Climate Feedback

Confusion on Climate Variability and Trends

(Posted by Olive on behalf of Roger)

Even the venerable New York Times is prone to completely botching a discussion of the science of climate change. In a front page article today, the NYT reports on how the National Arbor Day Foundation has updated plant hardiness maps to reflect recent changes in climate. (A plant hardiness map presents the lowest annual temperature as a guideline to what plants will thrive in what climate zones.) The NYT misrepresents understandings of variability and trend and in the process confuse more than clarify.

The new map updates a 1990 USDA map based on 1974-1986 data, and replaces it with <a href = “http://www.arborday.org/media/zones.cfm”>data from 1990-2006. In most places the range of increased average minimum temperature has moved north as can be seen from a difference map between the two time periods. The difference map, shown here, has the horizontal lines because the zones used are so broad — 10 degrees — that the differences are only noticeable at the margins of the zones.

The New York Times reports that these differences can all be attributed to human-caused climate change, using the case of Atlanta as an illustration:

Using data from the National Oceanic and Atmospheric Administration, the Arbor Day map indicates that many bands of the country are a full zone warmer, and a few spots are two zones warmer, than they were in 1990, when the map was last updated.

Atlanta, which was in Zone 7 in 1990, is now in Zone 8, along with the rest of northern Georgia. That means that areas in the northern half of the state where the average low temperature was zero to 10 degrees Fahrenheit are now in a zone where the average low is 10 to 20 degrees. A scientific consensus has concluded that this warming trend has largely been caused by the human production of heat-trapping gases.

Because the zones span 10 degrees (or 5 degrees in the case of the 1990 USDA map) and the largest change shown on the difference map is 2 zones (i.e., >10 degrees!, now corrected), then clearly no location has jumped 2

zones! This is just an error.

More important than this simple mistake is the claim in the NYT that the changes in temperature observed in Atlanta can be attributed to human-caused greenhouse gases. In fact, the IPCC argues that it needs 30 years of records to detect trends, much less make attribution. In fact, the IPCC report just out has reported that the U.S. southeast has actually cooled over the period of record as shown below.

usseipcc1901-2005.jpg

The underlying issue has to do with understanding the role of human-caused climate change in the context of climate variability on long time scales.


I spent last week at a meeting on the “Societal Impacts of Decadal Climate Variability” and it is clear that some scientists view decadal climate variability as an underappreciated subject. There is good reason for such concerns as many important societal decisions — like flood insurance, engineering design, and risk models — depend up understanding variability in the context of human-caused changes. The NYT article includes a passage that suggests that some people view the trend to be most important, while others think variability matters more:

The Agriculture Department is in the process of redoing the [planting guideline] map itself. But critics have taken issue with the department’s decision to use 30 years of temperature data, saying it will result in cooler averages and fail to reflect the warming climate. The 1990 U.S.D.A. map used 13 years of data; the Arbor Day map used 15 years ending in 2004.

Cameron P. Wake, a climatologist at the Climate Change Research Center of the University of New Hampshire, said “a 30-year period would include several cycles of multiyear effects like El Nino, with an underlying assumption that climate is stable and varies around a mean. Warming, on the other hand, is not variability, its a long term trend,” Dr. Wake said. “I would say the U.S.D.A. doesn’t want to acknowledge there’s been change.”

Kim Kaplan, a spokeswoman for the departments Agricultural Research Service, defended the decision to use 30 years, saying the longer time period would strike a balance between weather, which can vary greatly from year to year, and climate.

For planting your garden such debates may not be so relevant, but for decisions related to risk management and adaptation to climate more generally, this debate is central. Effective planning depends on knowing what range of possibilities to expect in the immediate and longer-term future. Use too long a record from the past and you may underestimate trends. Use too short a record and you miss out on longer time-scale variability. Throw in the politicization of the issue where trend partisans are associated with advocacy for action on greenhouse gas reductions and variability partisans are associated with those opposing action, and you have a very challenging decision making context indeed.

