2010 Global Temps

[iceicebyebye]

Well, depends on your definition of "signficant". Considering that GISS has increasingly diverged from the other temp sources over the past 10 years, that may or may not be signficant to you. And considering that a warming trend eventually requires more "warmest years on record", it may also be significant.

Sure, it's only the difference between #1 and #2 or #3...but in terms of trends and perception, it definitely means something.

not sure that 2nd or 3rd warmest years on record among the other sources would fit the description of a significantly lower result than GISS. and also not familiar with the requirement that 2010 be the warmest year on record across all sources to validate a warming trend. would certainly say that if a 97/98 level Nino occurs within the next few years, one might expect record level global temps from the major sources, absent any unexpected radical shift in other natural or HI negative forcings.

Nino values were stronger in the run up to & during 98, and the flip to Nina values has been more intense this year than the shift in 98. it's reasonable to expect the relative strength of these ensos would influence global temps in the short term.

[tacoman25]

not sure that 2nd or 3rd warmest years on record among the other sources would fit the description of a significantly lower result than GISS. and also not familiar with the requirement that 2010 be the warmest year on record across all sources to validate a warming trend. would certainly say that if a 97/98 level Nino occurs within the next few years, one might expect record level global temps from the major sources, absent any unexpected radical shift in other natural or HI negative forcings.

Nino values were stronger in the run up to & during 98, and the flip to Nina values has been more intense this year than the shift in 98. it's reasonable to expect the relative strength of these ensos would influence global temps in the short term.

I never said 2010 needed to be warmest on record to validate a warming trend...but given the supposed influence of GHG (.2C/decade), the odds should have been stacked in its favor.

And the flip to Nina this year has been pretty similar to 1998. This years Nina has been a little stronger in Sep/Oct/Nov, but prior to that was quite similar to 1998.

You apparently completely missed my explanation of the significance #1 compared to #2 or #3.

[cmichweather]

i guess i'm a noob for the climate data sources, so the GISS/NASA figures have a warm bias and aren't to be trusted? What are the other sources that people are using?

[tacoman25]

i guess i'm a noob for the climate data sources, so the GISS/NASA figures have a warm bias and aren't to be trusted? What are the other sources that people are using?

There are three other main global temp sources.

HadCRU: Surface-based like GISS, but doesn't included extrapolated Arctic temps

UAH: Satellite-based atmospheric temps

RSS: Satellite-based atmospheric temps

These three all have very similar trends over the past 10-12 years, while GISS deviates with a warmer trend.

[skierinvermont]

i guess i'm a noob for the climate data sources, so the GISS/NASA figures have a warm bias and aren't to be trusted? What are the other sources that people are using?

Just to add onto tacoman's post above.. the 5 sources are HadCRUT, GISS, NOAA which are surface based, and UAH, RSS which are satellite based.

UAH and RSS begin in 1979 while the other 3 surface sources date back to the late 1800s. There is strong agreement between all 5 sources on the trend since 1979. There is also strong agreement between the 3 surface based records on the trend since 1900. The trends from 1900-present and 1979 to present are essentially identical between all 5 data sources, indicating strong agreement on longer timescales.

However, as tacoman said there is some disagreement between the 5 major sources on shorter timescales such as the past 10-15 years. Over the last 10-15 years the satellite based sources (UAH/RSS) have shown less warming than GISS and NOAA. HadCRUT falls in the middle.

As a result, skeptics tend to prefer UAH/RSS while AGWers prefer GISS/NOAA.

Personally, I believe that the different sources have different strengths and weaknesses. The satellites are good because they have near-global coverage and you don't have concerns over urban heat island contamination and other problems (although the fact that the satellites corroborate the 3 surface data sources for the period 1979-2010 tells me that these problems are fairly insignificant). The surface records are good because they tell us what is happening on the ground, while the satellites are really reporting temperatures at around 15,000 feet. There is supposed to be slightly more warming at the surface than at 15,000 feet due to AGW. The difference between HadCRUT and GISS is mostly that GISS extrapolates a longer distance from individual temperature stations, which can lead to both true and false extrapolations. GISS has been extrapolating very warm temperatures across the arctic for the past decade. I believe that much of this extrapolation is correct because I think there is good independent evidence of extreme warming in the high arctic above 70N over the last decade. Because HadCRUT doesn't extrapolate nearly as far, they leave much of these regions that have experienced extreme warming as no data zones. Some of the extrapolation on GISS may be a little extreme however, so perhaps an average of GISS and HadCRUT is preferable.

