Wintertime Temperature Forecasting - PowerPoint PPT Presentation

1 / 26
About This Presentation
Title:

Wintertime Temperature Forecasting

Description:

Equations tend to account for local effects and mesoscale weather features fairly well ... .edu/weather) Nocturnal Wind Surges. A local example: Tallahassee ... – PowerPoint PPT presentation

Number of Views:44
Avg rating:3.0/5.0
Slides: 27
Provided by: Clark82
Category:

less

Transcript and Presenter's Notes

Title: Wintertime Temperature Forecasting


1
Wintertime Temperature Forecasting
  • Current Weather Discussion
  • 13 February 2008

2
General Temperature Forecasting
  • Various approaches can be taken to general
    temperature forecasting
  • Climatology long-term averages
  • Persistence what has happened over the past few
    days/hours
  • Upstream observations
  • Experience local knowledge
  • Model guidance raw, model output statistics
    (MOS), other statistical formulas
  • The best forecasters use a combination of all of
    these factors and possibly more when making
    their temperature forecasts.

3
Climatology
  • Many climate stations have long-term average and
    record temperatures readily available.
  • Average 30 year mean (currently 1970-2000) give
    an idea of the normally expected values
  • Records as long as the station has existed give
    an idea for the potential spread
  • Example Tallahassee
  • 2/13 normals high 68, low 41
  • 2/13 records high 84, low -2

4
Persistence
  • In times where air masses are largely unchanging
    (advection small, modification small),
    persistence may be used as a first-guess forecast
  • Simple definition todays temperatures will be
    the same as yesterdays values
  • Example Tallahassee, 2/13/08
  • 2/12/08 high 71, low 51
  • Perhaps a first guess temperature forecast may be
    the same as yesterdays values

5
Climatology / Persistence
  • Where will climatology and persistence fail?
  • Climatology where any significant departures
    from normal take place
  • Truthfully, pretty much any weather-related event
  • Persistence where any significant air mass
    modification or change takes place
  • Cold fronts, oceanic influences, etc.

6
Upstream Observations
  • Best when air masses arent modifying but are
    advecting
  • Can work for both lows and highs
  • Typical length/time scales to look at distance
    wind can influence in one day
  • Improves upon persistence by taking into account
    air mass movement
  • Can inherently account for common meteorological
    events
  • Mixing strength/magnitude
  • Temperature/moisture advection

(http//www.rap.ucar.edu/weather)
7
Upstream Observations
  • Not just limited to surface observations
  • What is going on in the boundary layer (e.g. 850
    hPa temperatures)?
  • Also not just limited to temperatures themselves
  • Satellites cloud cover
  • Dewpoint temperature moisture
  • Soundings mixing, vertical wind profiles

(http//www.rap.ucar.edu/weather)
8
Model Guidance
  • Two methods raw output and model output
    statistics (MOS)
  • Raw output taking the model temperature grids at
    face value
  • Will not account for local effects (e.g. urban
    heat island, soil character)
  • Will tend to under-represent mixing and other
    mesoscale weather influences
  • Model output statistics using developed and
    tested equations to obtain temperature forecasts
  • Equations tend to account for local effects and
    mesoscale weather features fairly well
  • Only a mathematical technique, however! Will not
    be perfect!

9
Experience and Advanced Tools
  • The best way to get a handle for general
    temperature forecasting is to practice it and
    learn from your good and bad forecasts.
  • Over time, youll get a grasp on some of the more
    advanced tools and tricks that you can use to
    forecast temperatures
  • Two examples one for high, one for low
    temperatures
  • High mixing from 850-925 hPa to the surface (dry
    adiabatic lapse rate temperature profile)
  • Low under calm/clear conditions, temperature
    will fall to near or below the midday dewpoint
    temperature

10
Mixing Example
  • 2/13/08 0000 UTC Dallas, TX
  • Note mixed layer from 900 hPa to the surface
  • Accompanied by dry adiabatic lapse rate
  • If you use early morning 850 or 925 hPa
    temperatures, you can get an idea for the maximum
    possible temperature!
  • Height to use dependent upon season, sky cover,
    etc.
  • Amount to add to upper level temperature
    dependent upon surface pressure (not reduced to
    sea level)
  • Has limitations
  • Only implicitly accounts for advective processes
  • Not always able to determine maximum mixing
    height

(http//www.spc.noaa.gov/exper/soundings)
11
Why Is This Important?
  • Specifically, what is so special about wintertime
    temperature forecasting?
  • Heavily influenced by mesoscale features that are
    not as prevalent in warmer seasons!
  • Snow cover impacts
  • Cloud cover impacts
  • Topographical effects
  • Nocturnal wind surges

12
Snow Cover Effects
  • Snow is an excellent radiator of heat energy
  • Albedo of 0.8-0.9
  • Compare typically 0.3
  • Results in influences on maximum and mimimum
    temperatures!
  • More pronounced at night, particularly under
    clear and calm conditions
  • Makes nighttime temperatures especially tough to
    predict
  • Model guidance often busts under those clear,
    calm nighttime scenarios

(http//www.rap.ucar.edu/weather)
13
Snow Cover Effects
  • Surface Map 1100 UTC 15 January 2008, N. Plains
  • Where might snow cover be impacting low
    temperatures?
  • A similar effect later that day can be shown as
    well, but is not as pronounced.

(http//www.rap.ucar.edu/weather)
14
Snow Cover Rules of Thumb
  • First, gauge cloud cover and winds
  • Is any cloud cover moving in for the night?
  • How windy does it look to be overnight?
  • Secondly, What happened under similar conditions
    last night, either here or upstream?
  • Finally, what are the midday dewpoint temperature
    and model guidance (MOS) forecast lows?
  • If it is projected to be calm and clear, lean on
    persistence and undercut the MOS forecasts and
    midday dewpoint temperatures.
  • Otherwise, model guidance might do reasonably
    well. Keep it and what happened past nights under
    consideration.

