Welcome everyone. Good evening, this is the second of the 2018
Science for Alaska Lecture Series. I’m Sue Mitchell, I’m the public information
manager for the Geophysical Institute. At the University of Alaska Fairbanks. Our lecture series is sponsored by the Geophysical
Institute and the Alaska EPSCoR program and is brought to you by a generous donation from
the Triplehorn family. As well this year from the Lifewater Engineering
Company and Class 5 Boatworks. Thank you very much for your donation. Now, I’d like to announce tonight’s speaker. Rick Thoman grew up in the Amish country of
Pennsylvania. He dreamt of life in Alaska. He’s worked in the weather and climate business
for the past 35 years in both public and private sectors. He realized his childhood dream when he joined
the National Weather Service Alaska region in 1988. And after 30 years he’s now comfortable calling
himself an Alaskan. He currently works as the National Weather
Service Alaska Region Climate Science and Services manager. And he’s the author or co-author on a dozen
or more papers on aspects of Alaska climate variability including basic climate analysis
wildfires and marine heatwaves. He has a Bachelor of Science in meteorology
from Penn State and I was interested to find this out, a masters in Athabascan linguistics
from UAF. He’s also says he’s working on a PhD but that’ll
be a retirement project for him. His pastimes over the years have included
dog mushing, hosting a medieval music show on KUAC, and being the historian for the Thoman
family. So welcome Rick Thoman. All right. Hopefully you can hear me now. All right well thank you very much. Boy what a honor to be here. Quite a gust gathering of Fairbanksans here
so. I’m very humbled. And thanks to Sue and Fritz for inviting me
in and making this happen. So as Sue said I am I work these days as a
climate science and services manager for weather service Alaska region. And you might be wondering what the climate
science and services manager does. Well. I have a couple of main parts of my job or
at least a couple of the most interesting parts. I spend a fair bit of time putting current
weather events into historical context. So I’ve had several inquiries how unusual
is this. Today was the 15th day in a row of 20 below
or lower in Fairbanks. How unusual is that. Let’s say a big storm is forecast in the Bering
Sea. When was the last time we had a storm like
that. So that’s one part of my job and maybe what
you might think of as climate. The other part in the weather service, climate
means any forecast beyond one week from now. So I have all of the past. And all of the future except for one week. So I’ve gotten most of the time line in my
job description. Tonight we’re going to be mostly talking about
the past part. So I’m going to start out talking about the
people behind weather and climate data in Alaska. I give you a little bit of a history of Alaska
weather and climate observations. Where and how they’ve been taken. And then just to illustrate some uses of what
we can do when we have decades worth of high quality data. And of course since we’re all Fairbanksans
and it’s winter we’re going to focus on interior stuff and the winter and a few other odds
and ends as well. So hopefully you find this enjoyable. So if you thought you were coming to a presentation
with lots of charts and graphs and maps you’re at the right place. But we’re not going to start out there. We’re going to start out with tipping our
hat to the thousands and thousands of people over the last five generations that have taken
the time sometimes paid sometimes volunteer to gather and collect these weather and climate
observations that make up really our priceless heritage. I mean if all of a sudden tomorrow the president
decided that climate research needed a trillion dollars. That’d be great. But you know what we can’t get. No amount of money can get us that old data
that we don’t have. No amount of money. So we have people like Melvin Summers one
of the very first meteorologist with the weather bureau in Juneau who wrote the first data
driven climate analysis of Alaska. We had hundreds of men and women during World
War II taking observations all around the Territory of Alaska. We had people like Corinne Gogue to hear it
was with the weather bureau in Fairbanks here in the 50s and early 60s. I do hope that she dressed up for the news-miner
photograph here launching the weather balloon in a dress and high heels I hope she wasn’t
doing that in July or in January. We have people like Ted Fathauer many of you
will remember Ted not only was he the head of the weather service forecast office here
for almost 20 years but he gathered weather data at his home on Cheen or ridge that are
fundamental to understanding what goes on with our inversion that’s so important here
in Alaska. And then there’s people like on the bottom
there holding the big plaque John Borg the cooperative observer in Eagle. He’s been by just because he’s interested
he’s been taking daily weather and climate observations every day for more than 40 years
