Ramblings of Meteorologist Jim Sullivan

Woah! What’s that? It looks a lot like a hurricane not too far from Los Cabos and La Paz over the southern Gulf of California. Well, as of 2PM PDT it was a strong tropical storm, however it is close to becoming a hurricane. I didn’t talk much of Manuel yesterday because it looked like this would remain below hurricane status and drift west towards La Paz over the next few days. However events today have significantly changed that thinking.

Instead of drifting west and only slowly intensifying resurrected Manuel has been slowly drifting north and has a very impressive radar presentation, complete with a well formed eye and eyewall. This system is expected to become a hurricane later this evening. So, what has caused this sudden forecast change? First and foremost, the system became much stronger than normal. When looking at the upper level winds near Manuel, conditions are very favorable for development with winds in the upper levels of the atmosphere fanning away from Manuel. This is called an upper level anticyclone (or high pressure) and helps promote rising air beneath it and helps maintain low wind shear over the tropical storm:

This favorable upper level wind pattern existed yesterday however this rapid intensification didn’t begin occurring until this morning. In addition, most of our forecast models and the National Hurricane Center did not expect this rapid intensification to occur today. Why not? The NHC cited downsloping winds off of the mountains in mainland Mexico as a possible hindrance to intensification. Downsloping winds aren’t good for a tropical cyclone because they feed much drier air into the circulation and disrupt thunderstorm development. However, we still had very intense convection occur in and around Manuel’s circulation today, and as of this writing they aren’t showing any signs of dissipating. My theory is that because Manuel is a relatively small cyclone with its circulation really only grazing land that downsloping isn’t having much of a negative affect:

Note how the lower dew points over inland Mexico…in the 50’s and 60’s…have NOT filtered down to the coastline with dew points in the 70’s around Manuel. This is one possible explanation for why Manuel has over performed today. Manuel being stronger than expected appears to be tugging the tropical storm north, which is different than the more westerly track expect just this morning. Here’s why:

When looking at the flow between about 5,000-10,000 feet up in the atmosphere over the eastern Pacific…and this is the flow that steers weaker tropical storms, like Manuel was yesterday…one can see that Manuel is stuck in weak steering currents between two areas of high pressure (note the two areas of clockwise winds to the east and west of Manuel). It was expected that Manuel would remain in this area of weaker wind flows and sit over the southern Gulf of California for a couple of days before the high pressure to the west extended east and began steering Manuel towards the west in the general direction of La Paz. However Manuel has been moving due north or perhaps a tad bit east of due north today, which was not forecasted. Let’s look at other layers of the atmosphere:

When looking at the flow between about 5,000 feet above the ground to about 23,000 feet above the ground over the eastern Pacific we can see some subtle differences…most notably being a better defined trough over the Rockies. While not a huge difference, the fact that Manuel has become stronger and is now being steering by the flow farther aloft allows it to “feel” the trough to its north a bit better, causing the northward motion we’ve seen today. While at the end of the day the forecast track will only be off by 50-100 miles compared to yesterday, the Gulf of California is less than 150 miles wide at its widest, which basically means Manuel will have direct impacts on the mainland of Mexico now as opposed to Baja.

So the forecast for Manuel:

Continue to drift along the western coast of Mexico for the next 36 hours. The system may make landfall at anytime between tonight and Friday between Culiacan and Ahome. Even if the system does not make a classic landfall anywhere along this stretch of coast, winds may gust over hurricane force at times over the next couple of days. Heavy rains will also be likely. As we head towards Friday and Saturday high pressure is still expected to briefly build in from the north and begin turning Manuel to the west. From this point it may threaten Baja between Loreto and La Paz. Fortunately the area between those two cities is extremely sparsely populated.

