One of the largest roles of a
Meteorologist is weather forecasting. Weather forecasts involve making predictions about the weather and atmospheric conditions at some future time. To make these predictions, meteorologists consider several pieces of important information, such as the current weather and atmospheric conditions, and both short-term and long-term trends. There are several different methods one can use when forecasting. Some of them are unique to weather forecasting, and others - like those found in
this book - can be used for forecasting any sort of future patterns or trends.
We will examine some of the most common forecasting methods unique to weather forecasting below:
Climatology
A climatology forecast is a forecast that the weather conditions will be representative of the climactic conditions for a given location. For example: If you were predicting the weather February 1st in Anchorage Alaska, you might predict that the mean daily temperature would be below freezing a high likelihood of snow. On the other hand, if you were predicting the weather for the same date in Australia, you might predict the main daily temperature to be much higher, These predictions would be based on the normal conditions for these regions during that time of year.
Climatology forecasts are more specific than this, however. It's common to take a 10 or 15 year average of the daily maximum, minimum, or mean temperatures for a given date, and to forecast this average as the predicted temperature. This type of climatology forecasts are excellent for predicting conditions on very long time scales.
The most famous source for climatology predictions are the
Old Farmer's Almanac and the
Farmers' Almanac, both of which make predictions for weather conditions and growing seasons many decades in advance.
Persistence
A persistence forecast is a forecast that the current weather conditions will remain exactly the same in the future.
Persistence forecasts are typically highly accurate on very short time scales. It's typically accurate to say, "It's raining now, so I can predict that it will be raining five minutes from now."
Persistence forecasts lose their accuracy, however, as the forecasting period increases. A similar statement of "It's raining now, so I can predict that it will be raining tomorrow" tends to be significantly less likely to be accurate. This forecasting method should not be used for forecasting periods of more than an hour or so.
Operational Forecasting
The accuracy of forecasts made by persistence can be extended for longer time periods by considering the current conditions and using them to predict future conditions based on known trends, rather than assuming conditions to remain exactly the same as present ones. For example: if there is a storm located to your east and your location has easterly winds, you can predict that your location will have stormy conditions in the coming hours or day. This type of prediction is the basis for operational forecasting.
Operational Weather Forecasting, sometimes also called Synoptic Forecasting involves examining the current weather conditions, thinking about physical and atmospheric phenomena, and using both of these sets of data to predict future weather.
The most widely-used source for obtaining this data is
NCAR Real-Time Weather Data, operated by the National Center for Atmospheric Research. This site provides satellite imagery, surface and upper air maps, and radar.
Dr John Gordon at the National Weather Service also maintains a fairly thorough listing of
relevant research papers in Operational Forecasting.
Numerical Modeling
Because making operational forecasts can often require looking-at and considering many separate atmospheric conditions, a computer can be used to speed-up the process. When a computer makes predictions for the future weather, based-on coded models of the physical atmospheric phenomena, and a set of current weather conditions; this is called Numerical Modeling.
The most widely-used numerical weather prediction system available for public use for making forecasts for anywhere in the US is
MOS (Model Output Statistics). An individual location of predictions in MOS would look something like this:
MOS for Albertville Airport in Albertivlle, AL.
The NCAR site, referenced earlier, also has a section called "
Forecasts," which displays the MOS predictions visually on a map and also allows you to plot specific data from MOS.
Verification and Forecast Accuracy
The last thing to consider when using any of these methods is that no forecast is ever perfect. The best one can hope for is a high degree of accuracy. There is an entire area of this discipline related to determining the accuracy of forecasters. The process of doing so is Forecast Verification. There are several methods for verification, just as there are several methods for forecasting, and many of them involve statistical analysis. Verification of some sort is performed at almost every weather forecasting entity - both public and private; although frequently the accuracy findings are not made publicly available.
To fulfill our tutoring mission of online education, our college homework help and online tutoring centers are standing by 24/7, ready to assist college students who need homework help with all aspects of weather forecasting. Our earth science tutors can help with all your projects, large or small, and we challenge you to find better online weather forecasting tutoring anywhere.
Read More