Top Four Misconceptions About Weather Forecasting

Whenever I tell someone that I aspire to be a meteorologist as my future career goal, I'm usually faced with a lot of the same questions and statements, ranging from the rather insulting to the more curious inquiries:

Weather forecasting is the only job where you can be wrong 50% of the time and still keep your job.

Meteorologists are never right.

It must be easy having their job.

Do meteorologists predict meteor showers (or earthquakes, tsunamis, eclipses, and other physical phenomena)?

How can anyone predict the future? 

And there are many more statements that I have had asked to me [and about other meteorologists] over the past several years. In order to help remedy this situation, I thought it would be nice to write an article explaining what forecasting is and isn't, as well as clearing up some common misconceptions. While I won't necessarily address all of the concerns listed above, I will be addressing more common [and reasonable] sentiments about the field of weather forecasting.

1. Meteorologists are hardly ever accurate.

This is the common profile of the meteorologist as known and expressed by the general public. Meteorologists are generally the punchline of many a joke, especially those that make light of our inaccurate forecasts.

However, this is not necessarily true. Yes, it is true that there are many forecasts that we screw up and there are many times when we fail to admit that we've screwed up. But for the average operational forecaster, either working for a private firm, a broadcast station, or for the U.S. Department of Commerce, there is generally a lot more forecast skill and accuracy than one would expect (as a side note, I would like to point out that there is a major distinction between 'skill' and 'accuracy' which won't be addressed here, but it is possible for a forecaster to have high accuracy, but low skill in their forecasting, just as it is possible to have high skill, but lower accuracy). What many do not often account for is the fact that forecasters do a lot more than just tell us when the next storm will hit or how much snowfall we will get from a winter weather event, but also make daily forecasts. Daily forecasts generally consist of the high temperature, the low temperature, wind speeds/direction, cloud cover, and the chance of precipitation. For the vast majority of situations, forecasters are generally fairly accurate. It is only when abnormal weather shows up that forecast skill and accuracy begins to vary.

Even then, however, forecast accuracy has strongly improved over the past several decades. Studies and assessments from the last ten years ultimately show that forecast skill and accuracy has markedly improved, especially in forecasting events that occur in the short term, ranging from six hours in the future to as much as two days. This is especially true for rapidly developing severe weather situations and mesoscale winter weather features. In all likelihood, this is due to the development of high-resolution short-range numerical weather models that are specifically designed to forecast small-scale phenomena which 'larger' global models cannot handle. Computer models will be addressed below.

In all, forecasts are generally not as inaccurate as everyone supposes. Most only remember 'that one time' when the forecaster screwed up on a major weather event. The general public usually doesn't remember when the forecaster actually got the forecast right, which is probably a majority of the time.

Note: There is still some debate about whether forecast accuracy has actually improved over the last ten years [from 2005-2015], but I can't find any sources on the issue. What I do know, however, is that the National Weather Service has come under heavy criticism from various meteorologists about their training process and their accuracy.

2. Forecasters just make up numbers from nothing. Anyone can do what they do. 

Although I haven't heard this one from anyone in particular, I think there is a general confusion from a lot of people about what exactly forecasters do and how the forecasts are even made. I will address this issue here and in the next section.

Here's the problem: A lot of time and effort often goes into the making of a forecast [or at least it should for any good forecaster] and is really more complicated than what one imagines. There are times when it certainly feels like forecasters are simply making up numbers just for the sake of getting a forecast out. There are even times when this is all a forecaster can do. When the weather pattern is relatively stable and nothing much is changing, this is really all the forecaster can do.

When a particularly dangerous and abnormal situation comes up, this is when forecasting skills are best tested for their accuracy. A forecaster must learn to draw upon not only his conceptual and scientific knowledge of the atmosphere. They must also be able to learn from their experience and apply what they learn to new situations. Finally, the forecaster must rely upon his or her intuition in order to make the final decision in the forecast. When dealing with intense situations, like major winter storms or imminent severe weather, these decisions have to be made regardless of what the actual data shows. The data only shows a part of the whole story; the forecaster has to be able to connect the data with their own experience and make a quick decision.

This goes along with general forecast accuracy; sometimes the forecast is inaccurate because of the precautions that the forecaster personally took when publishing the forecast. It is always better to remain on the safe side during a dangerous situation than to not say anything at all.

