Things that have always been issues

Here are some things that have always been issues, except for when they haven't been an issue at all. I'd first like to show the google timeline that I'm using is a generally good representation of what's important at a given time.

The search I did was for 'War' You can pick out the major wars on this ol' graph. Bigger bars mean more was written about it at the time. A bar twice as big is twice as written about. The bars don't scale to the same height, so what might be a blip on one graph, if plotted on another graph would be massive.

Interestingly, we're less interested in war these days than we were back in WW1 and WW2. Then again, in the WWs, the whole country was effectively mobilized to fight it.

Here's something that was a major point of contention in the 2000 election, but no one talks about anymore.

Here's something from the 2000 election that has periodically cropped up, but is mostly ignored between long stretches of the US nationally fixating on it.

I cropped all these timelines so they covered 1900 to today, but it's interesting to know that the last one was also a major issue periodically in the 1800s. Somehow, the nation avoided collapse then.


Here's something that was major in the 2008 election, that we were told was a brand spanking new issue that we must tackle now. Does the 80s count as new?

And here are two issues that I've been told have always been important, and people have always cared deeply about it, and they must be addressed now. 

And here's Hanging Chad. The most important thing of its time, and something we really don't care about now.

And here's a pet rock.

Not a very scientific analysis, but I guess what I'm saying is that there sure is a lot of what looks like fads in politics. Most of which never reach any form of resolution. And yet the world doesn't come quite to an end, like the politicians predicted if we didn't do what they wanted to do, right that moment..

Alaska Census changes, 2000 to 2010

Data time, people! Behold, tables:

 

Borough or Census area	2000 estimate	2010 estimate	% change
Yukon-Koyukuk Census Area	6,551	5,588	-14.70%
Ketchikan Gateway Borough	14,070	13,477	-4.21%
Prince of Wales-Hyder Census Area	6,146	5,559	-9.55%
Valdez-Cordova Census Area	10,195	9,636	-5.48%
Kodiak Island Borough	13,913	13,592	-2.31%
Bristol Bay Borough	1,258	997	-20.75%
Lake and Peninsula Borough	1,823	1,631	-10.53%
Yakutat City and Borough	808	662	-18.07%
Dillingham Census Area	4,922	4,847	-1.52%
Denali Borough	1,893	1,826	-3.54%
Sitka City and Borough	8,835	8,881	0.52%
Aleutians West Census Area	5,465	5,561	1.76%
Haines Borough	2,392	2,508	4.85%
Nome Census Area	9,196	9,492	3.22%
Northwest Arctic Borough	7,208	7,523	4.37%
Wade Hampton Census Area	7,028	7,459	6.13%
Aleutians East Borough	2,697	3,141	16.46%
Juneau City and Borough	30,711	31,275	1.84%
Southeast Fairbanks Census Area	6,174	7,029	13.85%
Bethel Census Area	16,006	17,013	6.29%
North Slope Borough	7,385	9,430	27.69%
Kenai Peninsula Borough	49,691	55,400	11.49%
Fairbanks North Star Borough	82,840	97,581	17.79%
Matanuska-Susitna Borough	59,322	88,995	50.02%
Anchorage Municipality	260,283	291,826	12.12%
Hoonah-Angoon Census Area	3,436	2,150	-14.79%
Skagway Municipality		968
Wrangell City and Borough	6,684	2,369	-21.11%
Petersburg Census Area		3,815
Statewide	626,932	710,231	13.29%

I have the table organized by the number of citizens gained or lost. Note that Hoonah/Skagway and Wrangell/Petersberg is reported differently between 2010 and 2000, so they take up two lines (Only one value in 2000 per either pair). What leaps out at me is most of the change comes from the MatSu - a wopping 50% change in population! Yikes! Fairbanks has also grown at an accelerated rate, but not nearly as much. I was wrong about the North Slope - it has a greater % change than the North West Borough. I was correct about the YK, but too conservative about the rate of growth since both major census districts topped 6% growth. I was too conservative about how horribly hard SE is getting hammered. Aside from Bristol Bay, the biggest percent changes were in SE. The biggest absolute change, though, was in the Yukon-Koyukuk area:

Borough or Census area	Absolute Change	Percent of State Change
Yukon-Koyukuk Census Area	-963	-1.16%
Ketchikan Gateway Borough	-593	-0.71%
Prince of Wales-Hyder Census Area	-587	-0.70%
Valdez-Cordova Census Area	-559	-0.67%
Kodiak Island Borough	-321	-0.39%
Bristol Bay Borough	-261	-0.31%
Lake and Peninsula Borough	-192	-0.23%
Yakutat City and Borough	-146	-0.18%
Dillingham Census Area	-75	-0.09%
Denali Borough	-67	-0.08%
Sitka City and Borough	46	0.06%
Aleutians West Census Area	96	0.12%
Haines Borough	116	0.14%
Nome Census Area	296	0.36%
Northwest Arctic Borough	315	0.38%
Wade Hampton Census Area	431	0.52%
Aleutians East Borough	444	0.53%
Juneau City and Borough	564	0.68%
Southeast Fairbanks Census Area	855	1.03%
Bethel Census Area	1,007	1.21%
North Slope Borough	2,045	2.46%
Kenai Peninsula Borough	5,709	6.85%
Fairbanks North Star Borough	14,741	17.70%
Matanuska-Susitna Borough	29,673	35.62%
Anchorage Municipality	31,543	37.87%
Hoonah-Angoon Census Area	-318	-0.38%
Skagway Municipality
Wrangell City and Borough	-500	-0.60%
Petersburg Census Area
Statewide	83,299	n/a

Anchorage had the greatest growth (Gold star for me) followed by MatSu and Fairbanks and the Kenai. Juneau, as I anticipated, remains essentially flat. I might have to double back on the Rural-to-Urban migration idea. There's 214 more rural residents than there were in 2010, which is a minuscule increase. Either there's more immigration than emigration, but elevated death to compensate, or births are higher than deaths, and there's a net emigration. The second scenario seems more likely.

Why didn't this show up in the school enrolment records in Anchorage? Well, it could be that people more likely to move don't have children yet, or maybe they're not all moving to Anchorage. Or, what I think is most likely - there's serious flaws in how we calculate enrolment in schools, where students from the bush are more likely to skip classes (true) and therefore not be reflected in the annual counts. This downward biases the enrolment numbers, and therefore funding, of schools with the students who need the most help.

But, that's navel gazing and guesswork. The real way to address this is to follow individuals, not counts of individuals. Census summaries we have access to can't show that that +1 person in the Kenai came from Juneau as opposed to Kansas. You need to do a more detailed breakdown than that to find those sorts of trends.

Finally, on the diversity front, diversity increased slightly (69% White to 66% White descent) mostly to an additional 1 percent more people of Asian descent, and a grab-bag of other.

There's a lot of information here, and I've barely played around with it. Hopefully, I'll get more time to dig deep into this well of information in the next little bit.

Where's the stastics?

After reading over and digesting Nelson's (1998) "Development of migratory behaviour in northern white-tailed deer" the group had gradually come to the conclusion that the paper did not really belong in the otherwise good Canadian Journal of Zoology. The paper is very little more than a case study of the fates of 36 collared white-tail fawns. Case studies have their value, but the problem arose when they tried to translate this into hypothesis testing.

The study, in short, radio collared the affronted 36 fawns in the winter and followed their fates. They found that the majority of them returned with their mothers to their 'natal summer ranges' and a portion of those back-migrated to the same winter ground. Some deer dispersed to new summer grounds, and would back-migrate to a winter ground. Everything else in the paper is individual case summaries too tedious to recreate here.

The most glaring omission throughout this paper was that there was almost no statistics to speak of. I could only find one instance where I knew a confidence interval, and it was on a trivial value. Everything else is presented as percents, without any way of knowing if the differing value of percentages means the effects were significant. Secondly, we have no way of knowing whether the deer truly returned to their natal ranges - the author just assumes where they went after their first winter was their natal range. They collared the fawns at the wrong time of year to really wrap their head around the phenomenon. Finally, in the discussion, the author included unpublished data to bolster their own hypothesis. We were not overly impressed by this.