I also think much of the disagreement over the last 10-15 years is because it is just too short of a timescale.

[tacoman25]

Just to add onto tacoman's post above.. the 5 sources are HadCRUT, GISS, NOAA which are surface based, and UAH, RSS which are satellite based.

UAH and RSS begin in 1979 while the other 3 surface sources date back to the late 1800s. There is strong agreement between all 5 sources on the trend since 1979. There is also strong agreement between the 3 surface based records on the trend since 1900. The trends from 1900-present and 1979 to present are essentially identical between all 5 data sources, indicating strong agreement on longer timescales.

However, as tacoman said there is some disagreement between the 5 major sources on shorter timescales such as the past 10-15 years. Over the last 10-15 years the satellite based sources (UAH/RSS) have shown less warming than GISS and NOAA. HadCRUT falls in the middle.

As a result, skeptics tend to prefer UAH/RSS while AGWers prefer GISS/NOAA.

Personally, I believe that the different sources have different strengths and weaknesses. The satellites are good because they have near-global coverage and you don't have concerns over urban heat island contamination and other problems (although the fact that the satellites corroborate the 3 surface data sources for the period 1979-2010 tells me that these problems are fairly insignificant). The surface records are good because they tell us what is happening on the ground, while the satellites are really reporting temperatures at around 15,000 feet. There is supposed to be slightly more warming at the surface than at 15,000 feet due to AGW. The difference between HadCRUT and GISS is mostly that GISS extrapolates a longer distance from individual temperature stations, which can lead to both true and false extrapolations. GISS has been extrapolating very warm temperatures across the arctic for the past decade. I believe that much of this extrapolation is correct because I think there is good independent evidence of extreme warming in the high arctic above 70N over the last decade. Because HadCRUT doesn't extrapolate nearly as far, they leave much of these regions that have experienced extreme warming as no data zones. Some of the extrapolation on GISS may be a little extreme however, so perhaps an average of GISS and HadCRUT is preferable.

I also think much of the disagreement over the last 10-15 years is because it is just too short of a timescale.

Couple points...

There is no previous deviation like GISS has shown over the past 10-15 years.

Also, NOAA and GISS share a lot of the same data and so tend to be pretty similar.

And finally, HadCRUT doesn't really "fall in the middle" over the past 10-12 years. It's trend is much closer to the satellites than GISS.

[The_Global_Warmer]

If the arcitc extrapolation is to warm.

how much warming would it take to ruin the ice sheet.

which is what we have seen.

unless the emphasis is going to be on Weather Patterns and wind.

seems like it would need to be well above normal temps wise as well to accomplish this task.

[tacoman25]

If the arcitc extrapolation is to warm.

how much warming would it take to ruin the ice sheet.

which is what we have seen.

unless the emphasis is going to be on Weather Patterns and wind.

seems like it would need to be well above normal temps wise as well to accomplish this task.

I don't know, ask the 1930s and 1940s.

[skierinvermont]

I don't know, ask the 1930s and 1940s.

I don't think ice was nearly this low in the 30s and 40s. And I know about the NW passage navigations and some of the other anecdotal evidence. Compilations of the data tend to show a low period, but nothing like what we are in now.

[skierinvermont]

Couple points...

There is no previous deviation like GISS has shown over the past 10-15 years.

Also, NOAA and GISS share a lot of the same data and so tend to be pretty similar.

And finally, HadCRUT doesn't really "fall in the middle" over the past 10-12 years. It's trend is much closer to the satellites than GISS.

Regarding previous deviations between the data sources over 10-15 year periods. I haven't done the math but clearly GISS must deviate as much from HadCRUT 1980-1995 on the cold side as it does on the warm side 1995-2010 since the trends from 1980-2010 are the same.

[tacoman25]

I don't think ice was nearly this low in the 30s and 40s. And I know about the NW passage navigations and some of the other anecdotal evidence. Compilations of the data tend to show a low period, but nothing like what we are in now.

