New and Improved Forecasts

MWA updated its proprietary 30-day global weather forecast ensemble to the MPAS model (Model Prediction Across Scales), which was collaboratively developed by the National Center for Atmospheric Research (NCAR) and the climate modeling group at Los Alamos National Laboratory (COSIM). MWA modeling staff selected the latest update (Version 5) of this unique research model for conversion to an operational forecast platform based on several key features distinguishing MPAS from all other weather forecast models.  Initial verification analyses show the model is highly stable and more accurate than the GFS model at forecast time scales beyond day-5.

The model is comprised of individual simulation components specific to atmosphere, ocean, land ice, and sea ice employing a unique hexagonal grid system especially suitable for higher resolution applications over any geographic area. Model output available to standard subscriptions include 30 day global forecasts of upper atmosphere and surface meteorological parameters; displayed graphically (including 5 day composites of temperature anomaly and precipitation), as well as in numerical form for over 200 cities. Forecasts specific to any city or global region can be customized to greatly enhanced resolution to satisfy individual user requirements

 

30 Day Weather Forecast:

 

This weather forecast newsletter is prepared by MWA meteorologists 3 times per week on Sunday, Tuesday and Thursday, and is often very different and in some cases totally contrary to direct computer model output.  Each forecast update summarizes primary weather events expected to be of significance to the energy industry during the next 30 days, with an emphasis on forecasting substantial pattern changes that move markets.  This format is designed to identify major trading opportunities in which markets may be moving in the wrong direction, based on government or private forecasts that are primarily repackaged computer model forecasts.

The format of the newsletter is designed to address a wide audience of traders and meteorologists. As such the broad based forecast presents a 30-day graphic in the 1st page, which depicts U.S. temperatures anomalies during the next 1-5 day, 6-10 day, 11-15 day and 16-30 day periods for easy comparison to computer models and other forecast providers.  The rest of the document includes detailed discussion of the forecast basis with supporting weather maps.

 

90 Day Weather Forecast:

This long range weather forecast newsletter is released once per month and includes detailed assessment of the main atmospheric indices of interest to seasonal weather conditions.  The forecast horizon is a minimum if 90-days, but can extends as much as 180 days depending on predictability of specific atmospheric patterns.

Similar to the 30 day newsletter (MWA 30) seasonal forecast maps and primary forecast basis are presented in a straightforward format in the 1st page. Detailed assessment follows describing major atmospheric features expected to drive weather conditions in an understandable format that provides the reader with the necessary tools to continually monitor the degree to which the long range forecast is verifying until the next update is issued. By describing the expected evolution of major atmospheric features expected to drive an upcoming season’s weather, the reader is equipped with the ability to most effectively utilize the forecast for trading decisions.

 

 

February 19 - Long Range Summary

The coldest air of the season in place across the central U.S. (35°-40° below average) is quickly spreading south and east into the Deep South and the Northeast with only minor modification (weakening) over widespread snowpack which is also still expanding behind 2 distinct winter storms lined up to track from the Rockies to the mid Atlantic coast through the end of this week. NWS Extreme Cold Warnings and Advisories valid today across nearly the entire Eastern half of the U.S. (including the western and central Gulf Coast) are certain to be extended as all models maintain extreme double-digit below average anomalies through the end of this workweek. While this intense level of cold air at the surface can only be slow to moderate through the coming weekend (Feb 22-23) faster pattern change aloft is certain to intensify above average warmth already building across the Desert Southwest (70s-80s) near record levels next week (90s), and spread this warmth east of the Rockies in a progressively weaker form during the final week of February. In this scenario cold temperatures across the Plains from North Dakota to Texas reverse fastest and greatest, to double-digit above average anomalies by Monday, Feb 24. In contrast slower and far less intense warming across the Midwest and East (mainly 3°-6° above average) peaking later next week will also remain briefest and most prone to periodic moderate cooling into early March under lower heights aloft.  Progressively greater forecast divergence separates extended range models with GFS forecasts currently coldest deep into the Midwest and East especially by the 11-15 day period. Unlike prior extended range modeling which attempted to substantially warm all of Canada all latest runs shifted notably colder north of the border, adding confidence to ultimate establishment of a north-south temperature gradient across the U.S. typical of La Niña. Longer range forecasts of both the CFSv2 model and 30-day MWA ensemble are latching onto this as leading to a markedly colder 2nd week of March featuring winter-like temperatures (highs mainly in the 30s) across the Great Lakes, Ohio Valley, mid Atlantic region, and Northeast strongly suggesting significant delay to the start of spring across the northeast quadrant of the U.S.


If your business or career depends on correctly predicting the weather, you can follow the pack or you can get ahead with MWA’s proprietary models and expert forecasts.
  • Contact Us

    Melita Weather Associates, 138 Northeast
    Circle Durango CO 81301

    david@melitaweather.com
    direct 970.385.8695
    fax 970.385.8398