Zeke alerted me that GISTemp posted. I’ve got an appointment to get my hair cut and then go out for my 26th wedding anniversary, but I have just enough time to post the current status in Q&A format:
- Monthly averages: Is this a record for hottest month anomaly in the record? Not even close; below you can see that many monthly anomalies lie above the red dashed line indicating this months temperature anomaly.
Is this a record for May? It’s close but no. May 1998 and May 2010 tie for hottest May in the record.
Is the trend computed since 1980 above or below “0.2C/decade”? Despite ending near the the top of El Nino, the trend is still less than “0.2 C/decade”.
- Is this the highest 12 month running average in the record? Yes. (I think it’s a safe bet it will continue to be break that record for at least 3 months.)
Is the 12 month lagging average temperature above or below the 12 month average of the multi-model mean extended using the models forced using the A1B SRES? At what appears to be a few months after the top of El Nino, the 12 month temperature is just above the multi-model mean. I’m guessing it will be there for at least a few months. After that, we’ll see.
Happy anniversary Lucia.
The model “ensemble” in figure two, I can help but look at the behavior it exhibits with respect to the Pinatubo eruption. Some of the models don’t have volcanic forcings, right? I wonder if this might make the model dip following the eruption appear smaller and less persistent than if all models had that forcing. I say this because I am used to seeing the models appear to cool too much and for too long after Pinatubo (maybe in part or in whole due to ENSO? I think the sensitivity/time response also plays a key part here), but in the above it looks like the opposite.
While the eruption is not the focus of the graph, I would want to see what models forced by volcanoes look like without adding in those without the volcanoes. at least in the vicinity of Pinatubo. I think you may have done this before, but I don’t recall.
Couple questions for lucia:
1. Isn’t a tie for the record by definition a record? Not a new record high, to be sure, but a record high nonetheless? Minor usage quibble.
2. Where do you get 0.2C/decade? I hear mostly 0.15C-0.2C, attributed to the IPCC, although I don’t know exactly how that figure is arrived at either.
I’m not sure. I admit that when I was writing, I was thinking “break the record”. A tie does not “break” the record.
Here is a definition: break the record phrases / idioms
1. Surpass a previous achievement, as in He was determined to break the record for the high jump. This usage is applied primarily to sports of various kinds. [1880s]’
The IPCC AR4 says “about 0.2 C/decade”. The TAR– which it supercedes, says 0.15. So, I compare to the newer document. The actual multimodel mean in the AR4 — which the text describes as the basis for the dark, solid lines in figures and values in tables — generally comes indicates more warming is projected by the AR4 than in the TAR.
You can verify this by reading the AR4 and the TAR themselves. (In fact, the AR4 written after the TAR discusses the difference in the magnitude of the projected warming.)
There are “some” (some of whom were involved in writing the AR4) who like to continue to compare data to the TAR– which is NOT the most recent document.
I’m comparing to the most recent nominal projections, not the lower estimates of warming in the documents that were superceded the AR4.
Shouldn’t we expect record global temperatures when El Nino events are placed atop the AMO?
All other things being equal (which they may not be), ifthe multi-model mean of the ensemble was not biased, we would exect the observations to lie above the multi-model mean during a real-earth El Nino and to lie below it during a real-earth La Nina.
The A1B SRES model projection, I’m assuming that’s ignoring nino? Is it also ignoring solar? In my toy model, solar influences cause temperatures to be about -0.05C below the 1951-1980 average. Nino is about +0.06C above the 1951-1980 average with regard to the average of the last 12 months.