now. And here he is getting the very prestigious
Thomas Jefferson Award. It’s people like this since the 1800s that
really allow me to give this presentation. So brief history. Well let’s start out of course with the non
western weather and climate information of course traditional knowledge for thousands
of years people living traditional lifestyles in Alaska of course have an innate deep understanding
of weather climate not just the averages but how those things can be expected to vary every
season of course can be different. And of course that’s not unique to Alaska. All traditional societies whether Athabascan
in interior Alaska or my ancestors, subsistence farmers on the Rhine River valley in Europe. Intimate knowledge of weather and climate. So we’ll jump ahead now to kind of the Western
society view. There actually were weather observations gathered
at Sitka during the Russian era from the 1820s until 1867 there actually a surprisingly good
data especially for precipitation. But after the United States purchased Alaska
we kind of went into the climate dark ages the Russians left as many of you know the
US rapidly lost interest in Alaska after 1867 and except for a very few special observations
for instance at the town that would become known as Barrow during the first international
polar year in the early 1880s. There are very few weather observations in
Alaska from 1867 until the early 80s 90s when interest in Alaska in the lower 40, what was
it the lower 45 then? Started to increase. So the weather bureau out of Seattle and San
Francisco started to send weather equipment to interested people and this is the cooperatives
of observers in the 1890s and that continued through the goldrush era. The first paid weather bureau staff arrived
in Juneau in 1917. That was the house that they soon soon moved
into. And during this time increased this cooperative
network cooperative observers are people like John Boere that do this because they’re interested
and then also established what were called second order stations you can think of those
as contract observers. They got additional equipment they got some
training they got a small amount of money and a lot of that was actually done in support
of the growing aviation industry. In the 1920s the Weather Bureau itself started
to expand opening offices in several locations but then the Great Depression came and there
was actually quite a contraction in the Weather Observation Network. Then of course we got to World War II dramatic
expansion almost overnight. And that sets the stage for our modern Weather
and Climate Network in Alaska. So here’s here’s what this looks like in map
form. So I’m Fairbanks first observations in 1904
Eagle and and on 1901. The earliest dates are of course in southeast
where where the early activity in Alaska was. You can see like Bethel, Utqiagvik in the
1920s that was part of that second order expansion, Nome gold rush 1906. You can see the ones in red here, this is
when the data starts but they’re significant either breaks or station moves that that make
the data less than as clean as we would like the ones in black are pretty solid. Since those dates there so. You can see that we’ve got a fair number of
sites now with around a century of observations. Notice though. In that the now population center of Alaska
there’s actually we have surprisingly poor climate records. Fairbanks was a well-established town by the
time Anchorage was even born. And because of Cook Inlet slow even small
station moves make a big difference so we don’t have as good a climate record in south
central as we like. For our part of the world though we have about
as good as anywhere in the state. Some of you will recognize this. This is the AgFarm now at UAF. We just nominated this site as a world meteorological
Centennial observing site. It’s the only one in Alaska this picture is
undated but probably from the 1920s and I’ve circled the little weather instrument shelter
right there and in fact the observation has been within about 200 yards of where it’s
at now since July 1911. Now the observations are not perfect. There’s a few breaks but this is actually
by far the longest nearly unbroken unmoved site in the state of Alaska. So we’re really lucky to have that here. For Fairbanks and the weather bureau a guy
named Howard Thompson was dispatched from Juneau in the spring of 1929. He showed up in Fairbanks rented some office
space in the Horseshoe building on 1st Avenue. His weather equipment arrived during summer
and fall and he was good to go by that November moved into the new federal building actually
before it was formally opened, moved in in 1933. Stayed there till the war years. Briefly out at Ladd field then the weather
bureau at Weeks field which of course were practically on right. Arctic bowl was a hanger for weeks field,
and then out to the international airport in 1951 where it’s at yet today. These are where the observations are of course
at some point here in the 1970s the forecast office was separated from the observation
