May 2017 Observations and Model Discussion for the Pacific Northwest

[Front Ranger]

Up to 4.77" of rain on the month. Slightly above the monthly average.

 

Looking hot early next week. Could be the warmest weather of the summer

1983!

[wx_statman]

1983!

 

2016!

[Front Ranger]

2016!

 

 

At PDX, the hottest weather of 2016 occurred in June and August.

[wx_statman]

 

At PDX, the hottest weather of 2016 occurred in June and August.

 

100 on 5/28/1983

100 on 6/5/2016

 

8 days

[Front Ranger]

100 on 5/28/1983

100 on 6/5/2016

 

8 days

 

True. But 1983 holds the distinction of having the warmest temp of the year occur in May, and it was not equaled later in the summer as 6/5/16 was.

[Phil]

Up to 4.77" of rain on the month. Slightly above the monthly average.

 

Looking hot early next week. Could be the warmest weather of the summer

Already 96*F here as of 1pm. Beat that.

[Guest happ]

Already 96*F here as of 1pm. Beat that.

 

That is wretched.

[Jesse]

Already 96*F here as of 1pm. Beat that.

We don't want to.

[wx_statman]

True. But 1983 holds the distinction of having the warmest temp of the year occur in May, and it was not equaled later in the summer as 6/5/16 was.

 

That's true. It was interesting, to me, how close we came to matching that feat last year. 

 

Although to be fair, PDX should have definitely topped 100 in August. We underperformed on 8/19-8/20.

[Kayla]

Already 96*F here as of 1pm. Beat that.

 

30º at noon here! Glad to be on the downstream out this way. 

[BLI snowman]

That's true. It was interesting, to me, how close we came to matching that feat last year. 

 

Although to be fair, PDX should have definitely topped 100 in August. We underperformed on 8/19-8/20.

 

The 2nd place May temp of 99 on 5/29/1887 was also the hottest of that year by a decent margin.

[Phil]

That is wretched.

Actually not too horrible with dews only in the mid-60s. Still early enough for less swamp and more evening cooling.

 

A few months from now this would be a 100/80 afternoon.

[Phil]

This is pretty cool.

 

 

Great example of cold front overtaking dryline, augmenting low-level convergence and aiding convection initiation! @VORTEXJeff @ScottSteiger pic.twitter.com/4RK9yKslGz— Jake Mulholland (@JakeMulholland1) May 17, 2017

That is absolutely sick.

[Phil]

True. But 1983 holds the distinction of having the warmest temp of the year occur in May, and it was not equaled later in the summer as 6/5/16 was.

Speaking of 1983, it's the top analog on the 06z GFS based CPC super-ensemble, followed closely by 1993, which also featured a torchy May in the PNW for similar reasons (ISM genesis igniting MJO, following the start of a huge -NAO).

 

http://www.cpc.ncep.noaa.gov/products/predictions/short_range/tools/gifs/500hgt_comp_06gfs814.gif

[Front Ranger]

Speaking of 1983, it's the top analog on the 06z GFS based CPC super-ensemble, followed closely by 1993, which also featured a torchy May in the PNW for similar reasons (ISM initiation igniting MJO, following the start of a huge -NAO).

 

 

Indeed.

 

Though from what I've seen, those analogs are only helpful for the upcoming pattern 2-3 weeks out. Not much predictive value beyond that.

[Phil]

Indeed.

 

Though from what I've seen, those analogs are only helpful for the upcoming pattern 2-3 weeks out. Not much predictive value beyond that.

What makes you say that? The top-three scoring analogs (since February) are 1993, 1980, and 1983, respectively.

 

Years with low frequency 160E-to-dateline subsidence will perform the best this year, in the long run. The fact that the timing of the intraseasonal forcings is roughly on par with 1993/1983 is just a lucky fluke, but their low frequency (background) states are very similar to 2017.

[Front Ranger]

What makes you say that? The top-three scoring analogs (since February) are 1993, 1980, and 1983, respectively.

 

Years with low frequency 160E-to-dateline subsidence will perform the best this year, in the long run. The fact that the timing of the intraseasonal forcings is roughly on par with 1993/1983 is just a lucky fluke, but their low frequency (background) states are very similar to 2017.

 

Watching them over the years.

[Phil]

Watching them over the years.

Well, that's why we separate "pattern analogs" from "seasonal analogs". In the long run, the best seasonal analogs will manifest more frequently as pattern analogs, too, which has been the case with years like 1980, 1981, 1983, 1991, 1993, and 1995.

 

There are still some differences amongst those years, due to other intraseasonal and peripheral forcings, but the general theme is similar from a multi-month standpoint.

[WeatherArchive]

What kind of temps/humidity do we see under retrogression this time of year? Do we see colder or warmer weather and what about nights?

[TT-SEA]

Well, that's why we separate "pattern analogs" from "seasonal analogs". In the long run, the best seasonal analogs will manifest more frequently as pattern analogs, too, which has been the case with years like 1980, 1981, 1983, 1991, 1993, and 1995.