Out of this, one thing seems clear — improved decision making is unlikely to occur if basic issues of climate science are simplified to such a degree that they are misrepresented, especially by an authoritative source like the New York Times.

Roger Pielke, Jr.

Center for Science and Technology Policy Research

University of Colorado

Comments

  1. Report this comment

    Eli Rabett said:

    Except that if you look at trends from say 1980 to 1990 or 1990 to 2007 in the southeast US you get a significant warming trend vs a standard 1951-1980 period.

    Most of the individual stations in the Southeast US were warm in the beginning of the 20th century, followed by a cooling to about 1970 and then a warming.

    For planting a garden probably the most important thing is the date of the last freeze. If you click on the stations you can see the historical record for several Georgia cities.

    So the question occurs, why use a map of anomalies from 1901? What was the point?

  2. Report this comment

    ecothreat.com said:

    arctic ice melts faster than predicted

    Arctic Ocean sea ice is melting faster than even the most advanced climate change models predicted, a new study concludes. The work, published today in the journal Geophysical Research Letters, used the current models to retroactively predict sea-ice d…

  3. Report this comment

    Deltoid said:

    Nature climate blog off to rocky start

    Nature has started Climate Feedback, a blog on climate change. One of the first posts is by Roger Pielke Jr, who claims Even the venerable New York Times is prone to completely botching a discussion of the science of climate…

  4. Report this comment

    richard said:

    from Deltoid:

    However, it is Pielke Jr who has completely botched the discussion.

    He writes:

    The new map updates a 1990 USDA map based on 1974-1986 data, and replaces it with data from 1990-2006. In most places the range of increased average minimum temperature has moved north as can be seen from a difference map between the two time periods. The difference map, shown here, has the horizontal lines because the zones used are so broad — 10 degrees — that the differences are only noticeable at the margins of the zones.

    The New York Times reports that these differences can all be attributed to human-caused climate change, using the case of Atlanta as an illustration:

    Using data from the National Oceanic and Atmospheric Administration, the Arbor Day map indicates that many bands of the country are a full zone warmer, and a few spots are two zones warmer, than they were in 1990, when the map was last updated.

    Atlanta, which was in Zone 7 in 1990, is now in Zone 8, along with the rest of northern Georgia. That means that areas in the northern half of the state where the average low temperature was zero to 10 degrees Fahrenheit are now in a zone where the average low is 10 to 20 degrees. A scientific consensus has concluded that this warming trend has largely been caused by the human production of heat-trapping gases.

    Because the zones span 10 degrees (or 5 degrees in the case of the 1990 USDA map) and the largest change shown on the difference map is 2 degrees, then clearly no location has jumped 2 zones! This is just an error.

    Indeed it is, but the error is Pielke Jr’s. The difference map shows not the change in temperature, but the change in zones (Hint: the legend says “Zone Change”). The largest change is not 2 degrees as Pielke Jr thinks but 2 zones. NYT 1 Pielke Jr 0.

  5. Report this comment

    Roger Pielke, Jr. said:

    Thanks for the comments! A few replies:

    1. Yes, I did write degrees when I should have said zones, apologies.

    2. Eli- The focus on the long term is because this is the time-period that the IPCC uses for attribution. I do not doubt that the climate has changed since the 1970s in the US southeast. The question is whether this decadal change can be attributed to human emissions of GHGs, as presented by the NYT. According to the IPCC on this short-time scale at this regional level, it cannot be. As I said in the piece the choice of what base period to use is tricky and creates errors whether you use a short or long record.

    3. Richard. Tim Lambert suggests on his blog that the NYT story indicates that average minimum temperatures in the US have increased by 5 degrees!! This is exactly the misinterpretation that I wrote of. An increase in 5 degrees nationally over two decades would indeed be a big deal. However, it has not occurred. I correct this and other misinterpretations over at that blog.