site. Nobody cares where the forecast offices. We only care where the [observations] are
from. So here I put them on a map for you. And you can see right there is downtown right
there is about where we are. 1911 to ’29 the AgFarm is the only weather
observation in the Fairbanks area. All of these are basically flood plain observation
sites except the farm. It’s about what 70 feet above the Chena river. So that can be important especially for temperatures
in the wintertime. So this isn’t the perfect record for the official
Fairbanks records it’s moved around a lot and we do have this elevation change possibly
significant but. That’s what we’ve got right. Thought you might be interested in this. This is the very first. Observation The Weather and Climate observation
here on September the 4th 1904 and some of you will recognize the person that signed
this form. The Reverend Hudson Stuck. Very first weather observer in Fairbanks. And so we’ve gone through lots of changes
during the the last hundred and fifteen years. And here’s the current weather observation. This person’s name is ASOS. The automated surface observation system this
is currently-. This platform is off the south end of the
runway. If you’re on the right side of the aircraft
landing coming in from the West. You can see it just before the wheels touched
down off on the right hand side inside the fence there. So it’s not too far from the Tanana river. So a pretty cold place. So I’ve just kind of summarized some of the
things I’ve just touched on here so even even for places like the AgFarm but everywhere
that we have century scale records there’s both type and protocol changes. So that could be from a cooperative station
to the weather bureau or the CAA taking over CAA being the forerunner of the FAA. Location changes of course instrumentation
changes nobody in the United States is using the same kind of equipment to do weather observations
that they were a hundred years ago. And that’s probably a good thing. Electronic thermometers started making their
appearance in the 1960s and are now virtually universal manual precipitation is still done
although increasingly automated there’s oddball things like time zone changes can have an
effect. The 1983 time zone change in Alaska. Probably not a big deal for Fairbanks it was
only one hour in the western part of the state though Kotzebue, Nome, Bethel that was all
that were on baring time as a two hour change so could be important during World War II. There was wartime and then for the cooperative
sites these days generally observe Daylight Time so that can be a confound too. But not all these changes are equally meaningful. Some like at the least the floodplain moves
of Fairbanks probably aren’t a big deal for temperatures but of course even small elevation
changes can be important. OK so. Now the part you’ve all been waiting for. Graphs. Data. Enough of that history crap. Okay so since we’re talking about observations
moving around. And and it’s winter and it’s Fairbanks. So let’s talk about low temperatures so this
is a plot. Every year since the winter of 1905-06 at
each dot is just the lowest temperature recorded that winter. And so what I put on here the bars are those
different observation sites that we just looked at. So back here in the Hudson Stuck era when
the observations were at the Episcopal Church you can see the blue line here was the average
low temperature during those years and you can see it’s much colder than any of the other
times now it’s just a cold few years. Observation of course practically on the banks
of the Chena river. Pretty cold place. And you can see immediately when we when the
observations moved out to the AgFarm. Considerably warmer, now is that a factor
of just the elevation difference were those years warmer. Probably some of both right. 1930 to nineteen forty three observation is
back on Cushman’s street. Federal Building and it definitely colder
than the era of the of the AgFarm brief time at week’s field. This is probably mostly a function of warm
winters that is higher and then the. The last line their last two lines are out
at the airport and I broke these up here. Not based on any weather factor but, pipeline. So is this just a function of the pipeline
era and then the growing urban sprawl of Fairbanks with these warmer temperatures here compared
to the 1950s to the early 70s. What weather and climate factors are going
on in there as well. All of these things are part of the mix right. We’re not going to find, there’s not going
to be one cause right. You’ve got all the interplay of changing urbanization
we’ve got different weather patterns different instrumentation. Different practices. So luckily for Fairbanks we can control for
some of that right. We’re blessed to have that century of observations
at the AgFarm where the observation has been almost in exactly the same site. And lo and behold this works out very well. So from the same site. Out at the Agricultural Experiment Station
1911-12 up through this winter, it’s of course, still we could still be lower than this but