The big disappointment is that this topic is of general interest. Migratory behaviour in ungulates is understudied compared to birds, and many species that engage in migratory behaviour are of considerable human importance. It would be fascinating to repeat this study with fawns collared at birth (or soon after) to tease apart how white-tails establish their migratory behaviour in more mobile populations. This does not shed much, if any, light on the issue.
(X-posted)

Your vote doesn't matter (redux)

It's that time of year again, when we scribble in bits of paper and then complain about how other people scribbled in their bits of paper for the next two years. Yes, it's time to vote! Alaska's had a wild and crazy senate race which most people are sick to death of at this point, but all signs point to it not being over. If Lisa comes close to the percents she's polling at, she'll be tired up in court for the next decade as Miller and McAdams sue eachother silly. Much like Minnesota spent eons wrangling over votes, except with the extra joy of write in ballots.

So, with that cheery note, it's my turn to remind everyone that their vote doesn't matter, unless they're feeling lucky. Most economists would tell you that your vote is useless, and you'd be better off earning extra wages and donating the money to a political group in the time it takes to vote. You can watch this wonderful video where a talented GMU professor outlines why democracy is fundamentally flawed anyhow (Spoiler alert: It's bundling, and the illusion of consensus). So stay home! Have a nice glass of tea. Especially if you plan on voting against me. :)

Me? I'm feeling lucky. ;)

Can we agree F-st has run its course?

Other scientists out there! Hi. Can we agree that Fst, as wonderful as it's been, has run its course? It was a good idea - a great first crack at population genetics. When Wright came up with it, it was a wonderful idea. And for some applications - those where heterozygosity is generally low (I'm looking at you allozymes) - it works quite nicely. But once you're outside the Hs of .4 to .6, your Fst value starts becoming highly constrained. Fst = (Ht - Hs)/Ht. If Hs is large, it doesn't matter how much skewed your heterozygostity partitioning is, Fst will be small. There may be a way around this in Jost's D. I'm not a math biologist, so I'm not qualified to review his equations. But Fst is deader than a door nail. It's been a good run

Not convinced? Fair enough. Consider this figure from Gerlach et al 2010.

Why should diversity have any effect on population sub-structure? This makes no sense. Imagine two herds of caribou, one in Alaska, and the other in Quebec. All the herds between them are wiped out by Caribou flu (and you thought pig flu was bad!), so there's no gene flow. They both diversify, generating new genotypes - Alaska generates Alleles A, B, [...] L, M. Quebec generates alleles N, O, L [...] Y, Z. Both populations lose their ancestral form. All alleles are represented at equal frequencies. Neither population has a single allele in common. Divergence is total. And Fst is only 0.04* for this population pair. Fst will only get lower as you add more unique alleles to each population. This is absurd. Adding alleles does nothing to alter the fact that there is no gene flow from Quebec to Alaska in my example.

Jost's D would calculate differentiation as 1.0, which I think is a more accurate reflection of the fact that they have no diversity in common. But, IANAMB**.

Two reasons I bring this up. First, because I'm reading Gerlach et al 2010, which took the approach of using a variety of datasets to show that Fst*** does not reflect true levels of differentiation. It's a heap of data that show that when corrected, Jost's D neatly tracks true population divergence while Fst... well, it's flogging a dead horse at this point. The second reason will become clear in a moment.

I propose the following. You're allowed to use Fst in your publications for one more year. But at the end of 2011, that's it. Either move to other metrics, or take up under-water basket weaving. I've got two manuscripts I've got in prep. that I really hate having to report both Fst and Jost's D within. And in my case, Fst is awful because my species have high heterozygosity all around. I'm to be told that subspecies on different continents have an Fst of ≥0.05. If I were to take the most simpleminded, naïve interpretation of this, I would be to believe that I have around 4 migrants successfully swimming the arctic ocean to Eurasia each generation.

How about not?