Manuel will likely become a hurricane this evening. Although the system is quickly intensifying as of this writing if it moves much closer to land the intensification will cease. Thus, I have a hard time forecasting anything stronger than a category 1 hurricane out of Manuel over the next couple of days. Most of our models show enough land interaction as Manuel moves up the increasingly narrow Gulf of California…and potentially some upwelling of cooler waters and Manuel constantly churns up the sea over the same area for the next few days…to significantly weaken Manuel by Saturday. This mucks up the track forecast somewhat over the weekend…as another trough is expected to dive into the western US. A stronger Manuel would likely be picked up by the trough while a weaker Manuel would likely continue to slowly drift near Baja. At this point we’ll again try to go with a weaker solution, causing Maneul to possibly make landfall on Baja this weekend as a weakening tropical storm that eventually dissipates. A look at our forecast track models for Manuel show how several different solutions are being portrayed by our various models:

Elsewhere in the tropics we continue to monitor an area of low pressure that has recently emerged over the Bay of Campeche after spending some time over the Yucatan Peninsula. Shower and thunderstorm activity has begun to become more concentrated near this rather potent low pressure center and a tropical depression will likely form within the next 24 hours if this trend continues:

The center of circulation appears to be near the northern edge of the ball of convection on the above satellite imagery, due to some moderate northerly wind shear. As we discussed yesterday I expect shear to begin diminishing over this disturbance as an upper level anti-cyclone builds over it. This appears to favor a strengthening system over the next couple of days and we should have Tropical Storm Jerry in the southwestern Gulf of Mexico by the weekend.

From there the forecast remains highly uncertain as a trough over the central US possibly interacts with this area of low pressure. The models show the trough getting far enough south into the Gulf to pick up a stronger cyclone but leave behind the weaker one. You can see how borderline it is on this afternoon’s run of the National Weather Service’s GFS model:

IF this system remains weak it may get left behind by the trough and either dissipate or make landfall in Mexico. A stronger system may get picked up by the trough and get sling shot towards the Northern Gulf Coast. I still favor a weaker system staying south and moving into Mexico Friday or Saturday however if this low pressure system becomes more robust…oh say like Manuel over the past day…there could be a threat to the Northern Gulf Coast. Continue to stay tuned as we monitor this disturbance.

Looks like the tropics will remain relatively active close to home for the next few days. Satellite shows a well-defined low pressure system just north of the Mexico/Belize coastline this evening with some thunderstorms firing near this low pressure. This system is tracking towards the WNW and should move into the Bay of Campeche on Wednesday.

Environmental conditions are currently marginally conducive for development. An upper level low just west of the low pressure may cause modest northerly shear over the system for the next day or two…however that upper level low is shown by all modeling to weaken/move west. A weak upper level anti-cyclone (high pressure) exists near the low pressure and this is helping to vent the thunderstorms associated with the low pressure…note the winds to the northeast of the low pressure rushing away. Our forecast models show this kind of upper level wind pattern continuing, if not improving over the next few days.

The above image shows winds fanning away in the upper levels over the low pressure…at this point in the Bay of Campeche. In addition, note how the vorticity, or low level spin (organge lines on second image) is consolidated with this low pressure system. Unlike with hurricane Ingrid, there is not going to be a tropical cyclone on the other side of Mexico in the Pacific to compete with this low pressure system. This promotes an organizing system. Although waters are a bit cooler in the Bay of Campeche than they were before Ingrid…they are still marginally favorable for tropical cyclone development. If this low pressure can get out of the southern Bay of Campeche…waters are warmer and are warm to a great enough depth to promote an intensifying tropical cyclone.

So the question is…where will this low pressure system go? While there is currently a mid-level ridge of high pressure over much of the Gulf of Mexico that will limit how far north this low pressure system can track over the next couple of days…this ridge is expected to be eroded over the next 2-4 days by a trough dropping into the central US (above image). If this low pressure system is still weak in 3-4 days it will likely move into Mexico before this hole in the ridging allows the system to possibly turn north. If this low pressure system deepens into a decent tropical storm or hurricane within the next 3-4 days this trough should be able to pick up this low pressure system and threaten the northern Gulf Coast.

Which option is more likely? It’s tough to tell as most of our forecast models are rather split. What is often noted as the most accurate model, the ECM, actually picks up half of this system with the trough and slings it towards Florida in about 5-6 days and leaves about half of the system back over the southwestern Gulf of Mexico.