There really is nothing to back this up. It's very difficult to say one way or the other exactly how that period compared to today...but my point was that Arctic temps skyrocketed during that period as well. Arctic temperature variation has always been considerably greater than lower latitudes...which means that it would be very easy to read too much into how much the Arctic warmed over the past couple decades.

[skierinvermont]

There really is nothing to back this up. It's very difficult to say one way or the other exactly how that period compared to today...but my point was that Arctic temps skyrocketed during that period as well. Arctic temperature variation has always been considerably greater than lower latitudes...which means that it would be very easy to read too much into how much the Arctic warmed over the past couple decades.

I have read several studies which say it was higher, NW passage navigations notwithstanding. Not the strongest most complete data sets but better than anecdotes from single locations.

[tacoman25]

I have read several studies which say it was higher, NW passage navigations notwithstanding. Not the strongest most complete data sets but better than anecdotes from single locations.

Yeah, but they are still just estimates and different sources will give you different answers.

Again, my point was more about Arctic temp fluctuations.

[nzucker]

There really is nothing to back this up. It's very difficult to say one way or the other exactly how that period compared to today...but my point was that Arctic temps skyrocketed during that period as well. Arctic temperature variation has always been considerably greater than lower latitudes...which means that it would be very easy to read too much into how much the Arctic warmed over the past couple decades.

Yes, the Arctic will cool down a lot with all the La Niñas coming in this next -PDO cycle. I'm sure we'll see some good changes up there as we get deeper into the new Pacific regime. Pacific Ocean is really brutally cold right now all the way from the GoA to the Baja cold pool to ENSO regions to the Humboldt Current down by the Southern Ocean. Should have enormous effects on global temperatures and the 500mb pattern that lead to cooling in the Arctic.

I have read several studies which say it was higher, NW passage navigations notwithstanding. Not the strongest most complete data sets but better than anecdotes from single locations.

I think it's safe to say the 1940s had more ice up there, but probably not that much. We also didn't have the same strength of +PDO in the 30s/40s as the 80s/90s.

[iceicebyebye]

Not much chance to beat 1998 on the satellites. We needed UAH and RSS to average around .4C for November/December and considering both were under that for November and are cooler now, it'll be pretty much impossible.

UAH would need to reach .435 to exceed 98, which as several have noted is very unlikely. as spencer observes:

"December will determine the outcome, but remember that the difference between the two years is not statistically significant."

Ch5 has risen back to above average the last few days, so curious to see if Dec on UAH even reaches the 98 figure of .277...probably a stretch.

[skierinvermont]

UAH would need to reach .435 to exceed 98, which as several have noted is very unlikely. as spencer observes:

"December will determine the outcome, but remember that the difference between the two years is not statistically significant."

Ch5 has risen back to above average the last few days, so curious to see if Dec on UAH even reaches the 98 figure of .277...probably a stretch.

Yeah it probably will be a little warmer than the .15 I suggested at the start of the month unless CH5 starts dropping soon

[nzucker]

Yeah it probably will be a little warmer than the .15 I suggested at the start of the month unless CH5 starts dropping soon

Channel 5 had started a downward trend on AMSU as of 12/7. We'll see if it continues but it usually does for at least a few days. I think .15 or so is a good guess. Should start seeing some negative values show up in 2011. Big cooldown occurring with the strong Niña...

[tacoman25]

Looking at the patterns from previous months, I think there's a good chance we see another significant drop this month. Dec 2010 probably won't finish close to Dec 1998.

[iceicebyebye]

ok, looks like some predictions shaping up....i'll be sure to check in on the dec calls here, and then i suppose in LEK's 2011 HADCRUT Global Temp thread regarding the expectation for negative UAH values....

[iceicebyebye]

GISS .74 for Nov

[BethesdaWX]

GISS .74 for Nov

Well, we can eliminate that one!

Even HADCRUT laughs at GISS.

[skierinvermont]

Well, we can eliminate that one!

Even HADCRUT laughs at GISS.

So we can throw it out simply because it is warmer than the others and you don't like that? How about some reasoning. Also I don't even think HadCRUT is out for November yet.

Let's take a look at WHY GISS was so much warmer than HADCRUT in November. GISS shows extreme warmth over northern Canada, northern Siberia, and the Arctic Ocean. HadCRUT has these regions as no data.

Let's take a look at the weather pattern the past month. Unsurprisingly it was warmest in the areas for which HadCRUT has no data. Right where GISS said it was warm.