*If I got my exact math wrong, you have permission to beat me with a stick. My point stands, though.**I am not a math biologist, so I'm not sure if Jost D's derivation is completely correct.
***Technically Gst, but Fst and Gst are used interchangeably, so 'nuff said.

GERLACH, G., JUETERBOCK, A., KRAEMER, P., DEPPERMANN, J., & HARMAND, P. (2010). Calculations of population differentiation based on GST and D: forget GST but not all of statistics! Molecular Ecology, 19 (18), 3845-3852 DOI: 10.1111/j.1365-294X.2010.04784.x

Croudsourcing consumer expenses.

Very particular consumer expenses. Since I'm apparently in a controversial mood today, there's a neat link I found called priceofweed.com. You can probably guess what it's about. While I don't do them, I find drug economics very interesting. The products are illegal, expensive, and in some cases dangerous to acquire, and yet according to the Author of Freakanomics, crack dealing has much the same business organization as McDonalds.

Anyhow, back to weed pricing: you can't trust police estimates of drug worth, since they have a vested interest in exaggerating the prices both to dissuade use and to make them look like they've accomplished something. When you spent several hundred thousand dollars, you need to justify it by saying you stopped some equal or greater amount of illegal activity, and the greater the estimate of the activity stopped, the more cost effective you can claim to be. "You should fund me, because for every one dollar, we can stop six thousand dollars of drugs." It gets so absurd that on the TV, I once heard police claim they had interdicted something along the lines of 1% of the US's GDP in drugs. 100 Billion dollars worth of drugs is... well... in 2008, it would have been the equivalent of trying to smuggle in the net worth of IBM. I imagined somewhere, an economist saw that on TV and promptly choked in his own spittal over the absurdity of the claim.

Will we ever catch up with Cuba?

Marginal Revolution, a wonderful economics blog, reproduces this following graph.

I'll take it down if they want me to - it's their work. Two things strike me when I look at this. First, the bulk of the wold is around the gender parity mark for education. Except for one massive exception, which is India. India has a large population, and falls on the left side of the parity line, meaning that females have less than males. You could move all the other outliers over into the main cluster, and you wouldn't budge the world average much.

Secondly, Muslim Countries are at one some of the best and worst nations when it comes to parity. Pakistan isn't some third world nation - it's a major political and economic player. The quality of living is fairly high there. And yet they're very low in terms of parity. On the other hand, Iran is not known for its womens rights (although it is no Afghanistan), and Palestine there are systematic abuses of women in the name of Moral policing. And yet they're both healthily above the parity line.  I wonder if culture is a major cause of variance in this graph... I'd do the analysis myself, but I wouldn't know how to lump the countries.

Beer Vs. Cosmos!

This is a cool tool! It's called 'Lexalist.com' Of course, the first thing I have to do, is find out who drinks what, where!

Oregon sure loves their beer! Not surprisingly... they have some of the most microbreweries of anywhere in the country!

On fMRI

So, supposedly a study using brain scans found that 4 of 27  patients in a vegetative state could actually answer questions. We're invited to feel horror at the thought of being locked in. Indeed, that seems horrible.

But, on the other hand, they found that a dead salmon did a lot of thinking in an fMRI, too.

These people, in the fMRI world, have what we like to call "Problems" with their statistical methods.

I forgot! Annoyance Degrees C

I forgot to say what annoyed me about `Half of 40°C`! How absurd that phrase is becomes apparent when you ask "What is half of 40°C?" You might guess 20°C. Okay, what's half of -40°C? -20°C? 

The problem is 0°C and F are both not absolute zeros. You can't just take the magnitude from them and apply a fraction, and have it even remotely meaningful. .5 * 40 is not half the temperature or energy of 40 (in either C or F). You have to convert to a Scientific unit called "Kelvin," a system where the number does represent the magnitude in that way. 0K is 0K - no energy, whatsoever. So, half the temperature of 40°C is actually -116.575°C. Those are some seriously cold Hummingbirds!

I'm sure someone will correct my correction. But I know that 20° is not half 40°! :p

Graph for the day

Vaccines: They work.