An experimental suite of ensemble guidance for tropical cyclone track…that has shown promise the last two seasons…shows this uncertainty. Most of these models actually show the low pressure system getting stuck in the Gulf of Mexico in 5-6 days. I think this is unlikely but illustraits how this track forecast has much uncertainty…although a system getting stuck in the Gulf of Mexico seems unlikely to me. Given the favorable environment that seems to await this low pressure system in the Gulf of Mexico and the fact that a stronger cyclone would be more likely to be picked up by the trough moving across the central US late this week…I have a sneaky feeling that this system won’t go quietly into Mexico. Either way…stay tuned…

So it’s getting to be that time of year where the days are just a little shorter…the high temperatures just a little cooler…and school getting back in session. Winter weather is still months away from most US locations…but it is starting to get to the point where we can look at some trends around the globe in both the atmosphere and oceans and try to take a stab at what this winter will bring. Here is a look at my final winter outlook from last year:

https://ohwxramblings.wordpress.com/2012/10/18/2012-2013-winter-outlook/

Pacific/Indian Ocean Water Temperatures:

Despite thoughts in the spring and even early summer than an El Nino may develop, the equatorial Pacific as of mid-August is running a bit cooler than normal, especially closer to South America. The most recent two winters have featured a warm Indian Ocean…as of mid-August however, the Indian Ocean has cooled to near or a bit cooler than normal. We are in the early half of a multi-decadal “negative” “Pacific Decadal Oscillation” (henceforth PDO), however this summer the PDO has tilted closer to neutral. We’ll discuss what this all means below:

El Nino-Southern Oscillation (from now on, ENSO):

The state of the ENSO describes sea surface temperatures in the equatorial Pacific, with a “El Nino” occurring when these waters are warmer than normal and a “La Nina” occurring when these waters are cooler than normal. The winter of 2011-12 was a strong La Nina. Last winter was a borderline cool neutral/weak La Nina. The question is, will the current cooler than normal waters in the equatorial Pacific lead to another winter La Nina?

A “La Nina” officially occurs when the ENSO regions (shown above) average 0.5 degrees Fahrenheit or more below normal for three or more consecutive three month periods. As seen above on the breakdown of ENSO region temperatures through early August, we aren’t there yet. However, all of the ENSO regions have either cooled or held steady temperature wise since the beginning of June.

Looking just below the sea surface in the equatorial Pacific Ocean…one can see that the expanse of warm anomalies that had developed sub-surface to begin the summer have generally subsided. However, cold anomalies have had trouble taking hold outside of the far eastern Pacific.

The Southern Oscillation Index, or SOI, is a measure of the pressure difference between western South America and eastern Australia. When the index is positive (as it has been on average for quite some time of late) it indicates that the pressures are higher over South America, which favors stronger than normal trade winds over the Pacific and an upwelling of cooler sub-surface waters. When the index is negative (which it was on Friday) the opposite occurs. A positive SOI favors the development/persistence of La Nina and vice/versa. The long term SOI this summer favors the development of La Nina, however recent values suggest otherwise. Updated SOI data can always be found at this link:

http://www.longpaddock.qld.gov.au/seasonalclimateoutlook/southernoscillationindex/

Current trends all lean fairly strongly against El Nino development, however there hasn’t been a strong push towards La Nina like conditions recently either. The very cold eastern Pacific does point towards a La Nina and an average of the ENSO regions is certainly tilted towards La Nina but isn’t quite there yet. Given only a modest tilt of the aforementioned factors towards La Nina development, I think that if a La Nina develops this winter it will be weak…if not, the ENSO will be “neutral” but certainly not warm.

Pacific Decadal Oscillation (PDO):

As mentioned above the PDO is in a multi-decadal “negative” phase since the late 1990’s…and has been especially negative since 2008. For reference, here are what the two phases of the PDO look like, courtesy of the University of Washington (warm/positive on left, cold/negative on right):

As seen at the top of the article, despite the long term cold PDO, the waters off the west coast of North America are rather warm at the present, which gives the current PDO a more neutral look. This is likely due to abnormally strong ridging and quiet weather over Alaska and western North America since mid-May, which has contributed to the rather rainy and at times cool summer over the central US:

The forecast models and expected placement of tropical convection this week suggest that troughing is likely to occur over northwestern North America over at least the next two weeks…and into September. If this occurs, those warmer waters will quickly begin to cool and the PDO will take on a more convincingly negative look. I expect this to occur and feel fairly confident in a negative PDO winter.