    The bottom line is that making sense of decadal variability in the context of long-term trends for adaptive decision making can be a tricky proposition.

    Thanks all!

  6. Report this comment

    richard said:

    In other words, you owe the NYT an apology.

  7. Report this comment

    Roger Pielke, Jr. said:

    Richard- Um, no.

    For anyone interested in the science:

    On trends in minimum temperatures:

    Vose, R. S., D. R. Easterling, and B. Gleason (2005), Maximum and minimum temperature trends for the globe: An update through 2004, Geophys. Res. Lett., 32, L23822, doi:10.1029/2005GL024379. http://www.agu.org/journals/gl/gl0523/2005GL024379/

    If you don’t have access to GRL: http://ams.confex.com/ams/pdfpapers/100744.pdf

    That should settle the question of the magnitude of minimum temperature increases in the US.

  8. Report this comment

    Hank Roberts said:

    Shame, you should cite with a link to the actual source of anything not to your own copy on your own page.

    Please post the link to the original IPCC page from which you took this chart:

    “In fact, the IPCC report just out has reported that the U.S. southeast has actually cooled over the period of record as shown below.”

    Rule one of database sanity: one record, many pointers. Otherwise you’ll get people posting claims that aren’t accurate because they can make it hard for people to look up the original source.

  9. Report this comment

    Eli Rabett said:

    As an avid, although very amateur gardener, I have to say that the average temperature over a winter is a very different animal than the average minimum temperature in any particular winter. A point that J. Hamilton made over at Deltoid. It is the latter that kills your plants and I have a lot of dead plants to show for it.

    In other words, why are you assuming that DOA is a bunch of carrots?

  10. Report this comment

    bubba said:

    [comment removed at commenter’s request]

  11. Report this comment

    Jim Angel said:

    Thanks for the though-provoking article Roger. As a climatologist and gardener, I find this a very interesting subject. Watching my own yard, I would say that both the zone concept and date of last freeze don’t come close to measuring the impact of weather/climate on plants.

    In Illinois, replacing the 1974-1986 data with data from 1990-2006 means you leave out our really cold and snowy winters in late 70s and early 80s. We have been experiencing a winter warming trend ever since. However, during the milder 1990-2006 period, we set or tied the record low at several station in January 1999. This includes a -25F in Champaign-Urbana in what is now considered a Zone 6. In addition, we have had a couple of episodes in recent years where temperatures have remained well below freezing throughout the day for several days in a row without any protective snow cover. I’ve seen a lot of winter dieback on shrubs as a result. These particular episodes occurred in warmer than normal winters (based on the 1971-2000 data).

    Considering the odds of extreme temperatures or joint probabilities such as cold temperatures and no snow are probably more useful in plant selection. In addition, my tolerance of risk is different between planting a small perennial or a large tree. In fact, with a large tree you might be better off examining the longer time period to get a better sense of the risk of failure. Personally, I lean towards native species since they have proven their robustness to Illinois climate.

    This is a classic challenge in how to use climate data for decision making when you have both climate change and variability (as well as data limitations). I deal with this constantly in my line of work. Roger, I’ll get back to you when I have all this figured out.

    Jim Angel

    Illinois state climatologist

  12. Report this comment

    Hank Roberts said:

    Roger, why did you show only half of their Fig. 3.9 from p. 250?

    The caption says:

    Figure 3.9. Linear trend of annual temperatures for 1901 to 2005 (left; °C per century) and 1979 to 2005 (right; °C per decade). …."

    The white + marks indicate areas where the data is considered significant. Anything without the + isn’t.

    You’re showing the left half but not the right half of the image inline here.

    This is why links to original sources matter, to read in context.

    Not everyone uses Firefox yet; let’s see if I can share one of its advantages, the “pdfdownload” extension.