Marginally, my model expects +0.020C per year. However, it is +0.016C when averaged over the last 30 years. I think the GCM models anticipate some level of acceleration in the global temperature growth rate. Hence, assuming the current growth rate is equal to the average 30 year growth rate, may be false, and the 30 year results may be in line with SRES models.
lucia (Comment#45546): The reason for my question: Since Nov 1981, the linear trend of North Atlantic SST anomalies (the Reynolds OI.v2 SST anomalies used by GISS) more than doubles the linear trends of the other major ocean basins. The second highest is the North Pacific:
http://i50.tinypic.com/iddpvd.png
The high linear trend of the North Atlantic should be caused by the AMO. And the AMO also influences land surface temperature anomalies in the Northern Hemisphere. RealClimate notes about the AMO, “This pattern is believed to describe some of the observed early 20th century (1920s-1930s) high-latitude Northern Hemisphere warming and some, but not all, of the high-latitude warming observed in the late 20th century.”
http://www.realclimate.org/index.php/archives/2004/11/atlantic-multidecadal-oscillation-amo/
So, with that in mind, I’ll ask my question again, Shouldn’t we expect record global temperatures when El Nino events are placed atop the AMO?
Add that to the bias at high latitudes GISS creates with their replacement of (lower trend) SST data with (higher trend) land surface data and it’s no wonder they’re finding record global temperatures. GISS deletes Arctic and Southern Ocean SST data in areas with seasonal sea ice and extends land surface data out over the open oceans when SST data is available. Discussed in this post:
http://bobtisdale.blogspot.com/2010/05/giss-deletes-arctic-and-southern-ocean.html
Congratulations – I hope you have/have had a lovely evening!
Congratulations Lucia (and your better half).
26yrs together this day n age (and still celebrating no less) says a lot about the both of you. My respects.
Re: Todd F (Jun 10 23:29),
This is a multi-model mean involving 22 AOGCMs some with multiple runs of one model. The individual models attempt to incorporate the physics that would create El Nino inside runs. However, none of the models try to time when El Nino would occur on earth. So, in some runs, El Nino might be near it’s peak in May 2010, in others it might be a deep La Nina. Averaging over an infinite number of runs and or models should result in an the “expected” trend in a statistical sense and it some sense, this means the multi-model mean does not account for El Nino or La Nina. In principle, the spread of models runs might, and so if the variation across the 55 or so runs was assumed to describe the range of earth weather, you’d expect the realizations (i.e. model runs) experiencing El Nino in May 2010 to be near the top of the spread– and lie above the average– and those experiencing La Nina to lie near the bottom of the spread and lie below the average.
The earth realization– which we know is experiencing El Nino– happens to be near the middle of the spread.
As for solar: The modelers were permitted to include or exclude solar. A few models included that going forward. I think most did not. I don’t know why different modeling groups made different decisions. To the extent that solar is missing, the projections could be expected to differ from temperature trends.
Will the earth temperature tend to lie above the mean when solar forcing increases owing to the solar cycle? Maybe. We’ll see, right?
If you examine the multi-model mean during the first 30 years, warming does not really accelerate during that period. For some SRES it does accelerate; for some it decelerates, but there isn’t non-linearity until later in the century.
Bob T,
The problem with your whole “AMO done it” argument is that the AMO is merely moving energy around the globe. As more energy is retained in the system, the AMO and other oscillations may sometimes appear to be adding energy when they are merely shuffling it around. Since AGW is heating the oceans, we’d expect oscillations to distribute that heat (though not necessarily in a way that we can predict).
As for expecting global records because of the AMO, I don’t think it is causing the increase in SH temperatures.
Happy Anniversary Lucia!
When was the last time you had to take or wear a sweater on your anniversary?
http://image.weather.com/images/maps/current/acttemp_600x405.jpg
Never in my whole life has it been too cold in Southern California for a backyard, bbq- pool party on June 12th (my birthday).
Liza–
I always take sweaters to restaurants. Most are air conditioned and you can’t be sure you might not end up with cold air cascading down on you.
I don’t think it’s particularly cold for June here. We’ve had cool days and warm days– just as we always do.
