WEATHER FORECASTING
Weather forecasting is both a science and an art. Several methods are involved in the weather forecasting. The method used in a forecasting depends upon a number of factors like the amount of information available, the level of difficulty that the forecast situation presents and the experience of the forecaster.
Following are the methods involved in the weather forecasting:
Persistence Method/ Traditional Method
Persistence method is the simplest method of forecasting. This method assumes that tomorrow's weather will be the same as today's i.e. the conditions at the time of the forecast will not change. The persistence method works well when weather patterns change very little and features on the weather maps move very slowly. It works well in places where summer weather conditions vary little from day to day. But if weather conditions change considerably (are hyper dynamic) from day to day, the persistence method usually does not work and is not the best forecasting method to use for such places.
Numerical Weather Prediction/Computer
Numerical Weather Prediction (NWP) uses the power of computers to make a forecast. Complex computer programs, also known as forecast models, run on supercomputers and provide predictions on many atmospheric variables such as temperature, pressure, wind, and rainfall. A forecaster examines how the features predicted by the computer will interact to produce the day's weather. The NWP methods also have some flaw in them, as the equations used by the models to simulate the atmosphere are not precise. This leads to some error in the predictions. In addition, there are many gaps in the initial data, as one does not receive many weather observations from areas in the mountains or over the ocean. Despite these flaws, the NWP method is the best forecasting method for the day-to-day weather changes.
Trends Method
Another important techniques for making a forecast is through weather maps. The trends method involves determining the speed and direction of movement for air masses, fronts, pressure systems, and areas of clouds and precipitation. Using this information, the forecaster can predict where will be the above features in the future. Differences in latitude, possible acceleration/deceleration of storm systems, local effects such as topography, bodies of water, and the heat of island affect the forecasting, so these factors are taken into consideration.
Climatology
The Climatology Method is another simple way of predicting the forecast. The first thing to be known about the place of forecasting is its climatology (average high/low temperature or precipitation). This method involves averaging weather statistics accumulated over many years to make the forecast. If the forecast is for temperature and precipitation, then use the recorded weather data to compute the averages for temperature and precipitation. Climatology is rarely a correct forecast for a given day, as the forecast may vary high/low than the average . The climatology method only works well when the weather pattern is similar to that expected for the chosen time of year. If the pattern is quite unusual for the given time of year, the climatology method will fail.
Analog Method
The Analog Method involves examining today's weather scenario and comparing a day in the past when the weather scenario looked very similar (finding an analog). Weather prediction is than arrived at that, the weather will behave the same way as it did in the past for similar conditions. For example, suppose today is very warm, but rains are approaching in an area. A similar weather conditions can be examined from the past weather data for that particular area and depending upon the past experiences or data, forecast can be done. The analog method is difficult to use because it is virtually impossible to find a perfect analogy (day with similar weather conditions). Various weather features rarely line up in the same locations as they were in the previous time. Even small differences between the current time and the comparing past data can lead to very different results.
FORECASTING USING TEXT
The text gives an overview of the significant weather over the past 24 hours as well as a significant indication of weather occurring during the next 24 hours. The previous day climatic data summary for the city is examined which gives the precipitation, high and low temperatures. Maps are then plot over several hours to get the trend of the motion of air masses, weather systems. These maps are analyzed and weather for future is predicted. Forecasting by using text is difficult, as one has to forecast using only the weather text information available.