Hat tip: Bad Astronomy. Graph source: CDC

Carbon pawprints revisited.

Remember this qimulvak? He's been getting a ruff shake. People are barking up a storm about how he's a bad boy for having such a big carbon pawprint. The twitter version of the story is that medium sized dogs have bigger carbon emissions than a typical SUV. I linked an ABC version, but there are other agencies who reported on the story, too. I was somewhat sceptical when I read the story, but I figured most dogs up here are safe. Aside from the urbanites and their dogs, most dogs I know around here are heavily supplemented with fish and meat. Some eat nothing but locally harvested fish. The paw print on that is minuscule - just the oil burnt in getting out there to jig.

But most people I know bought into it without much critical review - even I gave some of the claims a pass, because they shook well with what I knew about the effects of agriculture on carbon emissions. Well, it turns out the numbers were much more worse-case than they probably realistically are. Clark Williams-Derry, a director for the Spotlight Institute, a pro-green environmental think-tank, spends some ink on explaining how the footprint of SUVs really is that bad, and how Landrovers are, in fact, worse than Rover.

Read the whole article. It's very well written, and entertaining (with even more dog puns than my post). But if you're too lazy, the big flaws are here:

• They underestimate how much driving people do.
• They underestimate how much oil is involved in car production and maintinance.
• They assume dogs get high quality meat and cereal, when dogs get trimmings meat in most dog chow. 

They also look top down, to show that practically, dogs have little impact on food from real buying habits.

So sleep easy, dog-outside-the-Turtle-Club. You're not worse than my truck.

Rejected hypothesis

This is the way science works. You make a hypothesis, you form a prediction, you test the prediction, and then evaluate your hypothesis.

We are forced to reject the hypothesis of massive rural influx into urban areas as the data does not support it. I, myself, was wrong. This graph pretty much says it all:

More at the ADN, where the story is from.

Not only do we need to revise our hypothesis, but it appears the data supports an nearly opposite mechanism! a) Students come from primarily urban areas and b) Among rural areas, those worse off economically appear to export less students! Speculating, this may be because households worse off economically cannot afford to move their households. Or, they tend to value living in situations like their current situation more than the additional leap in economic opportunity.

No one else would put it in economics speak like that, so here's an example: C lives in Alakanuk. With commercial fishing having crashed in the area (and subsistence having been slashed), C decides to move his family. C compares other places - Bethel, Dillingham, Anchorage, Nunapitchuk, Kwigillingok etc - for characteristics including 1) economic oppertunity, 2) current relations 3) similarity to C's preferred ideal of a village/city/whatever. Among those options, he makes a decision where more than immediate economic oppertunity factors in. ("Anchorage has more jobs, but I've got family in Dillingham, and he can get me work on a boat"). 

You'd test this by finding out who moved, who moved where, and whether presence of relatives or some other similarity factors influence movements more than chance. Sounds like a project for some young aspiring masters student out there. ;)

The Parnell plan for reducing Domestic violence contrasted

When Governors announce that they'll reduce Domestic violence, rape, incest, or anything of that sort, it feels a bit like when teachers from the states come here and announce that they'll solve all our problems - you know they mean well, but they don't have a clue what they're doing. Parnell, bless him, says he'll be the man to reduce domestic violence rates.

• They plan to aggressively pursue stiff sentencing for offenders.
• The sex offender list will become unavoidable through plea bargains.
• New sexual assault and domestic violence investagors will be added to the roll.
• Increased funding to shelters.
• An expansion to the VPSO project.
  

I'm afraid the end result will not be a reduction in the real rate, but that the real rate will remain unchanged, and we'll see an increase in the observed rate thanks to the increase in enforcement. Domestic violence is not a pre-meditated thing. It's typically a crime of passion, where deterrence just doesn't work. In fact, there's been a study on just this subject, where Sherman et al 1991 found, "Contrary to deterrence theories, arrest had no overall crime reduction effect in either the official or victim interview measures of repeat domestic violence." In fact, there's a chapter in a book, titled, "Does Arrest Deter Domestic Violence." The answer seems to be 'no.'