Hansen et al. 2010:
“We use ocean temperature change only in regions that are ice-free all year (a map of this area is included in Supplementary Material), because our data set is intended to be temperature change of surface air. Surface air temperature (SAT), measured at heights of 1.25-2 meters at meteorological stations, is of most practical significance to humans and it is usually SAT change that is reported in climate model studies. Change of sea surface temperature (SST) is a good approximation to change of SAT in ice-free ocean areas; climate model simulations [Hansen et al., 2007] suggest that long-term SAT change over ice-free ocean is only slightly larger than SST change. However, ocean water temperature does not go below the freezing point of water, while surface air temperature over sea ice can be much colder. As a result SST change underestimates SAT change when sea ice cover changes. Indeed, most climate models find that the largest SAT changes with global warming occur in regions of sea ice [IPCC, 2007]. Thus we estimate SAT changes in sea ice regions by extrapolating actual SAT measurements on nearby land or islands; if there are no stations within 1200 km we leave the temperature change undefined.”
http://data.giss.nasa.gov/gistemp/paper/gistemp2010_draft0601.pdf
Re: Andrew_FL (Jun 10 15:02),
Some of the models don’t have volcanic forcing. If you create a mean based only on models that include forcing the already distinct dips and recoveries after Pinatubo and other eruptions are more distinct.
Lets see what the other sources say.
UAH
1998 5 0.65
2010 5 0.54
RSS
1998 5 0.668
2010 5 0.588
So, clearly not the hottest May on record.
I think the question of whether or not May 2010 is a record high comes down to what level of accuracy you ascribe to the data. I believe that CRU claims that HADCRUT3 annual values are accurate to +/- 0.05°C, but I’m not sure what NASA claims for GISTemp. A few years back Gavin used the concept of an “unambiguous” record while discussing the accuracy of IPCC models in this post:
http://www.realclimate.org/index.php/archives/2008/05/what-the-ipcc-models-really-say/
In this post he claims that according to the models we should see a new record high that exceeds the previous record high by 0.1°C at least every 18 years (CI=95%). Regardless of whether or not 2010 turns out to be a nominal record, I’m pretty sure it will not beat the 1998 record by 0.1°C. This means that if we do not see an unambiguous record sometime in the next 6 years, even Gavin will have to admit the models have over estimated the rate of warming.
Time will tell.
PaulM (Comment#45595) June 11th, 2010 at 11:00 am
Lets see what the other sources say.
UAH
1998 5 0.65
2010 5 0.54
RSS
1998 5 0.668
2010 5 0.588
So, clearly not the hottest May on record.
.
no t the hottest. but we have a term for it: SECOND HOTTEST.
.
i can introduce you to the second best boxer. he will explain the concept to you.
.
.
————————————
.
In this post he claims that according to the models we should see a new record high that exceeds the previous record high by 0.1°C at least every 18 years (CI=95%). Regardless of whether or not 2010 turns out to be a nominal record, I’m pretty sure it will not beat the 1998 record by 0.1°C. This means that if we do not see an unambiguous record sometime in the next 6 years, even Gavin will have to admit the models have over estimated the rate of warming.
Time will tell.
.
while time will tell us, how hot 2010 was, time has already told us, that you didn t bother to check the facts.
.
http://data.giss.nasa.gov/gistemp/graphs/
.
January to May anomaly:
2010: 0.72°C
1998: 0.61°C
.
so at the moment, 2010 actually is more than 0.1°C warmer than 1998.
so what information about the future, makes you sure, that this will not be true for the whole year?
So, where’s all this May heat geographically?
It’s certainly not been in the UK.
Adam Gallon (Comment#45619) June 11th, 2010 at 1:50 pm
So, where’s all this May heat geographically?
It’s certainly not been in the UK.
.
GISS provides a tool. click the “make map” button.
.
http://data.giss.nasa.gov/gistemp/maps/
My problem, I guess, is that I think more like a chemical engineer than a statistician; and so I tend to be more focused on energy balances. So when I look at the graph of the 12 month moving average global temperature anomaly in comparison with the model runs, and compared with an overlay of Lucia’s interpretation of the model forecasts, I see it in a different way than Lucia does.
.
I see the models (which take into account energy balances) reasonably matching up with global temperature records, although clearly not capturing the variation due to ENSO and solar cycles. I believe the match would be even better, if the models were allowed to adjust the interchange of energy between the ocean and the atmosphere based on some factors of the relatively unpredictable (on a short term basis) ENSO cycle, and of the cyclical solar cycle.
