Posted Thursday 08/07/25 @ 5:18 PM — High pressure to our northeast remains in control of our weather through the weekend with easterly winds and near average temperatures.
The high pressure ridge that noses down from the northeast is blocking hotter and more humid temperatures from reaching the Philadelphia area—
Canadian GDPS with “AI Spectral Nudging” captures the weather setup. High pressure (blue) is blocking moisture and heat from reaching our area. The 500-1000mb thickness line at 760 decimeters (orange) gives a three dimensional outline of the high pressure air mass. South and west of the 760 line (orange), it is very hot and humid. That 760 line might reach us by next Monday or Tuesday. (Click on image for a larger view.)
With sunny skies, the only change will be a gradual increase in temperatures through Sunday.
Today’s 18zNBM temperatures and dew point meteorgram for Blue Bell PA. Philadelphia runs 1-2º higher. (Click on image for a larger view.)
Thursday and Friday
Posted Wednesday 08/06/25 @ 8:04 PM — We had the clouds and the smoke today. Showers broke out on schedule late this afternoon and current radar shows light activity outside of the immediate PHL area.
Radar Image 8 PM (Click on image for a larger view.)
For Thursday and Friday (and the weekend as well) clouds will diminish and both days will have mostly sunny skies. Highs in the low 80s and dew points similar to today, in the low 60s.
Blocking high pressure in the Atlantic will keep this pattern through the weekend, but expect a gradual increase in temperature and humidity, which will become apparent by Sunday. Things change early next week with more heat and humidity as the high pressure system breaks down and retreats.
Wednesday Forecast Change
Posted Wednesday 08/06/25 @ 10:06 AM — As another test for the REFS model under development (and expected release in 2026), the latest REFS is a bit more emphatic about some shower activity Wednesday, late afternoon and evening. A past known model flaw has been an over-prediction of convective precipitation; we’ll see if this has been solved.
Today’s 06z REFS forecast at 5 PM (previous 1 hour rainfall) The showers will move northwestward instead of eastward. (Click on image for a larger view.)
Wednesday- Cloudy and Smokey
Posted Tuesday 08/05/25 @ 7:54 PM — High pressure to our northeast will keep an easterly flow over our area. Dew points remain in the low 60s, not “tropically humid” but not all that dry. Meanwhile, smoke will circulate into our area and will be noticeable.
18z HRRR smoke model forecast for Wednesday afternoon. Contours are low level smoke. (Click on image for a larger view.)
An area of moisture from a disturbance off the southeastern coast will move northward. A cloudy day with smoke. A few light, very widely scattered showers are possible Wednesday evening, most models have these showers to the west of Philadelphia, although the Canadian HRDPS has some showers near Philadelphia and into NJ.
Less cloudiness on Thursday, but a chance of a shower in the afternoon. No distinct sign of tropical development off the southeastern US.
Tuesday Update
Posted Monday 08/04/25 @ 8:12 PM — Tuesday will likely have the most significant smoke at ground level that we’ve seen recently. On top of an increasing cloud cover, especially from the city southward, sunlight will appear dimmed.
18z HRRR smoke forecast near ground level around 4 PM. Smoke will be noticeable. (Click on image for a larger view.)
It appears that any rain from the southeastern coastal system will be suppressed to just south of the city late Wednesday into Thursday. Cape May county will likely see some showers.
A diffuse area of unsettled weather off the southeastern US coast later in the week still not showing much sign of organization. The ECMWF was the most on-board for tropical like development, but the AI models keep the system disorganized.
Humidity remains relatively low through the end of the week moving just into the low 60s. That may change with any tropical development.
Monday Update
Posted Monday 08/04/25 @ 8:37 AM — The wildfire haze will likely decrease a bit over the daytime hours Monday, according to the HRRR-Smoke model.
As for the uncertainty regarding the storm and moisture along the southeastern US coast, the GFS-AI, ECMWF-AI and the Canadian GDPS with “AI Spectral Nudging” have high pressure suppressing the rain just to our south Thursday and Friday. No longer is there a distinct sign of tropical storm formation.
However, a fair amount of cloudiness is expected Tuesday and Wednesday due to moisture moving up aloft.
