2026 Mid-Atlantic Severe Storm General Discussion

[yoda]

BULLETIN - IMMEDIATE BROADCAST REQUESTED
Severe Thunderstorm Warning
National Weather Service Baltimore MD/Washington DC
213 PM EDT Thu Jul 9 2026

The National Weather Service in Sterling Virginia has issued a

* Severe Thunderstorm Warning for...
  Southwestern Carroll County in north central Maryland...
  Northwestern Howard County in central Maryland...
  Central Montgomery County in central Maryland...
  Southeastern Frederick County in north central Maryland...
  Central Loudoun County in northern Virginia...
  Northern Fairfax County in northern Virginia...

* Until 245 PM EDT.

* At 213 PM EDT, severe thunderstorms were located along a line
  extending from near Linganore-Bartonsville to near Point Of Rocks
  to 6 miles north of Poolesville to near Leesburg to Purcellville,
  moving east at 25 mph.

  HAZARD...60 mph wind gusts and quarter size hail.

  SOURCE...Radar indicated.

  IMPACT...Damaging winds will cause some trees and large branches
           to fall. This could injure those outdoors, as well as
           damage homes and vehicles. Roadways may become blocked by
           downed trees. Localized power outages are possible.
           Unsecured light objects may become projectiles.

* Locations impacted include...
  Germantown, Rockville, Gaithersburg, Leesburg, Olney, Damascus,
  Lansdowne, Lowes Island, Poolesville, Ashburn, Montgomery Village,
  North Potomac, Redland, Countryside, Purcellville, Round Hill,
  Boyds, Green Valley, Clarksburg, and Darnestown.

PRECAUTIONARY/PREPAREDNESS ACTIONS...

For your protection move to an interior room on the lowest floor of a
building.

Large hail, damaging wind, and continuous cloud to ground lightning
are occurring with these storms. Move indoors immediately. Lightning
is one of nature`s leading killers. Remember, if you can hear
thunder, you are close enough to be struck by lightning.

Torrential rainfall is occurring with these storms, and may lead to
flash flooding. Do not drive your vehicle through flooded roadways.

 

[yoda]

Watch up until 10pm

[SnowenOutThere]

That MD cell has some storm rotation. Not tight or anything but I’m sure someone will get a nice storm picture of a sculpted base or something. 

[SnowenOutThere]

There’s now a well defined hook on radar. 

[BlizzardNole]

Those purples on radar over Germantown look rough.  Might be some trees down there

[frd]

 

https://forecast.weather.gov/product.php?site=NWS&issuedby=PHI&product=AFD&format=CI&version=1&glossary=1&highlight=off

 

While probability
outlooks could still change significantly being 7 days out, early ML
& AI outlooks have currently identified a 15%-30% probability for
severe weather occurrence associated with this system. As
details on this system become more defined, will have to
continue to monitor synoptic fluctuations in model guidance to help
identify any hazards associated with this system in the coming
days.

[high risk]

This has all of the makings of a "pity MD" day.

[SnowenOutThere]

5 minutes ago, high risk said:

This has all of the makings of a "pity MD" day.

What about on the edge of an actual MD day? I always find those worse. 

[high risk]

The rare Day 7 SPC outlook for our area!

[frd]

https://www.spc.noaa.gov/products/exper/day4-8/

... Friday/Day 7 and Beyond -- Upper Great Lakes into the
   Mid-Atlantic ...

   Strong northwest flow will be established between the previously
   described western US ridge and eastern Canadian trough. Rich
   boundary layer moisture is expected to be present to the south of a
   residual cold front that will be draped somewhere from the Great
   Lakes southeast into the Mid-Atlantic/Southern New England. Strong
   diurnal heating each afternoon will support scattered thunderstorm
   development capable of producing damaging outflows. Severe
   probabilities were introduced for Day 7 with this outlook given
   increasing confidence in thunderstorms developing within surface
   troughing east of the Appalachians. The exact location of these
   highlights will likely be refined with subsequent outlooks.

   Additionally, the overall large-scale pattern seems conducive for
   one or more shortwave troughs to move through the broader
   northwesterly flow across this region. Given the (seasonably)
   strongly unstable and sheared environment, one or more southeastward
   moving MCSs may be possible during the Day 7 - Day 10 period.
   Ensemble guidance varies significantly as to the timing and location
   of these potential short-wave troughs, but the overall pattern would
   suggest additional severe weather potential at the end of the
   forecast period and in the days that follow.

