WeatherMatrix (Jesse Ferrell)

What Caused the Johnstown Floods?

By Jesse Ferrell, Meteorologist/Community Director
7/20/2010, 3:22:12 AM

Share this article:

Warning: Educational (EDU) Lesson Ahead!

Last month we talked about three famous floods that have struck Johnstown, Pennsylvania, the anniversary of one (1977) being Friday. But what, weather-wise, caused the floods?

The Historic Flood of May 31, 1889

First let's look at circumstantial evidence on the 1889 flood (2,209 killed, $17m damage). Most Internet records concentrate on the aftermath and don't give any insight into what meteorological conditions led to the flood, and some disagree on when the rain fell. For example, This site says that the dam broke around 4 PM on May 31, after "a night of heavy rains" but the New York Times and WikiPedia say the dam break followed "days of rain" and adds the following crucial information (from an uncited source): "On May 28, 1889, a storm formed over Nebraska and Kansas, moving east. When the storm struck the Johnstown-South Fork area two days later it was the worst downpour that had ever been recorded in that section of the country. The U.S. Army Signal Corps estimated that 6 to 10 inches (150 to 250 mm) of rain fell in 24 hours over the entire section."...


SURFACE WEATHER MAP MAY 30TH 1889, 8 AM (ENLARGE | WHOLE U.S.) says that businesses dismissed their employees to secure their belongings at home (due to the flooding, which was routine in the city) when "heavy rain began to fall on May 30, 1889." USCD confirms, saying "May 30 storm struck western Pennsylvania; worst downpour ever recorded locally."


But the best record of individual rainfalls and amounts is stored in the Monthly Weather Review from May 1889, which Google has been kind enough to scan in. The amounts, many between 4 and 8 inches, appeared to have mostly fallen over a 36-hour period beginning on the afternoon of May 30th (see begin times map). On Page 117 [See The Flood Report | download whole PDF] begins a complete report with tables of rainfall and river gauge observations. It does not, however, state what synoptic weather system caused the heavy rain, other than saying that the heaviest rain was northeast of a low pressure system which was in Cincinnati at 8 AM on May 30th (see below). The heaviest rain measured the closest to Johnstown was 7.9" at Blue Knob (the Johnstown rain gauge was washed away after 2 inches fell), even though more than 10 inches fell to the northeast, in fact on the official rainfall chart at right, Johnstown is shown in only the 4 to 6 inch area.

Many people don't know that the rain amount itself may not have even caused the dam to collapse (though I would argue it could have been the duration of the high water). The New York Times said the next day:

"The cause of the calamity, it is admitted by the President of the South Fork Fishing Club, the proprietor of the artificial Conemaugh Lake, was the weakness of the dam alone. No cloudburst or waterspout occurred to compel it - the frailty of the dam and the tremendous pressure of water behind it was the only cause of the catastrophe."

And the Monthly Weather Review confirms:

"From this it may be inferred that the rainfall-water passing at the time, though great, did not have much significance in causing the disaster, as it was probably not more than one-tenth of that from the reservoir. This great rainfall, 1.1 of a cubic mile in the Susquehanna Valley and 1.0 cubic mile in the Potomac Valley, occurred to the northeast of a definite low area of barometer, that designated as No. viii. The lowest barometer (29.58) was at Cincinnati at 8 a.m. of the 30th."

Looking at the surface weather map above, and reading the text above which seems to blame the heavy rain on the low pressure system, I'm going to say that the 1889 flooding was caused by an unusually strong Spring low pressure system moving towards the East Coast and pulling in large amounts of moisture from the Atlantic, possibly moving the air up the eastern slopes of the Appalachians, causing heavy "upsloping" rain. Note the similarities between the 1889 rainfall map and this map from Hurricane Ivan which involved upsloping conditions. It would be hard to tell without more data.

P.S.: For a 24-hour forecast 120 years ago, they didn't do too bad. Below is the actual forecast from the May 30th Surface Map:



- Commentary Video by Johnstown Resident BeckyMarie


- Johnstown Flood National Memorial Website
- Wikipedia Johnstown Flood 1889
- Google Newspaper Archives: 1889 | Later
- Photos from the Museum and Johnstown by Becky and Local Storm Chaser Ron Shawley (see below)

The Killer Flood of March 17, 1936

The cause of the 1936 flood (24 killed, $43m damage) seems fairly straightforward -- the Johnstown Flood Museum website says that "On March 17, 1936, Johnstown experienced a devastating flood caused by heavy runoff from melting snow and three days of rain.". Indeed, a look at the daily weather map shows that the 1936 flood was caused by a strong late-winter storm below 29.30" pressure over the Carolinas, spreading 50-degree temperatures and heavy rain behind a warm front into Central Pennsylvania, causing rapid melting of snow, while heavy rain fell.

NEW: Click here for a new blog entry I wrote detailing the rainfall and snow pack in the 1936 storm.



- Commentary Video by Johnstown Resident BeckyMarie on 1977 & 1936 Floods


- WikiPedia: 1936 Pittsburgh Flood
- Google Newspaper Archives on 1936 Flood

The Deadly Flood of July 20, 1977

Wikipedia states that the 1977 flood (85 deaths, $300m damage), was "a relentless storm reminiscent of 1889" but through the research below, I found out that they were completely separate meteorological causes. I note that both storms occurred in Spring and Summer, and through Google research I now believe that both were probably caused by a Mesoscale Convective Complex (MCC [JessePedia]).

A WikiPedia draft by MesoZ states that "One of the most recent notable MCCs occurred overnight on 19-20 July 1977 in western Pennsylvania. The MCC resulted in heavy rainfall which led to the disastrous flooding of Johnstown, Pennsylvania. The complex was tracked 96 hours back to South Dakota and produced copious amounts of rain throughout the northern United States before producing up to 12 inches of rain in Johnstown."



An AMS (American Meteorological Society) paper [PDF] by Lance Bosart and Frederick Sanders confirms this, saying (in the abstract) "The heavy rains responsible for the disastrous flash floods near Johnstown, Pennsylvania on the night of 19-20 July 1977 are shown to be part of a large quasi-circular mesoscale convective complex." They say the MCC started in South Dakota 4 days earlier (which I would say is a pretty impressive time frame). Above is a satellite shot from the morning of July 20th, showing the approximate location of Johnstown as the small "x". That picture confirms that the 1977 flood was caused by an MCC. That being the case, it's no wonder that many people remember the intense lightning (see my recent blog entry featuring a lightning animation from an MCC).



The surface map above doesn't offer much information after those other resources but you can see what's left of the MCC over the Delmarva Peninsula.


- Commentary Video by Johnstown Resident BeckyMarie on 1977 & 1936 Floods


- Google Newspaper Archives On 1977 Flood
- Photos of flood damage by Local Storm Chaser Ron Shawley (see random photos below, reload for more)

Surface maps on this page were provided by the NOAA Central Library U.S. Daily Weather Maps Project. Death tolls and damage estimates taken from The Flood Memorial website.

The views expressed are those of the author and not necessarily those of AccuWeather, Inc. or


Comments that don't add to the conversation may be automatically or manually removed by Facebook or AccuWeather. Profanity, personal attacks, and spam will not be tolerated.

WeatherMatrix (Jesse Ferrell)