I’m going out on a limb here. 2014 has been a very warm year. We’ve had a number of record setting months. But, a couple of months were also coolish, and November was one of them. December started out cool (like November ended) globally, but actually over the last few days the global average temperature has been going up. But, unless December gets really warm really fast, is is probably true that we will break some records but not all. This entire discussion, however, is problematic for a number of reasons.
How much does one year matter?
How warm or cold a given year is does not matter much for the overall trend. The upward march of global surface temperature is squiggly, but on average upward. Expect variation. Decadal trends are more important and more relevant. Global warming is continuing at the surface (sea surface and land based thermometers).
The graph above is a quick and dirty depiction of warming, using NASA’s GISS temperature record. The Y axis scale is anomaly in hundreds of degrees C, the X axis is months since the beginning of a 12 month moving average from January 1980. The point is just to show a) the increase in temperature just over the last few decades and b) how it squiggles up and down. We appear to be in an upward squiggle at the moment.
There are multiple temperature records
There are multiple “records” and they are assembled in slightly different ways and thus have slightly different data. They all show the same long term warming, and they all tend to correlate with each other. But they are slightly different. Some databases probably under-sample certain regions, for example. If we have a year that is very warm in relation to the most recent “hottest” year, it is unlikely to be so much warmer that it blows the previous record away by a huge number. Actually, that could happen, but it is more likely that some of the data sets are going to break the record while others do not.
Warmest year since when?
Not so much related to this specific year but important to keep in mind: This is the instrumental record. When we speak of record breaking years, we are usually comparing a particular year (like 2014) to each and every other year in a database that has been assembled from instrumental measurements. These databases variously go back in time to some point in the 19th century. They all start after CO2 was being release into the atmosphere at levels that probably matter, but way before the huge increase that has caused our present climate crisis. So, the instrumental records do measure, and as it turns out, demonstrate global warming. When we try to extend this record back in time, we lose track of variation in two ways. First, the proxyindicators (indirect measurements) used to estimate what the instruments would say were there instruments (and a time machine, presumably) have their own variation, so a number from ancient times is not perfectly comparable to a measurement from, say, 2013. Second, there is variation and conflation across time. We generally can’t point to a particular datum on a long term squiggle of global temperature from ancient times and say it represents a particular year.
What we can say is that for a particular period of time in the past the likely range of annual temperatures then was such that a given number (like, for example, this year’s annual global average) would likely be outside that range. When we do this, all of the recent years of global surface temperature are very very unlikely to have been exceeded by any actual annual temperature since the last interglacial (over 100,000 years ago). Most of the last one to two million years have seen mostly glacial and occasional interglacial conditions, but with the difference between those to climate settings increasing more recently and being less in the more distant past. It is possible that some years during interglacials over the last one or two million years exceeded our current warm temperatures (of the last couple of decades) but not many. As you go back in time, the chances of that increase because it was a bit warmer. For various reasons we are more confident about the last 800,000 years or so (as having few if any warmer years). When you get back to two to three million years there were time periods that may well have had lots of years warmer than the 21st century to date average. To get to consistent temperatures, for most years, warmer than present, you probably have to go back father.
So, we have this sentence: “2014 is the first or second warmest year since _______ .” That will likely be what we can say in a few weeks, after the data are measured, collected, processed, and made available. Or, we may be able to change that sentence to “2014 is the warmest since _____ in all of our instrumental records” or perhaps “2014 is the warmest since _____ in X out of Y of our instrumental records.”
Filling in the blank involves inserting the first year of the relevant instrumental records (such as “1880”) but it can also be filled in with older dated depending on how we feel about variation in the older, proxyindicator records. But it should also be rewritten a bit to include the probabilistic component.
This is just the surface temperature and does not reflect the totality of planetary warming
Personally, I think we should try to refer to these numbers as “surface warming” or the “surface temperature” and continuously remind people that this is only part of the story. How do you measure your body temperature? A thermometer stuck in an orifice will do. Or one of those magical strips on the forehead. Or an ear thermoeter. But these are all surface measurements of your body and are subject to error or variation. The better measurement is the one the medical examiner uses in estimating time of death; stick the thermometer into the liver (they have special pointy thermometers for this purpose, and only do it on dead people.)
