Sunday, January 25, 2026 Snow Review

Meteorology is a science, which means it should follow the scientific method.

We make a prediction, Mother Nature runs the “experiment”, we analyze the data. At least, that is how it should work.

From a purely snow-amount driven perspective, yesterday’s storm underperformed… majorly in some areas. Instead of a widespread 12-18 inches, it was more like 6-14 inches.

Using National Weather Service snow report summaries, I have compiled a map of snow reports across our region.

Reports colored Green met the 12-18 prediction. Orange reports at least had double-digit snow amounts, and red reports failed to reach 10 inches.

Why Less Snow?

In my post “Friday morning thoughts on Sunday’s winter storm” under the heading “Storm Impacts & Amounts”, I took a lot of effort to explain how snow flake structure can impact the snow-to-liquid ratio (SLR):

Still, we are talking about significant snowfall, regardless. Were this system all rain, we would likely see three-quarters to an inch of rain. A highly assumptive and usually inaccurate rule of thumb is to multiply that by 10. Our climatology says multiplying by 12 or 13 is a very slightly better method. These methods alone push snow totals toward double-digit territory.

 

However, with the cold air in place, multiplying by 15 or 20 is a bit more accurate for this particular system. You can see how that drastically changes the picture, pushing snow totals well over a foot.

 

The truth of the matter is that this snow-to-liquid ratio is not constant throughout the storm. At times it may be over 20, while other times it may drop below 15. It all depends on the snowflake formation and shapes.

The impacts of snow flake structure on snow amounts is something I’ve noticed in these types of events over the last few years… which is why I wanted to bring it up two days before the storm arrived.

It hasn’t been easy to get accurate liquid equivalent measurements to match up with snowfall amounts, but I was able to collect some data from the COCORAHS spotter network. Instead of ratios of 15 or 20… or even 10… many of the ratios I calculated were under 10. That is to say that an inch of rainfall-equivalent would produce less than 10 inches of snow.

When I said “were this system all rain, we would likely see three-quarters to an inch of rain”, that turned out to be true. There were some areas that saw less than that 0.75 inch threshold, many places got over that mark.

So, the primary culprit at play here was the way-off-the-mark SLRs indicated by the models.

One comment on my blog post this morning said:

It was snowing hard – tiny white needles. Maybe 1/4 or 1/3″ long. They were light and dry – easy to shake off my coat. But they fell to the ground and packed into a rather dense layer of snow. Not wet and heavy, but dry and dense.

Other comments I had compared it to granulated sugar or walking through sand.

I’ll include here once again the snow flake structure diagram I included Friday and in an update Sunday afternoon. Click the image to enlarge it.

In my Sunday afternoon update, I pointed out how some limited data from the NYS Mesonet indicated temperatures in the prime area of the clouds for snow development was too warm for large, fluffy flakes. The observation of needle like structures, which I also mentioned on Friday, confirms that temperatures were in fact too warm aloft, despite temperatures at the surface being so cold.

Accumulating snowflakes shaped like needs is like stacking sheets of paper. Take 10 sheets and lay them on top of each other… it barely is noticeable. Take those same 10 sheets and crumple them into a ball, and the vertical volume, aka the accumulation of large, fluffy “dendrites”, is vastly different.

Doing some further reading on how the weather models (numerical weather prediction models, or NWP models) try to estimate SLR both before and after the storm, and looking at the results of this storm, it is evident that our current modeling capability is… not very good.

From Pivotal Weather, a company that takes weather model data and makes it accessible to meteorologists like me:

From a physics perspective, SLR comes down to the structure and density of the snow crystals, the formation mechanisms of which are quite complex (see Takahashi et al. 1991). Like any such pinpoint-small detail, though, current NWP models can only parameterize (estimate) this based on larger-scale variables like the predicted air temperature, moisture, and wind. Within the model, these variables could theoretically be used in a nuanced way to estimate SLR with considerable accuracy, but this is rarely done in current operational NWP. Instead, external users like Pivotal Weather must estimate SLR themselves based on the more limited data provided publicly.

In other words, the complex formation of snowflakes is more than modern weather modeling can handle.

For many years, I have harped on models overdoing the cold in situations where sleet mixing in is a possibility. I’ve accurately made predictions far below my contemporaries in these situations… often receiving flak on social media for having predictions so much lower than everyone else, but ultimately being vindicated.

This situation looks no different. Thankfully, more model data at more layers in the atmosphere is becoming available, so this will be something I will be placing even more emphasis on in my forecasting process from now on.

Automated phone apps certainly won’t be taking this into account in the future. Weather pages, media, private sector, or hobby pages that simply repost model data and call it their forecast won’t catch this either. Mark it down as another example of why having a local expert science based meteorologist is a vital necessity.

 

Service Analysis

Despite the raw snowfall projections being off, I am still very proud of the work I have done.

Did you know that last year, I made zero snow maps. This year, I’ve made two.

Even when I did post this storm’s snow map, I did so in a very deliberate way:

The problem with snow maps, or most weather graphics, is the way people use them. In our social media driven, mirco-attention span culture, a snow map is far too often the only piece of the weather forecast that people see. Now, this is a bit like preaching the choir… Finger Lakes Weather users are not the average consumer! You already know that there is much more to the forecast than a single map can tell.