we’re we’re not out of time but we’re almost out of time. But basically we’ve got some cold low temperatures
we’ve got some not so cold low temperatures but there’s obviously not a lot of trend starting
in the 1980s. Notice the Agfarm has not been below 55 below
since the 1970s. We know that this is not caused by a difference
in location. I would argue that what we’re seeing here
is the growing urban effect. There’s also some weather effect here but
especially the lack of really cold here is probably reflecting urbanization. These really warm winter lows are really interesting
these are probably not the result of growing urbanization. The growing urbanization mostly comes into
play. Not entirely but mostly comes into play with
ice fog but when’s the last time we had any good ice fog. Been awhile. So that’s probably not. That’s probably not. This is probably another signal. OK so population in Fairbanks has grown a
lot in the last hundred years especially in the last 50 years. So let’s go out to Tanana. Tanana has as long a record weather and climate
as Fairbanks it’s not as clean. There’s a few breaks population though is
almost exactly the same now as it was 100 years ago. And so here instead of looking at low temperatures
I thought well let’s look at something that’s not likely to be influenced by ice fog or
small changes in elevation. So this is each dot is a plot of how many
days that winter was a high of 20 below or lower. So that’s not an extreme event by interior
standards. And you can see here the blue line here kind
of represents the upper bound of typical for those that care that would be this is the
seventy fifth percentile line and the best fit here is about a little over 20 days up
into the 1970s and it’s come crashing down since then. However if we ask what’s kind of the bottom
end of the low range that has not changed in the last hundred years. Around 6. So here we’ve controlled for both population. And we picked a not so extreme value. And again we see pretty dramatic decrease
in the number of cold days at hand and on. Okay feel free to throw your tomatoes. Okay. So this I thought this would be fun. So the math underlying map here by Dr. Brian
Brettschneider at UAF and this is the lowest temperature for whatever the period of record
we have for various stations. There’s the 66 below in Fairbanks and okk
in January 1934. There’s your 80 below, Prospect Creek 1971. 81 below at Snag, Yukon. 47 75 at Tannacross across the same day. Okay who wants to beat this. Okay so I put some off here to the side just
because these numbers were getting crowded. There is no reasonable chance that if we had
a long time series from chicken that the low would only be 72 below 40 my country is incredibly
cold in the winter, deep valleys high elevation. We’ve only got about 25 years of data from
chicken and none of them are during the epic cold snaps. So if somebody anybody here ever live in chicken
in the winter. Well if we do it’s not hard to find people
that say, oh, it was 85 below it’s 90 below. All right well is that just a tail grows taller
thing. Well it might be some of that but it might
be true too. You can see the coastal effect very nicely
here with them of course. Extremes are not nearly as high or as low
on the coast the interior is the land of. Is the land of 60s and 70s below for sure. One below all time record low in Sitka. OK. So temperatures are one thing. How about snowfall. Well snowfall actually shows no trend at all. Over the last 80 years. In Fairbanks you can see this is just a plot
of the seasonal total and there are some big spikes here. The 1937 some here in the 60s the- several
big years there in the 90s. Last year was pretty respectable but overall
there’s no trend. There’s a few low years but snowfall represents
one of those quantities does not. The when you plot how these fall doesn’t fall
into that typical bell shaped curve or the stereotypical bell shape curve you can see
it’s got a long right tail. There’s a there’s a few ways to get a whole
lot of snow in Fairbanks in the wintertime. It’s pretty tough to get less than about 25
inches. So what can we do with this kind of high quality
snowfall data though 85 years worth. Well one of the things we can do is ask how
much could it snow. So you maybe have seen this kind of thing
or heard this kind of thing with heavy rainfalls during Hurricane Harvey this past summer. You may have seen estimates are one in a thousand
year rainfall kind of thing. Well we can do the same thing with snow right. So this is just a plot. So using that snowfall data. So this is this is the return period down
here on the horizontal axis. And this is just snowfall and inches here. And so right now there are maximum 24 hour
snowfall in Fairbanks is around 20 inches and that comes out just about to be a one
in 100 year event. On the other hand the about 73 inches of snow
that fell in 31 days there in that terrible January of February of 1937. That has a return period of almost 150 years. So this is the kind of thing we can do with
