Tag Archives: National Blend of Models (NBM)

WEEKEND WEATHER FORECAST

High pressure builds in for the weekend as the upper air jet stream flow takes a dip.

For Saturday, high pressure builds in with a cool, northwesterly wind.

Saturday—

  • Sunny
  • High temp 64.5º sd 1.5º.
  • Light winds WNW  2-10 mph.

On Sunday, a frontal system in the Great Lakes area brings clouds from the west while a developing low pressure system far off the coast brings clouds in from the southeast.  We will be sandwiched in between, but the clouds will win out.

Sunday—

  • Partly sunny in the morning.
  • Increasingly cloudy by afternoon.  clouds hold off until late afternoon or evening.
  • High temp  66.9º sd 1.0º
  • Winds SSE 10-15 mph, increasing late in the day.

A few days ago, the models had this dip highly amplified, suggesting a deep coastal low formation for Monday.  The latest models have the jet dip less amplified and positively tilted, resulting in a weaker low pressure system forming late Sunday that remains considerably east of us. (red circle).

However, some models show some light rain for Monday.

 

GEFS forecast for Saturday afternoon.  (Click on image for a larger view.)

 

The new model blend (NBM v 4.0) which just became operational this past week keeps the rain in NJ.

NBM Precip forecast for Monday noon. (Click on image for a larger view.)

 

Last week saw the implementation of the newest version of the National Blend of Models (NBM), version 4.0.

The NBM statistically blends, weighs and combines many models and re-calculates its forecast based on updates from the short range hourly models and actual observations (“Real Time Mesoscale Analysis”) , and re-interpolates it based on the accuracy of previous 6 hour actual observations (“Unrestricted Mesoscale Analysis”).

The models included in the blend are the : GFS, GDAS, GEFS, CMC, ECMWF, FNMOC, ACCESSG, NAM, SREF, RAP, HRRR, HIRESW, WAVE, GFSMOS, NAMMOS, EKDMOS, LMP, GLMP, HWRF, HMON

I’ve been using the the NBM prelease 4.0 for high temperature forecasts over the past several months.  The NBM’s high temperature forecasts are based on a blend of the above models and includes the standard deviation (sd) which gives an idea of the spread between the various models and confidence in the the mean (average) forecast high.

Last year, the older NBM v 3.2 was highly accurate for predicting snowfall amounts and precipitation types, rain, freezing rain, sleet, snow. I expect the new NBM to be even better.

WEEKEND WEATHER FORECAST

Yesterday’s posted Weekend Weather Outlook remains unchanged— this weekend will feature high pressure that brings sunshine, light winds and moderating temperatures, especially by Sunday.

Saturday will be sunny with moderating temperatures by afternoon. The high will be 47 (Blue Bell  Ft Washington) reached about 3 PM.  (The EKDMOS has a high of 51.)

Sunday will also be sunny.  Light winds.  High 54. (The EKDMOS has a high near 58.)

Here’s the current EKDMOS high temperature forecast for the coming week—

Ensemble Kernel Density Model Output Statistics (EKDMOS)   Red is statistical mean value, green is is the lower 10 percentile value and blue is the upper 90 percentile value.

The National Weather Service this past week updated the National Blend of Models (NBM) to version 3.2 and I’ve based my high temperature forecast on this new NBM data for Fort Washington.

It’s also been announced that they will soon be eliminating the EDKMOS forecasts with the idea that the NBM will take its place.

There are significant differences with the NBM and EKDMOS high temperature forecasts for this weekend.  Let’s see which turns out to be correct.