The GFS-AI maintains the furthest northern extent of the showers on Thursday; it’s likely too far northward—
00z GFS-Graphcast AI model forecast for Thursday. Likely too far north with its showers. (Click on image for a larger view.)
Originally Posted Sun 8:31 PM —Except for the smoke plume that may be with us at daybreak, Monday promises to be another very nice day with low dew points.
The forecast for the balance of the week is more uncertain than you might believe from the broadcast weather people.
There remains significant questions about a large disturbance along the southeastern US. Some of the moisture may move in as clouds as early as Tuesday afternoon.
There is even some question of potential tropical storm development in the same general area. The RRFS, German ICON and ECMWF-AI have some of the moisture moving up along the coast late Wednesday into Thursday. The period beyond Thursday is even more uncertain.
Here’s the latest ECMWF-AI model forecast for Wednesday morning, showing clouds and rain trying to move northward, but somewhat blocked by high pressure. How far that advances is still ‘up in the air’.
18z ECMWF-AI forecast for Wednesday morning. Maroon shading is mid level cloud cover. Rain is blocked from moving northward by high pressure in the northeast. (Click on image for a larger view.)
Posted Sunday 08/03/25 @ 9:54 AM — Another beautiful day expected today, but an increasing flow of smoke from Canadian wildfires will put an increasing layer of haze over us. Near ground smoke concentrations may reach levels that give the air a distinct wood fire smell; something that has occurred a few times since Friday.
By tonight into Monday morning, a smell of smoke should be apparent to sensitive noses—
12z HRRR smoke model forecast: vertically integrated smoke with smoke levels 8 m above ground. (Click on image for a larger view.)
Blocking high pressure should keep low pressure suppressed to our south through at least late Wednesday. The forecasts diverge at that point, with the GFS-AI and the ECMWF-AI models bringing clouds and rain up the coast by Thursday, while the conventional numerical prediction models (GFS, ECMWF, GDPS) keep it suppressed to our south. The German ICON model is somewhat closer to the AI models.
This will be a good test for the AI models.
Today’s 06z GFS Forecast for Wednesday night. The AI models are very different, bringing the low pressure system inland and up the coast by Wednesday night A good test for the AI models. (Click on image for a larger view.)
Originally Posted Fri 4:28 PM —Sprawling high pressure moves in for the weekend, as the heat and humidity is suppressed far to our south.
00z GDPS with “AI Spectral Nudging” forecast for Friday evening. Black-grey is cloud cover. Orange contour is the 576 thickness line (1000-500 mb thickness) The 576 line is a dividing line between tropical air and more pleasant temperatures and humidity A weak tropical system move out to our southeast. (Click on image for a larger view.)
Saturday Forecast
Sunny with light winds, becoming somewhat breezy after noontime.
NBM high temperatures: Blue Bell, PA 81 º Philadelphia, PA 82º Low Uncertainty (based on standard deviation): ± 0.9º
Sunday Forecast
Sunny, less windy
NBM high temperatures: Blue Bell, PA 84º Philadelphia, PA 85º Low Uncertainty High Confidence (based on standard deviation): ± 0.9º
Posted Thursday 07/31/25 @ 3:02 PM — Update: Storms developed in Chester County and westward about 11 AM and have coalesced and continue to develop. A very strong area of upward motion just west of the city will continue to feed thunderstorm development.
RADAR at 2:47 PM with RAP model vertical motion (omega) yellow contours. Yellow Box shows very strong upward motion just to our west. (Click on image for a larger view.)
The latest RRFS hourly shows the rain ending west to east about 6-7 PM in the Philadelphia area. Clouds linger with a chance of a shower later on.
Posted Thursday 07/31/25 @ 8:06 AM — I’ve reviewed last nights 2AM (06z) models and here are the trends and features of today’s expected storms—
Storms won’t “move in”, but will develop in-place beginning as early as noontime in our area. “Pop-up” storms captures it.
Storms become more numerous and coalesce into a large group of storms as the afternoon progresses.
Peak period is the 4 PM to 8 PM time frame
The threat of tornadoes, if any, is more likely in Lancaster county and south, despite yesterday’s NBM forecast.