    If it works, this link should work for 24 hours to get the whole image, for anyone:

    http://pdfdownload.04340.com/070505/tmp-4lLoLs/260302016.png

    This for a while will get the whole paper, quick-changed to HTML by Pdfdownload:

    http://pdfdownload.04340.com/pdf2html.php?url=http://ipcc-wg1.ucar.edu/wg1/Report/AR4WG1_Ch03.pdf&images=yes

    (Images are deleted from the server at the pdfdownload site after 24 hours)

  13. Report this comment

    Eli Rabett said:

    Bubba, will you please introduce me to my paymaster? I could use the carrots.

    In any case I found this blog through Planet Fleck. There is something non standard about its html (are you using microsoft extensions or something) as there is a bunch of &&&&& interspersed in the text there.

  14. Report this comment

    Eli Rabett said:

    Hey folks, if you are going to run a blog, how about paying attention. Bubba’s post has been up for almost a day. Eli posted this morning asking for a pointer to his paymaster so he could get some carrots. The bunnies are hungry.

  15. Report this comment

    Roger Pielke, Jr. said:

    Thanks Jim for the comment. I know that the state climatologists have to grapple with this challenge just about every day!

    Thanks Hank for the question. The answer is that I am not challenging the trend from 1979-2005, so that figure is repetitious. I am challenging the issue of attribution and the fact that the notion of decadal variability – yes it does exist and it does matter — was overlooked in the NYT article.

    The IPCC is quite clear that “trends” over less than 30 years, particularly for specific regions and locales cannot be readily attributed to greenhouse gas emissions. The southeast US is a particularly good example of this.

    Please note that this in no way challenges the consensus of a human effect of the climate system. It is in fact the consensus!

    Those who think that climate will simply get steadily and incrementally warmer everywhere as the future unfolds will set themselves up for some nasty surprises.

    Thanks!

  16. Report this comment

    Aaron said:

    Roger Pielke, Jr. wrote, “That should settle the question of the magnitude of minimum temperature increases in the US.

    Does it really? Because it would only suggest to me that you are confusing the distinction between annual minimum and average daily minimums temperatures.

  17. Report this comment

    Eli Rabett said:

    So Bubelah’s post has been up there all day and Eli STILL has not heard where his paymaster is. This is getting serious folks, especially for a wanna be high class blorg. The bunnies are getting VERY cranky without their carrots.

  18. Report this comment

    bubba said:

    [comment removed at commenter’s request]

  19. Report this comment

    Eli Rabett said:

    Allow me to point out the serious implications of what Roger said

    Those who think that climate will simply get steadily and incrementally warmer everywhere as the future unfolds will set themselves up for some nasty surprises."

    Now we are very sure that one of the places it will get much warmer is the Arctic including Greenland. That has 6 m or so of nasty surprises. But more urgently the fact that it will get much warmer some places much faster, coupled with some bad luck on natural variability means that dealing with climate change is a very urgent problem.

  20. Report this comment

    Aaron said:

    bubba wrote, I suppose it might be explained by your having attained tenure at an institution with lax requirements, but I can think of no other rational explanation for the prodigious volume of posts you litter across the internet.

    I’m not sure if you would consider it rational or not, but some people really do care passionately about the fate of our planet. So much so that the would spend hours of their own free time expounding on the subject. It’s revealing that you would accuse Rabett of “quibbling over minutiae and semantics” while Pielke Jr. has been caught mid-chicane. Pielke Jr. confuses zones with degrees and then tries to cover his tracks by citing incommensurable average annual minimum temperature data. On the Deltoid blog, he even went so far as to compare the credibility of the National Arbor Day Foundation with that of Ahmed Chalabi. So let’s not quibble about Eli Rabbet’s style of prose, shall we?

  21. Report this comment

    Roger Pielke, Jr. said:

    Eli- I agree! While me may disagree on how best to deal with this challenge, we should all agree that our discussions are improved with some degree of mutual respect and common courtesy. Best wishes …

  22. Report this comment

    Ike Solem said:

    The issue here is really one of how data is reported and presented – especially as it relates to the reporting of anomalies as data – and yes, as this shows, the baselines DO matter.