Indian Ocean:

The Indian Ocean Dipole (or IOD) is a measure of the east to west temperature gradient across the Indian Ocean. A negative IOD, as we are in currently, is indicative of warmer waters near Australia and cooler waters farther west near Africa. This summer’s current cool waters over the western Indian Ocean have at times created enough of a temperature difference across the ocean to meet the threshold of a “cold” dipole event. This is in stark contrast to the abnormally warmer waters over the western portions of the basin over the last two summers, which at times were warm enough to meet the threshold of a warm dipole event.

http://www.jamstec.go.jp/frcgc/research/d1/iod/DATA/dmi_HadISST.txt

Based on this data set provided by Jamstec, I have found 9 cases since 1959 that featured a cool Indian Ocean Dipole over the course of December, January and February that may be similar to what we will see this winter. In general I looked for an average DJF index of -0.3 to -0.8, and threw out a few cases that matched this based on the index not being anywhere near this year’s negative index. Here is a look at where convection was more/less active than normal in the tropics during the years I selected during the DJF timeframe:

Note how unlike last winter, convection is generally less active than normal over the Indian Ocean and more active east of Australia and even east of the dateline. The composite SST anomalies of these winters shows a cool/neutral ENSO, however a couple of these winters were El Nino’s that may have skewed up precipitation amounts some east of the Dateline. A key here is however that we likely won’t see persistent convection as far west of the Dateline as we did last winter, which resulted in an abnormally strong Asian/Pacific jet and rather persistent ridging well off the western US coast. Below is a look at the SST in the winters looked at above:

Summing up the Pacific/Indian Ocean:

I expect a cool neutral ENSO this winter, with some chance at a weak La Nina. I expect the waters off the west coast of North America to cool, giving the PDO the more negative look we have been accustomed to in recent years. I expect the Indian Ocean Dipole to remain cool and perhaps become a true cold dipole event. In general one would expect convection to be most active from Australia to the dateline based on this SST configuration, putting ridging near/ just off the west coast of the US to the northeast of the convection. This is in contrast to last winter when ridging was strongest south of Alaska due to a persistently strong Asian/Pacific jet, especially early in the winter.

I have thumbed through DJF reanalysis SST’s over the Indian Ocean/Pacific since 1949, and found 3 years (coincidentally all since 2000) that reasonably match what I expect out of this winter’s SSTs:

The mean 500mb pattern in those winters is close to what I described above…with ridging persisting a bit off the western North America coast, a bit east than where it was centered last winter:

With cool waters in the equatorial Pacific and ridging just off the west coast of the US, one would expect a weaker than normal sub-tropical jet promoting drier and warmer conditions than normal across the southern US this week…especially the southeastern US. With a possible southeast ridge and cooler than normal air east of the eastern Pacific ridge over the Rockies and central US…conditions could be rather wet across portions of the Mississippi Valley if the Atlantic cooperates…speaking of which:

Atlantic Ocean:

This summer has featured persistently warm waters off of the New England and Nova Scotia/Newfoundland coastlines, along with cooler waters off of the south coast of Greenland. The 500mb height anomalies have almost matched this configuration through the summer, with abnormally high heights from June 1 through mid-August off of the New England/southeastern Canada coastline, and lower than normal heights over the southern tip of Greenland. The question is, will this positive NAO pattern continue into winter?

I have thumbed through all summers since 1949, seeking out other June-July-August periods that featured similar Atlantic sea surface temperature configurations…with emphasis on warmth over the NW Atlantic and cooler waters over southern Greenland…I also looked for summers that didn’t feature scorching waters in the equatorial Atlantic. Here is the mean SST anomalies from those JJA periods:

It turns out, when looking forward, that in most of those winters the warm anomaly off of the southeastern Canada coast went away when averaged out:

And for better or for worse, the mean 500mb height pattern over the Atlantic those winters featured a clear +NAO signal with eastern US ridging and lower than normal heights over much of the Northern Atlantic:

What is scary is a look at the Pacific and western North America 500mb height anomalies these winters is somewhat close to what I expect this winter, with ridging just off the western North American coast and trough over the central US, with perhaps some southeast ridging….although the above “Atlantic analog” years do not all match the “Indian Ocean/Pacific analog” years used earlier.

