Documents
Resources
Learning Center
Upload
Plans & pricing Sign in
Sign Out

ppt

VIEWS: 28 PAGES: 64

									        Chapter 12

Weather Analysis and Forecasting
           A dangerous Job?
• Francisco Arias Olivera was a popular TV
  personality in a small Peruvian town (Sicuani)
  Population (21K). One day Francisco forecast a
  two-inch rainfall but instead they got 19 inches of
  rain. The local river flooded the town washing
  away 250 homes and killing 17. Outraged
  citizens stormed the TV station, and lynched
  Francisco. Six people charged with his murder
  were released after pleading justifiable
  homicide. Rumor has it that the TV station still
  has not hired another Weather Forecaster.
      Why make a forecast?
• Forecasts are issued to save lives, save
  property and crops and to tell us what to
  expect in our atmospheric environment.
• Who among us does not look at the
  weather at some time or other during the
  week?
              Science or Art?
• Weather forecasters are responsible for predicting
  weather accurately so thousands (if not millions) of
  people will know what to expect in the weather.
• But weather forecasting is not an exact science…every
  now and then we fail to get it right. There is hard
  science behind every weather forecast: primitive
  equations, computer models, calibrated equipment etc.
  However, there is an art in making a forecast,
  interpreting the current and forecast conditions. The
  atmosphere is not a “Closed” system. We cannot track
  every short wave or minor perturbation in the
  atmosphere.
Acquisition of Weather Information



• Over 130 nations
• Over 10,000 land sites and hundreds of ship
  observations report each day
• All in a common format
• WMO is responsible for the international
  exchange of weather data and certifies that
  observation procedures do not vary among
  nations
• Weather info from all over
  the world is transmitted
  electronically to NCEP
  located in Camp Springs
  Maryland.
• NCEP relays the weather
  info to various public and
  private agencies for use
  in preparing local,
  regional and global
  forecasts.
                               NCEP
•   Comprised of 9 centers which provide a wide variety of national and
    international weather guidance products to National Weather Service
    field offices, government agencies, emergency managers, private
    sector meteorologists, and meteorological organizations and societies
    throughout the world. NCEP is a critical national resource in national
    and global weather prediction. NCEP is the starting point for nearly all
    weather forecasts in the United States.

• Office of the Director
  Aviation Weather Center
  Climate Prediction Center
  Environmental Modeling Center
  Hydrometeorological Prediction Center
  NCEP Central Operations
  Ocean Prediction Center
  Space Environment Center
  Storm Prediction Center
  Tropical Prediction Center
Climate Prediction Center
      The Modern Forecaster
• Has access to hundreds of maps, charts, vertical
  atmospheric profiles (soundings), satellite
  images, Doppler Radar, etc.
• Many radio and TV stations hire private
  meteorologists or meteorological companies to
  make their own forecasts with the aid of NCEP
  products.
• Some stations use untrained announcers to
  simply read forecasts from the National Weather
  Service or private companies (Accu Weather)
Forecasting Methods and Tools
• In the 50s many short ranged forecasts where
  made by moving the existing systems along at a
  steady rate (up two over one)
• In late 70s and into the mid 80s many weather
  maps and charts were still plotted and analyzed
  by hand
• Modern Computers and present observing
  techniques make today’s forecasts much easier
  and better than those of the past.
 Numerical Weather Prediction
• Each day many thousands of observations
  transmitted to NCEP are fed into high speed
  computers that are used to prepare Numerical
  Weather Predictions.
• Computers analyze data and use it to predict the
  weather.
• Routine daily forecasting of weather by the
  computer has come to be known as Numerical
  Weather Prediction
• Atmospheric models – simulation of the
  atmosphere’s behavior by mathematical
  equations
          SOME HISTORY
• 1755 – Euler develops first equation of fluid
  mechanics.
• 1827 and 1845 – Navier-Stokes add
  molecular viscosity terms to fluid equations.
• 1904 – Bjerknes proposes a paradigm shift
  from empirical forecasts to ones applying
  basic laws of physics.
• 1922 – L. F. Richardson calculates using
  crude method first numerical forecast using
  the primitive equations (PE).
  – 6 weeks to crank out a 6 hour forecast
  – Pressure calc an order of magnitude in error
Richardson’s vision and today




                 UK MET Office NEC
                 supercomputer
                                  L.F. Richardson, 1922