    I raised this issue with Roger Pielke Jr. before, as it relates to the use of baselines over at NOAA:

    As far as the NOAA issue goes, the use of a baseline to calculate temperatue anomalies relates to the issue of what is meant by ‘anomaly’. Now, in 2000 NOAA decided to start using the time period 1971-2000 as the baseline for calculating their anomalies, in contrast to the widely accepted use of the 1961-1990 time period for their baseline.

    The differences in the two anomalies are fairly dramatic; see for example

    using NOAA’s 1971-2000 baseline, summer 2006:

    http://www.emc.ncep.noaa.gov/research/cmb/sst_analysis/images/archive/monthly_anomaly/monanomv2_200606.png

    Using the 1961-1990 baseline: summer 2006

    http://www.bom.gov.au/bmrc/ocean/results/SST_anals/SSTA_20060625.gif

    Also, NOAA uses the 1971-2000 baseline for their 2005 Arctic Climate Report, but does not explicitly discuss this. Obvious, this gives a perception that warming in the Arctic is less severe then it actually is.

    When the word ‘anomaly’ is used in public discourse, it is taken to mean ‘deviations from normal behavior’. So, does this issue represent ‘cherry-picking’ or deliberate manipulation of scientific data for political purposes, in your expert opinion?

    Pielke response: “#65 Ike- Thanks for the clarification . . . the choice of base period for looking at anomalies makes no scientific difference, which may be what you are getting at. It does as you suggest convey a perceptual difference when the graphs are displayed. So were I a political strategist I’d suggest using an older period to emphasize trends and a newer period to downplay them. Such a decision necessarily must be made on nonscientific factors. There is no purely scientific answer to the question of what baseline to use.”

    What seems to be going on in the above post is that Pielke is choosing a baseline that reduces the trend (i.e. back to 1901) – in other words, as a political stragist, he is following his own advice and choosing the baseline that supports his political goal – which is apparently to get attention as a ‘noted climate skeptic’.

  23. Report this comment

    Ike Solem said:

    Actuallt, for anyone “interested in the science”, the following paper is more useful than the one referenced by RPJr.

    Large-scale changes in observed daily maximum and minimum temperatures: Creation and analysis of a new gridded data set, Caeser et al GRL 2006

    In particular, look at figure 6. Looks like there is indeed an upward trend in minimum temperatures, with regional variation – i.e., there are no problems with the NYT story.

  24. Report this comment

    Hank Roberts said:

    > I am challenging ….

    but Dr. Peilke you draw a big, red circle on the image.

    Inside that there is only a tiny area marked with the white + symbols indicating the data is considered valid.

    Where do you explain this to readers who stop with your main posting? Nobody reads the Comments, who’s not already engaged in digging for details.

    It’s the people who don’t chase down these hairsplitting threads who ought to be given the clearest possible info, eh?

    Almost the entire circle you’re pointing to is one where they say the very pale gray indicates nothing of significance.

    Where’s the challenge? The only cooling in the image that’s marked as significant is a few counties along the Gulf Coast.

  25. Report this comment

    Hank Roberts said:

    Look, who at this blog actually read the caption?

    Quoting it in full:

    ""Figure 3.9. Linear trend of annual temperatures for 1901 to 2005 (left; °C per century) and 1979 to 2005 (right; °C per decade). Areas in grey have insufficient data to produce reliable trends. The minimum number of years needed to calculate a trend value is 66 years for 1901 to 2005 and 18 years for 1979 to 2005. An annual value is available if there are 10 valid monthly temperature anomaly values. The data set used was produced by NCDC from Smith and Reynolds (2005). Trends significant at the 5% level are indicated by white + marks."