This year, the monthly NAO values have ranged from +0.5-+1.0 for each month April-August. When thumbing through monthly NAO values provided on NOAA’s Climate Prediction Center’s website (ftp://ftp.cpc.ncep.noaa.gov/wd52dg/data/indices/nao_index.tim) I found 6 years (since 1950) that had spring-summer NAOs that fairly closely resembled this year’s monthly values…and I had to get a bit generous with which years were included on the last at times just to get a data set larger than one or two years, evidently modestly but consistently positive monthly NAO values in spring-summer don’t happen often. The years found are: 1959, 1965, 1976, 1989, 1992, 2002. In those six years, four winters featured a predominantly negative NAO (1959, 1965, 1976, 2002) and two featured a predominantly positive NAO (1989 and 1992). Talk about conflicting data in more than one sense of the word! Here is a look at those winters from a level of about 500mb, on the mean:

As encouraging as this looks, the mean ENSO in these years was warm with a warm PDO…and well below normal waters off the coast of New England and southeastern Canada, which seems unlikely to occur this year. So, as encouraging as this may be for the eastern US, I’m not sure I’ll be giving it much weight.

Quasi-Biennial Oscillation:

A look at the upper levels of the atmosphere over the tropics suggests that the zonal winds are blowing stronger than normal out of the west, and strengthening in the stratosphere. Lower level temperatures and winds (500/200mb respectively) have not responded drastically, but it appears as though the QBO will be positive or westerly this winter, and perhaps moderately so. Since 1979, per this data-set using the 30mb level as a proxy (http://www.cpc.ncep.noaa.gov/data/indices/qbo.u30.index), I have found 5 years that featured the QBO rising to fairly high positive values in the summer that persisted into winter…they are 1982, 1985, 1990, 2008, 2010. When the QBO is in a “negative” or easterly phase sudden stratospheric warming events and potentially severe blocking are more likely. When the QBO is in a “positive” or westerly phase sudden stratospheric warming events tend to be less likely. In the years above, here is a look at the winter time 500mb heights anomalies over the upper latitudes in the Northern Hemisphere:

There are signs of high latitude blocking, especially from eastern Siberia to Alaska (which I’m not sure I agree with based on the Pacific pattern discussed above) and signs of a modest –NAO…although they aren’t strong.

Years with similar summer patterns:

This summer (through mid-August) has featured positive height anomalies from Alaska east across northern Canada with lower than normal heights over southern Greenland…and above average heights off of the New England and southeastern Canada coastline. The western US has seen modestly higher than normal 500mb heights with modestly below normal heights over the Mississippi Valley and Great Lakes.

Looking over the North American summers since 1949, I found a few summers that matched enough of the above mentioned features for me to take a closer look. 1994 was the best match by a fair margin and was given extra weight. The 500mb height anomalies in those summers looked like this, not a terrible match:

The composite 500mb height anomalies for the following winters aren’t pretty across the whole US:

The mean SST anomalies from these winters featured a west-based El Nino, which appears unlikely this winter. There were some similarities in a mean cold PDO…similar Atlantic SST configuration…and not too far off Indian Ocean configuration. Although I won’t give the above composite a ton of weight, it may need to be watched a bit if the La Nina falls apart. At this time it appears unlikely, but it is still just August. Here is a look at the SST anomalies from those winters in the tropics:

Putting it all together:

Given everything discussed above, I feel fairly confident in ridging just off the west coast of the US and a trough over the Rockies and central US. I also feel fairly confident in a southeast ridge appearing at times with generally below average precipitation across the southern US from coast to coast due to a cool ENSO. I feel somewhat confident in the NAO averaging somewhat positive but not strongly so, which doesn’t bode particularly well for the east coast outside of New England.