 “Imagine a large hall like a theatre, except that the circles and the galleries go right
  round through the space usually occupied by the stage. The walls of this chamber
     are painted to form a map of the globe. The ceiling represents the north polar
      regions, the tropics in the upper circle, and the antarctic in the pit. A myriad
 computers are at work upon the weather of the part of the map where each sits, but
each computer attends only to one equation or part of an equation. The work of each
    region is co-ordinated by an official of higher rank. Numerous little “night signs”
   display the instantaneous values so that neighbouring computers can read them.
From the floor of the pit a tall pillar rises to half the height of the hall. It carries a large
pulpit on its top. In this sits the man in charge of the whole theatre. One of his duties
is to maintain a uniform speed of progress in all parts of the globe. In this respect he
   is like the conductor of an orchestra in which the instruments are slide-rules and
  calculating machines. But instead of waving a baton he turns a beam of rosy light
   upon any region that is running ahead of the rest, and a beam of blue light upon
                                 those who are behindhand.

Four senior clerks at the central pulpit are collecting the future weather as its is being
 computed, and despatching it by pneumatic carrier to a quiet room. There it will be
               coded and telephoned to the radio transmitting stations”
          THE COMPUTER AGE
• Computers arrive in 1940’s
• Von Neumann, Charney, Rossby, Eliassen, and Platzman attempt to
  use computer resources to solve PE.
    – Realized need to simplify due to scarce computer resources.
• 1950 – first successful numerical forecast
    – Took 24 hours to crank out a 24 hour forecast
• 1952 – Computer technology advances cut 24 hour forecast time to
  5 minutes
• 1954 – Services for JNWPU. Attempt baroclinic model run and not
  successful (no skill over empirical methods)
• 1963 – First successful baroclinic model 6-level PE.
• 1976 – Limited Fine Mesh model started up. Some skill, phased out
  in 1994.
             THE COMPUTER
              REVOLUTION
• 1980 – 12-level Global Spectral Model
  – Consists of 2 parts
     • Aviation – forecasts in support of aircraft out to 5 days
     • MRF – Medium Range Forecast out to 15 days
         – 14 minutes runtime/day = 3.5 hours for 15 day forecast
• 1990 – Nested Grid Model (NGM) introduced
• Early 90’s – ETA Model introduced
  – 32 km resolution and 45 layers
• 1999 – Meso-ETA introduced
  – 10 km horiz grid, 60 layers
  Numerical Weather Prediction
• How is a forecast made today?
          The forecast process
• 1. Gather observations for the globe to define the
  “current state” of the atmosphere
     Collect observations
     Perform quality control
• 2. Use these observations in a model that describes how
  the the atmosphere changes with time
     Data assimilation
• 3. Take this as the “current” state of the atmosphere and
  run the same model into the future
     Stop after 24, 48, 72 hours and interpret weather
     forecast!
The Forecast Process
     Users of weather forecasts
• Media- Television, radio,       • Entertainment- outdoor
  newspapers, internet              events
   – Widely use in public and        – Winds, rain (Hogmanay)
     private sectors
                                  • Agriculture & Fishing
• Transport
   – Airlines- fog, visibility,      – Frost, rainfall; winds,
     severe weather                    visibility
   – Ships- winds, storms,        • Recreation- mountaineering,
     visibility                     sailing
   – Railways- winds, flooding       – Winds, rain, cloud level;
   – Roads- snow (gritting),           winds visibility
     flooding
                                  • Military
• Energy
                                     – Visibility, winds
   – Heat waves/cold spells
     when energy demand levels    • Construction
     change                          – Rainfall, winds
The forecast cycle over 24 hours
                       00
Data gathering

Quality control

                  18          06
Assimilation

Forecast


                       12
Observational data gathering
Observation types-surface data:
   Temperature, humidity, pressure, wind
       Observation types-
         radiosondes:
wind, temperature, humidity at vertical levels
Observation types-aircraft:
 temperature and wind along route
Observation types-satellite:
   geostationary images: winds
Observation types-satellite:
 polar orbiting: temperature profiles
   Observations- important points
• In a 6-hour period receive >100,000 observations
  of:
  – Temperature, wind speed and direction,surface pressure,
    humidity
• For some data types transformation performed
  – e.g. radiances  temperatures
• Data types have different geographical coverage,
  vertical structure and temporal distribution
  – Surface observations are sparser coverage in SH
  – Only certain observations types give vertical profiles
  – Surface stations report at ~6 hourly intervals. Satellite data
    are continuous
  Quality control- 1.check data
• Check all data have reasonable values
  – No 100oC temperatures or negative winds
• Check values are consistent
  – T >= Td
• Check displacement of ships and buoys
  – e.g.-ship data over land
• Check for agreement with neighbours:
  – Compare with adjacent stations-buddy check
  – Performed relative to background forecast- to account
    for different weather systems at two locations
 Reject bad data
 Quality control –2. assign errors
• Assign errors to data retained
• Calculate observation error which
  depends on data type/instrument
• Calculate background error= error at a
  given location dependent on synoptic
  situation (fast or slow moving systems)
  and data coverage
• Observational error and background
  error combined  error estimate
          Data assimilation

