Apr 09 2008
March 2008 Temperatures: Related to La Nina?
The most recent round of temperature indexes were released, with the exception of the Hadley dataset. GISS, RSS, and UAH all show increases in the average global temperature. Skeptical Science postulates that it is because the La Nina event in the tropical Pacific Ocean has started to dissipate. In general, a La Nina will bring cooler global temperature, just as an El Nino will bring warmer global temperatures. He posts a graph which shows the latest 15 months of temperature measured by GISS and the SOI, a scalar indicator of the strength of the ENSO event.
There is general agreenment between the two curves. As SOI goes up, temperatures go down. His right axis is reverse, with positive values at the lower end and negative values at the top end. This is just to improve readability. However, the SOI has only recovered to its value in January. That is, the strength of the current La Nina should be about the same as it was in January. But the GISS global temperature anomaly in January was only around +0.1, while the March anomaly is +0.65 (I just eyeballed his graph.)
The “reversal” of the SOI does seem to suggest that the La Nina conditions in the tropical Pacific are weakening. And that would have a small effect on the global temperatures. But it should not have caused the large (~0.4C) jump in the temperature index. Instead of looking at just one value for the month, RSS and GISS provide monthly maps of the temperature anomaly.
Global Temperature Maps
|
This is the RSS temperature anomaly for January 2008. Notice the large area of blue in the tropical Pacific, indicative of the large positive anomaly in the SOI. The United States seems to be mostly covered with blue as well. But there are regions of the globe that have positive temperature anomalies even in a strong La Nina month. Both the northern and southern Pacific Ocean have slight positive anomalies. And there is a huge positive anomaly near Scandinavia and western Russia.
|
February 2008 looks almost exactly like January 2008. There is still a cold anomaly in the tropical Pacific, and the spatial pattern of the anomalies in general qualitiatively look the same.
|
March 2008 is a different story. Again, there is a cold anomaly in the tropical Pacific, and temperature patterns over much of the Americas are the same, that is not the dominant feature of this map. The difference from February to March is the extremely large anomaly, in terms of both temperature and spatial area, over most of Asia.
|
This is also shown in the GISS temperature data. But another interesting feature is found by comparing these two maps. The GISS temperature has no data for much of southern Africa. The RSS temperatures have no data for a chunk of what appears to by China and Mongolia (the Gobi Desert?). The area that GISS has no data shows a negative anomaly in the RSS data, and the area that RSS has no data shows a positive anomaly in the GISS data.
I’m unsure how or if the RSS and GISS global temperature metrics account for the missing data. If they are not somehow included in the global temperatures published recently, it would appear the the GISS temperature would need to be corrected to a slightly lower value and the RSS temperature corrected to a slightly higher value.
Conclusions
I don’t think that the temperature increase as seen in the globally averaged temperature is because of the “reversal” of the La Nina. Instead, it appears to be due to a large positive anomaly over much of Asia; most likely a regional weather feature. Small errors in the magnitude of the current anomaly could be caused by missing data in both GISS and RSS.
Map sources:
RSS maps from here and GISS map from here.
(No Ratings Yet)
Loading ...Related Posts:
17 Responses to “March 2008 Temperatures: Related to La Nina?”
To reduce spam, comments are automatically closed 30 days after the last comment. If you would like to comment on any closed thread, please use the contact form at the top of this page.
Even after subtracting all possible cyclical effects, there is still noise. Imagine how surpirsed we would be if there weren’t.
IIRC that missing piece of RSS is the Tibetan Plateau (since it’s high enough to interfere with the TLT channel).
Steve Bloom is correct that it is the Tibetan Plateau and is generally not included in the RSS stats. Since this data is not generally included, no adjustment is required on the RSS data to account for its absence.
On the other hand, can anyone confirm whether the missing chunks in the GISS information are generally included or excluded from the data set? Otherwise I would tend to agree with Atmoz that GISS needs to adjust for missing data.
Quick add, still haven’t seen the NCDC datasets on their webpage yet either.
