theweatherprediction.com
[--MAIN HOME--] [--ALL HABYHINTS--] [--FACEBOOK PAGE--]

BEING YOUR OWN FORECASTER

METEOROLOGIST JEFF HABY

Learning to become a skilled forecaster does not occur over night nor do two forecasters predict weather with the same method. Just as a thunderstorm develops, learning to forecast is an ongoing process. The best forecasters use five allies to gradually improve their forecasting skill. They are time, concentration, a solid knowledge base, questioning and communication skills.

Forecasting knowledge and accuracy will not improve unless time is spent EACH DAY studying relevant meteorological data. Uninformed reading of MOS data and "copying" NWS forecasts will not develop forecasting skill since a person is depending on someone else's forecast. In this case it would be NWS employees and computers. It is best to use MOS and the NWS as forecasting TOOLS, but they should not be the sole basis of a forecast.

Time should be spent looking at meteorological data BEFORE reading NWS or MOS data. MOS data and NWS data should be examined AFTER you have come to your own conclusions of what weather to expect. The forecast can then be adjusted if a trend in the MOS data or forecast information covered by the NWS differs from your initial "trial" forecast.

Time spent looking at data will vary with the type of weather in the area and the forecast responsibility incumbent on a person. Some of the most important data to study before formulating the forecast include: (1)Analysis charts for each level in the atmosphere, (2)Thermodynamic soundings nearest the forecast region, (3)Comparing model initiation to the analysis charts, (4)4panel NGM/NAM/GFS and other model data, and (5)current synoptic scale surface charts (temperature, dewpoint, pressure, wind, frontal position, and so forth). It should take at least 20 minutes to examine these data.

Concentration is critical to producing timely accurate forecasts. How often does someone say they spent 2 hours doing homework or working out when in fact they may have only spent 30 minutes due to other extraneous activities such as watching TV at the same time or talking with friends at the gym? Most forecasting has to be performed by a deadline. Concentration will streamline the forecast process and will limit forecast errors.

Having a solid forecasting knowledge base grows by experience and meteorology coursework. It takes money to make money; similarly, it takes weather knowledge to develop new weather knowledge. If a forecast is missed, it is critical to learn from the mistake in order to prevent the mistake from occurring again. Many people do not take this step since it is time consuming. Half of becoming a better forecaster is not just making forecasts, but researching the past forecasts for accuracy.

A degree in meteorology or courses in meteorology build the educational base. It is easier to develop forecasting skill when the individual has knowledge of atmospheric dynamics and thermodynamics. However, even a meteorology degree does not guarantee one will become a good forecaster. The other elements discussed in this section plus forecasting techniques must be included with the meteorological knowledge base.

Forecasting skill often grows more rapidly with on the job forecasters such as NWS interns, broadcast meteorologists and beginning consulting meteorologists as compared to student forecasters that have no job experience. It is amazing how forecast skill improves when you are forced to forecast and have a forecasting method to follow. The best way to learn to drive a car is to drive; The best way to learn to forecast is to forecast! The forecasting knowledge base grows with effortful forecasting along with the studying of past forecast mistakes and successes.

Questioning is perhaps the most overlooked of the 5 elements. It is just so much easier to believe the MOS data and other forecasters than to make your own forecast. It is best to take time and research the conclusions of other forecasters. For example, if the NWS states "the thermodynamics look favorable for severe thunderstorms" a good forecaster would go through thermodynamic diagrams, the models, and indices charts to determine if in fact it looks to be a situation favorable for severe weather.

It is much easier to take a MOS low temperature of 34 and "copy" this number into a forecast than to research what the predicted low temperature could be. RULE: Don't look at MOS until AFTER making your own initial temperature forecast. One can not learn to forecast by "copying" numbers. A meteorology student with a sound foundation in forecast principles WILL out-forecast MOS over the long term. Use learned meteorology principles such as temperature advection, cloud cover, dewpoint, mesoscale effects and so forth to develop the forecast. This is much more painful and time consuming than "copying" numbers, but it must be done in order to make one a better forecaster. As a wise man once said, "The path to JOY and KNOWLEDGE is through PAIN and COMMITMENT".

Communication skills are the final of the five elements. Forecasts are nearly meaningless unless they are used for some money saving, life saving, or preparation purpose. Forecasts should be understood by the target audience they are intended for.

Communication between forecasters can be equally as important. Many forecasts are performed in teams. This has the benefit of limiting individual error while having the forecasters learn from each other. Forecasting with teams develops forecasting skill more rapidly than on your own primarily due to the questioning factor. For example, FORECASTER 1: "Notice how much stronger the warm air advection is on the NAM as compared to the NGM." FORECASTER 2: "You are right, I may have to bump my high temperature up a degree or two. But wait, the NAM solution could produce more cloud cover also, which will reduce solar heating." FORECASTER 1: "The increased cloud cover may offset the warm air advection." Two or more forecasters will be able to examine more weather information and share the information among the group while coming to a forecast consensus.

Obviously there is much one can learn if paired with an experienced forecaster. FACT OF LIFE: the way to get smarter is to hang around people that are smarter than you. Whether through books, meteorology course work, meteorology professors, a chief meteorologist, NWS employees and so forth, you must use these resources to more quickly advance your forecasting skill.

The diagram below shows the relationship between Forecast Effort and Forecast Success.



Low effort over the long term will produce low forecast accuracy. Low efforts (shown as #1) manifests itself through a small amounts of time studying weather data, a lack of concentration, a small amount of meteorology knowledge, little questioning of other forecasters and a lack of communication skills. These forecasters are often termed "rip and read", "MOS forecasters" or "at a glance forecasters". Their forecasting knowledge and skill is stagnant because they do not put the time nor effort into becoming better forecasters. Averaged over the long term they make poor forecasters. When MOS data and NWS forecasts are not available, they are helpless.