• Observations and their error
  estimates are “assimilated” into
  the model:
  1. Interpolate observations onto
     model horizontal and vertical grid
  2. Combine latest observations with
     previous=background forecast
  3. Perform adjustments
              1. Interpolation
• Incorporate irregular observations into model grid
• Model grid has many shells with regular horizontal grid in
  each shell
OPTIMUM INTERPOLATION
                                                                          X
  X                                                         X
                                                X


Each ob must be corrected for:
Time
Location
                                                        X
Quality
Compared to those around it and to previous forecasts       O = Model Grid Points
This is done for over 6 million obs per day
                                                            X = Observations
2. Combine new observations with
       previous forecast
• adjusts the model background field -the forecast
  from the previous model run- towards the new
  data received from observations
• Include observational errors to determine how
  reliable these new data are
• Process is very complex (adjustments often
  needed) and known as variational analysis
• Data assimilation can take 30% of the
  computational effort
Data assimilation- schematic
             3. Adjustments
• Observations and background forecast are not
  always consistent:
  – pressure gradient from background forecast may imply
    a different wind from that observed
  – horizontal temperature gradients from background
    forecast may imply a different thermal wind (wind
    shear)
• Failure to deal with these discrepancies causes
  models to fail catastrophically
• An adjustment process is used to iteratively (over
  short time-steps) reduce discrepancies to a
  minimum
  An “analysis” of the current
   state of the atmosphere
• We now have the best possible estimate of
  the current state of the atmosphere on a
  regular grid over the whole world.

• This is called an analysis

• The process of making the forecast can
  now begin
                    ECMWF
European Center for Medium Range Weather Forecast
                 Reading England
Vertical coordinate- All models use discrete
vertical levels, just like horizontal grid points.
The obvious choice is for a vertical coordinate,
height or pressure, is not in common use because
it is numerically difficult to cope with a sloping
lower boundary. Instead, modelers generally use
some variation of sigma coordinates (name after
the greek letter), with the lowest level following
the Earth’s surface and higher model levels
defined as being some fraction of a distance (in
pressure) from the ground to the top of the
atmosphere.
       Pronostic Chart (PROG)
• NWP provides a series of
  charts or numerical
  output for a specified
  period, 12, 24, 36, 48, 72
  hours etc out to 7 to 10
  days in some cases
  longer 16-30 days.
• The final forecast chart
  representing the
  atmosphere for a
  specified time in the
  future is called a PROG
Two surface pressure and precipitation progs for 2000 EST, September 29, 2003 –
48 hours into the future. Prog on left is Navy Operational Global Atmospheric
Prediction System (NOGAPS) model, whereas prog on right is the Global Forecast
System (GFS) model from NCEP
  Why Forecasts go Wrong@#!!
• Flaws in computer models limit the accuracy of the
  forecast
• Models idealize the atmosphere, making certain
  assumptions that may be correct in some instances, but
  wrong in others.
• Data sparse regions of the world leave gaps in our
  observations
• Models cannot accurately interpret many factors that
  influence weather, may not handle terrain features well,
  may not have the resolution to “get the local weather
  right”
• Chaos Theory – a butterfly flaps his wings in a rainforest
  in South America and your forecast in Asheville goes to
  pot. Small disturbances in the atmosphere, not picked up
  by the model, can ruin your forecast
Tools for Forecasting the
        Weather
      Advanced Weather Interactive Processing System - AWIPS




Can process DOPPLER, ASOS, ingest models, etc. Handles and displays obs
and weather maps.
           WSR-88 Doppler Radar
•   Doppler radar data from
    Melbourne, Florida, on March 25,
    1992, during a severe hailstorm
    that caused $60 million in
    damages in the Orlando area. In
    the table near the top of the
    display, the hail algorithm
    determined that there was 100
    percent probability that the storm
    was producing hail and severe
    hail. The algorithm also estimated
    the maximum size of the
    hailstones to be greater than 3
    inches. A forecaster can project
    the movement of the storm and
    adequately warn those areas in
    the immediate path of severe
    weather.
NEXRAD
 WSR-88D
Doppler Radar
Profiler winds for California
              Upper air sounding
• A sounding of air temperature,
  dew point, and winds at
  Pittsburgh, PA, on January 14,
  1999. Looking at this
  sounding, a forecaster would
  see that saturated air extends
  up to about 820 mb. The
  forecaster would also observe
  that below-freezing
  temperatures only exist in a
  shallow layer near the surface
  and that the freezing rain
  presently falling over the
  Pittsburgh area would continue
  or possibly change to rain, as
  cold easterly surface winds are
  swinging around to warmer
  southwesterly winds aloft
Major Types of Satellites
Geostationary               Polar Orbiting




