70s return; MN drought deepens; How “outlier” model nailed “Drought 2012” forecast

41F degrees low temp at MSP this morning

30F degrees at Lakeville this morning

+11F “Urban Heat Island Effect” this morning from city to outer suburbs

80F possible in SW metro by Friday?

New U.S. Drought Monitor issued this morning

77.45% of Minnesota now in “drought.”

64.11% last week

+13% in one week

35% of Minnesota now in “severe” or “extreme” drought

ClimateCast: “Outlier” models accurately predicted drought & heat of 2012 – in January 2012 -details below

MSP quick look forecast:

119 ql.PNG

Source: Twin Cities NWS

Classic “Urban Heat Island Effect” (UHI) today:

This is textbook stuff.

Conditions this morning were perfect for a pronounced UHI effect around the metro. The magnitude was about 11 degrees…it was 41 at MSP Airport and 30 at Lakeville this morning.

(click image to enlarge)


Source: NOAA/University of Utah

That’s the difference between a solid frost and no frost. The UHI has “saved” the metro inner core from the season’s 1st frost several times this month. With milder days in the forecast, it looks like the inner metro core will escape frost (well?) into October again this year.

70s and 80s return:

The center of seasonably cool Canadian high pressure is sliding east today. As winds begin to turn south, temps will warm the next few days.

119 wxs.png

Source: Twin Cities NWS

Look for 70 in southern Minnesota and the metro today, and temps will make a run at 80 tomorrow, especially southwest of the metro. A “retrograding” low pressure system to the east may keep the temp rise in check this weekend, but temps in the mid 70s will be a good +10 degrees vs. average this weekend.

Drought deepens: Reaching critical phase now

This is no longer about throwing a little extra water on your lawn in Minnesota.

Drought is now deepening to “severe” and “extreme” levels in Minnesota, according to today’s updated U.S. Drought Monitor.

119 DM.PNG

Source: http://droughtmonitor.unl.edu/

77.45% of Minnesota is now in drought, up 13% in just one week. Only the tip of Minnesota’s Arrowhead is staving off drought…and not for long.

The real concern is areas on northwest and southern Minnesota that are now pushing into the “extreme” drought category. Areas surrounding Thief River Falls in northwest Minnesota, and along I-90 is the south from Luverne to Worthington and up to Mankato have slipped into the “extreme drought” category.

I am growing increasingly concerned about our chances to recover from this expanding drought before the soil freeze up in December. If we don’t see multi inch rainfall events between now and then, much of Minnesota is going to be precariously dry heading into Spring 2013.

This could have a huge impact on farmers next year if weather patterns persist.

After a wet spring & early summer, extremely dry soils, plummeting lake & pond levels, and tinder dry forests vulnerable to fire are again becoming the weather story in Minnesota this fall.

The good news? There’s still about 8 weeks for weather patterns to change, and for traditional fall storms to wind up and dump heavy rains (and some snow) in Minnesota in October and November.

I’m not ready to give up on our chances for drought easing rains this fall, but the overall weather pattern does not look favorable for the next 2-3 weeks…and the window is growing shorter by the week.

ClimateCast: How did “outlier” model nail the drought of 2012?

That’s a question many at NOAA are trying to answer these days.

Most of the “ensemble” models used to make long range seasonal forecasts missed the massive drought and intense heat wave of 2012 in the central USA.

One model, NOAA’s GFDL (Geophysical Fluyd Dynamics Laboratory) basically nailed the notion fo drought and heat in the central USA this summer as early as January 2012.

119 GFDL.png

Source: NOAA & Climate Central

But, the GFDL was laregly discounted as an “outlier” solution. The big question is why?

NOAA will be grappling with that question for the next few months, and trying to come up with a way to decide which models can provide the highest “forecast skill” in the future.

Andrew Freedman from Climate Central elaborates:

Although official drought outlooks failed to provide Americans with advanced notice of one of the worst droughts to strike the U.S. since the Dust Bowl-era — a drought that is still ongoing — there were some computer models that got the forecast right. Viewed as outliers at the time by climate forecasters tasked with making seasonal forecasts, such models look downright prescient with hindsight.

In the wake of the flawed forecasts, climate researchers are seeking to understand what enabled certain computer models to anticipate the drought and intense heat that affected much of the U.S. beginning in March, in order to recognize the early warning signs the next time around. Their task is a complex one, since models show varying levels of skill depending on the initial climate conditions and time of year when a forecast is made.

Had NOAA been able to put confidence into the eventual accuracy of the GFDL, the implications are huge.

Farmers would have been able to steer toward “drought tolerant” seeds. Commodities brokers and food companies would have anticipated record corn prices. State and local governments and shipping concerns could have planned for rapidly falling water supplies and low water on the Mississippi River.

Sometimes an “outlier” is a more valuable forecast tool than the “consensus.” The trick is knowing when.


  • ML

    Really interesting post as always, Paul. Thanks.

  • Chris

    So the outlier was right. Trusting the concensus was not. And this was just a short term model run with I’m guessing fewer parameters and better understood parameters than say, oh I don’t know, global climate change. Maybe the outliers are right and we’ll be Venus soon enough, maybe the other outliers are right and in the end change occurs but not that much of a GLOBAL impact. Or maybe the concensus is right.

    Point is be careful when humans are predicting anything…the vast majority will be wrong. Those that are right are just a little luckier, not necessarily better. In the end we are human. And the biggest point is computers only calculate what humans tell them to calculate. It’s a glorified calculator, nothing more.

  • Andy

    The map showing UHI is really interesting, but not at all surprising to those of us who pedal to work early in the morning this time of year (and come spring). From my home in South Minneapolis to my job in Crystal, I feel a significant temperature change (especially on bare skin). I leave home a little chilled, but by the time I’m pedaling past the Guthrie, I’m noticeably warmer (and comfortable). Then, by the time I’m at Plymouth and Emerson in North Minneapolis, the temp is dropping again, even as the sun is starting to show it’s light. The coldest spot on my ride, though, is an interesting experience with micro-climates: I go underneath Broadway/Bottineau where Lowry and Memorial Parkway cross. I can go from warm enough to shivering in less than 30 seconds on some days as I descend into the little bowl of much colder air. From there, the temp feels pretty stable, but if I’ve gotten too cold, there’s no way I’ll warm up for the last two miles.

    The ride home is similar, but not as pronounced. Downtown is always warmest, and then as soon as I cross south of Lake Street on Minnehaha, the temp drops again. Even in January, I can feel the warmth of the downtown core as I pedal through.

    I wish I had a small digital thermometer to mount on my bike to watch the change as I ride, but I’m guessing that I experience as much as a 10 degree swing from downtown to Crystal on some days with 5 degrees being a pretty consistent spread.