weatherwiz

That December pattern is a pretty classic EL Nino look, especially with some of the more recent EL Nino's. That is a great evolution into January though, but my concern would be suppression and perhaps moreso, a rather inactive storm track.

I know the depth of the warm waters don't seem particularly deep, at least enough to make me feel like we are headed towards EL Nino. We'll have to see if we can maybe get some strong WWB's as well. After a rapid transition into EL Nino, however, the degree of warming has certainly leveled off and maybe even slightly decreased.

Something else I wanted to add to this. It's also extremely interesting the gap between weak EL Nino events. After the 1979-1980 weak EL Nino (which isn't even defined via the ENS-ONI) there was not a weak EL Nino until 2004-2005 (which also was not defined via the ENS-ONI).

I'm not even sure if a quick turn around would have much of an influence.

I've spent much of the weekend constructing my thoughts on the upcoming winter and really hammering into things and there is one thing I've noticed. There has been an absolutely glaring shift in atmospheric circulations between older EL Nino events and newer EL Nino events (say from like the 1970's or early 1980's onwards). Now it is interesting to note that it is possible data sources and data sets could be a big driver here. While there are ENSO categorizations dating back into the mid-to-late 1800's it's obviously important to understand there may be questions with the validity and accuracy of the data. I know Ray only focury900-onwards. The NCEP/NCAR R1 dataset contains data from 1948-Present and the 20th Century Reanalysis V3 dataset contains data from 1836-2015. So for the earlier ENSO years I used the CRV3 dataset (years 1900-1980) and NCEP/NCAR from 1981-Present. anyways, check this out. Note: These are just weak EL Nino's. (some of the earlier ones were borderline moderate so I removed them and re-plotted, trying to see if it maybe would skew more towards later EL Nino years and no. Obviously the major thing here is a very weak (perhaps even reversed) Aleutian Low. Anyways though certainly could be other drivers (PDO state which I'm starting to dig into as well).

Are you not able to look at previous dates anymore? There used to be a whole thing at the bottom where you could look at different dates going back like several months or even year. https://www.ospo.noaa.gov/data/cb/ssta/ssta.daily.current.png

Yeah the amount of resources and work would be very large and quite complex. Way above anything my noggin can do

Agreed - good point. I really wish we had a great dataset of like near-miss events. I know the KU books has some cases on these but having such a dataset I think would expand skill set significantly. Of course, there would have to be some definitions to define "near misses". I mean the GFS or CMC or some random model from someone's basement showing a blizzard 10 days out which doesn't verify does not count as a "near miss".

I've never been a fan of snowfall predictions in seasonal outlooks. IMO, something such as snowfall or even precipitation isn't necessarily tied into the pattern, but more so deviations which happen during pattern evolutions. When you look at seasonal snowfall totals and compare to ENSO, NAO, AO, etc., sure you may see some "correlation" but the correlations aren't very strong in that the spread is still relatively high. While a pattern such as WC ridge and EC trough can be associated with increased snow chances, that doesn't necessarily mean it is going to happen or will occur (which I think everyone understands). I think what's more important is how the pattern is evolving which is very tough to create in composites. You can create H5 anomalies for snowiest and least snowiest winters and certainly draw a quick correlation, however, what that won't tell you is how things evolved leading up to the event. That is more important than just WC ridge/EC trough in the composite.

I am hoping to spend a good chunk of this weekend formulating some thoughts on the upcoming winter. I've always found seasonal forecasting to be quite intriguing, albeit stressful but haven't really been able to dabble much into it. Started getting back into it last fall.

Came across this chart. Not sure how old it is. But it looks like values closer to 1 indicate it has a strong impact on the pattern and values closer to 10 indicate less importance https://www.cpc.ncep.noaa.gov/data/teledoc/teletab.gif

I found a bit of a work away after going at it for some time. 1) I did not realize most web browsers no longer support ftp. 2) I don't know what made me try this but within the url I just removed ftp and added http://www. and it worked!!! I had no other clue of accessing TNH data...looked everywhere. Though my brief start at the TNH (which I haven't really looked at in like 10 years) is it is very chaotic in nature.

Does anyone know why on the CPC indices pages when you click "Historical Index" to view the data this pop up happens?

Crazy morning here. Will have like a 45 second shower and have had periods where it's pretty bright with sun peaking through.

Missing the sunset pics

What kind of spider is this

I've been debating on just focusing on 1950 onwards as well, however, we do have enough reliable data to incorporate years prior to 1950, it's just there has to be an understanding that the data be used with caution. The issue with just going from 1950 onwards is you're really limiting your dataset and eliminating potential outcomes or probabilities.

I've been reading through some papers and doing some additional research. What I'm going to mention below is something I already knew but it always helps to just verify what you're thinking. When dealing with these early years there needs to be extreme caution given the uncertainties which exist with the viability of the data. There also needs to be an understanding of the strengthens and weaknesses of the dataset being used. For example, with my SST stuff, my source is the HADISST dataset. There appears to be some question marks with this dataset after the 1980's due to a cold bias. Also, the earlier years there is big question marks within the northern latitude oceans (which I think is reflected very well using an 1881-1910 / 1891-1920 climo period). I'm just further trying to understand all the biases, strengthens, and weaknesses because at some point when I start blogging about this more or making a website I want to be sure I am properly communicating this and explaining any results I am presenting are just based on my choosing of the dataset and periods and results could differ slightly if using other methods.

This falls into the category of just finding someway to argue for a cold/snowy winter.

Figured today would be a misty/drizzly day in places. MOS/NBM profiles across the region yesterday were showing some low visibility numbers with BR in the obs. Sometimes you'll see the NAM go wonky with that stuff but GFS/NBM were like that too so consensus was there.

Maybe a fishing spider. I wouldn’t sleep down there either

Looks like a wolf spider?

I'm really unsure how Saturday unfolds. I do think western areas see a quite a bit of rain from eastern NY through VT, western MA, and western CT. Strong cold front with great dynamics and plenty of moisture advection. This seems like a fairly high confidence forecast. Going to central and eastern areas is a bit wild card. There may be quite a bit of subsidence between the shortwave energy associated with the approaching trough and Philippe off to the east, however, if Philippe were to interact with this front and trough the stage would be set for some heavy rain region wide, though the greatest totals should still be west.

I'm wondering if I made a terrible mistake comparing variables to their respective climatology periods, especially with sea-surface temperatures. Per PDO dataset, the PDO phase was negative during ASON of 1921 (averaged value = -1.48). However, here are the SSTA's for that period using a climo period of 1891-1920. Clearly, the PDO signal is very weak. Well I was doing some reading and I believe the current version of the PDO still uses a base climatology of 1971-2000. So here is the SSTA's compared to 1971-2000 Clearly this exhibits a negative PDO look. Now when dealing with these older years reliability of the data needs to be considered and the data set you use (HADISST, COBE, NOAA ERSST, etc). But the discrepancies above yield a significant question; How do you go about conducting analysis on these earlier years? I mean I suppose there is no right away. I mean if comparing to it's respective climo period, the PDO signal is very weak, but if comparing to a different period, maybe that yields a stronger signal. But the truth of the matter is you can't (I don't think compare a year to a set of years which hasn't happened yet. But it is very possible too that the data from that period is very scarce so would it be better to compare to a more recent period? Reading through literature there are questions regarding the reliability of the data from the 1800's to even like 1940. AHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH

We use Ac in the bedroom at night. Bedroom is super warm. I don't mind it cold as long as I can wrap myself up. I usually end up looking like a caterpillar getting ready to become a butterfly. One of these days I expect to wake up as a butterfly and flutter around the house