    Look inside the big red circle on the part of the figure reproduced in the original post. What do you see? Gray.

    “… Areas in grey have insufficient data to produce reliable trends.”

    That’s for the illustration labeled at the top of this thread.

    Your readers here will mistakenly think the word “Cooling!” pointing to the big red circle means you have data reliably showing cooling inside the big red circle.

    Where is it? The gray means insufficient data, the lack of + marks means no trend significant at the 5% level. What is inside the big red circle?

    Could you maybe enlarge it for the readers and include the caption?

    Maybe this does make sense, but I think your explanation ought to be edited in the original post, because it’s not clear.

    I also don’t get the argument for ignoring the later half of the figure. Okay, say the IPCC states (cite, please?) that it needs 30 years of data to attribute a trend to climate change in the Southeastern US I’d like to look at that original page too, I haven’t found it.

    Once we can compare what the IPCC says about the data to what the NCDC says about the data actually posted here,

    If the NCDC does have good enough statistics to say there’s a trend using 18 years – their claim –– that ought to be shown.

    If the IPCC says the trend can’t be attributed with less than 30 years of data to climate change that’s a different issue.

    A trend needs to be noticed first, then explanations can be considered, whether deforestation, city smog levels, a bad batch of instruments bought from a lowest-price bidder for the entire region, or whatever else.

    Please, cites and explanations that fit what’s being pictured and claimed about the pictures.

    There are Hoagland Mars pictures reminiscent of the Big Red Circle thing.

    I’m sorry to be grouchy, but I know people struggle with this. Be helpful.

    Be more helpful. I make this request generally to the entire editorial board for this blog:

    “Trust — but verify.” — R. Reagan

  26. Report this comment

    Eli Rabett said:

    Quite the high quality blog over here. Now Eli is quite flexible and will play by whatever the local rules are or, as Roger will tell you, if he doesn’t like the rules, he will simply leave. I take it Bubbles is hitting the outer limit of this place. If not, please let me know what game we are playing here so that I can participate. It’s your blog folks.

    However, back to what counts. The point about variability was brought home to me several years ago in a paper which in the context of global surface temperature put the question, given random variability from what point is it most likely to hit a new maximum. The answer is from a maximum. Thus random variability on top of a rising trend will produce new maxima. Same goes in spades for local variability, thus even relatively modest global increases can lead to large local increases. Since we live locally, watch out.

  27. Report this comment

    bubba said:

    Greetings Aaron,

    I respectfully decline your invitation to not quibble. While obviously a learned soul, Mr. Rabett’s prose is, well, rather strange.

    It detracts from his ability to drive a point home with anyone but an acolyte accustomed to sitting at his knee.

    And I assure you, sensitive one, that I too care about the fate of this planet. That is why I surf these sites while drinking a mid-morning cup of coffee or while the fiancee is being consumed by the latest episode of American Idol.

    Gaia’s fate is our fate. I’m with you on that, buddy.

    And, as a point of order, Dr. Pielke did not " compare the credibility of the National Arbor Day Foundation with that of Ahmed Chalabi". You did, my dear Aaron. Your words, your phrasing, and a very, very distinct difference.

    Very naughty, and disingenious, of you to suggest otherwise.

  28. Report this comment

    M. Simon said:

    Uh,

    Shouldn’t the temperatures be based on an 11 or 22 year cycle to better average out solar cycle variations?

    Or would some other averaging period be better?

    The trouble, of course, is that there are so many non-harmically related cycles going on that the choice of averaging period is difficult.

    It would be interesting if papers on the subject divulged the reasons for their choice of averaging period and how different choices might affect the outcome.

  29. Report this comment

    M. Simon said:

    I agree with Eli about the need to deal with climate change.

    Some folks in the sun watchers category are predicting a decline in solar output in 10 to 20 years.

    Suppose all this effort going into warming is misdirected?