[BethesdaWX]

Obviously, As it should. I really don't think we disagree much actually....at least not on the basics of the cluster.

My point was Not that NOV wasn't much above avg, because it certainly is much above avg, as it should be at this point....but its definitely not coming in at +0.7C. GISS is the only model that has strayed almost 0.35C from the others, which is why I usually don't even bother reading anything about it, and what it says. Lets not forget that GISS actually has less Data then HADCRUT, and also deleted all the Arctic data for an unknown reason. True, we cannot go out and just say "Its wrong because its straying".. and I realize that. However, I would prefer to go with the higher agreement, and I'm sure you would too.

Is there a reason why we can't just use Satellite data? I mean, it has much more coverage, cleaner data, no extrapolations, and it can cover just as much,if not more, of the poles than GISS/HADCRUT do. Not to mention we have 31 years of observations by satellites in the N & S pole.

Taxpayer money is what is funding these GISS/HADCRUT projects, so I see no reason for NASA to refuse FOI requests.

[skierinvermont]

Obviously, As it should. I really don't think we disagree much actually....at least not on the basics of the cluster.

My point was Not that NOV wasn't much above avg, because it certainly is much above avg, as it should be at this point....but its definitely not coming in at +0.7C. GISS is the only model that has strayed almost 0.35C from the others, which is why I usually don't even bother reading anything about it, and what it says. Lets not forget that GISS actually has less Data then HADCRUT, and also deleted all the Arctic data for an unknown reason. True, we cannot go out and just say "Its wrong because its straying".. and I realize that. However, I would prefer to go with the higher agreement, and I'm sure you would too.

Is there a reason why we can't just use Satellite data? I mean, it has much more coverage, cleaner data, no extrapolations, and it can cover just as much,if not more, of the poles than GISS/HADCRUT do. Not to mention we have 31 years of observations by satellites in the N & S pole.

Taxpayer money is what is funding these GISS/HADCRUT projects, so I see no reason for NASA to refuse FOI requests.

I would think the satellites are preferable too, but I have heard that the surface is supposed to warm more than the lower troposphere, which is what the satellites are measuring. I'm not sure how big of a factor that is.

All I know is that HadCRUT and GISS pretty much agree on the areas they both cover, but HadCRUT is missing data where it has warmed the fastest. This tells me that at least some combination of GISS and HadCRUT is preferable.

For the surface, I would take an average of HadCRUT and GISS. I think the surface has warmed more than HadCRUT shows because HadCRUT doesn't cover the areas we know are warming more.

This sort of tells me that the satellites and surface have diverged slightly if we use UAH+RSS for the satellites and an HadCRUT+GISS for the surface.

[tacoman25]

So we can throw it out simply because it is warmer than the others and you don't like that? How about some reasoning. Also I don't even think HadCRUT is out for November yet.

Let's take a look at WHY GISS was so much warmer than HADCRUT in November. GISS shows extreme warmth over northern Canada, northern Siberia, and the Arctic Ocean. HadCRUT has these regions as no data.

Let's take a look at the weather pattern the past month. Unsurprisingly it was warmest in the areas for which HadCRUT has no data. Right where GISS said it was warm.

GISS looks warmer in Asia, shows more extensive/extreme warmth over the Arctic, warmer in South America, and warmer in Africa. About the only continents that seem to agree are Australia and North America.

As is often the case, the warmest anomalies are in the least populated regions. This is a consistent theme with GISS. It's actually kind of comical how you can see their massive extrapolations at work over huge areas...most of which have spotty/unrealiable historical records and sparse populations.

[skierinvermont]

GISS looks warmer in Asia, shows more extensive/extreme warmth over the Arctic, warmer in South America, and warmer in Africa. About the only continents that seem to agree are Australia and North America.

As is often the case, the warmest anomalies are in the least populated regions. This is a consistent theme with GISS. It's actually kind of comical how you can see their massive extrapolations at work over huge areas...most of which have spotty/unrealiable historical records and sparse populations.

Take off the glasses.. GISS looks colder in Asia to me. NCEP has a large area of +4C in Asia where GISS is only +2-4. NCEP has all of Greenland way above, mostly +4, GISS has the eastern third below. NCEP also has a little +4 in Africa, where GISS is only 2-4.