Even more problematic, it appears not only arrest appear to be ineffectual, but it may actually increase domestic violence in the long run as those penalized through stiff sentences increase the rate of battery. This may be due to the increased disruption in the batterers' leading to a poor economic situation, and this hypothesis would be supported by the fact that increased arrest increases the rate of domestic violence and sexual assault among the unemployed. It seems that the situation is positively hopeless!

The current situation is clearly not tenable, so what would a better plan be? The bush suffers differentially from domestic violence, with some areas it being positively epidemic. There are a few strong links to domestic violence that we could address, possibly with greater success. Most studies have focused on male to female violence, so please keep that in mind.

• Female highest level of education is negatively associated.
• Having multiple partners is positively associated.
• Conflict over his drinking is positively associated.
• Greater income inequality between partners is positively associated.
• Other conflict in their lives was positively associated.
• General increasing income equality was negatively associated.
• Family income is negatively associated.
• Living with more extended family members is positively associated.

I've only listed the things we societally could address. For example, the older you are, the less likely domestic violence is, but we can't go out and age people. In the light of this information, it should seem highly unsurprising that many bush communities have high rates of violence, because they are on the increased risk end of many of these factors. My arm-chair quarterback suggestions for reducing the rate of domestic violence in Alaska would be:

• Invest very heavily in rural economic development.
• Continue to invest in domestic violence shelters.
• Increase the availability and desirability of bush student scholarships.
• Increase the amount of available affordable, energy efficient housing in the bush.
  

In my last point, energy efficiency is important, because households (in my purely observational opinion) tend to consolidate over the inability to keep two houses running at once. Typically over heat. As a side note, I went to AZ once, and a lot of the old BIA houses are the exact same ones they use on the Navajo reservation. And they wonder why they leak heat!

The problem with the TwoYaks proposal is it'd be slow. It would take a long time to see the fruits we've planted, but I don't know of any method to give the state an immediate short-term drop in domestic violence. There just doesn't seem to be a quick and easy fix. But the TwoYaks proposal has the advantage of being based on sound science and observation, and also effecting other things we find societally undesirable too. For example, poor opportunities for employment is associated with high levels of murder and alcoholism, two things we also don't like.

Sadly, I'm not Governor, so I just make them start. That means it's time to hit the pen and paper, and write some letters to my congress critters.

2nd Grade Math.

Find your nearest 2nd grader and show them this. Maybe they can tell Fox News what they did wrong when making a piechart. :p

Graphs revisited; Ice puzzles revisited.

First, an answer: this graph...

is a windchill graph. The coloured regions are the onset of frostbite in exposed people. :) NOAA has a nicer version here.

Avery chimmed in on the question about the ice, and raised potentially good points. I knew that the explanation that I'd got when I was younger had considerable doubt cast on it, and caveated my post appropriately. Or, maybe not enough! But, luckily, I'm in fine company, as my incorrect answer is repeated in various textbooks. Here's the most correct answer, as we understand it today, from ScienceNOW, the news arm of the Journal Science:

Ice has always been a slippery subject. As simple as an ice cube may seem, scientists have long been baffled about why its surface is so slick. But an upcoming paper in Surface Science may give researchers a firmer grasp of ice's surface subtleties by hinting that its outermost molecules behave like a liquid.

That would give the surface layer drastically different properties from those of the bulk of the crystal, says Gabor Somorjai, a surface chemist at the Lawrence Berkeley National Laboratory. The liquidlike layer could explain, for instance, why it is more fun to skate on ice than on concrete. According to Somorjai's colleague, Michel van Hove, the popular conception that ice's slipperiness comes from pressure-induced melting is wrong. "It doesn't work out," says van Hove. "You put data into the formula, and there's not enough pressure." The slippery layer, he says, is there to start, even at very low temperatures.