The Earth’s liver is the ocean. Well, not exactly, but the majority of extra heat that happens because of the increased greenhouse effect caused mainly by human added CO2 ends up in the top 2,000 meters of the ocean (see this for a recent paper on the topic). At medium scales of time, the surface temperature does a good job of tracking the Earth’s temperature, but heat moves, to different degrees at different times, between the air and the sea, so on a year to year basis it is a rougher approximation. But it is the best approximation we’ve got, so we use it.
El Niño
Some of my colleagues have been snarking about changing the name “El Niño” to “El Annoyingo” or something like that. We are expecting an El Niño. We’ve been expecting it off and on for months. It has been a long time since a major El Niño, perhaps longer than we’ve ever had since good records have been kept. The Pacific Ocean looks very El Niño like in some ways but it is not an official El Niño. Whether or not you have an El Niño is something of a continuum.
It is generally felt that the effects of a coming El Niño are not particularly influencing the 2014 average global temperature, but if a real live El Niño emerges over the next few months, next year will be the record breaking year, as opposed to this year. Or both, one right after the other.
Other commentary
Most of the climate bloggers and publicly conversing scientists I know were probably planning to not talk about 2014 as a “warmest” or “second warmest” or “record breaking” year until after the data are in. But a couple of major news outlets have started talking about it, so now we are seeing some conversation on the topic and I’ve posted links below to some of that. Personally I think it was a mistake for major media to start talking about 2014 as a warmest year when close to 10% of the data were not in, the journos were looking only at one or two data sets, and to a large extent we are talking about weather not climate. Mark my words: If 2014 turns out to be second warmest in the majority of data sets, climate science deniers will make the claim that “they claimed it would be the warmest year, but it wasn’t. Checkmate, climate change!”
CNN’s premature claim: NOAA: 2014 is shaping up as hottest year on record
BBC’s premature claim: UN climate talks begin as global temperatures break records
Reuter’s premature claim: U.S., British data show 2014 could be hottest year on record
Sensible blogging on the topic:
2014 Headed Toward Hottest Year On Record — Here’s Why That’s Remarkable
Tallying 2014: Closing in on a Record?
from ScienceBlogs http://ift.tt/1rXJpbZ
I’m going out on a limb here. 2014 has been a very warm year. We’ve had a number of record setting months. But, a couple of months were also coolish, and November was one of them. December started out cool (like November ended) globally, but actually over the last few days the global average temperature has been going up. But, unless December gets really warm really fast, is is probably true that we will break some records but not all. This entire discussion, however, is problematic for a number of reasons.
How much does one year matter?
How warm or cold a given year is does not matter much for the overall trend. The upward march of global surface temperature is squiggly, but on average upward. Expect variation. Decadal trends are more important and more relevant. Global warming is continuing at the surface (sea surface and land based thermometers).
The graph above is a quick and dirty depiction of warming, using NASA’s GISS temperature record. The Y axis scale is anomaly in hundreds of degrees C, the X axis is months since the beginning of a 12 month moving average from January 1980. The point is just to show a) the increase in temperature just over the last few decades and b) how it squiggles up and down. We appear to be in an upward squiggle at the moment.
There are multiple temperature records
There are multiple “records” and they are assembled in slightly different ways and thus have slightly different data. They all show the same long term warming, and they all tend to correlate with each other. But they are slightly different. Some databases probably under-sample certain regions, for example. If we have a year that is very warm in relation to the most recent “hottest” year, it is unlikely to be so much warmer that it blows the previous record away by a huge number. Actually, that could happen, but it is more likely that some of the data sets are going to break the record while others do not.
Warmest year since when?
Not so much related to this specific year but important to keep in mind: This is the instrumental record. When we speak of record breaking years, we are usually comparing a particular year (like 2014) to each and every other year in a database that has been assembled from instrumental measurements. These databases variously go back in time to some point in the 19th century. They all start after CO2 was being release into the atmosphere at levels that probably matter, but way before the huge increase that has caused our present climate crisis. So, the instrumental records do measure, and as it turns out, demonstrate global warming. When we try to extend this record back in time, we lose track of variation in two ways. First, the proxyindicators (indirect measurements) used to estimate what the instruments would say were there instruments (and a time machine, presumably) have their own variation, so a number from ancient times is not perfectly comparable to a measurement from, say, 2013. Second, there is variation and conflation across time. We generally can’t point to a particular datum on a long term squiggle of global temperature from ancient times and say it represents a particular year.