.
Essentially the models are showing an imbalance in the planetary energy budget. The net thermal energy from the planetary budget imbalance distributes between the oceans (over 90% of the imbalance), ice sheets (4-5%), and land and soils heating (2%) with only about 1% of the thermal energy ending up as net heating of the atmosphere. Based on what we expect from a reasonable interpretation of GHG theory behind the model runs, along with the distribution of the net heating between the oceans, ice, land, and atmosphere; the energy imbalance will eventually distribute the net heat build such that the lowest portion of the atmosphere should heat about 0.2 deg C per decade. So when Lucia places an straight line of the slope 0.2 deg C per decade on the graph, she is making the assumption that the net heating is continuous and constant over the time period she has placed this line (forecast).
.
But AGW theory clearly show that aerosol forcing impacts are important, particularly aerosols due to tropical volcanic eruptions that release massive amounts of sulfates (like Pinatubo). The models show that the planetary energy imbalance shifts from a net gain of energy to a net loss of energy for at least a year after an eruption like Pinatubo, and the accumulated energy imbalance doesn’t get back to break-even with the day of the eruption for almost three years (see the dip after Pinatubo in the graph above). During this three year time period, the net accumulated heat in the heat sinks discussed above is zero.
.
The upwardly sloped 0.2 degC “forecast” line that Lucia has overlaid, should be interrupted for three years, and the upper section of the line should be displaced three years into the future at the time of the Pinatubo eruption. If readers print out the graph, and displace the upper section, they will see that the warming since the Pinatubo effects were washed out (in late 1994), is consistent (and likely has exceeded) the rate of 0.20 deg C per decade.
.
This more accurate graph clearly shows the models have been amazingly accurate, and that the current global temperature data contradict the lukewarm position that the rate of warming will be substantially lower than AGW theory and models predict. Comments anyone?
Re: Paul K2 (Jun 11 16:44),
There is no assumption; the straight line is a trend shown for comparison purposes. Putting the line on spares people from having to put their own on and see how the trend happens to compare the the nominal numerical value in the AR4.
If you are saying it would be better to plot the multi-model mean, which shows this feature, that’s plotted with the 12 month lagging average temperatures.
For what it’s worth, if you fit a trend to the multi-model mean starting in 1980 and ending now, the trend exceeds 0.2 C/decade slightly. If you fit the to each model, then average over those, that exceeds 0.2 C/decade. That happens even though Pinatubo is already reflected in the models runs.
The blue circle in the key below the graph is labelled “jan”. Shouldn’t that be “May”?
Tom–
Yes. For some reason I filled the wrong cell in the legend. The circled months are may. I’ll swap out figures tomorrow.
UAH and RSS represent temperature of the lower troposphere, while GISTEMP represents the temperature at the surface. A record for one does not require a record for the other.
Boris (Comment#45583): You wrote, “Since AGW is heating the oceans…”
It is? There is little to no evidence in the NODC (Levitus et al) OHC data of anthropogenic influence. All you have to do is divide the oceans into subsets to see when and why OHC rises. Refer to:
http://bobtisdale.blogspot.com/2009/09/enso-dominates-nodc-ocean-heat-content.html
And:
http://bobtisdale.blogspot.com/2009/10/north-atlantic-ocean-heat-content-0-700.html
And:
http://bobtisdale.blogspot.com/2009/12/north-pacific-ocean-heat-content-shift.html
And if you’re wondering why the NODC OHC data (0-700m) has been falling recently, it’s because the dataset with the highest linear trend since the mid-1950s, the North Atlantic, is no longer warming. In fact, it’s cooling and has been for the past few years, just as one would expect with an OHC dataset influenced by AMOC:
http://i50.tinypic.com/2eexa8w.png
The NODC OHC data is available through the KNMI Climate Explorer. The NODC allowed KNMI to post it there so it could be examined:
http://climexp.knmi.nl/selectfield_obs.cgi?someone@somewhere
If and when any of the other groups who have recently produced papers about OHC want to upload a long-term OHC dataset to KNMI, we can then divide it into subsets and see what influences its variations.