    Warming will certainly affect a lot of things. However, I’m told that with a bit of effort you can grow plants in an ocean. I’m told the effort required to grow plants under ice is significantly larger.

  30. Report this comment

    Aaron said:

    bubba wrote, And, as a point of order, Dr. Pielke did not " compare the credibility of the National Arbor Day Foundation with that of Ahmed Chalabi". You did, my dear Aaron. Your words, your phrasing, and a very, very distinct difference.

    He compared the NY Times reprinting of the plant hardiness map to Judith Miller’s atrocious reporting. The implication is obvious. If you had paid any attention to Miller’s reporting, you would know that the she uncritically transcribed propaganda from sources deemed unreliable by the CIA and German intelligence. Pielke Jr.‘s comparison implies that the National Arbor Day Foundation’s map should have been considered equally questionable. That is an outrageous accusation, especially when his specious interpretations have been shown to be inaccurate. Pielke Jr. should be ashamed of himself for attacking a perfectly good piece of science writing. He certainly could do no better.

  31. Report this comment

    Aaron said:

    Hank Roberts wrote, Where is it? The gray means insufficient data, the lack of + marks means no trend significant at the 5% level. What is inside the big red circle?

    Could you maybe enlarge it for the readers and include the caption?

    Good questions. Here is an enlargement. There is sufficient data for the area inside the red circle.

  32. Report this comment

    Jim Angel said:

    I have a comment about the original map that Roger used and the confusion about the caption. The phrase in the caption says, “areas in grey have insufficient data to provide reliable trends”. If you zoom in on the original Figure 3.9, I believe the grey areas actually refer to the lack of data in the polar regions. The area in the southeast US certainly contains enough data and is actually a light shade of blue and represents cooling trends on the order of -0.2 to -0.5C in the lightest shade and -0.5 to -0.8 in the next darkest shade. It’s unfortunate that they chose such ambiguous colors for these two important distinctions. Three of the pixels show a statistically significant cooling trend during this time period.

    BTW, this so-called “warming hole” in the 20th century has been addressed in a few papers. For example,

    Kunkel, K.E., X.-Z. Liang, J. Zhu, and Y. Lin, 2006: Can CGCMs simulate the Twentieth Century “warming hole” in the central United States. J. Climate, 19, 4137–4153.

  33. Report this comment

    Charles Drummond said:

    The type of pseudo-science presented here by ‘Dr.’ Pielke Jr. is very disturbing, and frankly, insulting.

    Nature should consider imposing some sort of review process upon posts by authors, like this one, who establish a record of non-credible writings.

  34. Report this comment

    Hank Roberts said:

    Could one of the editors here who’s good with math sum up what’s inside the big red circle? Higher resolution image courtesy of Aaron:

    http://www.flickr.com/photos/wossname/490746298/

    It looks to me like adding up the values inside the big red circle comes to a slight warming, not a cooling. That’d include all the amber grid squares or portions thereof that have plus marks, eh?

    A much smaller red rectangle around the three blue squares with white plus marks would be correct to label as showing a trend (95% confidence level) of cooling, albeit a very small temperature change per the caption.

  35. Report this comment

    bubba said:

    And one would add up the values for all grid cells within the red circle for what reason?

    I think the post is mainly about the discrepancy between the the USDA’s and the Arbor Day group’s methodologies for formulating a plant hardiness map with specific emphasis on the Atlanta region.

    Including grid cells from the Yucatan and Upper Great Plains, which lie within the circle, would be of little to help Northern Georgia’s, or elsewhere in the U.S. southeast, gardener’s and arborealst.

    As far as the “very small temperature change” in the blue cells with significant trends, at between -.2 to -.8 centrigrade, I suppose they are as insignificant as the +.2 to +.8 exhibited by the cells within the circle that show a significant warming trend? Or perhaps I’ve misinterpreted you point.

    And why a “smaller red rectangle”? Aren’t grid cells supposed to be of uniform size?

Comments are closed.