   ..Marsh.. 07/11/2026

[Kmlwx]

If we are going by weenie superstitions that is a kiss of death. 

[CAPE]

[CAPE]

A quarter inch of rain would be cool. Not asking for much. That would get me to a little over an inch and a half for the month, while continuing the pathetic monthly totals since March.

[vortex95]

6 hours ago, high risk said:

The rare Day 7 SPC outlook for our area!

Yes, esp. this time of year when large-scale dynamic forcing and strong low pressure areas are less, yet enough of a signal in the probs for SPC to outlook an area.

NW flow aloft can lead to some of the best tornado events due to excellent directional shear.  We have good speed shear very often, but strong directional shear is much less common. NW flow promotes lee troughing which can back the low-level winds, so you can over 90 deg turning sfc to 500 mb.

And you tend to get solid EMLs for more CAPE overall.

[frd]

54 minutes ago, CAPE said:

Interesting boundary from the storms in South Jersey is moving SW through Northern DE into NE MD,  causing storms to fire. I just missed one to my South.  

[SnowenOutThere]

57 minutes ago, CAPE said:

This is the most pity of pity MDs released. 

[CAPE]

6 minutes ago, frd said:

Interesting boundary from the storms in South Jersey is moving SW through Northern DE into NE MD,  causing storms to fire. I just missed one to my South.  

Yeah I mentioned that outflow boundary in the other thread.

[frd]

[frd]

https://forecast.weather.gov/product.php?site=PHI&issuedby=PHI&product=AFD&format=CI&version=1&glossary=1&highlight=off

KEY MESSAGE 4...Friday could be active and unsettled with a
chance for showers and thunderstorms, some of which could be
severe.

We are still several days out but we are monitoring the severe
weather setup on Friday. The Storm Prediction Center (SPC) has
highlighted most of the area in a Slight Risk (15% chance of
severe thunderstorms 25 miles from any point). Storms look to be
driven by a strong cold front swinging in from the northwest
and with ample daytime heating and instability, the ingredients
are certainly there for strong to severe thunderstorms and
damaging wind gusts.

In their long term outlook, SPC did mention some small MCS
potential during the day 7 to 10 period as some shortwaves are
likely to ripple through the northwest flow. Again, still
several days out, but we will have to monitor the severe weather
potential for the end of the week and possibly next weekend.

[frd]

 

https://www.spc.noaa.gov/products/exper/day4-8/#

 

 

... Friday/Day 6 -- Mid-Atlantic Region ...

   This evening's deterministic guidance suite (20260712 00 UTC), along
   with yesterday morning's deterministic guidance suite (20260711 12
   UTC), is faster moving the surface boundary south than last night's
   guidance suite (20260711 00 UTC). However, the overall ensemble
   guidance continues to suggest severe potential across the region,
   owing to differences in the position of the surface boundary and the
   timing of mid-level vorticity maxima. Severe probabilities (15%)
   have been retained with this forecast in a nod to the ensemble
   guidance suite, but if the current trend of pushing the surface
   boundary farther south on Thursday (Day 5) continues, severe weather
   probabilities may need to be removed as the surface boundary will
   push south of the better deep-layer shear.

   ..Marsh.. 07/12/2026

 

[Kmlwx]

Yep - it’s always the most likely scenario when outlooked that far out. Things will almost always trend earlier or later and either direction  can of course be worse for timing. 

[yoda]

Good disco from Marsh at SPC in the day 4-8 range

Day 4-8 Convective Outlook  
   NWS Storm Prediction Center Norman OK
   0405 AM CDT Mon Jul 13 2026

   Valid 161200Z - 211200Z

   ...DISCUSSION...
   All ensemble guidance suites are in broad agreement that the central
   US ridge will once again build westward in response to the
   development of a eastern North America long-wave trough during the
   forecast period. However, significant differences exist within and
   between ensembles regarding the timing and exact nature of this
   transition. 