Somorjai and van Hove discovered this layer when they probed the surface of thin layers of ice with low-energy electron diffraction, a technique that uses electrons to determine the surface structure of a crystal in the same way as x-ray diffraction reveals the crystal structure of a solid. The researchers expected to see the scattering signature of the first three layers of ice molecules, but they only saw two. After determining that the invisible top layer did, indeed, exist, the researchers hypothesized that its water molecules were vibrating three or four times faster than those in the lower layers--blurring its diffraction pattern to invisibility. Although the water molecules are bound in the lattice like a solid, says Somorjai, "the vibrational amplitude is like a liquid."

Besides making ice slippery, says Somorjai, the liquidlike layer could help explain how ice crystals in the upper atmosphere help catalyze the chemical reactions that deplete ozone. The finding, says Steve George, a chemist at the University of Colorado, "illustrates how we don't understand the simplest things we know about."

Can you recognize this graph?

As the title. I made this graph after an experiment in its contents. Hit me with your best guesses!

Payall Bidding to explain political lobbying.

I haven't the time for a proper post this morning. So instead, brazenly stolen from Marginal Revolutions: Payall Bidding to explain political lobbying.




I wonder at which point you you are indifferent to bidding. Is there even such a point? In fixed increment bidding, the point is probably equal to the increment you win plus the bid in terms of the indifference point. But don't quote me on that.

Warm winter?

I'm not sure why I even read the news-story comments on ADN and Minor-News. I think its because I want to remember that there are idiots out there. Very, very vocal idiots. For example, whenever a weather related story comes up, you have people pop out of the wood work to go `Hey Al Gore, where's your global warming now, HMMMMMMM?!?` Like the fact that they put on a sweater that morning proves that carbon dioxide doesn't absorb the infrared wavelengths. Anyhow.

So, I mention this because the ADN did a story about `Can Alaska believe the forecast? NATIONAL OUTLOOK: Government predicts a warmer winter for us.` Actually, it was an AP story they picked up and put the title on. As normal, people are out talking about how the `gubmint dunno nothing 'bou nothin'. Even the title of the story suggests the newspaper thinks it knows better than a team of dedicated Alaska Climatologists.

So, to discuss this, I've put the 3 month outlook off to the side. The deepening areas of colour reflect the probability of deviations from statistical average temperatures. This is what prompted the story. They publish these on a regular basis, and most of the time updates don't seem to merit a story. I think the drought data is why AP wrote it, and ADN just picked it up to fill space.

On any normal given `average` day, the odds of below average, above average, or average might be something like 20%-20%-60% (numbers I've just invented, because I'm too lazy to look up the real numbers). The bars represent one event becoming more probable on average - such as 40% chance of above average, 10% chance of below, and 50% of average (Again, numbers I just invented. I do recall that the opposite category shrinks faster than N).

So when I read comments like Denseyler's

The rest of the states having equal chances of being warmer or cooler than normal??? Now there's a forecast!!!!!!

I have to roll my eyes a bit. Yes, it is a forecast. It's an important one, too. Climatically, it represents a normal distribution of above and below average temperatures, which is important to know.

People forget that NOAA doesn't just exist to tell them how warm it'll get on their way to work. It also informs farmers on what to expect from the following year, it informs the Navy and Coast Guard about important weather events and trends, it provides information for what might be happening with wildlife... the list goes on. It's easy to remember that one day the forecasters said it'd be sunny, and it was cloudy all day. It's easy to forget the other 360 days out of the year they were spot on.

So, sorry ADN, and sorry commentator. I'm going to listen more to the meteorologists on this one.

PoliSci trivia for people outside.

Nationally, Rural areas tend to be more supportive of Republicans than Democrats. Nationally, Urban areas tend to be the opposite. So it blows the minds of quite a few people outside when they find out that AK, yet again, is a weird, different place. Here's the prelim results from Stevens/Begich. It's from the ADN's wonderful figure.
Most of those rural areas are not merely plurality blue, but fairly deep blue. You can also see this reflected in the make up of the state offices, when looking at the location of each district. - I'll hasten to add that it's not a perfect correspondence with this (E.g., Kodiak), and there's slightly more republican representation at the state level. Still, the averages remain: most of Alaska's Republican lean comes from urban areas, especially Anchorage and Mat-Su.