Storms will be slow moving, capable of dropping heavy rain.
There is some suggestion from the models that the immediate Philadelphia will be shielded from the heaviest activity. This is a low confidence prediction.
Many storms look to be severe. Here’s my current severity table for today’s storms. (I’m in the process of transitioning to the RRFS from the HRRR.)
Severity Parameter (HRRR)
Sept 1 , 2021 Tornado Outbreak (example of highly severe)
Severity Parameters Impact: ⚑ indicates favors Severity⚐ indicates Possible Severity ↓ indicates works against Severity ⇩ Significant, but less impact
Wild Weather Thursday
Posted Wednesday 07/30/25 @ 8:04 PM — A quick update based on the afternoon model runs (18z). The GFS and the RGEM show scattered storms as early as noon in our area. In fact, the 18z GFS shows scattered/isolated storms breaking out as early as 9 AM! Not sure about that. It’s going to be an interesting day weather-wise tomorrow. Stay tuned. Updates tomorrow morning.
Posted Wednesday 07/30/25 @ 5:41 PM — Before we get to Thursday, it’s possible that some isolated thunderstorms may make it into western areas of Chester/Montco/Bucks Counties this evening. Current radar shows a few storms with some strong upward vertical motion in their paths—
Radar at 5:30 PM with superimposed RAP model vertical velocity pressure. (Click on image for a larger view.)
For Thursday, many models are showing severity parameters to be leaning towards the high range. Regarding coverage of thunderstorms, the NBM is showing much of our area to have maximum (4/4) coverage by 3-4 PM—
NBM “thunderstorm coverage” 1-4. Much of the area is a “4” at 4 PM. I don’t often see this degree of thunderstorm coverage. (Click on image for a larger view.)
Some scattered thunderstorm activity is possible earlier, as early as 12-2 PM.
I’ve found that the morning models are the most reliable when it comes to severe weather in the afternoon. I’ll update tomorrow morning.
Wednesday and Wild Weather Thursday
Posted Wednesday 07/30/25 @ 9:41 AM — With temperatures close to the highs reached on Tuesday, but with humidity (dew points) possibly higher, today (Wednesday) will be very hot again. Here’s the latest 12z Model Blend (NBM) just available—
NBM (Model Blend) forecast highs for Wednesday. (Click on image for a larger view.)
Apparent Temperatures (Heat Index) will be in the 101º-105º range, peaking about 2:30-3 PM.
As for Thursday, a cold front will approach. There are signs that severe weather will be a possibility. Timing is currently 2-7 PM. Thunderstorms will be most numerous about 4:30 PM-6 PM. This timing may change.
Specifically, the NBM hourly “tornado probability” parameter is registering in the 1.2% range for the immediate PHL area. That may not sound very high, but on previous days this entire season, its value hasn’t exceeded 0.1% in our region. So it’s 12x higher than previous days this season. Stay tuned.
Tuesday Forecast Review & Wednesday Outlook
Posted Tuesday 07/29/25 @ 5:18 PM — Well, the high at Philadelphia airport was 98º about 3:54 PM and the dew point was 70º. (The transient 99º will likely be ignored.)
What’s interesting is the RTMA, at 3:45 PM (runs every 15 minutes) shows the same temperature 97 as the official record, but lower dew points.
RTMA temps at 3:45 (Click on image for a larger view.)
RTMA dew points at 3:45 PM
RTMA dew points at 3:45 PM. Significantly lower than the actual recorded dew point. I’m not sure why they happens. Another model, the URMA will be available 6 hours later and may recalibrate the dew points to the actual observations. (Click on image for a larger view.)
Here’s the forecast highs for Wednesday (NBM)
18z NBM forecast high temps Wednesday. Dew points similar to today. (Click on image for a larger view.)
Tuesday Heat
Posted Tuesday 07/29/25 @ 9:32 AM — Today, we’ll likely see temperatures reach areas that are very hot, but I would still reserve the term “extreme” for actual temperatures in excess of 100º.
Today’s 12z NBM, just available, shows high temperatures in the upper 90s. (Click on image for a larger view.)
