March Snowstorms in Alabama - What are the Odds?
National Weather Service
National Weather Service
Ahhhhhhh, Spring's arrived for good, hasn't it? Alabama is ablaze with forsythias and daffodils; the air's replete with the sweet aroma of blossoms, of damp, freshly-spaded earth. Winter's holding on by a toenail now, isn't it? Surely we can forget about the prospect of icy bridges, school closings, and crazed mobs storming the bread racks of area supermarkets. We've dodged another bullet until next year, right? Well.....probably. But here in Alabama, it's impossible to know what nature has in store for us from week to week, especially this time of the year. The weather is so capricious during March; temperatures can tumble 25 degrees from one day to the next. Warm, muggy air can envelop the Gulf states, fostering the formation of violent tornadoes, which in turn can be followed by the blast of chilly arctic winds and the sting of wet snowflakes.
Over the years, staff members at the National Weather Service in Birmingham have published a number of articles concerning the climatology and synoptic patterns associated with significant snowfall events in Alabama. Another informal study of monthly Alabama snowfall climatologies is now underway, and should be completed this summer. This article presents a few findings from this current investigation, and focuses on the climatology of late season snow events - i.e., those occurring on or after March 1st. The primary question we seek to address is this: What are the long-term odds of receiving a measurable late season snowfall in north and central Alabama? And are those odds affected this year by the presence of our friend, El Niño?
Where did the data come from?
As part of this study, snowfall totals were tabulated for ten sites in north and central Alabama, with the goal of establishing monthly frequencies of occurrence for varying snowfall amounts. To narrow the scope of the current discussion, results are presented only for Huntsville, Birmingham, and Montgomery. Furthermore, since multiple snowfall events after March 1st are rare anywhere in Alabama, we'll consider only the probabilities of occurrences of measurable snow, regardless of amount. Data for this study were obtained from the National Climate Data Center publications Alabama Monthly Climatological Data, and Local Climate Data.
It should be clearly noted that this is an informal study, utilizing only records on hand at the National Weather Service office in Alabaster, and that the findings here are preliminary. The limited availability of older issues of the necessary climate publications resulted in a wide variation in the length of the snowfall records for the ten sites. For example, over 100 years of data were available for Birmingham, while the duration of records for several other sites was only 38 years. It is anticipated that the records for some of the sites will be augmented by additional snowfall data before this study is completed.
Table 1 depicts the total number of times measurable snow has occurred in each winter month at Birmingham, Huntsville, and Montgomery. In addition, the respective duration of the records, and the resulting frequencies of occurrence are listed. If we consider Birmingham's data first, we see that measurable snow has been recorded in March and April 13 times in the past 101 years, resulting in a long-term frequency of 12.9%. This value is small compared to the totals for January and February, which (as expected) represent the prime months for snow in the Deep South.
Table 1. Occurrences of measurable snow by month.
|Number of years in record||39||101||66|
However, an examination of Table 2, which lists the years in which late season snows occurred, reveals some interesting trends. During the 36 year period from 1943 to 1979, Birmingham experienced only 1 measurable March snowfall! Our recent Springs have been a lot snowier, however. Since 1980, the Magic City has recorded measurable snow after March 1st in 5 of the past 18 years, which works out to one snow every 3.6 years. In fact, two of Birmingham's biggest and most memorable snows have been recent late season affairs. The April 3rd, 1988 snow dumped 5 inches on the city and raised considerable havoc, while the 13 inches deposited by the March 13th Superstorm effectively paralyzed the metropolitan area for 3 days.
Table 2. Years in which measurable snowfall occurred. Monthly totals are in parenthesis.
|1960 (2.9)||1896 (1.3)||1932 (0.1)|
|1965 (2.3)||1915 (2.8)||1987 (0.8)|
|1968 (2.1)||1924 (6.5)||1993 (3.8)|
|1980 (2.0)||1931 (0.1)|
|1982 (0.3)||1932 (0.3)|
|1993 (7.3)||1934 (1.2)|
|1995 (0.2)||1942 (2.5)|
|1996 (0.3)||1957 (0.7)|
Owing to its northern location, Huntsville receives considerably more snowfalls than other parts of the state. This trend holds true for March/April events as well. As seen in Table 1, late season snows have occurred in 8 out of the past 38 years - which results in a frequency of 21%. Another look at Table 2 indicates that March/April snowfalls in Huntsville, like those in Birmingham, have become more common in recent years. Indeed, measurable snow has occurred in this Tennessee Valley city in 5 of the past 18 years.
Further south, snow at any time of the year in Montgomery is quite uncommon, and March/April events are rarer still. Table 1 indicates that measurable late season snows have occurred in only 3 out of the past 66 years, resulting in a frequency of only 4.5%. The 0.8 inches recorded in April 1987 represented the first measurable post-February snow in Montgomery since 1932, while the 3.8 inches received during the March 1993 Superstorm was one of the largest snowfalls to occur in the city in the past 100 years.
But wait...this is an El Niño year, so all bets must be off!