15
Cloud Cover Effects
  • Clouds are an excellent insulator of energy
  • Warmer temperatures at night (less outgoing
    longwave radiation)
  • Cooler temperatures in the day (less incoming
    solar radiation)
  • Level(s) and coverage of clouds important
  • Thinner, higher clouds (e.g. cirrus) less
    impacts
  • Thicker, lower clouds (e.g. stratus) more impacts

(http//www.rap.ucar.edu/weather)
16
Cloud Cover Effects
  • At upper left infrared satellite image from 1000
    UTC 14 January 2008 across the Ohio Valley
  • At lower left surface observations across the
    same region at the same time
  • Note the difference between Indiana and Illinois
  • Indiana more clouds, warmer temperatures (upper
    20s)
  • Illinois less cloudy, cooler temperatures
    (low-mid 20s)

(http//www.rap.ucar.edu/weather)
17
Cloud Cover Rules of Thumb
  • First, use satellite (visible and infrared) loops
    to determine current cloud cover and projected
    cloud cover movement
  • Secondly, use satellite loops and surface sky
    cover observations to determine current and
    projected cloud types
  • Thirdly, use model output soundings and
    isobaric analyses to determine where cloud
    cover might develop
  • Particularly at upper levels (200-400 hPa) and
    near the surface
  • Finally, consider persistence, model guidance,
    and upstream observations. Temper your maximum
    temperature forecasts on the low side and your
    minimum temperature forecasts on the high side.

18
Topographical Effects
  • Primarily manifest in three ways
  • Chinook-like downsloping winds
  • Localized mountain/valley flows
  • Larger-scale wedging events
  • Effects are seen with each with both minimum and
    maximum temperatures
  • All are influenced by snow and cloud cover
    effects as well, however

(http//fermi.jhuapl.edu/states)
19
Downsloping Winds
  • Found on the downwind side of mountain ranges
  • Descent is a dry adiabatic (compressional)
    process
  • Parcels warm with the dry adiabatic lapse rate
  • The end result is often very warm, dry,
    occasionally windy conditions
  • Note the warm tongue over Montana and Alberta

(http//www.nco.ncep.noaa.gov)
20
Mountain/Valley Flows
  • Localized features specific to the terrain of the
    region
  • For one, valleys often tend to see colder
    temperatures at night than higher elevations!
  • Cold air drainage
  • Calm winds due to boundary layer decoupling
  • Best methods to forecast under these situations
    experience and persistence

(http//www.whistlerweather.org)
21
Wedging Events
  • Occur on the east side of significant mountain
    ranges
  • Most prevalent along the east side of the
    Appalachians
  • Features a surface high pressure system along the
    mountain range
  • Flow along/east of the range is north to
    northeast
  • Abuts the mountain range but cannot go up and
    over
  • Thus, air mass is forced to spill southward along
    the mountain range

(http//www.nco.ncep.noaa.gov)
22
Wedging Events
  • These events result in cooler, drier air
    funneling well to the south
  • Often extends all the way to the Gulf Coast
  • Models tend to underdo the coolness/dryness of
    the air
  • These features are of critical importance for
    frozen precipitation!
  • Southeast located over central Appalachians
  • Northeast blocked over northern Appalachians

H
(http//www.rap.ucar.edu/weather)
23
Wedge Words of Wisdom
  • Do not use model guidance verbatim for
    temperatures, cloud cover, or winds!
  • Instead, use their depictions of the large-scale
    flow, particularly at the surface with the
    surface ridge.
  • Nine times out of ten, as long as the ridge
    remains in place and it tends to remain in
    place the cooler, drier air will hold on longer
    than model forecasts.
  • This leads to great bust potential with your
    forecasts!
  • Events that dont follow this tend to be ones
    with very intense forcing for an area of low
    pressure in the Gulf of Mexico
  • Isentropic upglide atop the wedge will often lead
    to light-moderate precipitation occurring in the
    cooler, drier air
  • This helps to reinforce the wedge via evaporative
    cooling

24
Nocturnal Wind Surges
  • Primarily driven by two sources
  • Synoptic-scale forcing (low level jet) ahead of
    an area of low pressure
  • Inertial instabilities, often south of a surface
    ridge
  • Both wind surges are primarily focused above the
    surface (1000 ft-850 hPa)
  • Accompanying mixing helps extend their influences
    to the surface
  • This results in warmer nighttime temperatures as
    boundary layer cannot decouple

(http//www.rap.ucar.edu/weather)
25
Nocturnal Wind Surges
  • A local example Tallahassee-area surges
  • Occur when a surface ridge sets up to our north
    to northeast
  • An easterly jet develops 1000-2000 ft above the
    surface
  • Wind speeds within the nocturnal jet 15-35 kt
  • 20 kt seems to be the threshold for surface
    impacts
  • 20 kt winds 3-5 mph surface winds, keeping
    temperatures up
  • Best way to monitor radar wind profiler data

(http//weather.cod.edu/analysis)
26
Summary
  • There are many ways that you, as a forecaster,
    can improve upon model guidance.
  • Knowledge of situations where model guidance
    should be treated with caution gives you an upper
    hand.
  • The best way to learn from these examples is from
    practice, practice, practice.
  • Even the best forecasters bust on many of these
    situations
  • But, they wont bust twice on the same situation
    they learn from their mistakes through practice
    and analysis
Write a Comment
User Comments (0)
About PowerShow.com