The latest NBM forecasts dew points that are likely (possibly) too high—
12z NBM dew points at 3 PM. Likely (possibly) too high again. (Click on image for a larger view.)
Today’s deterministic numerical models, particularly the NOAA models, keep the dew points in the 70s. However, the AI models have the dew points falling into the upper 60s after noontime, which will keep the over-hyped “Heat Index” or “Feels Like” temperatures lower.
The AI models were correct in yesterday’s dew point forecasts.
As I said yesterday, it’s not unusual to have dew point temperature forecasts be inaccurate, and that’s a reason why “Feels Like” temperature forecasts can sensationalize and be inaccurate since “apparent temperature” forecasts depends on the forecast accuracy of two parameters, humidity and temperature, not just one.
I won’t be posting the NBM “Apparent Temperatures”, aka “Heat Index”, since today’s forecast dew points are a bit on the high side.
Yesterday’s experimental Canadian GDPS with “AI Spectral Nudging” did particularly well in its temperature forecast. Here it is for today—
00z GDPS with “AI Spectral Nudging” high temps (Click on image for a larger view.)
Here’s the GDPS-AI dew point temp forecast for the same time, 3 PM—
00z GDPS with “AI Spectral Nudging” dew points at the same time, 3 PM (Click on image for a larger view.)
Monday
Posted Monday 07/28/25 @ 4:02 PM — I didn’t want to say this, but this “heat wave” seems just a bit over-hyped by the media. Today’s dew points were significantly lower than had been forecast resulting in heat indices that were lower as a result. It’s summer, it’s 93º.
If actual temperatures were approaching 100º, I could understand the use of “extreme” in the forecasts , but..
Here’s the official data from PHL airport—
Dew points dropped below the forecast 70º range in the early afternoon resulting in just a regular summer hot day. (Click on image for a larger view.) Source: https://www.weather.gov/wrh/timeseries?site=kphl
That said, actual temperatures are expected to be higher on Tuesday and Wednesday. But it’s not unusual to have dew point temperature forecasts go wrong and that’s another reason why “Feels Like” temperature forecasts can sensationalize and be inaccurate to boot.
BTW, over past years, I’ve been asked which is my favorite weather app for mobile. Many of them are quite ok and I don’t use any particular one.
I really use my mobile web app (Safari) and and I really like this official web site for local conditions: NWS IDS Forecasts
For live radar, I like something that’s truly live. Many of the apps are delayed images. I’ve subscribed to WeatherTap for over 26 years. They’re not cheap, but I like their radar images on the go.
I have four computers running 24 hours a day automatically downloading and processing weather models. I access these on mobile via a vpn. For excellent weather model data visualization, I use LuckGrib to display my own downloaded data. There is no better app than this for weather grib data visualization, available for iPhone, iPad or Mac.
Posted Monday 07/28/25 @ 9:10 AM — No news here. It’s going to be hot and humid. Here’s the model blend (NBM) forecast highs and heat indices—
06z NBM forecast high temps (± 1.6º) (Click on image for a larger view.)06z NBM forecast heat indices (max) at 3 PM (Click on image for a larger view.)
Originally Posted Sun 8:40 PM —A very hot and humid flow of air returns for first half of this week. The upper level high pressure “heat dome” will be centered over the Southeastern US.
Water vapor imagery shows the upper circulation and the RAP model thickness over South Carolina is 582 dm which is quite high, indicating a very hot air mass. The clouds to our south are associated with the showers that moved through today—
Current satellite water vapor image with superimposed RAP model 500-1000 mb thickness lines (yellow), Mean Sea Level Pressure isobars (black contours) potential vorticity (fine violet contours) with superimposed MRMS RADAR. Water Vapor Imagery reveals emitted water vapor infrared TEMPERATURE, NOT REFLECTED LIGHT from CLOUDS. (Click on image for a larger view.)
High temperatures over the next few days will be in the 90s with heat indices exceeding 100º here. Here’s the latest NBM forecast temps and dew points for Blue Bell—
Model Blend (NBM) Heat index (called “apparent temperature” in the models) Temperatures and dew points with standard deviations. (Click on image for a larger view.)