There may be a sudden stratospheric warming event and a period of strong –AO blocking, however I don’t feel like the QBO supports this, which may result in a predominantly positive AO winter. The best shot at persistent high latitude blocking may be over the EPO region (Alaska/western Canada) where I feel like ridging will be most consistently impinging on the polar vortex. Given where I think trough is most likely across the central US a favored storm track may be from the central/southern Plains east/northeast towards the Great Lakes and New England…which may favor above average snow across the central/upper Mississippi Valley…central/upper Great Lakes…and inland New England. Areas like the Ohio Valley…lower Lakes…upper Mid-Atlantic and southern/coastal New England may see many messy storms and even some warmer/rainy storms.

Trends among sea surface temperatures worldwide and other factors such as fall snowcover will give us a better idea of what to expect as we head towards October and November, but here is a look at what I expect this winter across the US and southern Canada…on a mean for December-February. If time permits I will update this outlook as needed and add monthly maps in about two months. Thanks for reading!

My current thoughts on DJF temperatures:

Thoughts on DJF precip:

Based on my interpretation of the pattern, storm tracks that may be common:

This will be article 1 or 4 discussing moderate to in some cases extreme snowfall that has occurred over Northern Ohio in April. Because my memory and in some cases archives of these events become quite fuzzy the farther back we go, I will be examining for decently recent events, from 2005 onward, of heavy April Snowfall.

Synopsis/Observations:

The winter of 2004-2005 was already a notable one across much of Northeastern Ohio, with Cleveland Hopkins International Airport, henceforth CLE, the flagship official climate station for the Cleveland metro, approaching record snowfall for a single winter. April snowfall does occur from time to time across Northeastern Ohio, and every now and then more than just a light dusting does fall, especially in the Snowbelt. However, with CLE only averaging 3.3” of snow during the month of April (using 1981-2010 averages), one is never able to firmly count on April to deliver significant snowfall, even after a winter like 2004-2005 during which the atmosphere seems to produce excessive amounts of snow ad-nauseam across Northern Ohio.

However, the atmosphere would put to doubt whether or not 2004-2005 would become the snowiest winter on record in Cleveland during the first three days of April:

Although wavelengths tend to shorten in the spring, similar principles still apply. As we look at the 500mb chart, representing the weather pattern at 18,000 feet on 4/1/05, it is evident that there is ample western US ridging for trough amplification over the central US. This, combined with higher heights than normal over eastern Canada and Greenland, which aren’t readily visible on this map, ensured that the trough would not be in any hurry to move east and allow mild air to return to the Midwest, Great Lakes, and Ohio Valley in early April of 2005.

A shortwave seen rounding the base of the trough on April 1 sparked off a surface cyclone over Texas that tracked east-northeast during the first and second of the month towards the Tennessee Valley:

A look at the surface map for the evening of April 1 shows our surface cyclone tracking well south of the region. However, due to the fact that it’s April, there was not much cold air to the north of the system yet, meaning much of the precipitation falling over the Great Lakes and eastern Midwest was rain. So, we would have to wait for colder air to arrive for the snow to begin.

The wait would end as we trudged towards April 2, as the 500mb trough (shown above the morning of the 2^nd) would close off, allowing dynamical cooling to slowly cool the lower levels of the atmosphere over the northern Ohio Valley and lower Great Lakes. Some cooler air was also pulled down from Canada during this process.

Another look at the morning of April 2^nd off of the Penn State e-wall shows the freezing line at 850mb, or about 5000 feet above mean sea level, plunging into the Deep South. In April this usually does not correlate well with the rain/snow line; at the elevation of northern Ohio (600 feet at the Lake Erie shoreline to over 1100 feet in some of the hills in far Northeastern Ohio) 850mb temperatures usually need to dip below -5C/23F to get accumulating snow in April. Although the -5C isotherm is not near Ohio yet on this image, the air was cold enough and precipitation heavy enough for snowfall to begin reaching the ground over Northwestern Ohio in the few hours proceeding this image. CLE began mixing with snow around 4am on the 2^nd, changed to snow by 4:30am, and remarkably reported snow in every observation for the rest of the day, with visibilities at times dropping to ¼ of a mile, signifying heavy snow.