3. There's no way you can predict the future! 

While this is technically true, I think the objection misunderstands the nature of the practice of weather forecasting. Weather forecasting isn't so much about 'foretelling the future' as it is simulating the future state of the atmosphere based upon known data from the present. It isn't so much predicting what will happen as it is trying to establish connections between weather phenomena.

As mentioned above, the main reason why modern forecasting has made great strides in forecasting is due to the advent and use of computer technology in forecasting. Because of the impressive advances in computer power, calculation, and storage, meteorologists have used these computers to simulate the future state of the atmosphere.

How is this possible? In the early 1900's, scientists, who became the pioneers for the modern field of meteorology and atmospheric science [this is one of the newer sciences], derived a series of equations dependent upon Newton's laws of motion and thermodynamics that were able to 'describe' (in a sense) certain scalar and vector quantities in the atmosphere. These included, but are not limited to, variables such as temperature, moisture, wind, pressure, etc. Some of these scientists proposed that, in theory, if one could solve these equations and then duplicate this process, one could forecast the state of the atmosphere at a given future time period. One particular scientist attempted this process by hand, trying to forecast the weather for a location in France for 48 hours out; it took this particular scientist nearly two years before he could finally come up with a forecast! This might seem like a huge failure, but in reality, it was one of the biggest breakthroughs in modern scientific thought. This precipitated action by other scientists to try to solve this dilemma by using different methods for solving these equations.

Finally, with the advent of computers, they were utilized to attempt to address this issue. The first model 'forecasts' were run in the early 1950's on some of the first supercomputers. The forecast was crude and very inaccurate, but it was a breakthrough in modern forecasting. Over the next decade, forecast accuracy improved remarkably with these primitive equation models (as they were called). While they were not used to forecast variables such as surface temperature and wind directly, they were used to forecast the general upper level weather pattern, which allowed forecasters to make more reasonable predictions about imminent weather.

Now, forecast models are run on the world's fastest supercomputers, making billions upon billions of calculations in a very short period of time. These model forecasts, in the United States at least, are produced every 6 hours [at 00, 06, 12, and 18 GMT], and can span either the entire world [as in the infamous Global Forecasting System] or simply North America [as in the North American Model or the NAM]. Forecasters then take this data and integrate them into their own decision making process for the final forecast product.

Unfortunately, numerical weather prediction doesn't solve all the troubles of the forecaster. Model forecasts are limited by the data they ingest, their own physical parameters and methods of solving the equations (which differ from model to model). Plus, there is simply too much chaotic motion in the atmosphere for any model simulator to deal with. This is why human forecasters will always be needed. Unlike computers, humans possess the necessary memory and intuition needed for the final forecast to be successful.

4. Meteorology is an easy career path
 

Unfortunately, it isn't very easy at all. For those who become meteorologists by going the traditional route, by attending college and acquiring a B.S., the road isn't easy. Meteorology is a field of science that draws upon many other different fields. In a way, it is very interdisciplinary in comparison to other fields of science. Because of this, meteorology majors are required to be rigorous in mathematics and must be able to have computer and programming skills. They must also have additional advanced knowledge in fields like physics and chemistry, both very important fields for understanding atmospheric processes. Meteorologists must also have a working knowledge of geography and topography. They must be familiar with social media, have writing and communication skills, and should be familiar with broadcast journalism.

Even for the meteorologist that does not take the traditional college route, they must have a working knowledge of all the fields mentioned above. Furthermore, a meteorologist who is not college educated has to pass the American Meteorological Society test that allows one to become a certified meteorologist. Thus, they must be able to demonstrate that they have working experience and knowledge in operational forecasting [which is the most practical application of the field].

Even after all the educational requirements are completed, it's often very difficult for students to find jobs in the field of meteorology. In years past, the federal government has been the largest employer of meteorologists, but with freezes in spending and the cutting of funds to branches like the National Weather Service, there isn't as much hiring anymore. "Tenured' or experienced meteorologists generally remain with the weather service for long periods of time which doesn't allow for a lot of flexibility for incoming meteorologists.

Fortunately, the private forecasting field is expanding for incoming meteorology students. Energy corporations, private firms, insurance companies, and others are now looking for more meteorologists to hire. These companies benefit from both the short term and long-term forecasts that meteorologists can give, allowing them to make better and wiser decisions.