   At the start of the forecast period, a frontal boundary will be
   pushing south and west in the wake of a strong short-wave trough
   that moved across New England in the prior days. The trend over the
   past 48 hours is for this front to push farther south and west,
   bringing a continental dry airmass into New England and portions of
   the Mid-Atlantic. For example, several days ago ensemble guidance
   was in good agreement that the front would be draped across the
   Mid-Atlantic on Friday. Many, but not all, of the recent runs now
   push the front south into southern Virginia and northern North
   Carolina. This results in significant uncertainty regarding the
   potential severe weather environment on Friday (Day 5) across the
   Mid-Atlantic. For example, the GEFS mean temperature for Washington
   D.C. is 93F on Friday afternoon, while the EPS mean temperature is
   102F. However, both ensembles have dewpoint temperatures in the
   50Fs.

   Given the increasing uncertainty as to the location of the surface
   boundary on Friday and the overall quality of the severe weather
   environment, unconditional 15% severe probabilities for Friday are
   removed with this forecast. It is possible that severe probabilities
   may be reintroduced in later outlooks should it become apparent that
   the surface boundary washes out earlier than expected and quality
   low-level moisture is able to return sooner.

   That said, the front should begin to dissipate by Saturday in
   response to a mid-level trough digging southeast across the Great
   Lakes. This will allow rich boundary-layer moisture to advect
   northeastward into the Mid-Atlantic and southern New England. This
   increase in boundary-layer moisture and strong diurnal heating will
   allow for afternoon MUCAPE values between 1000-2000 J/kg, perhaps
   locally higher on Saturday and into Sunday. 

   This unstable and vertically sheared environment will be ripe for
   thunderstorm/severe thunderstorm development ahead of any digging
   mid-level trough. However, as alluded to previously, the timing,
   magnitude, and number of these troughs is highly variable within the
   ensemble guidance, leading to too little confidence to pinpoint
   where and when unconditional severe probabilities should be added.
   As confidence in the timing, track, and strength of these embedded
   troughs becomes clear, severe probabilities will likely be needed
   somewhere across the region stretching from the Upper Midwest
   southeast into the Mid-Atlantic for one or more days.

   ..Marsh.. 07/13/2026

[Kmlwx]

I'll be out of town on Saturday (still in the slight, but still) - so I expect a massive severe line to wipe out the DC area. 

[midatlanticweather]

Slight risk Saturday. Guess it depends how much clearing we get and heating ahead of the energy and way too many unknowns. 
 

ZCZC SPCSWODY3 ALL

ACUS03 KWNS 160734

SPC AC 160734

Day 3 Convective Outlook

NWS Storm Prediction Center Norman OK

0234 AM CDT Thu Jul 16 2026

Valid 181200Z - 191200Z

...THERE IS A SLIGHT RISK OF SEVERE THUNDERSTORMS ACROSS PORTIONS OF THE OHIO VALLEY INTO THE MID-ATLANTIC...

...SUMMARY...

Severe thunderstorms will be possible across portions of the Ohio Valley during the late afternoon and evening hours before moving southeast toward the Mid-Atlantic overnight. Additional strong-to-severe thunderstorms will be possible during the afternoon across portions of the Mid-Atlantic.

... Synopsis ...

The CONUS-wide mid-level ridge centered over the north-central Plains for much of the past week will have retrograded west by Saturday and be located across the US and Canadian Rockies. Strong mid-level westerlies will persist on the ridge's northwest, north, and northeast periphery. Embedded within the westerlies, a series of short-wave troughs and smaller-scale vorticity/speed maxima will move across the Great Lakes and Northeast.

At the start of the forecast period, the main short-wave trough will be located across central Ontario, with its associated jet streak located across the Upper Midwest. This feature will dig southeast on Saturday, ending up in the vicinity of northern New England by Sunday morning. Ahead of this main wave, the 20260716/00 UTC guidance suite shows multiple speed/vorticity maxima moving through the west-northwesterly flow, each of which will be capable of aiding the development of scattered elevated thunderstorms.

At the surface, a weak surface low should develop Saturday morning across southern Ontario/the northern Great Lakes. This low will slowly deepen during the day as it digs southeast, reaching upstate New York by evening. Ahead of the surface low, southwesterly low-level flow will transport a very moist airmass northeast into the Lower Great lakes, with 70F surface dewpoints possible as far northeast as western New York.

