Weather Forecasting Study Notes

CH 13 - WEATHER FORECASTING

Definition of Weather Forecasting

  • Weather forecasting is the process of predicting how the current state of the atmosphere will change over a defined period of time.

Time Scales in Weather Forecasting

  • Nowcasting: Present - 6 hours
  • Short Range: 1-3 days
  • Medium Range: 3-5 days
  • Long Range: 5-10 days
  • Extended: Up to 90 days

Importance of Weather Forecasting

  • Influences decisions on:
    • Clothing Choices: People dress according to the weather forecast.
    • Planning Activities: Outdoor events, travel plans, agricultural tasks.
    • Agricultural Information: Farmers rely on forecasts for planting and harvesting schedules.

Forecasting Techniques

Folklore Forecasting
  • Uses traditional sayings, some reliable, some not. Examples include:
    • "Red sky at night, sailor's delight" - Reliable saying indicating favorable weather.
    • Groundhog Day - Unreliable prediction method.
Persistence Forecasting
  • Assumes that weather patterns will remain consistent.
    • Definition: What happened yesterday will likely happen today.
    • Limitations:
    • Accuracy may decline with frequent weather changes.
    • Overall effectiveness is location-dependent; works best in tropics, summer, and winter.
    • Stays accurate during periods of blocked weather patterns.
Climatology Forecasting
  • Utilizes long-term averages of weather conditions.
    • Limitations:
    • Ineffective during record-setting weather events.
    • May provide acceptable accuracy but could be improved.
Trend Method
  • Assumes the speed and direction of weather systems won’t change.
    • Uses short-term trends from nearby observations of temperature or precipitation.
    • Example: If a cold front drops temperatures by 20 degrees in Denver, a similar effect is expected when it reaches another location.
    • Limitations: Effectiveness decreases over time.
Analogue Method
  • Involves matching current weather maps to those from the past.
    • Based on pattern matching.
    • Effectiveness relies on the forecaster’s experience; usually applies well to severe and winter weather events.
Numerical Method

  • Employs numerical equations to predict the future state of weather variables such as:

    • Pressure
    • Temperature
    • Winds
    • Humidity
    • Clouds
    • Precipitation

  • Utilizes sophisticated programs on supercomputers for calculations.

Numerical Methods

  • Each model integrates a set of basic equations that describe the atmosphere's dynamics based on the Second Law of Motion:
    • Accounts for:
    • Pressure variations
    • Gravity
    • Friction
    • Heat and moisture transfer
    • Phase changes in water
Model Grids
  • Present data enters the computer model; since the atmosphere is 3D, numerical weather prediction (NWP) requires data from various atmospheric layers.
    • Calculations Example:
    • A typical forecast model might include 700 x 800 grid points per level and may extend to around 60 atmospheric levels.
    • Each level necessitates solving for approximately 210 variables, resulting in:
      • 336 million data points for one time frame
      • Continuous calculations over 48 hours for a single location.
Different Weather Models
  • Various models cater to distinct atmospheric scenarios:
    • NAM: North American Mesoscale
    • OFS: Global Forecast System
    • ECMWF: European Model
    • SREF: Short-Range Ensemble Forecast
    • HRRR: High-Resolution Rapid Refresh Model
    • Wef: Weather Research & Forecast Model

Reliability of Numerical Models

  • A successful numerical model forecast hinges on:
    • Providing accurate initial conditions for the computer program; hence, it is only as good as the initial observations.
    • The adage "garbage in, garbage out" applies, emphasizing data quality.
Spaghetti Plots
  • Each weather model produces unique forecasts illustrating possible atmospheric states for specified time frames.
Ensemble Forecast
  • Involves multiple iterations of a single model, each adjusted for different scenarios or equations.
National Blend of Models (NBM)
  • Provides a comprehensive forecast blending numerical models into various formats including charts, graphs, and probabilities.

Forecasting Process

  • The overall forecasting workflow includes:

    • Observations:

    • Global data collection of weather conditions.

    • Analysis of current weather:

      • What is happening now?
      • What weather systems are situated to the west and northwest?

    • Analysis:

    • Identification of storm systems, analyzing cold and warm fronts and high and low-pressure areas.

    • Questions posed regarding current conditions and progression of storms.

    • Prediction:

    • Determining which model provides the best representation of the current situation based on historical performances.

    • Post Processing:

    • Evaluate forecasting accuracy, aiming for an acceptable range of prediction accuracy.

    • A benchmark of being 80% accurate or higher indicates proficiency in weather forecasting.

Weather Maps & Station Models

  • Station Model Plots:
    • Represent a cluster of current weather data from a specific weather reporting site.