 

There are still some differences amongst those years, due to other intraseasonal and peripheral forcings, but the general theme is similar from a multi-month standpoint.

 

I have to give you major props regarding those years... they have continued to pop up over and over since you started tracking it almost 2 months ago.  

 

It will be interesting to see how June plays out.    

 

On the one hand... those years were WAY drier in the spring months here (Feb-May) and a wet June and July seemed more likely than this year which has been record setting wet in that period.   Its seems unlikely that we continue with a very pattern all the way through July after coming off a record setting wet spring.   That has not happened before.  A very wet spring typically means a shift to dry and warm in June.

 

But here we have a late May hot spell approaching and that was a hallmark of 1983... and 1993 had the same thing about 2 weeks earlier.  

 

Do we continue to follow analog years for such a long time?   In the past, it seems like we have tracked one analog year for awhile and then completely break away to something different. Personally, I am also hoping that the volcanic influence played a role in those summers and the result might be different (drier and sunnier) here even its below normal.   

 

Time will tell.   I cannot argue against those analogs now even though they represent two of the worst June/July periods ever here in terms of persistent rain.  

[Phil]

I have to give you major props regarding those years... they have continued to pop up over and over since you started tracking it almost 2 months ago.

 

It will be interesting to see how June plays out.

 

On the one hand... those years were WAY drier in the spring months here (Feb-May) and a wet June and July seemed more likely than this year which has been record setting wet in that period. Its seems unlikely that we continue with a very pattern all the way through July after coming off a record setting wet spring. That has not happened before.

 

But here we have a late May hot spell approaching and that was a hallmark of 1983... and 1993 had the same thing about 2 weeks earlier.

 

Do we continue to follow analog years for such a long time? Seems like we can track one year for awhile and then completely break away to something different. Personally, I am also hoping that the volcanic influence played a role in those summers and the result might be different (drier and sunnier) here even its below normal.

 

Time will tell. I cannot argue against those analogs now even though they represent two of the worst June/July periods ever here in terms of persistent rain.

Thanks for the kind words.

 

Despite overall structural similarities, I highly doubt we'll match 1983 or 1993 as far as the UV200 jet is concerned. Both of those years were off-the-charts intense in that regard, mostly thanks to weakened Asian monsoons and/or the residual +QBO/lowered tropopause. The volcanic forcing then enhanced the overall static-stability in the tropics in those years, which was already present thanks to the residual +QBO at 50mb, so we entered into a perpetual constructive interference regime in both of those eras. This was partially, if not mostly responsible for the persistent ENSO behavior through those two interdecadal stretches (middle 1980s and early/middle 1990s).

 

This year, we have the residual +QBO and modest Asian monsoons, as well as the weaker zonal SST/convection gradient across the tropical Pacific. However, it's definitely hasn't reached 1983/1993 levels yet.

[Front Ranger]

I have to give you major props regarding those years... they have continued to pop up over and over since you started tracking it almost 2 months ago.  

 

It will be interesting to see how June plays out.    

 

On the one hand... those years were WAY drier in the spring months here (Feb-May) and a wet June and July seemed more likely than this year which has been record setting wet in that period.   Its seems unlikely that we continue with a very pattern all the way through July after coming off a record setting wet spring.   That has not happened before.  A very wet spring typically means a shift to dry and warm in June.

 

But here we have a late May hot spell approaching and that was a hallmark of 1983... and 1993 had the same thing about 2 weeks earlier.  

 

Do we continue to follow analog years for such a long time?   In the past, it seems like we have tracked one analog year for awhile and then completely break away to something different. Personally, I am also hoping that the volcanic influence played a role in those summers and the result might be different (drier and sunnier) here even its below normal.   

 

Time will tell.   I cannot argue against those analogs now even though they represent two of the worst June/July periods ever here in terms of persistent rain.  

 

And that's just it. Sure, they've been good large-scale analogs for the spring, but that's still resulted in very different results for the smaller-scale details. And then beyond that, spring is a transitional season...you'll find limited correlations between dominant Mar-May patterns and the summers that follow.

 

Then you have the fact that this spring is transitional from an ENSO perspective (going from -ENSO to +ENSO), and I think the odds are even lower that the same spring analogs will continue dominating through the summer.

[Front Ranger]

12z Euro is less potent with the clipper and resulting crash late next week.

 

Looking like May 2017 could end up quite warm.

[TT-SEA]

12z Euro is less potent with the clipper and resulting crash late next week.

 

Looking like May 2017 could end up quite warm.

 

The EPS mean is also less robust.

 

But the 12Z ECMWF control run is quite robust and it actually cuts it back over the PNW where it sits for several days into Memorial Day weekend.

[Phil]

And that's just it. Sure, they've been good large-scale analogs for the spring, but that's still resulted in very different results for the smaller-scale details. And then beyond that, spring is a transitional season...you'll find limited correlations between dominant Mar-May patterns and the summers that follow.