To the southwest of the surface low, a surface cold front will serve as an initiating boundary for thunderstorms during the afternoon. An unstable and strongly sheared airmass will be in place by later afternoon across Ohio northeast into western New York. As thunderstorms initiate along the front, large hail will be possible before thunderstorm outflows eventually congeal into one or more linear clusters posing a threat for damaging winds. This line of storms will push south and east during the evening and overnight with a continued threat of strong, damaging thunderstorm winds.

Farther southeast across the Mid-Atlantic, most model guidance indicates the development of a surface trough to the east of the Appalachians during the late afternoon. To the east of this trough axis, temperatures will warm into the 90Fs, although there is some uncertainty how far north the hot boundary layer will extend. Given the presence of this surface trough and strong diurnal heating, several clusters of thunderstorms will be possible during the afternoon, especially if any of the previously discussed vorticity/speed maxima can interact with the low-level trough. Strong, damaging winds will be possible with any thunderstorm complex during the afternoon. Additional severe potential will also exist overnight as the remnants of the convection along the cold front approaches the region. Gusty thunderstorm winds will be the likely threat with these storms.

..Marsh.. 07/16/2026

[WxUSAF]

Why has the NAM just looked like shit for any shower/storm day we’ve had this year from any range >6 hours??

[midatlanticweather]

16 minutes ago, WxUSAF said:

Why has the NAM just looked like shit for any shower/storm day we’ve had this year from any range >6 hours??

I asked Chat GPT - Here is what it thinks (LOL) Waste of time, but I could not resist!

-------------------------
 

I've noticed the same thing this warm season, especially across the Mid-Atlantic. The NAM (both 12 km and 3 km Nest) has had a habit of looking terrible beyond about 6–12 hours for convective days, then suddenly "finding" storms in the final few runs. A few reasons:

1. It's heavily driven by its own physics

   • The NAM tends to overmix the boundary layer during the day.

   • That lowers dewpoints too much.

   • CAPE gets underdone and convective initiation is delayed or suppressed.

2. Small errors explode in summertime

   • July convection in Virginia often depends on:

     • a leftover outflow boundary

     • a weak differential heating zone

     • a tiny vorticity maximum

     • terrain circulations along the Blue Ridge

   • If the model is off by 20 miles or an hour on any of those, the entire simulated radar looks different.

3. Data assimilation doesn't capture morning boundaries well
   Once radar, satellite, surface observations, and morning soundings are ingested, the 6–12 hour forecasts improve dramatically because the model finally "knows" where the atmosphere actually is.

4. The FV3-based CAMs have generally been outperforming it
   This year I've found:

   • RRFS

   • HRRR

   • WoFS (short range)

   • even the HRDPS at times

   have handled scattered convection better than the NAM Nest.

5. The pattern we've been in is especially unfriendly
   We've had a lot of:

   • weak northwest flow

   • pulse convection

   • little synoptic forcing

   • abundant moisture

   That's almost the worst-case scenario for deterministic CAMs. The difference between "nothing" and "10 storms" can literally be one county's outflow boundary.

For Loudoun County specifically

The NAM has repeatedly shown this sequence:

• 24–36 hr: mostly dry

• 18 hr: isolated showers

• 12 hr: scattered storms

• 6 hr: widespread afternoon convection

• Reality: the 6-hour run ends up being closest.

For your forecasting, I'd probably trust the NAM mainly for:

• temperature trends

• dewpoint trends

• low-level winds

• large-scale forcing

...and lean much more on the HRRR/RRFS for actual storm placement once you're inside about 18 hours.

The other thing I've noticed this summer is that the ECMWF AI (AIFS) has often had a better "feel" for whether the day would end up active than the NAM did 24–48 hours out. It obviously can't tell you exactly where a cell will pop, but its broader convective signal has been surprisingly consistent.

[yoda]

Hmmmm... sounds like Saturday could be quite interesting... afternoon AFD from LWX on the threat

@wxmeddler @Eskimo Joe @Kmlwx @high risk

KEY MESSAGE 2...A strong cold front brings the threat of severe
weather Saturday, possibly lingering into Sunday to the south.

