are provided on their university web pages) should help with spotting any obvious errors or bias.

 

I realise that Kriging involves more than 2 points; I was trying to make the point obvious and understandable to a lay reader. Using multiple points still is an analogue of linear trends - the main point being, as I said, it is a linear solution to a non-linear problem - lines can, and do exist in multiple dimensions. Anyone familiar with any sort of vector arithmetic, for instance, would understand that you need to use Pythagoras' in the quantity of dimensions that match the cardinality of the vector - crucially: it's still linear, no matter how many dimensions you include.

 

Do you really think that we could find the surface temperature of a location in the English channel if we used points from Cornwall, Kent, France and Belgium?? Maybe even more points - shall we Krig a daytime temperature as an experiment and see how we go? This is rather like saying that if we know the CET (three points) we can determine the temperature at Whitehall.

 

Kriging has it's uses, such as determining probabilities spatially: they tend to move smoothly (such as the probability of finding gold or oil) Temperatures tend not to be smoothly distributed - ie the temperature average inside the M25 against the temperature average of Kent, whilst related, are likely to be associated by some non-linear relationship that Kriging is unlikely to be able to resolve.

Edited April 17, 2015 by Sparkicle