Conclusion
 

With how difficult forecasting and the career field can be, why does anyone bother becoming a meteorologist in the first place? Mainly because most of us enjoy what we do. I have talked to many meteorologists and nearly every single one that I have talked to have always had a passion for the weather. One thing that characterizes many meteorologists is the 'in-born' passion that many of us had as children. This is one field that if you don't truly love it you shouldn't go into it. It isn't for the light-hearted and it certainly isn't for those who only give lip service to the passion. But for anyone who truly loves nature and the outdoors, meteorology is certainly a field that is both practical and awe-inspiring to study and admire.

Winter is Arriving... Finally

After waiting anxiously for well over a month, snow lovers are gasping a sigh of relief as forecast models are finally depicting a better and improved winter weather pattern across much of the United States.

While I don't have too many details at this time, it appears that we will likely see progressively colder weather affect the area over the next two weeks. The first wave of colder weather will hit the area later this week, as highs struggle to make it above 30 degrees by the opening days of 2016.  Fortunately [or unfortunately, depending upon your perspective], temperatures will once again moderate as upper-level ridging across Canada allows warmer air to filter into the region. There are, however, consistent long-range signals from forecast models and ongoing observations/trends that suggest a far more substantial pattern shift is ongoing. In fact, in the January 10-15th time period, we could finally see an active winter pattern across the East, with cold Arctic air and the potential for snow coming into the picture. I will have more updates on this possibility in the next few days.

Thunderstorms and Warm Summer-like Weather On the Way

After several consecutive months of below normal temperatures, wet, snowy conditions, and overall dismal weather, things are finally turning around for the month of May. And it looks to stick around for a good portion of the next two to three weeks, if not longer.

Over the past two or three days, the area has experienced temperatures in the mid to upper 70's, which is above to well above normal for this time of year. With sunny skies and light winds, this has allowed for nearly optimal conditions for prom weekend across the region. However, this nice weather doesn't look to stick around. Thunderstorms are moving back into the forecast and they look to impact the area on more than a few days this week and the following weekend.

A surge of modest moisture from the Gulf of Mexico, a shifting frontal boundary and relatively warm conditions will allow for the development of scattered numerous thunderstorms on Monday afternoon and evening. While instability is not expected to be all that incredible, it will be enough for thunderstorm development and even a small chance for isolated severe weather. Thunderstorms will become more widespread Monday night into Tuesday before ending Tuesday afternoon. Rainfall amounts could be up to a half inch or more across the area.

By Wednesday, drier, warmer conditions will move into the region, with highs approaching the lower 80's under partly sunny skies. By Thursday and Friday, temperatures may approach the mid-80's with slightly higher humidity. By Saturday and Sunday, thunderstorms look to be a threat once again, especially as humidity levels increase. That being said, temperatures should remain relatively warm in the 80's, which is well above normal for this time of year. This type of pattern is expected to continue well into next week as weak upper-level ridging remains across the eastern United States.

I will have more updates later.

Post Storm Overview

This is a revised version of an article I wrote that will be published in the West Bend News this coming Wednesday. I will be posting a more detailed analysis of the storm later this week.

After a winter that has been relatively cold and snowy across the area, I think many of us are ready for the nice, warm, and even wet weather of March and April. This is especially true after the area was recently hit by a strong blast of winter weather, blasting the area with heavy snow, ice, and gusty winds. Now we’re all hoping that winter ends soon, even if Punxatawney Phil disagrees.

On January 31st through February 2nd, the lower Great Lakes region, including northwest Ohio, northeast Indiana, and southern Michigan, was hit by a crippling winter-storm. Stretching from Omaha, Nebraska to Albany, New York, heavy snowfall and freezing rain shut down major highways and cities, with up to 20” of snow observed in some locations. The I-80/I-90 turnpike had travel restrictions due to the dangerous conditions. Hundreds of flights were cancelled at Chicago O’Hare and Midway airports, as well as at Fort Wayne, Detroit, and Cleveland. Many schools were closed on February 2nd due to snow-covered and in some cases, impassable roads.

Freshly-fallen snow outside of my house in the early morning February 2nd.