An unseasonably strong upper trough and embedded shortwave/jet
streak will pivot near and north of the area Saturday. Strong
height falls associated with this trough, coupled with high
low-level moisture content and modest mid-level lapse rates will
likely result in strong instability, perhaps even if cloud
breaks are muted by early day precip. The strong deep layer flow
will lead to effective shear magnitudes of 30-40 kts, helping to
organize storms that develop in the vicinity of a pre-frontal
surface trough in the wake of any morning activity. Low-level
flow will be a bit weaker during the day, so although transient
supercell structures are possible, the main risk should be
damaging wind gusts and perhaps large hail as opposed to
tornadoes. This could change if the effective warm front
lingers overhead.

Depending on storm coverage, temps could reach well into the 90s
Saturday. With high humidity, heat indices may approach 100.

Heading into Saturday evening, low-level flow increases again
ahead of the approaching synoptic cold front associated with
the aforementioned upper trough. This sends PWATs hurdling to
over 2.2" and keeps CAPE elevated well into the evening. The
expectation is for a strongly forced line of storms to develop
in the unstable airmass from the eastern Great Lakes into the
upper Ohio Valley (i.e. western NY/PA into OH/WV). This line of
storms may be slow to weaken as it treks into the area from the
northwest through the evening, especially if there is moderate
to strong instability left (depends on earlier activity). The
main threat at this point would appear to be gusty to damaging
winds and heavy rain due to multiple rounds of storms. High
instantaneous/sub-hourly rain rates would be plausible in this
setup. If any discrete activity develops in the evening ahead of
the line as the LLJ increases, there may be a risk for large
hail as well as a couple tornadoes, particularly in areas for
surface flow is backed (i.e. near/east of I-95 near any bay
breeze boundaries, increasing SRH). Some guidance such as the
NAM12/RRFS-A has the SigTor parameter in the 1-3 range Saturday
evening in this environment.

Activity should wane during the overnight hours Saturday as the
front and associated convection drop south and the atmosphere
gradually stabilizes. The front will probably end up near or
south of I-64, with any remaining severe threat later on Sunday
ending up near and south of the effective frontal boundary.

 

[SnowenOutThere]

3 hours ago, midatlanticweather said:

I asked Chat GPT - Here is what it thinks (LOL) Waste of time, but I could not resist!

-------------------------
 

I've noticed the same thing this warm season, especially across the Mid-Atlantic. The NAM (both 12 km and 3 km Nest) has had a habit of looking terrible beyond about 6–12 hours for convective days, then suddenly "finding" storms in the final few runs. A few reasons:

1. It's heavily driven by its own physics

   • The NAM tends to overmix the boundary layer during the day.

   • That lowers dewpoints too much.

   • CAPE gets underdone and convective initiation is delayed or suppressed.

2. Small errors explode in summertime

   • July convection in Virginia often depends on:

     • a leftover outflow boundary

     • a weak differential heating zone

     • a tiny vorticity maximum

     • terrain circulations along the Blue Ridge

   • If the model is off by 20 miles or an hour on any of those, the entire simulated radar looks different.

3. Data assimilation doesn't capture morning boundaries well
   Once radar, satellite, surface observations, and morning soundings are ingested, the 6–12 hour forecasts improve dramatically because the model finally "knows" where the atmosphere actually is.

4. The FV3-based CAMs have generally been outperforming it
   This year I've found:

   • RRFS

   • HRRR

   • WoFS (short range)

   • even the HRDPS at times

   have handled scattered convection better than the NAM Nest.

5. The pattern we've been in is especially unfriendly
   We've had a lot of:

   • weak northwest flow

   • pulse convection

   • little synoptic forcing

   • abundant moisture

   That's almost the worst-case scenario for deterministic CAMs. The difference between "nothing" and "10 storms" can literally be one county's outflow boundary.

For Loudoun County specifically

The NAM has repeatedly shown this sequence:

• 24–36 hr: mostly dry

• 18 hr: isolated showers

• 12 hr: scattered storms

• 6 hr: widespread afternoon convection

• Reality: the 6-hour run ends up being closest.

For your forecasting, I'd probably trust the NAM mainly for:

• temperature trends

• dewpoint trends

• low-level winds

• large-scale forcing

...and lean much more on the HRRR/RRFS for actual storm placement once you're inside about 18 hours.

The other thing I've noticed this summer is that the ECMWF AI (AIFS) has often had a better "feel" for whether the day would end up active than the NAM did 24–48 hours out. It obviously can't tell you exactly where a cell will pop, but its broader convective signal has been surprisingly consistent.