Also, GISS uses a 1951-1980 baseline while NCEP is using 1968-1996 which makes the anomalies on GISS larger since 1951-1980 was a cold period.

All areas that appear as white or orange below should appear as the darkest red on GISS. You can see GISS is too cold in northern Alaska, in the Middle East, around the black sea, and in Greenland. The extrapolations across Greenland, northern Canada, and the Arctic Ocean are largely correct and if anything miss some of the warmth in Greenland.

Also I know for a fact that when people have taken ECMWF or NCEP reanalysis data, it corroborates GISS and its extrapolations, regardless of whether or not you think it does for this particular month.

[BethesdaWX]

I would think the satellites are preferable too, but I have heard that the surface is supposed to warm more than the lower troposphere, which is what the satellites are measuring. I'm not sure how big of a factor that is.

All I know is that HadCRUT and GISS pretty much agree on the areas they both cover, but HadCRUT is missing data where it has warmed the fastest. This tells me that at least some combination of GISS and HadCRUT is preferable.

For the surface, I would take an average of HadCRUT and GISS. I think the surface has warmed more than HadCRUT shows because HadCRUT doesn't cover the areas we know are warming more.

This sort of tells me that the satellites and surface have diverged slightly if we use UAH+RSS for the satellites and an HadCRUT+GISS for the surface.

I see your point, and I do somewhat agree. My only subtle difference being that there are satellites that can measure surface temperatures, as well as LT.

Also, yes, HADCRUT is missing data where alot of "warming" has occured recently, but at the same time, it has more coverage than GISS, the arctic, sahara/africa, & asia being a few examples. Of course, that alone cannot explain the deviating trend 100%, so there must be a difference in measuring technique somewhere, or maybe extrapolation size? I personally prefer HADCRUT, based on the fact it has more Data, and Matches up well with satellites, albeit a little warmer.

[skierinvermont]

I see your point, and I do somewhat agree. My only subtle difference being that there are satellites that can measure surface temperatures, as well as LT.

Also, yes, HADCRUT is missing data where alot of "warming" has occured recently, but at the same time, it has more coverage than GISS, the arctic, sahara/africa, & asia being a few examples. Of course, that alone cannot explain the deviating trend 100%, so there must be a difference in measuring technique somewhere, or maybe extrapolation size? I personally prefer HADCRUT, based on the fact it has more Data, and Matches up well with satellites, albeit a little warmer.

I think 100% of the difference is due to the extrapolation distance. GISS extrapolates much farther. If you get rid of the extrapolations, it shows basically the same thing as HadCRUT. So it really comes down to: are the extrapolations correct or not?

By and large, on average, I think they are. They might overdo the warmth slightly, but HadCRUT is clearly too low given it leaves large parts of Canada, Siberia and the arctic ocean blank and we know they have warmed rapidly the last 15 years.

I know satellites can measure SSTs.. but what satellites measure land surface?

[salbers]

There really is nothing to back this up. It's very difficult to say one way or the other exactly how that period compared to today...but my point was that Arctic temps skyrocketed during that period as well. Arctic temperature variation has always been considerably greater than lower latitudes...which means that it would be very easy to read too much into how much the Arctic warmed over the past couple decades.

As for sea ice, I did post some records back to the 1950s over in Eastern Wx...

[BethesdaWX]

I think 100% of the difference is due to the extrapolation distance. GISS extrapolates much farther. If you get rid of the extrapolations, it shows basically the same thing as HadCRUT. So it really comes down to: are the extrapolations correct or not?

By and large, on average, I think they are. They might overdo the warmth slightly, but HadCRUT is clearly too low given it leaves large parts of Canada, Siberia and the arctic ocean blank and we know they have warmed rapidly the last 15 years.

I know satellites can measure SSTs.. but what satellites measure land surface?

This NASA satellite measures land surface temperature. Its faily new (only since 2000), but it seems to be pretty detailed. It is true that the satellite does not measure actual anoms.......but if we can measure land surface, I don;t see why we cannot develop something that would give 30M temos or something along those lines.

Unlike alot of satellites, it heard rumor that it measures without lag. Either way, I like this one.

Notice how it covers the entire globe

http://earthobservat...?d1=MOD_LSTAD_M

Latest Image (SEP 2010)