So if a snow map isn’t useful for the entire forecast… neither is it useful as the only point of reference for seeing how well a forecast did. So, here are some forecasting wins that I had with this event:

Storm timing and danger

While most models had the snow not completely overspreading the Finger Lakes until later in the morning, I pointed out that the models are often too slow and predicted snow would start around sunrise in the south and overspread the region completely by mid morning. It did.

The roads DID become extremely snowy, icy, and dangerous. Travel Sunday would have been difficult if not nearly impossible. The impacts of the storm that I outlined still matched up almost exactly to the impacts that were realized.

Historic Context

In my Saturday morning post, under the heading “How bad will it really be”, I said: “Considering just the snow amounts, this is the type of storm we get probably 3 or 4 times every 5 years. Significant, but not one for the record books. In recent memory, storms in December 2020 and March 2017 will remain ahead of this one in terms of snow and overall impacts. One thing that makes this storm a bit more concerning that the “average” yearly winter storm is the deep cold. That will make treating the roads more difficult and will pose more of a risk when outside for any reason.” I would argue this was exactly on spot. It certainly wasn’t something I was using big inflammatory adjectives on.

Snow map redesign

Not only was I shooting for something accessible to the color blind and those with greyscale screens… but I purposefully chose blue colors to reduce the perceived hype surrounding snow amounts.

To assist me in the color scale of this redesign, I utilized some Artificial Intelligence (Claude, by Anthropic). It tried to tell me my decisions were boring and wouldn’t stand out! I had to point out that it would actually stand out because it would be different than the alarming colors everyone else uses. After pointing that out, it understood better.

It is a fine balance, weighing the risks of a weather event and deciding how to communicate that with the public. Ultimately, I do not think that bright, anxiety inducing colors belong on a snow map for an area that sees as much snow as we get each year. My goal is information sharing… not generating clicks for a social media feed algorithm.

Update Frequency

There are some weather events, like severe thunderstorms, where the absolute latest information can be vital. Severe thunderstorms change by the minute, which is why I live-blog those events.

Winter storms, however, are slow moving and change slowly, relatively speaking.

Two and three days before the storm, I had a single post in the morning that focused mostly but not exclusively on the storm. One day before, my morning post took the form of storm FAQs. I also had 1 special evening post dedicated to the snow map. It was a short post, and most of it was explaining the changes I made to the style of the map.

My policy is to produce one single snow map for an event. This prevents confusion in the forecast, especially since most other forecasts change multiple times a day or even hourly. These forecasts chase the latest model trends. My forecasts supersede the model noise and account for all of the information I have, which includes my knowledge and experience banks.

Post dozens of times, doing hourly live updates, making snow map after snow map… I believe that it all serves to hype up an event. I do not think this is a conscious decision others make, but more a product of our social-media driven information culture. Social media algorithms thrive on and therefore promote hype. If you don’t feed the algorithm, it leaves you behind. I know this is as fact, since my traffic from social media sources has cratered from over 50% of my traffic a couple years ago to roughly 20% of my traffic now.

 

Conclusions

In conclusion, since I’ve rambled on long enough now:

• Taking my forecast as a whole, based on the impacts and context of the storm, I felt it was highly accurate and by reading my work, most people should have been well prepared but not overprepared for this event.
• I am an independent meteorologist, and I approach weather reporting drastically different from the mainstream media, from social media, and from traditional methodologies in an effort to help people know what to actually expect from the weather.
• This storm was a momentous learning opportunity when it comes to understanding how models deal with snow-liquid ratios, supplementing my experience and knowledge base with model biases in winter storms where temperatures are warmer and sleet may mix in.
• It is amazing how complex the atmosphere is and how a difference of a couple degrees can totally change the outcome of a storm, all because the shape of the snow crystals changes.
• It is equally awe-inspiring how perfectly everything has to come together in the atmosphere to produce the truly big storms… and how difficult it can be to get all those factors to line up.

Thank you for taking the time to read these reflections. I hope they give you a greater understanding and appreciation for the complexities of both the science of meteorology… and the sociology of communicating the weather to the public.

And a final tremendous thank you to all of my donors, who make my work possible. Thank you to the nearly 100 people that contributed to Finger Lakes Weather over the last few days, and the several hundred others who have support me with monthly donations or have supported me in the past.

Independent local media is so important, and I am blessed to have such a dedicated donor base to support me.

Side note: Please accept my apologies for any typos or grammatical errors. I’ve been working on this post for a couple of hours and my family is waiting at the dinner table for me to wrap it up… so I don’t have the time to do my usual proofreading!

 

Support FLX Weather

Finger Lakes Weather relies on support from you, its users, as its primary source of funding. Want to support Finger Lakes Weather and keep the no-hype, local information flowing?
Please consider making a contribution using the forms below or sponsoring FLX Weather with your business. Checks may also be mailed to: Finger Lakes Weather LLC, PO Box 326, Groton, NY 13073

Thank you for your continued support of Honest, Accurate, Accessible Weather!