Isn't this blatantly wrong? It's the HRRR that always has the overmixing problem, just in the past couple setups the soundings from the Nam may show 3000 CAPE while HRRR is out with 1000 or less. This is a good reminder that AI is bullshit; as in it is extremely confident but has no actual checks on if it is true which is my personal definition of bullshit. It always sounds "correct" until it is your field of expertise you ask it on. 

[biodhokie]

1 minute ago, SnowenOutThere said:

Isn't this blatantly wrong? It's the HRRR that always has the overmixing problem, just in the past couple setups the soundings from the Nam may show 3000 CAPE while HRRR is out with 1000 or less. This is a good reminder that AI is bullshit; as in it is extremely confident but has no actual checks on if it is true which is my personal definition of bullshit. It always sounds "correct" until it is your field of expertise you ask it on. 

Depends on how you prompt the AI and what model you use. If you use the most complex model and give it a detailed prompt and question, you'll get a detailed answer in return. It's still simply a tool rather than an absolute source. Have I used it in my own work? Yes, and it's inadvertently given me academic papers that I or my coworkers are on as sources and quoted it correctly so YMMV. I personally compare the quickest, and usually free versions of the AI to be the same as chatting with friends about random topics while hammered.

[SnowenOutThere]

33 minutes ago, yoda said:

Hmmmm... sounds like Saturday could be quite interesting... afternoon AFD from LWX on the threat

@wxmeddler @Eskimo Joe @Kmlwx @high risk

KEY MESSAGE 2...A strong cold front brings the threat of severe
weather Saturday, possibly lingering into Sunday to the south.

An unseasonably strong upper trough and embedded shortwave/jet
streak will pivot near and north of the area Saturday. Strong
height falls associated with this trough, coupled with high
low-level moisture content and modest mid-level lapse rates will
likely result in strong instability, perhaps even if cloud
breaks are muted by early day precip. The strong deep layer flow
will lead to effective shear magnitudes of 30-40 kts, helping to
organize storms that develop in the vicinity of a pre-frontal
surface trough in the wake of any morning activity. Low-level
flow will be a bit weaker during the day, so although transient
supercell structures are possible, the main risk should be
damaging wind gusts and perhaps large hail as opposed to
tornadoes. This could change if the effective warm front
lingers overhead.

Depending on storm coverage, temps could reach well into the 90s
Saturday. With high humidity, heat indices may approach 100.

Heading into Saturday evening, low-level flow increases again
ahead of the approaching synoptic cold front associated with
the aforementioned upper trough. This sends PWATs hurdling to
over 2.2" and keeps CAPE elevated well into the evening. The
expectation is for a strongly forced line of storms to develop
in the unstable airmass from the eastern Great Lakes into the
upper Ohio Valley (i.e. western NY/PA into OH/WV). This line of
storms may be slow to weaken as it treks into the area from the
northwest through the evening, especially if there is moderate
to strong instability left (depends on earlier activity). The
main threat at this point would appear to be gusty to damaging
winds and heavy rain due to multiple rounds of storms. High
instantaneous/sub-hourly rain rates would be plausible in this
setup. If any discrete activity develops in the evening ahead of
the line as the LLJ increases, there may be a risk for large
hail as well as a couple tornadoes, particularly in areas for
surface flow is backed (i.e. near/east of I-95 near any bay
breeze boundaries, increasing SRH). Some guidance such as the
NAM12/RRFS-A has the SigTor parameter in the 1-3 range Saturday
evening in this environment.

Activity should wane during the overnight hours Saturday as the
front and associated convection drop south and the atmosphere
gradually stabilizes. The front will probably end up near or
south of I-64, with any remaining severe threat later on Sunday
ending up near and south of the effective frontal boundary.

 

This is one of the more legitimate discussions we've gotten this year. Not too often large hail is mentioned so often for us. 3k NAM into the future has CAPE nearly reaching 5000 in ffx county as the first round of storms roll through. Has appreciable supercell/tor composite values thanks to the shear mentioned in the AFD and some turning with height alongside the directional stuff. As mentioned, even after it gets dark we stay with impressive CAPE values of 3000+! Question for our mets, but could the D-word get tossed around with that line of storms overnight?