Monday, February 15, 2010

Weather ≠ Climate Meme

The "Science Is Settled" pro-GW crowd keeps trying to regain lost ground amongst the myriad of scandals and winter weather that keeps being uncooperative.

The last few days, we have been subjected to numerous "weather is not climate" stories. Example here, and, from blog follower Geri, here. I won't bother to list the other 151 articles that Google News says contain that phrase. Actually, they couldn't be more wrong.  Climate = weather.

Don't believe me?  I just pulled my trusty Oxford Dictionary of Current English off the shelf and found this definition:  Climate: Prevailing weather conditions of an area.  Not a technical enough source? Here is what the AMS's Glossary of Meteorology says:  Climate - the long-term manifestations of weather.

Of course, the 'weather' can be unusual and atypical of the climate of an area (i.e., the snow that accumulated in Florida last week), but that snow storm is still a part of Florida's "climate."

Besides, if climate is not "weather" what is it?  A hat, or a brooch or a pterodactlyl?
(My apologies to the movie, "Airplane!") 

And, from the NPR interview:  And Trenberth notes that you don't need very cold temperatures to get big snow. In fact, when the mercury drops too low, it may be too cold to snow. Too cold to snow? Looks like Dr. Trenberth needs to read this blog!


  1. Mike, I hate to disagree (again), but you are wrong when you say Climate = Weather. As I've said before in your comments, and you said you agreed, climate = E[weather]. (Climate is the expected value of weather - for those unfamiliar with the mathematics notation.) As you said above, a single weather event does not make a climate; using an equals sign to relate climate and weather would imply the opposite.

    And, even though Dr. Trenberth needs to be careful when throwing around phrases such as "too cold to snow", the underlying idea is true. When the atmosphere is colder, it holds less moisture. In the post you liked to, you mention a temperature of 10F, which is not all that different in the grand scheme of things than 30F. What about a temperature of -128.6F (the world record low)? As a meteorologist, tell me how much moisture (which would fall as snow) would fall in that environment? There is a reason why Antarctica is considered a desert. It's because most areas receive less than 2" of snow a year. It's because it is too cold for the atmosphere to hold much more moisture than that. So while it may not be "too cold to snow", it certainly can be too cold to have a significant snow.

    I ask (honestly) why you feel the need to reduce the complexities of our profession down to, what I would consider, misleading soundbites? Although I do not agree with every piece of your commentary during the Phil Jones interviews, those posts have been your best argument for your GW position.

  2. Patrick,

    You are free to disagree any time (this isn't!). I welcome constructive, respectful comments as yours always are.

    Actually, the linked post first mentioned significant snow falling at 0°F, which is quite different from 30°. That said, heavy snow is rare at very low temperatures because, as you point out, the air can hold less moisture. But there is no such thing as "too cold to snow" and I would contend that it is Dr. Trenberth that is spouting "misleading soundbites."

    That said, why do I sometimes do short, simplified posts along with long/detailed/technical posts? It is because people have different levels of interest and different technical aptitudes. You are a meteorologist, so you have higher levels of knowledge, which is why I suspect the annotated interview with Dr. Jones appealed to you.

    Have you read my 7-part analysis of Climategate? Have you gone back and read any of the other multipart posts (other than Dr. Jones and Climategate)? If you have, I believe you will find a wealth of information.

    But, because I want to inform with a little bit of entertainment and a light touch, I will post short, simplified items from time to time.


  3. Hi Mike,

    I have read all your posts, including all the Sons of Climategate posts. I agree Dr. Trenberth engaged in a sound bite, and I said he should be careful in doing such things. However, all other factors being equal, a thermal profile near 0F will result in less liquid water content than a thermal profile of 30F.

    In any event, you calling out Dr. Trenberth wasn't what prompted me to reply. He deserves what he gets for creating a poor, inaccurate sound bite. What really bothered me was the assertion you made that climate = weather. The definitions you pointed out in your post don't even support this assertion. To me, this was your misleading sound bite - which you did not discuss above.

    At the very least, I would love to see a post dedicated to explaining what you mean when you say climate = weather. Because I certainly don't see it.

  4. Patrick,

    In your first comment on this post you suggested that climate = the expected weather, E[weather].

    I have a different perspective, Climate = Σ[weather]. For non-math types, that is climate = the sum of the weather at a given location. And, I believe that definition fits the definition given in the Glossary of Meteorology.

    So, since climate is the sum of the weather, climate IS weather. Or, put another way, climate is long-duration weather.

    Hope that clears it up. Thanks for reading!



  5. I admit that when I first sat down to respond to your post, I was going to put that climate = Σ[weather]. But the more I thought about it, the more I realized that isn't correct. But I'll assume for a moment that it is...

    Let's say for the past five years, the temperature in Wichita, KS has been 50F, 30F, 40F, 20F, and 60F on 16 February. Using your definition provided above the climate of 16 February would then be Climate(16 February) = Σ[weather(16 February)] = 50F + 30F + 40F + 20F + 60F = 200F. Thus you are saying that the climate should be a value that has never been observed.

    Now using my definition that Climate = E[weather] we'd have Climate(16 February) = E[weather{16 February)] = Mean[50F, 30F, 40F, 20F, 60F] = 40F which is a much more reasonable answer.

    As for the AMS definition, I still argue that my definition is more accurate. Continuing the example above, lets say that the next five years the temperature on 16 February was 20, 25, 20, 30, 30. Using the summation argument you'd end up with a climate that now has a temperature of 325F. Whereas I'd have a climate with a temperature of 32.5F. In words, even through it has been colder the last five years your "climate" is warming. Using my definition, my the last 5 cool years has manifested itself by cooling the "climate". Climate cannot be a summation of events, lest you end up with a "climate" that is far removed from observations.

    Yes, I know I'm arguing semantics, but as you have implied throughout your blog posts, the words people choose are important.

  6. Oh, I forgot to say that in mathematics the only function I am aware of that is an indefinite summation of itself is a constant. Thus, the only way your definition can be accurate is if the climate is constant.


Note: Only a member of this blog may post a comment.