I disagree that because RSS routinely excludes the Tibetan Plateau it does not need to be adjusted. Consider the hypothetical where the Tibetan Plateau has a temperature anomaly of +10C. The rest of the globe has an anomaly of 0C. Because they can’t sample the Tibetan Plateau they are underestimating the global temperature anomaly.
There are usually portions of Africa missing from the GISS data, but this is larger than normal.
Good post. Just to further debunk my own post, the global temperature reversal seems to lag the La Nina reversal.
Oops, I meant to say the other way around - La Nina lags temperature
[Reply: I was going to mention that except that it doesn't make sense to me why this should be. La Nina is a change in the surface temperature of the ocean which will affect changes elsewhere through teleconnections (gravity waves?). But the process of transporting the energy away from the tropical Pacific will take time, so I would expect the global temperature changes to lag ENSO by a small amount. I think the lag seen in the current plot may be due to noise in the climate system. You'd have to look at other ENSO events to conclude one way or the other if La Nina lags temperature or vice versa.]
I’d say that if they have routinely and consistently excluded the Tibetan Plateau it makes more sense to avoid spurious adjustments. We also know that they do not sample Antarctica. From other sources we know that Antarctica landmass temperatures are MUCH lower over the past decade, yet there is no imperative to adjust the RSS temperature anomalies lower to compensate.
In this case, I prefer to let the data be the data without introducing additional adjustments which could potentially corrupt the record.
Actually it just occurred to me, the simplest way to handle the omission of the Tibetan Plateau and Antarctica from the RSS dataset is simply to include it in the error bars. It’s simplicity itself to do so, the error for the missing data simply becomes a function of the expected min/max anomalies for the unsampled areas times the areal extent as a percentage of the Earth’s surface. No need to adjust the sampled data at all.
If you take a look at March 2008 temperature Anomoly Map at GISS at the 250km Resolution instead of the 1200Km one you will see there is no Data from most of Canada and the Arctic where it was colder, and Southern Africa (was slightly cooler), but there is huge abundence of data for Russia. Where it was warmer.
I find it strange that no data for Canada, but on checkign the March totals have not been done even for the EC website, but the daily records are there and on numerous long/lat checks versus the 1200KM anomoloy map the data used does not appear accurate. Fort Nelson, BC was -2.5C from the baseline, but reports at .988C on GISS based on long/lat comparison, this is a consistent fairly isolated station, so no proximity adjustments, and has a complete record, even at 1200Km resolution this should appear.
Or am I barking up the wrong tree here and missing something?
When you create the GISTEMP map, change the base period to 1979 to 1998. That will correspond more closely to the RSS map.
I didn’t create the maps. I borrowed them from their respective websites.
[...] a quanto appena detto, sembrerebbe emergere una inversione nel fenomeno La Niña. Tuttavia, in queata interessante analisi, è stata fatta un’ulteriore ipotesi. Andando a confrontare le anomalie termiche per zone [...]
The GISTEMP website allows you to create maps based on whatever base period you wish. To more closely match RSS, you should use the 1979 to 1998 base period.
http://data.giss.nasa.gov/gistemp/maps/
I don’t get it!
The Northern Hemisphere, particularly China, just experienced its coldest winter for decades during Jan, Feb and March. Images of the Arctic shows healthy new sea ice and land snow cover. However NOAA has posted a land surface anomaly of 2.23 degrees for March 2008, almost a record high.
From what I understand the 1901-2000 period baseline is continually fiddled with. Once a baseline is set, it should be left alone otherwise true condition monitoring becomes out of wack.
Is this shifting of the baseline used to keep the AGW dream alive?
When I take my temperature, I put a thermometer in my mouth. My temperature may be warmer or cooler in other areas. I don’t take several readings at different locations for an average temperature. Isn’t there a way in the 21st century to take the earth’s “temperature”?
This does look odd in light of the news stories of all the record snow falls in China in Feb and March. This is showing Northern China far above normal right where they had all the extreme snow events?
Looking at the RSS maps (top 3), my interpretation is that January and February were at or below average temperatures in most of China. By March northern China had warmed, but southern and Central China (e.g. Hong Kong, Shanghai) were still at about the average temperature.