Yes, Alabama (like everybody else) is experiencing an El Niño this year - and a big one at that. So how does this affect the likelihood of a late season snowfall event in our state? To answer this question, let's look at what's happened during previous El Niño episodes. Table 3 displays the years between 1900 and 1997 in which El Niños have occurred - 27 events in all. Let's consider Birmingham first, since it has the longest period of record. In the Magic City, March/April snowfalls have occurred in 6 of the 27 El Niño years. This frequency (22.2%) exceeds the long- term frequency of 13%, thereby suggesting a slightly increased risk of late season snowfall during El Niño episodes. However, keep in mind that this finding is based on a small, statistically- insignificant sample size, and therefore should be considered inconclusive.
Huntsville, on the other hand, has recorded no March/April snow events during El Niño episodes from 1958 to present. One explanation for this may be that the frequency and magnitude of precipitation amounts in the Tennessee Valley tend to be reduced during El Niño events, as the primary storm systems remain further south near the Gulf of Mexico. Chances for El Niño snowstorms aren't any better south of Birmingham, either. In the past 66 years, Montgomery has received measurable snow only once during an El Niño episode - this on April 3rd, 1988. While this city may receive somewhat more precipitation during El Niño than its northern counterparts, the odds of the temperature being cold enough for snow during these events are apparently quite small.
Come on!! February was so warm...surely it can't snow!
One more interesting statistic we can consider is the relationship between average February temperatures and the occurrence of March/April snows. Table 4 depicts the average monthly temperatures for Huntsville, Birmingham, and Montgomery for each of the Februaries preceding the late season snowfall events listed in Table 2. Table 5 breaks these values down in relation to their corresponding 30-year means.
Table 4. Average monthly temperatures (and departures from normal) (degrees F) for Februaries preceding March or April snowfall events.
|1960: 39.5 (-5.6)||1896: 50.0 (+2.0)||1932: 60.6 (+10.3)|
|1965: 42.8 (-2.3)||1915: 47.8 (-1.8)||1973: 46.8 (-3.5)|
|1968: 34.6 (-10.5)||1924: 46.2 (-1.8)||1987: 50.2 (+0.3)|
|1980: 37.5 (-6.1)||1931: 49.7 (+1.7)||1993: 48.7 (-1.1)|
|1982: 44.6 (+1.0)||1932: 56.7 (+8.7)|
|1993: 41.8 (-1.3)||1934: 43.8 (-4.2)|
|1995: 42.9 (-0.2)||1942: 43.2 (-4.8)|
|1996: 42.1 (-1.0)||1957: 55.2 (+7.5)|
|1980: 42.3 (-4.6)|
|1983: 44.8 (-1.5)|
|1984: 46.8 (+0.5)|
|1987: 46.9 (+0.6)|
|1993: 44.9 (-0.8)|
Table 5. Frequencies of above normal, below normal, and near normal temperatures listed in Table 4.
|1 degree F or warmer above mean||1||4||1|
|1 degree F or colder below mean||6||5||2|
|within 1 degree F of mean||1||4||1|
In the case of Birmingham and Montgomery, no clear trends emerge, as these cities' early spring snow events were preceded by roughly equal numbers of warm, cool and average Februaries. Huntsville, on the other hand, experienced an overwhelming number of cool Februaries prior to the eight Marches in which it recorded measurable snow.
Again, the small sample sizes considered here make it difficult to form any definite conclusions about the relationships between February temperatures and the occurrence of late season snow. Nevertheless, it's hard not to evaluate the prospects for a March/April snow this year in light of these findings. The February 1998 departures from normal average temperature for Huntsville, Birmingham, and Montgomery are: +2.8, +2.3, and +1.5, respectively. The above normal monthly temperature for Huntsville suggests a reduced likelihood for an early spring snow in this city, all other factors excluded. Birmingham and Montgomery also experienced a warm February. However, the absence of a clear historic relationship between February temperatures and late season snows makes it impossible to form any conclusions between this year's averages and the possibility of a March/April late snow in these cities.
To recap...here are some of the findings of our study:
1. Long-term frequencies of occurrence of measurable March/April snow:
2. Frequencies of occurrence of measurable March/April snow since 1980:
3. Frequencies of occurrence of late season snows during El Niño:
4. Possible relationship between mean temperature of the preceding February and the occurrence of March/April snow:
The intent of this study is to present some simple observations on the probability of late season snowfall in northern and central Alabama. Climatologically, measurable snow is an infrequent visitor to the Deep South, even in the depths of winter. And those snowstorms that occur after March 1st are rare events, indeed. In Huntsville, the long-term odds are about 5 to 1 against seeing an accumulation of snow in March or April; in Birmingham and Montgomery, the chances are even more remote. Moreover, the presence of an El Niño episode or an abnormally cold or warm February overall doesn't appear to significantly increase the risk for these events. However, the threat is not non-existent. And as most of us can personally attest, when these winter weather events affect Alabama, they can have crippling implications for our state. So what does nature have in store for us this March? We'll just have to wait and see!