What empirical evidence supports Hansen’s model to justify making up data derived from the Arctic?
According to Polyakov et al, observations do not support the plastic banana science from GISS.
Observationally based assessment of polar ampli cation
of global warming
So why again is it that since 2009 GISS is showing runaway warming compared to HadCRUT?
sod-
January to May anomaly:
2010: 0.72°C
1998: 0.61°C
.
so at the moment, 2010 actually is more than 0.1°C warmer than 1998.
.
Yes, but only if you only look at GISTemp. If you look at any of the other datasets (HADCRUT, UAH, RSS) or even average all four together, 2010 is not warmer than 1998. Some might call only looking at the one dataset that supports your position cherrypicking.
Yes, but only if you only look at GISTemp. If you look at any of the other datasets (HADCRUT, UAH, RSS) or even average all four together, 2010 is not warmer than 1998. Some might call only looking at the one dataset that supports your position cherrypicking.
.
somebody made the claim, that 2010 would not be 0.1°C warmer than 1998. i pointed out, that at the moment it is, in the GISS dataset.
.
it is really strange that you attack me, and not the person who made an obviously false claim.
.
i am not sure, that you can transfer the claim that Gavin made, to the satellite datasets.
.
http://www.realclimate.org/index.php/archives/2008/05/what-the-ipcc-models-really-say/
.
but check it for yourself…
The record in question is GISTEMP, as the title of this thread makes clear. May 2010 tied the record for the hottest record in GISTEMP. The preceding 12 months is the warmest 12 month period in GISTEMP. The first 5 months of 2010 is the warmest first 5 months of any year in GISTEMP. If you want to discuss the non-recordness of the other analyses, you need only scroll down.
Re: cce (Jun 13 12:19),
You are correct that my discussions of records are whether or not something is the highest recorded value in GISTemp. They are not discussions of whether we can be confident that the highest recorded values in GISTemp correspond to the highest temperature experienced on earth during the period in the record. Both questions are worth discussing.
We can figure out the highest recorded value in GISTemp by inspecting that record– and that’s what I did when posting. Figuring out whether the highest recorded value in GISTemp corresponds to the hottest month or year ever is more difficult. I didn’t try to do that in my post and don’t ever plan to try when writing a post immediately after GISTemp updates.
Adam Gallon (Comment#45619) June 11th, 2010 at 1:50 pm
I’m with you!
I reject their reality and replace it with mine.;)
The “whole globe” wasn’t warming in May.
http://1.bp.blogspot.com/_6aSqRkDo2y4/S_DLolUocbI/AAAAAAAAAZU/Wq8h0gUN-Ps/s1600/Cal+temp+5-1+to+5-15+2010.gif
I love the graphic generated from sod’s link. Where you and I live it’s blue/white “cold” and places like Siberia are shown as “hot” (and there are less surface stations and people in these places) and “hot” is something like 27 degrees F instead of 23 degrees F. lol Antarctica- temperatures-bright purple (-9.3) and cold white/blue areas right next to raging “hot” (+3) red area.
The “whole globe” isn’t cooperating. 😉
Liza,
Go here:
http://www.remss.com/msu/msu_data_monthly.html
That will show you the relative temperatures of the lower troposphere as calculated by RSS.
If you click on “anomaly” you will see which areas of the world are warm or cold for this time of year, which is what we are interested in.
I have been thinking lately that I, since I am in my forties, seemingly have experienced unprecedented warming in my lifetime; an experience that no other generation apparently has felt. Oddly, I don’t feel it nor do I see it. O, yes we have had a very warm day, recently . But I skated last winter on the fenn. The ecosystems I walk through do not change very much, nor do they collaps. So if the 0.8 K change of temperature (or something like that) I have endured in my lifetime basically doesn’t show all that much effect, yet is considered unprecedented warming, how is it possible that seemingly far greater effects such as 1540 are considered to be normal.