these with these high quality data sets. The blue line here. Two week snowfall one in two hundred years
would give us about 68 inches of snow in two weeks. That’d be a lot of snow. But that’s entirely. That’s certainly within the realm of plausibility
if this were to happen sometime soon. This is the kind of information that I would
be using when I’m talking to media about all this is we broke the record by so much how
unusual is this. This is the kind of information that I’d be
using. I hate freezing rain. We’ve got a lot a lot a lot of questions about
freezing, we get a lot of questions about freezing rain because it’s so impactful. Unlike the lower 48 of course when it rains
here in November significantly it means secondary roads are going to be icy until breakup. Bad bad news for us. So. I trawled back through the historical data
since 1916 and I was looking for big rain events. More than a tenth of an inch. I picked this out not because that’s impactful
nowadays. Nowadays the smallest amount of rain is very
impactful but I picked this so that I could go back through that that daily data and have
a reasonable estimate and the results were very surprising to some people. For those of you that were here in the 1960s
pre-pipeline may remember this. There’s a big spike of freezing rain events
in the 1960s when many of us in this room or at least some of us in this room showed
up in Alaska in the 80s early 90s. There is a drought of freezing rain events
there weren’t very many of these big events. Not-. Neither are these quite compares to the 1930s. Late 20s and 30s when there were multiple
rain events during the winter winter here defined as November through March really quite
quite remarkable. If this were to happen today. I mean we’d be brought to a standstill. But as it is the main thing that we’re seeing
now it looks like is we’re on a long streak here. The green line is the just the events per
decade of people’s working memory and. Looks like work is getting to be about as
long as that 1920s streak. So very interesting. Did I mention I hate freezing rain. Just wanted to be sure but OK so now that
you’re chilled. Thought I’d bring up a couple of other examples
looking ahead so River breakup which I know many of you will take an interest in. A little bit about growing season. And then I thought I’d finish up with a couple
of slides on what’s going on in Utqiagvik. It’s been in the news. I figured some of you would ask. So I threw in a slide or two. OK so Tanana River at Nenana. Many of you have probably made this graph
yourself. And if you haven’t you’ve surely seen it. This is the Rick Thoman version. So in the green green band here is kind of
the near normal. Anything above the green band would be later
than normal below the green band earlier than normal. And as we all know Break-Up is very strongly
concentrated now significantly earlier than it was in the first two thirds of the 20th
century to the point where last year’s May 1st breakup was very close to significantly
later than normal in our era. What would have been significantly earlier
than normal 60 years ago. So and of course. Tanana at Nenana Break-Up is the climatologists
dream data set because it’s been made in the same place in the same funky way. For a hundred years the population of Nenana
hasn’t changed very much. The Tanana River is one channel there. It’s the ideal dataset. But of course it’s ideal but it’s still not
good enough right. I didn’t know this till I started researching
this. Break-Up on the Tanana at Nenana was not the
first breakup betting game in Alaska. That was it for Fairbanks the Chena at Fairbanks. And it turns out we’ve got very good breakup
dates on the Chena river since 1903. So this is just the winter after ET Barnett
got booted off the Lavelle Young. So what I did here and of course the Chena
river of the early 20th century is not the same Cheen a river that we know today prior
to the Dyking at Moose Creek that occurred in the early 1940s. The Chena river was regularly referred to
as Chena slough and it carried mostly, most of the year, it was mostly carrying Tanana
River water. So it was carried a lot more water than it
does now. So we’ve got about 25 years of overlap on
breakup dates on the Chena at Fairbanks before the dike was built. And at the same time we have the breakup dates. For the Tanana at Nenana. So I just correlated those dates and then
just use those as a prediction for, okay well given this correlation prior to the mid 1940s
what would have breakup dates have been on the Tanana at Nenana and do the dots are the
best guess the little lines there that’s the kind of the the 80 percent chance that the
date would fall somewhere on those lines and in some ways this was completely uninteresting
right. It look I mean it looks exactly like when
we have information but what it does it helps reinforce the idea, there’s nothing different
going on prior to 1917 back to 1903. We see exactly the same thing with the breakup
information on the Yukon from Dawson where breakup dates from 1898 up through the middle