 

Then you have the fact that this spring is transitional from an ENSO perspective (going from -ENSO to +ENSO), and I think the odds are even lower that the same spring analogs will continue dominating through the summer.

Eh, it's crucial to understand why these analogs have performed well, whether the boundary conditions in the analogs are representative of the present boundary conditions, and whether that will remain the case going forward.

 

You'll find limited success during the spring transition if you rely on linear statistical correlations, but if you account for lower frequency system states and pattern forcings, there is homogeneity in the transitions themselves as well as during the subsequent season.

 

As for ENSO, well, there really won't be a strong signal there this summer, and if anything, this weak +ENSO move is working to reinforce the analog signal for a dampened zonal SST gradient across the Pacific and weakened Asian monsoons. A commonality between 1980, 1983, and 1993 is a warm EPAC relative to the WPAC/IPWP.

 

As for the next several weeks, it's already quite evident as to where this pattern is going. The EAMT/jet cycle, which is tied to the MJO/ISM genesis and upcoming warm pattern in the west, will reverse phase in June after the MJO bifurcates from the ISM anomaly and orbits the WHEM. When it cycles back, the dateline subsidence resumes and the background state of GOA/west coast troughing will resume.

[Jesse]

A warm May overall would likely bode well for a cooler summer.

[TT-SEA]

A warm May overall would likely bode well for a cooler summer.

 

Seems to work that way usually.

[Jesse]

Seems to work that way usually.

Not that it's a slam dunk by any means. Truly cool summers have become a very rare breed around here in recent decades. But the warm May/cool summer script seems like a somewhat common one.

[BLI snowman]

Seems to work that way usually.

 

Only occasionally. Just as many years like 1958, 1972, 1987, 1992, and 2014 where a warm May kicks off summer going on forever.

[Phil]

Only occasionally. Just as many years like 1958, 1972, 1987, 1992, and 2014 where a warm May kicks off summer going on forever.

It's worth noting that all of those years featured deep El Niño circulations with strong dateline convection and weaker Walker Cells and trade winds.

 

This year is mostly opposite in that regard, almost a hybrid between an east-based Niño and west-based Niña. A tight Walker Cell and strong trade winds going forward will only act to reinforce this regime:

 

http://i724.photobucket.com/albums/ww243/phillywillie/Mobile%20Uploads/7B3365BF-4994-4D1A-9422-F51EE79BD416_zpsfcjneiqv.jpg

[Jesse]

Only occasionally. Just as many years like 1958, 1972, 1987, 1992, and 2014 where a warm May kicks off summer going on forever.

I was going to post how strong +ENSO often plays a role in years like these, but it looks like Phil beat me to it.

 

Also, even if this May ends up warm overall, it was not a wall to wall torch like some of those. Lots of ups and downs. But a generally warm last two weeks will likely clench it as an above average month.

[Jesse]

First real marine layer esque day of the season today.

[Front Ranger]

Seems to work that way usually.

 

Nah, I posted the correlation map yesterday. Overall, it's warmer than normal for June/July/August.

[Front Ranger]

I was going to post how strong +ENSO often plays a role in years like these, but it looks like Phil beat me to it.

 

Also, even if this May ends up warm overall, it was not a wall to wall torch like some of those. Lots of ups and downs. But a generally warm last two weeks will likely clench it as an above average month.

 

If the 12z Euro were to verify, good chance we're looking at a significantly warm May for the region...probably warmer than 2013, 2015, and possibly 2014. 2016 is probably out of reach.

 

And wall to wall torch in May is very, very rare. Most warm Mays have ups and downs.

[Phil]

Nah, I posted the correlation map yesterday. Overall, it's warmer than normal for June/July/August.

Eliminate nonsense ENSO analogs and that correlation reverses. Your analogs were all over the climatological map.

[Jesse]

If the 12z Euro were to verify, good chance we're looking at a significantly warm May for the region...probably warmer than 2013, 2015, and possibly 2014. 2016 is probably out of reach.

 

And wall to wall torch in May is very, very rare. Most warm Mays have ups and downs.

Thank you for the bounty of unnecessary caveats.

[Front Ranger]

Thank you for the bounty of unnecessary caveats.

 

The only caveat there was "if the 12z Euro were to verify". That's a fairly common one on this forum.

[Front Ranger]

Eliminate nonsense ENSO analogs and that correlation reverses. Your analogs were all over the climatological map.

 

ENSO is in flux. Only using the years you approve of would result in a very small sample size, which then significantly weakens any correlative power.

[epiceast]

The only caveat there was "if the 12z Euro were to verify". That's a fairly common one on this forum.

Big if true?!

[Phil]

ENSO is in flux. Only using the years you approve of would result in a very small sample size, which then significantly weakens any correlative power.

What on Earth? The ENSO describes the entire systematic state of operation. Warm Mays in the PNW describe nothing. Why rely on a faceless variable at the expense of the entire system state?

 

If you have an ENSO mismatch, you're relegated to comparing two entirely different system states, holding little to no collective homogeneity. Hence, the reason why your analogs were/are all over the place.