Low effort forecasters can occasionally make accurate forecasts (shown as #2). This is primarily due to the accuracy of an NWS forecasts or MOS guidance. "Copying" numbers and other's forecasts will not and can not make one a better forecaster. These are the forecasters who can NOT tell when MOS might be in error, and CANNOT distinguish which model may be producing a better solution, and do NOT research NWS forecasts.

On the opposite extreme we have high effort/ high accuracy forecasters (shown as #4). These are the forecasters who look over vasts amount of weather data in a short amount of time, have a high degree of meteorological weather knowledge/ experience, question and research much of weather data they come across, and communicate quickly and effectively with co-workers and other meteorologists. They DO know when MOS is in error and which MOS (NGM, NAM or GFS) will be most correct. They have a passion for researching the accuracy of their forecasts and come to forecast conclusions independently. They can see 3-Dimensional atmospheric motion and processes with each forecast model panel or satellite image they study. Over the long term they have a high degree of forecast accuracy.

Even the most brilliant forecasters will bust (shown as #3). Although most of #4's forecasts will have high accuracy, occasionally a forecast will not pan out as planned. No matter how much knowledge and experience one may have attained, the mystery of the atmosphere is never understood completely. While #4 forecasters bust due to the mystery of the atmosphere (unforeseen developments), #1 forecasters bust because someone else's forecast was incorrect. #4 forecasters can learn from the mistake while a #1 forecaster will be incapable. Over time, the knowledge gap between #4 and #1 forecasters grows.

Most forecasters lie in between #1 and #4 forecasters. The goal through time should be to "lean" toward the attributes of #4 forecasters which again include time pouring through weather data, concentrating, increasing meteorology knowledge (taking more and more BMP/OMP courses!), questioning and compiling weather data, and communicating effectively.

Improving forecasting skill manifests itself in a daily commitment to using all available forecasting tools. Below are 11 ways one can improve their forecasting skill.

1. Set up a map room. Have the forecast models, analysis charts and other meteorological data up to date and at your disposal to view at the same time.

2. Have forecast contests. These can be done through a national organization such as the National Collegiate Weather Forecasting Contest or through contests with co-workers. Don't get too carried away with betting on the weather though. I have seen friends lose lots of money on weather predictions that did not come true (e.g. location of hurricane landfall). A gentleperson's bet involving no money is best.

3. Surf Internet sites. Just about all weather data is available over the Internet. Many sites have help menus, that show how to interpret charts, the forecast models, Skew-T's, etc. People who are self-motivators at reading help menus benefit greatly.

4. Read AMS/NWA journals. Become members of the AMS and NWA and use their literature to improve forecasting knowledge

5. Library Research. Several AMS and other meteorology journals are available at the library. Just about any aspects of forecasting can be learned by researching back issues of forecasting and meteorology journals.

6. Re-read Introduction to Meteorology book. Without an understanding of the basics, it is impossible to learn the more complicated aspects of meteorology. A thorough knowledge of the Intro book will teach more about forecasting than is realized.

7. First hand experience. All weather occurs in the real world, not in a text book or on a computer. Storm spotting, weather observing, vacationing to different climates and watching the sky everyday will add to your textbook understanding of the atmosphere.

8. Mentors. Having people to "show" forecasting techniques and meteorology in general is extremely helpful. A short cut to knowledge is having people to talk to that can answer any meteorology question you come across.

9. Increase your computer knowledge/upgrade computer. Those with the best and most up to date weather information have the fastest and most state of the art computers. Like other fields, being a forecaster requires an understanding of computers. Don't be afraid of becoming a computer weather geek.

10. Keep a forecasting journal. Learning from mistakes is critical to becoming a better forecaster. It is easy to forget how well or bad a forecast goes each day if you can not reference what the forecast was.

11. Read the NWS forecast discussion each day. Read NWS text data and try to understand that they are saying. Research for yourself by looking at the charts to see the processes in the atmosphere they are discussing. Much about forecasting can be learned by reading the discussions.



There are four stages to becoming a respected forecaster. To become a good forecaster, it is important to build a foundation of meteorology knowledge. This knowledge is attained through learning from others (such as through a school of meteorology). Once you have the basics under your belt, research can be performed. This research and learning is done through reading books, reading journals, and finding information about meteorology that is not known to you. The next step is on the job experience. Perhaps the best teacher is experience. It is the opportunity to apply weather knowledge to the "real world". Once complete coursework in meteorology is attained with several years of on the job experience, you are well on your way to reaching the 4th stage which is a respected and experienced forecaster.

Many off-campus students go through these stages in a different order. First, with on the job experience and then meteorology coursework. This is an extremely effective way of learning since one can both practice in the real world and learn the meteorological course work AT THE SAME TIME. On the job experience is extremely important for finding the best jobs in meteorology. No matter the order of accomplishing the first 3 stages, the 4th stage is the ultimate goal.

There are also psychological factors that determine your success as a forecaster. It is important to have a positive attitude and perseverance. As soon as one gives up, one begins to tumble to the bottom of the mountain. All forecasters make mistakes, however, with experience and training these mistakes will be less numerous and less severe. All forecasters experience humility. The humility of completely blowing a forecast or realizing you need to know much more about forecasting. The key is to keep moving forward: continue to learn and learn from mistakes. Criticism and humility will try to break a forecaster, but don't let them.