part well, past the middle part of the 20th century, were were random scatter didn’t show
much trend. And then in recent decades are getting earlier. Alright some of you want to plant your garden. Big big hot topic every spring and fall is
growing season. So again we’re going to use our AgFarm are
perfect (not perfect) but our long century long observations from the unmoved site at
the AgFarm and the green bars are. This is just the. This is the start of the growing season. This is the end of the growing season, growing
season here just defined as the first, the last and first occurrence of a temperature
of 32 degrees. And so this is really interesting I think. So the best fit line here in the fall is just
this steady increase we’ve increased there what. From August 22nd to September 5th is the best
fit here for the median breakup date. Notice though on the first the last freeze
in the spring doesn’t show that nice steady trend but rather it’s flat here up until the
the late 1970s and then shows a fairly substantial decline since then. So that’s what we’ve lost- we’ve gained two
weeks of freeze free. In the median date here in the spring time. So that’s right. So overall we’ve gained about a month. Here at the AgFarm for the duration of days
with temperature remaining above 32. So I’ll let you in on a little secret here. A paper I’m working on right now so we’re
going to do this kind of analysis for the whole state. This one go from 1952 to through this past
fall. And curiously all the places where the growing
season is increasing are in red and there are some places where it’s actually decreasing. I picked 1952 because all of these sites have
had observations at the same place since then. The 10 is actually the Fairbanks Airport there
but it’s kind of interesting here. So both battles and Fort Greeley airport. The growing season has decreased slightly. It’s not statistically significant. That is to say could just as well be random
chance that it’s shorter on this area in southwest Alaska is interesting too only because it
looks spatially coherent there. So but really some really quite dramatic changes
over mainland Alaska. Thirty days at Eagle, 32 days at Kotzebue
with a pretty strong marine component. That’s probably reflecting changes in sea
ice and ocean temperatures in the fall. Okay last little bit. And I wasn’t going to put this in but I thought
that people would ask because it’s kind of different so you probably have heard in the
news this fall that temperatures at Utqiagvik have increased so much that some algorithms
run by NOAA have spit out the data as being-. They didn’t think it was realistic, a realistic
climate change. Now let me let me preface let me post this
by saying everyone knows the data is fine. It’s the algorithms that were having a problem
and we’ll see this right here. So this is the average temperature for the
month of October since 1920. And you can see the red dots are Octobers
that were warmer than the long term average significantly warmer. The black dots are near normal. The blue dots are significantly below normal. OK so exactly nothing interesting here right. So you’ve got some warm Octobers you got some
cold Octobers. Yeah. You know if you look here. So there’s a bunch of warm Octobers back here
in the 1930s and 40s. There’s kind of a dearth of warm Octobers
in the 1960s. But all in all doesn’t look very exciting. So that’s up through 2001. Well what’s happened since then. Ta-daa. Congratulations ladies and gentlemen you’ve
just witnessed the birth of a new climate. So this is of course is entirely driven by
the collapse of autumn sea ice. So not only since 2001 is every October significantly
warmer than normal there has been a complete collapse of the year to year variability that
we had back here. Now I say birth of a new climate it’s not
a new climate in the whole world. But Barrow’s gone from kind of a mixed Arctic
marine and Arctic continental depending on the ice that October to entirely an Arctic
marine climate. Stunning. And this is the kind of change that was throwing
off NOAA’s QC algorithm. And that’s because that algorithm was tuned
to look for changes that would be typical at middle attitude’s. While the Arctic is warming. Much faster. Than the rest of the world and places like
Utqiagvik, places like Svalbard that have the potential for a very strong marine influence. Those changes in ice. Have a dramatic, dramatic effect. So last wrap up slide. Another photo inadvertently containing a weather
station. Since the 1890s this talk is the work of thousands
of people. We’ve got a good number of sites now with
a century or more of observations and I hope I’ve given you a taste of what we can do. When we have high quality information over
longer than a human lifetime. It’s not just about climate change. That’s part of it. But there’s many other things that we can
do with that information that can help. As we build our state in our community for
the future. So. With that I would like to again thanks Fritz
and Sue and Ned Rozell who reviewed this and thanks for coming.

Author Since: Mar 11, 2019

Related Post