Once you got to the east side, the track of the 850mb low (shown in the bottom right panel) held back the rain snow line through a good chunk of the day. At Ashtabula County Regional Airport (henceforth HZY) the transition to snow did not occur until 1:25pm on April 2, and the air temperature did not drop below 35 degrees until after midnight, holding accumulations in check well east of Cleveland until after dark on the 2^nd.

Either way, snow was ongoing and accumulating moderately to heavily at times by late April 2 and continued right on into April 3 before tapering in the morning from west to east.

The snow on the west side, including at CLE, shutoff soon after 12am on the 3^rd as the event become more and more lake-effect driven thanks to the synoptic precipitation with the surface low pressure beginning to pull off to the east. However, on the east side, snow continued through much of the morning on the third:

Note how from Cleveland west, the mid-levels of the atmosphere are drying and high pressure at the surface is beginning to move in. East of Cleveland, the atmosphere remained moist and the western edges of a cyclonic flow helped insure large scale lift/convergence over the region. With 850mb temps dropping to near -5C in this timeframe, there was just enough instability added from Lake Erie (water temps approaching 40F/+5C (educated guess)) for snowfall amounts to be enhanced east of Cleveland. This resulted in many hours of moderate to heavy snow east of Cleveland.

At HZY, moderate to heavy snow occurred in consecutive observations from 4:53am on April 3 through 1:53pm on April 3, meaning visibilities were reduced to less than ¾ of a mile. There were 5 hourly observations (non-consecutive) in this timeframe during which heavy snow was reported, meaning visibilities were reduced to less than ¼ of a mile for a considerable portion of the event.

Despite all of the snow, CLE never dropped below 33F during the event, which occurred during the afternoon and evening hours on the 2^nd, and HZY, stunningly, never dropped below 33.8F during an hourly or intra-hourly observation, and was generally sitting at 34-36 degrees during the heaviest snowfall. CLE rebounded to 53 by late afternoon on the 3^rd and HZY rebounded to 46 by late afternoon on the 3^rd, despite recording “heavy snow” as late as 12:53PM and “snow” (with a visibility of half a mile) as late as 1:53PM. CLE reported 6.8” of snow with the event. Ashtabula (not formally specified as HZY in the official release) recorded 16.8” of snow.

Four other regional airports in Northern Ohio were affected in at least a moderate fashion by this early April snowstorm in 2005:

Mansfield (MFD):

Changed to snow at 7:30am on 4/2/05, and reported snow in every observation through 11:11PM when the snow summarily ended. Being farther inland from the slightly warmer waters of Lake Erie, MFD was able to cool to 32-33F during the majority of the heaviest snowfall. The visibility at MFD was held at or below 1 mile from 7:52am through 5:10PM on 4/2/05, with the visibility held at or below ¼ of a mile from 10:22am through 1:52pm, signifying over three hours of consistent heavy snowfall. Visibilities again were held to ½ to 1 mile from 6:45pm through 9:52pm, signifying another three plus hour period of heavier snowfall. MFD reported 8.0” of snowfall with the event.

Akron/Canton (CAK):

Rain did not completely change to snow until 1:51pm on 4/2/05 at CAK after two hours of rain/snow mixed. After the rain fully changed over to snow, snow was reported in every single observation through 5:50am on 4/3/05 (with another brief period of light snow around 8:00am on 4/3/05). Being inland from Lake Erie also helped temps at CAK sit at 32-34 degrees during the heaviest snowfall, which is conducive for accumulation once snow comes down at a heavier clip. At CAK, the visibility was held at or below 1 more from 2:04pm on 4/2/05 through at least 1:51am on 4/3/05 (although, it should be noted there were no observations between 1:51am and 4:50am on 4/3/05, potentially due to a power outage). Visibilities at CAK never dropped below ½ of a mile, however with many hours of moderate snowfall and temperatures at or just above freezing, some accumulation did occur. CAK reported 5.1” of snowfall with the event.

