OCTOBER OUTLOOK: Trick or Treat?

October is very much a transitional month weather wise in Colorado. We can, and regularly do, experience three seasons worth of weather in one month: summer, fall, and winter. Those who have lived here long enough know that the weather often changes on a dime: 80s one day, with a frost or freeze, or sometimes even snow, a day or two later. Here I'll endeavor to paint a picture of how we might expect our weather to behave over the month of October this year. I'll primarily be looking at past years that featured similar overall atmospheric conditions to those we're experiencing now to predict what we might expect; this is called looking at analogues

We've seen predominately hot and dry conditions since May, with the most 90° degree days (total and consecutive) on record; the most days above 95°, and the most 100° degree days on record. All that heat was accompanied by little in the way of regular moisture. In fact, most of the state of Colorado is experiencing moderate to severe drought conditions at present. That said, there are reasons to believe this pattern is in the process of breaking down and that October has the potential to feature periods of colder-than-normal conditions and the potential for accumulating snow earlier than usual. 

In making longer term outlooks, meteorologists look at many different overall atmospheric patterns and oscillations to provide guidance as to the major factors influencing the 'big picture' meteorologically. Right now we are experiencing weak El Nino conditions, which are expected to persist through the winter. We are also currently in the negative phase of the Pacific Decadal Oscillation (which functions a bit like the El Nino/La Nina cycle, but on a larger time frame), and we're currently in the positive phase of the Atlantic Multidecadal Oscillation. What does all that mean? It is constructive to go back and look at previous Octobers that featured these same overall conditions. The closest match happens to be quite recent -- October of 2009; there are other years that match up as well, though not as closely as 2009: these include 1951, 1957, 1963, 1965, 1968, 1972, 1976, and 2002.

The Climate Prediction Center (part of the National Weather Service) that produces longer-range outlooks is very hesitant in forecasting cold in any of their outlooks unless there is strong evidence to suggest a high likelihood. I know this because I used to work there! Why? Persistence suggests that we've switched into a longer-term cycle where warmer than normal conditions occur more regularly and with more intensity than periods of cold. Nonetheless, and somewhat surprisingly, they have a large swatch of colder than normal predicted for the eastern half of the US, with Colorado (at least east of the Divide) in the "equal chances" category, which to suggests, to them, that there is not strong evidence one way or another to support a forecast for warmer or colder than normal conditions in our neck of the woods. Respectfully, I disagree...


The analogues paint a slightly different picture -- one with more cold across most of the country including Colorado. First we'll take a look at all of the Octobers that match (to varying degrees) the setup we're seeing right now. That would suggest that almost the entire country has the potential to see colder than normal conditions in October. If we look at the best matchup -- which is October 2009 -- we see a similar picture, though with the cold signal positioned almost directly on top of Colorado. What we take from this is that the analogue years are suggesting that the current atmospheric oscillations and patterns are conducive to widespread cold developing this October.

Meteorologists can consult analogues for longer term forecasts, but it is helpful, indeed necessary, to also consult model guidance that tries to predict how weather systems will develop and progress. At least for the first half of October, we're seeing a cold signal as well -- this from the Coupled Forecast System (CFS). This predicts colder then normal temperatures through at least the middle of the month for a broad swath of the eastern two-thirds of the country. This model has pretty good accuracy within the Week 1-2 period, so we can say with a fairly high degree of confidence that colder than normal temperatures are likely to develop. In addition, shorter-range models also show a sharp cold front sweeping through the center part of the country by late in the day 10/3; and medium-range models suggest that the cold air this system is bringing with it is likely to last at least a weeks. This colder regime is likely to bring an end to the growing season, with a widespread frost or freeze likely by the morning of 10/4. Note, some models are also suggestive of the possibility of our season's first snowfall in the October 7-14 period -- about a week earlier than normal.

When we put all of this together, I can say with fairly high confidence that the first half of the month is likely to feature below normal temperatures with near normal precipitation. Predicting what will happen during the  second half of the month, since is further out, is a lower confidence proposition, but I think we can expect near or slightly-below normal temperatures with near normal precipitation. (There are some suggestions that the cold, relative to normals, may shift to the east and ease up somewhat over Colorado, though it's premature to predict this with any confidence.) Bottom line, though: I would be very surprised if we don't have at least one accumulating snowfall event before the month is up!

Finally -- some food for thought: Remember 2009, which I've said is the closest match to how the atmosphere happens to be behaving this year? That October saw a whopping 17.2" of snow in Denver, some 400% of our normal October tally of 4.2"! And, temperatures averaged a full 8° below normal -- super cold. (Interesting to note, and we'll tackle this a month from now, but that November also saw more than twice the normal amount of snowfall in Denver.) While I don't think this October will be as cold, or feature as much snow, as in 2009, the writing is on the wall regarding colder than normal conditions and increased possibilities for snow.