Meteorology Numerical Weather Prediction


The calculation of weather data

What is the weather going to be like tomorrow?

For a long time, people have tried to predict weather conditions using the hydrologic climate cycle.In the early 1920’s scientists were able to compile a six hour forecast, back then it took six weeks to analyse weather data collected at only two points in Europe and calculate, by hand, a useful illustrative model. Today, supercomputers are used to predict the weather for a period of several weeks. The complex modelling programs require several million data points for parameters such as temperature, humidity, pressure, vertical & horizontal wind velocity with time stamps and absolute coordinates. To create a correlation between the data and the environment, scientists “slice” the atmosphere virtually into smaller horizontal &vertical parts—this process is called discretisation. It is more useful to compute the chronological change of the parameters using this model.

watercycle-pageClockwise from top left: Map of the average temperature over 30 years. .Weather station on Mount Vesuvius. .Water cycle summary

Meteorological events that are too “small” such as a single thunderhead, layer clouds or smaller turbulence’s will be parameterised through variables. This parameterisation is a science of its own that aims to reduce uncertainties as best as possible. Every forecast calculation starts with the current weather conditions. The quality of this input is crucial for the accuracy of the final forecast. Meteorologists link the forecast of yesterday’s weather with the actual measured parameters. Only large data centres are capable of computing this data assimilation. The overall result is a best possible calculation basis predict the weather for the next day. If this groundwork is flawed the forecast may be incorrect. For example, it could report rain at the wrong location. Today’s meteorological mathematicians also take parameters into account that change extremely slowly compared to the other factors. Growth and the reduction of polar ice, or the temperature of the oceans are summarised as boundary values. After a model is run using all the available data, meteorologists process and customise reports for a wide range of target groups such as public authorities, flight control centres, energy producers, industries and many more. These reports also include specific weather warnings.

Why the need to measure humidity?

Atmosphere_composition_diagramAtmosphere composition diagram

As described above, the daily weather forecast relies on the precise measurement of weather parameters. The science of numerical weather prediction aims to describe the daily hydrologic cycle in numbers. Humidity plays an  important role. Typically, data errors will multiply during calculations. Humidity values influence weather calculations e.g. through the water vapour balance equation— this formula expresses the influence of humidity through rain & condensation, and vice versa. Incorrect measurement or incomplete humidity data directly leads to wrong predictions of a huge number of weather phenomena such as the condensation altitude of clouds, locations of hyetal regions, fog layers and storms. In 1999, incorrect data sent by a weather station in Nova Scotia, Canada led to a incorrect forecast for Hurricane Lothar two days before it hit Central Europe. Authorities were insufficiently prepared to alert people in time. The prediction of rain and snowfall is still challenging for meteorologists. Only more extensive networks of weather stations and enhanced mathematical models will reduce problems due to unknown factors.

Facts & Figures

  • 7 inches is the diameter of the largest hailstone ever recorded.
  • Sukkur City in Pakistan is one of the most humid places in the world with
    30 °C dew point & a felt air temperature of 65 °C.
  • A study showed that a small thunderstorm system holds more than 10 million tons of water.
  • No two weather patterns are completely alike.
  • Some weather models assimilate data obtained from more than 25,000 weather stations.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s