Youngstown (YNG):

Changeover to snow occurred by 1:51pm on 4/2/05, with snow reported in every observation through 1:51pm on 4/3/05, meaning YNG received 24 consecutive hours of snowfall in April. Being inland from Lake Erie also helped YNG hover between 30-33 degrees during the majority of the snowfall, allowing for snow accumulations to occur. At YNG the visibility had dipped below 1 mile by 6:38pm on 4/2/05 and did not rise above 1 mile again until 1:43pm on 4/3/05, representing over 19 consecutive hours of “high-end light” or moderate/heavy snowfall. In that timeframe YNG reported half mile or lower visibility from 5:51am on 4/3/05 through 12:51pm on 4/3/05 (7 hours) and reported quarter mile or lower visibility from 6:04am through 10:51am on 4/3/05 (just short of 5 hours). Through all of that, YNG reported 8.5” of snowfall.

Lorain/Elyria (LPR):

LPR changed over to snow by 3:51am on 4/2/05, and reported snow in every observation except for one through 7:53pm on 4/2/05, representing 16 hours of essentially persistent snowfall at LPR. Although LPR did not have any notably long periods of visibilities at or below one mile, there were a couple hours of moderate snow in the morning and again after noon, with visibilities bottoming out at ¼ of a mile at 2:53pm on 4/2/05. Temperatures generally sat between 33-35F during the snowfall at LPR. No official tally from LPR was available from what I could find, however, based on nearby snow totals and observations at the airport, it appears as though 4-6” of snow reasonably fell on 4/2/05 at LPR.

Societal Impacts:

Leaves were just beginning to open up across Northeastern Ohio when this snow fell, however winds gusting over 40MPH and the heavy nature of the snow (note, surface temperatures generally sat above freezing or fell just below it during the snow) allowed for widespread tree and power line damage from eastern Cuyahoga, northeastern Summit and northern Portage Counties points eastward through Geauga and Ashtabula Counties. Tens of thousands were left without power. Travel due to downed trees/wires and heavy snow accumulation on roadways was nearly impossible on the 3^rd across much of Geauga and Ashtabula Counties, and was hampered at times well west of there late on the 2^nd into the morning of the 3^rd where lighter, but still significant, snowfall totals were observed.

Comments:

The fact the Toledo never reported snowfall with the event and Findlay only reported a few hours of light snow or rain/snow mixed illustrates how much dynamical cooling thanks to the deep/closed 500mb low helped change rain to snow, and also shows how heavier precipitation rates helped the changeover as well, as Toledo and Findlay never got into heavy precipitation on 4/2/05. The slow movement of the 500mb trough/low allowed the deformation snows to sit over north-central and northeast Ohio for many hours, and added moisture/conditional instability from Lake Erie allowed the Snowbelt communities to see the heaviest snowfall totals.

2-day radar loop of event (times are in UTC, subtract 4 hours from time shown on radar to obtain EDT equivalent):

https://www.youtube.com/watch?v=6MqWPptZ-aM

Snowfall total highlights:

Ashtabula County:

22.0” in Conneaut along I-90

23.0” 2 miles SE of Pierpont

16.9” in Geneva

16.8” in Ashtabula

Cuyahoga County:

9.0” in Shaker Heights and North Royalton

6.8” at CLE

Geauga County:

24.8” 3 miles south of Thompson

22.9” 5 miles southwest of Thompson

20.5” in Hamden Township

17.0” in Chardon

Lorain County:

5.6” in North Ridgeville

Mahoning County:

8.0” in Austintown

Medina County:

5.8” in Hinckley

Portage County:

11.5” in Hiram Rapids

5.2” in Kent

Summit County:

5.1” at CAK

Trumbull County:

8.5” at YNG

*Colt Station in NW PA reported 29.0” of snowfall with the event.

I have the complete list of snowfall totals saved to my computer. If you would like a full copy of snow reports let me know.

Images 1-3 were obtained from this website:

http://locust.mmm.ucar.edu/case-selection/

Radar loop was also constructed using data obtained from that website.

Images 4-5 were obtained from this website:

http://www.meteo.psu.edu/~fxg1/NARR/2005.html

The archived PNS containing snowfall reports used above was obtained from this website:

http://has.ncdc.noaa.gov/pls/plhas/HAS.FileAppSelect?datasetname=9957ANX

The hourly observation data sited above for CLE, CAK, MFD, CAK, LPR and YNG were obtained through Wunderground.com