In other words, I see anecdotical evidence of far greater climatic change (drought in the early sixteenth century and so on) being considered normal cycles, yet this unprecedented climate change hardly produce anything that might be considered worth an anecdote.
cce (Comment#45828) June 14th, 2010 at 10:34 am
Do you believe the technology of observing the Earth from space was there for the base period of 1951-1980 ( used to determine those “anomalies” on the charts such as sod linked to for May) ? There’s a pull down menu at your link that only goes back to 1978. 1978 for May has “no data” so apparently there is only May 1979 and 1980 data you are using to decide what is “anomalous” is.
liza (Comment#45889) June 15th, 2010 at 6:26 am
cce (Comment#45828) June 14th, 2010 at 10:34 am
Do you believe the technology of observing the Earth from space was there for the base period of 1951-1980 ( used to determine those “anomalies†on the charts such as sod linked to for May) ? There’s a pull down menu at your link that only goes back to 1978. 1978 for May has “no data†so apparently there is only May 1979 and 1980 data you are using to decide what is “anomalous†is.
.
liza, you have been around a little, but seem to be completely unable to pick up any knowledge in the process.
.
my link was to a map by NASA GISS. i used that graph, because we are discussing GISS in this topic. and someone made the stupid claim, that he has not noticed the heat.
.
the map above is from a different source. RSS is satellite data. it has a different baseline. seems to be “averaging that month from 1979 through 1998”
.
——————–
.
In other words, I see anecdotical evidence of far greater climatic change (drought in the early sixteenth century and so on) being considered normal cycles, yet this unprecedented climate change hardly produce anything that might be considered worth an anecdote.
.
you are being confused by LOCAL climate changes from the past. but we are talking about GLOBAL changes here.
.
most people will NOT notice a temperature rise, because weather variability obfuscates the change, and we basically don t notice day minimum temperature at all. (we sleep at that time).
.
the people who look at evidence (glaciers, flowers, thermometers) and or depend on climate (farmers, ski resorts..)actually DO notice the change.
@sod
But those are mere statistical blips. When I disect all the claims off horrors of climate change they all turn out to be projections. We see milimeters of sea level change a year, yet project disasterous yards of sea level rise. We see a very small, though statistically significant, change in the days that this or that flower start to blossom, yet we project enormous extinction-waves. Mind you, there has been almost a whole degree of climate change: You would expect to see far more then this. Especially since this is unprecedented climate change.
Farmers? Well, if you consider every drought on earth the result of our unprecedented climate change you have huge impact. But then what is the difference with my considering LOCAL climate changes? I mean, the Sahel isn’t having a drought. Actually it’s doing quite well.
sod (Comment#45892) June 15th, 2010 at 6:58 am
I am asking cce a question about his graph; and he says “which is what we are interested in” not yours which is what apparently they/he/she/and the rest are not interested in when they says “we”.
As for the rest of your reply what a bunch of bologna with that “most” people stuff. Some of us keep sharing that we are cold; from different parts of the world too. And speaking of ski resorts here in California they are extending the season until July!
peeke
In other words, I see anecdotical evidence of far greater climatic change (drought in the early sixteenth century and so on) being considered normal cycles, yet this unprecedented climate change hardly produce anything that might be considered worth an anecdote.
= = = = = = = = = = =
Drought in your local region in 1540 meant no food to eat. Drought in 2010, in your local region, means farmers apply for state aid or insurance to cover their losses and your salad still flies in from 2000km away.
At a guess, others ofcourse will disagree.
Dorlomin–
It’s certainly not looking like drought in Illinois this year!
California’s breadbasket is now a basket case. Farm owners managed even in drought years but are now are standing in food bank lines because of a fish. Who needs drought when you’ve got the Delta Smelt? (and don’t forget the FREEZES we’ve had in the last couple of years too)
Dorlomin,
Maybe you are right. On the other hand, consequences of drought in 1540 also were fixable. There was intensive cereal trade back then.