What we can say is that for a particular period of time in the past the likely range of annual temperatures then was such that a given number (like, for example, this year’s annual global average) would likely be outside that range. When we do this, all of the recent years of global surface temperature are very very unlikely to have been exceeded by any actual annual temperature since the last interglacial (over 100,000 years ago). Most of the last one to two million years have seen mostly glacial and occasional interglacial conditions, but with the difference between those to climate settings increasing more recently and being less in the more distant past. It is possible that some years during interglacials over the last one or two million years exceeded our current warm temperatures (of the last couple of decades) but not many. As you go back in time, the chances of that increase because it was a bit warmer. For various reasons we are more confident about the last 800,000 years or so (as having few if any warmer years). When you get back to two to three million years there were time periods that may well have had lots of years warmer than the 21st century to date average. To get to consistent temperatures, for most years, warmer than present, you probably have to go back father.
So, we have this sentence: “2014 is the first or second warmest year since _______ .” That will likely be what we can say in a few weeks, after the data are measured, collected, processed, and made available. Or, we may be able to change that sentence to “2014 is the warmest since _____ in all of our instrumental records” or perhaps “2014 is the warmest since _____ in X out of Y of our instrumental records.”
Filling in the blank involves inserting the first year of the relevant instrumental records (such as “1880”) but it can also be filled in with older dated depending on how we feel about variation in the older, proxyindicator records. But it should also be rewritten a bit to include the probabilistic component.
This is just the surface temperature and does not reflect the totality of planetary warming
Personally, I think we should try to refer to these numbers as “surface warming” or the “surface temperature” and continuously remind people that this is only part of the story. How do you measure your body temperature? A thermometer stuck in an orifice will do. Or one of those magical strips on the forehead. Or an ear thermoeter. But these are all surface measurements of your body and are subject to error or variation. The better measurement is the one the medical examiner uses in estimating time of death; stick the thermometer into the liver (they have special pointy thermometers for this purpose, and only do it on dead people.)
The Earth’s liver is the ocean. Well, not exactly, but the majority of extra heat that happens because of the increased greenhouse effect caused mainly by human added CO2 ends up in the top 2,000 meters of the ocean (see this for a recent paper on the topic). At medium scales of time, the surface temperature does a good job of tracking the Earth’s temperature, but heat moves, to different degrees at different times, between the air and the sea, so on a year to year basis it is a rougher approximation. But it is the best approximation we’ve got, so we use it.
El Niño
Some of my colleagues have been snarking about changing the name “El Niño” to “El Annoyingo” or something like that. We are expecting an El Niño. We’ve been expecting it off and on for months. It has been a long time since a major El Niño, perhaps longer than we’ve ever had since good records have been kept. The Pacific Ocean looks very El Niño like in some ways but it is not an official El Niño. Whether or not you have an El Niño is something of a continuum.
It is generally felt that the effects of a coming El Niño are not particularly influencing the 2014 average global temperature, but if a real live El Niño emerges over the next few months, next year will be the record breaking year, as opposed to this year. Or both, one right after the other.
Other commentary
Most of the climate bloggers and publicly conversing scientists I know were probably planning to not talk about 2014 as a “warmest” or “second warmest” or “record breaking” year until after the data are in. But a couple of major news outlets have started talking about it, so now we are seeing some conversation on the topic and I’ve posted links below to some of that. Personally I think it was a mistake for major media to start talking about 2014 as a warmest year when close to 10% of the data were not in, the journos were looking only at one or two data sets, and to a large extent we are talking about weather not climate. Mark my words: If 2014 turns out to be second warmest in the majority of data sets, climate science deniers will make the claim that “they claimed it would be the warmest year, but it wasn’t. Checkmate, climate change!”
CNN’s premature claim: NOAA: 2014 is shaping up as hottest year on record
BBC’s premature claim: UN climate talks begin as global temperatures break records
Reuter’s premature claim: U.S., British data show 2014 could be hottest year on record
Sensible blogging on the topic:
2014 Headed Toward Hottest Year On Record — Here’s Why That’s Remarkable
Tallying 2014: Closing in on a Record?
from ScienceBlogs http://ift.tt/1rXJpbZ
Aucun commentaire:
Enregistrer un commentaire