The storm was just as significant across northeast Indiana and northwest Ohio. Reported total snowfall across the area ranged from as little as 6” in Putnam county to as much as 13” in Allen and Whitley County, Indiana. In Antwerp, 10.6” of snow was measured, while in Fort Wayne and Defiance, as much as a foot of snow was reported. Across far northern Indiana, snow totals ranged from 15-20”, making it the most significant snowstorm since the Groundhog’s Day blizzard of 2011 and the New Year’s blizzard of 1999. The storm also produced strong, gusty winds across the area, leading to significant blowing and drifting snow well after snowfall had ended on the night of February 1st and into the morning of February 2nd. Drifts as high as three to five feet were reported in some locations, especially in northern Indiana. Blowing and drifting lead to travel emergencies and restrictions area wide, with many roads becoming impossible to drive on or even impassable.

Across Northern Illinois and lower Michigan, snowfall amounts as much as 20" were reported. Chicago reported a snow total of 19.3", making it its fourth largest snowstorm on record. Detroit reported over 16" of snow, making the storm its third largest snowstorm on record. Many other locations reported similar totals.
Where did this storm come from? Why did it form? And can we expect more to come for the rest of the winter?

It began with an upper-level disturbance that moved onshore British Columbia Thursday night, January 29th. Forecast model guidance had suggested that the disturbance would dive southeast into the United States and produce a swath of light to moderate snow from the Central Plains to the East Coast. At most, snowfall of 4-6” was expected across parts of the Midwest, southern Indiana, Ohio, and into Maryland and Virginia. However, by early Thursday morning, forecast guidance began to shift the track northward, foreshadowing the development of a much more powerful storm system.

By January 30th, the development of the storm was fairly certain. The upper-level disturbance across British Columbia would still dive southeast, but it would also interact with a weak upper-level low across the American Southwest. This interaction would strengthen the disturbance, fueling it with abundant moisture from the Pacific Ocean and the Gulf of Mexico. Forecast models were then showing a significant snowstorm stretching from Chicago eastward into Pennsylvania. Snowfall totals of 16-20” of snow were indicated by forecast models, especially from Chicago to the Fort Wayne area. National Weather Service forecasters indicated that if 16-20” of snow were to fall, it would be one of the area’s heaviest snowstorms to ever impact the area. However, the NWS was not going to predict such high totals due to the uncertainty involved.

By Saturday, January 31st, forecast models were making some uneasy. While they had retained the record-breaking snowfall amounts, models were showing differing placement of the heaviest snow. They were now indicating that the heaviest snow would fall just to the north of our area. For this reason, the National Weather Service kept snowfall amounts in the 10-15” range when they issued the Winter Storm Warning. Not only this, but there was also the possibility that blizzard conditions would be present across the area. The National Weather Service refrained from issuing a blizzard warning due to the fact that the strongest winds were not expected to impact the area until after the snow had ended, but they did note that near blizzard conditions would be possible during the heaviest snowfall.

Widespread precipitation developed across the Central Plains by the afternoon and evening hours Saturday. By nightfall, light to moderate snowfall mixed with rain was impacting the Chicago area.
Snow began falling across our area by 11 p.m. that night. Light to moderate snow fell through most of the night, with most locations reporting 2-4” by Sunday morning. As the low-pressure system developed, however, a break in the snow was observed in many locations, especially east of I-69. Very light snow fell for most of the morning across Paulding and Defiance counties, with heavier snow just off to the west. At the same time, moderate to heavy snowfall continued across northern Indiana, extreme northwest Ohio, and southern Michigan, with snowfall totals quickly approaching a foot by early afternoon.

By noon, moderate to heavy snow had shifted back into the region, dropping visibilities and making roads hazardous.

Snow continued through most of the afternoon, mainly north of U.S. 30. With the strengthening low-pressure system, warmer air was drawn northward into the Lima and Van Wert areas, causing some of the snow to mix with or change over to rain. Surface temperatures rose into the mid and even upper 30’s across these locations. This was something that surprised forecasters, as the warm air was not expected to reach this far north. By late afternoon, cold, Arctic air seeped southward into the area, changing all precipitation to snow, and allowing another period of moderate to heavy snow to blast the area. By 9 p.m., only light snow was being reported across the area, with storm total accumulations approaching a foot in many locations. This was especially true north and west of U.S. 24.