                Very important for covering vast areas of
                globe. 70% of earth is water…very few
                surface observations over the oceans
                    GOES
    (Geostationary Operational Environmental
                    Satellite)

• Orbital period exactly
  matches the rotation
  of the earth
• Altitude of 35,800 km
• Appears to "hover"
  over one spot on the
  Earth's equator.
          Global GOES Coverage
GOES-10
GOES-WEST   GOES-CENTRAL   GOES-EAST
(GOES-10)   (GOES-11)      (GOES-8)
GOES Images
POES
  • Altitudes usually
    range from 700 to 800
    km
  • Satellites in this
    category include;
    NOAA, DMSP,
    Landsat, and SPOT
  • Slightly more than 14
    orbits in a single day.
POES ORBIT COVERAGE
        Meteorological Uses
• Cloud cover, types, obstructions to
  visibility
• Tropical Cyclones
• Low pressure systems at sea
• Winds velocity
• Rain rates
• Sea heights
    Other Forecasting Methods
• Persistence Forecast – easiest way to make a forecast.
  Simply predict future weather will be the same as
  present weather.
• Steady-state or trend Forecast – uses the principle
  that surface weather systems tend to move in the same
  direction and at approximately the same speed as they
  have been moving. Today using this method for very
  short duration forecasts is called “nowcasting”.
• Analogue Method – relies on the fact that the existing
  features on a weather chart (or series of charts) may
  strongly resemble features that produced certain
  forecast conditions in the past. “I’ve seen this situation
  before, and this is what happened”
   Other Forecasting Methods
• Ensemble forecasting – approach based
  on running several forecast models – or
  different versions (simulations) of a single
  model – each beginning with slightly
  different weather information to reflect
  errors inherent in the measurements. If at
  the end of a specified time, the models
  match each other fairly well, the forecaster
  can issue a forecast with high
  confidence…if they don’t match…well…
    Other Forecasting Methods
• Climatological forecast – a forecast based on
  the climatology (average weather) of a particular
  region. I.e. It rarely ever rains in Los Angeles in
  July and August. Stats from all past years
  indicated it is not prudent to forecast much rain
  in these months.
• Probability forecast – a forecast of the
  probability of occurrence of one or more of a
  mutually exclusive set of weather conditions.
          Probability Forecast
• Probability of a "White
  Christmas" - one inch
  or more of snow on
  the ground - based on
  a 30-year average.
  The probabilities do
  not include the
  mountainous areas in
  the western United
  States.
Accuracy and skill in forecasting
• 12-24 hour forecasts are quite accurate
• 1-3 day forecasts are fairly good
• 7-10 days are getting better but still often fall into the
  “smoke and mirrors” category.
• What qualifies as a right or wrong forecast? If I forecast
  65oF for a max temp tomorrow and we get 66oF – did I
  blow it??
• We say a forecast shows skill when it is more accurate
  than a forecast based only on persistence or climatology.
• Today we often accurately predict large scale weather
  events many days in advance (major snow storms,
  blizzards, hurricane strikes)
               Book of Signs
• Theophrastus, student of Aristotle, 300 BC
  compiled all sorts of weather indicators in his
  “Book of Signs”
• Ways to foretell the weather by examining
  natural signs: shape of clouds, intensity at which
  a fly bites.
• Our own folklore: “halo around the moon
  portends rain”
• Do you have some knowledge or experience
  that can help foretell weather in mid latitude?
  Weather forecasting using surface charts




• Surface weather map for 6:00 A.M. Tuesday. Dashed lines indicate
  positions of weather features six hours ago. Areas shaded green are
  receiving precipitation.
   Determining the Movement of
  Weather Systems (Rules of Thumb)
• For short intervals, storms and fronts tend to move in the
  same direction and at the approx. same speed as they
  did during the previous 6 hours – unless there is reason
  to indicate otherwise.
• Lows tend to move in the direction that parallels the
  isobars in the warm air ahead of the cold front
• Lows tend to move toward the region of greatest surface
  pressure drop, highs tend to move toward the region of
  greatest surface pressure rise.
• Surface pressure systems tend to move in the same
  direction as the wind at 5500 m – the 500 mb level. The
  speed at which the surface systems move is about ½ the
  speed of the winds at this level.

								
To top