This winter-storm will not likely be forgotten. Although we didn’t see nearly as much snowfall as areas north and west of us did, the 8-14” of snow that many of us did see was enough to shutdown schools and businesses across the area. This certainly reminded us that we are still in winter, and still in an area that is prone to major winter storms.

Winter-Weather and Potential Storm

As I said in an earlier post this month, I won't be posting as much because I'm working on a new Wordpress site, and I'm still trying to transfer articles from this page onto that one. But for now, I'll occasionally post on this blog and on my FB page. I am also a blogger on another significant weather page, so you can follow me at www.wildaboutweather.com.

I have a few concerns this morning.

First, the area had some light freezing rain last night, leading to icy surfaces just about everywhere. At this time, most schools across the area have closed and nearly all have delayed. Be careful out there! Temperatures should rise throughout the morning hours, so by noon, icy roads should be gone. The potential for snow, sleet and rain will continue into the afternoon, before tapering off later this evening.

After this, a weak cold-front will move through, dropping temperatures back into the 20's. This is when yet another system will begin to impact the area.

At this time, however, not much is known, and any rumor that you have heard about a "major snowstorm" is probably wrong. No forecast scenario shows a "major snowstorm" and many of the forecast solutions show only a short period of light-moderate snow on Sunday. However, there is a LOT of disagreement between forecast models at this point, and so we really can't rule anything out.

Going against the grain of the European and the American model, I believe there is enough evidence to support my growing confidence that a strengthening system will track through the Ohio Valley, and then transfer to the coast. This means that many areas from the Central Plains through the lower Great Lakes and Northeast could see some sort of winter-weather event. Whether it will be a "significant snowstorm" or a period of light snow has yet to be determined.

Model guidance such as the long-range NAM is showing the potential for a much more significant storm. So are the Canadian and British guidance [CMC and UKMET]. The GFS is taking the middle ground with only a moderate snow-event for areas from Chicago to Indianapolis, Detroit, and Cleveland [and everywhere in between]. However, given that there is very little confidence at this point, we really don't know what's going to happen. The only thing I can say is that I will be monitoring the potential for a winter-storm and it *COULD* be significant.

I will have more details over the next few days.

Icy Roads, Rain, Bitter Cold, Snow and More!

I apologize for not having posted in the last two months. I've mainly been posting updates on the FB page for the time being, as I plan on transitioning the blog to a Wordpress site [which is quite a bit more user-friendly than blogger].

Here's what I had posted on Facebook page earlier today:

After a rather messy set-up this morning [I had a lot of trouble just getting to work this morning!], all precipitation has transitioned over to rain with the exception of counties further north. Counties along the Indiana/Michigan/Ohio border COULD continue to see freezing rain for a little while longer before temperatures get warm enough to allow any remaining ice to melt. Even in areas where temperatures have warmed into the mid-30's, roads are still rather icy and dangerous, so exercise caution.

Rain will continue through the evening hours before tapering off overnight. A few showers will persist but nothing substantial until Sunday morning when sleet/freezing rain/snow should redevelop, with a complete transition over to snow for the afternoon, mainly west of I-69. Areas farther east will likely see some snow, but not before it tapers off. Snow accumulations will not be substantial, with the only real chance of accumulation being in the lake-effect zones.

By Sunday night, all precipitation will taper-off, leaving behind breezy conditions and falling temperatures. Temperatures will likely fall into the lower 10's and single digits Monday morning as the Arctic front sweeps into the region. Wind-chills will likely be below zero as well.

Temperatures will remain rather steady on Monday, with a very real possibility that some areas DO NOT make it above 10 degrees for a high. With slightly breezy conditions, bitterly cold wind-chills will be in place.

However, by Monday night, a clipper-system will approach the region rather quickly, bringing periods of light to moderate snow by the very early morning hours. Snowfall should be at its heaviest by the Tuesday morning commute. Based upon current model indications most of the snowfall should end by noon or early afternoon, but could linger a little while longer in areas east of I-69.

At this time, snowfall amounts are uncertain, as there is too much uncertainty regarding the intensity of the system and the amount of moisture involved. If more moisture is involved and the system actually develops a closed surface low, we might see more substantial snowfall. However, this does not appear likely. The National Weather Service has said that they are expecting rough amounts in between 3-6". I will keep you updated on this system!