INTRODUCTION A post analysis is looking back at the outcome of a weather situation in order to determine the strengths and weaknesses from the forecast. In forecasting, we typically think of forecasting the weather for the future. The “fore” in forecasting means to look forward and tell in advance. One way to improve in forecasting is to learn from mistakes made in previous forecasts by looking back. A post analysis can be done for any weather situation. Examples include high temperature forecasts, low temperature forecasts and the character of a precipitation event. Post analyses are popular when it comes to big weather events such as a severe thunderstorm outbreaks, tornado outbreaks and winter storm outbreaks. These big forecasts will be followed by many more people, thus a forecaster is often more cognizant of the successes and failures that occur with these forecasts. It is a good idea to keep a forecast journal that gives both your forecast and the results that occurred. Similar weather patterns will repeat in the future thus forecasting skill will typically grow when an effort is made to learn from each forecast made. Each piece of knowledge gained will help with developing improved forecasts in the future. LEARNING FROM MISTAKES One reason to do a post analysis is to learn from mistakes. Since there can always be a forecast to be made it can be difficult to find time to look back at a previous forecast and examine the specifics on what went right and what went wrong with the forecast. Looking back will help you learn from mistakes, especially if you have taken good notes. This is one reason that keeping a forecasting journal is important. Keeping a journal will help you remember the forecasted values and notes about the general weather pattern and important mesoscale features. It is a good idea to also keep track on the MOS numbers and compare those to your forecast and for what really happened. MOS stands for Model Output Statistics. They are computerized forecast numbers for a specific location. Of particular interest are the high temperature, low temperature and precipitation probability and character. MOS values have improved over time and it is becoming more and more difficult to outforecast the MOS values. With experience, outforecasting MOS can be done, especially in changing and big weather situations. Below is example of information to put into a forecast journal. Take note of these values and also add post analysis notes for why you think the forecast ended up turning out like it did. This journal will focus on the next day forecast. What are the important synoptic scale and mesoscale features that you see when looking through weather data: MOS values of high and low temperature: Your forecast high and low temperature: Actual high and low temperature: MOS POP (Probability of Precipitation) values: Expected precipitation character from MOS (i.e. type, severe, intensity, spread, convective, dynamic, etc.): Your expected precipitation POP and character: Actual precipitation character and approximate areal coverage: Post Analysis: What went right and wrong with the forecast?: WEATHER PATTERNS REPEAT Another advantage to regularly logging a post analysis and keeping a forecast journal is that over time you will begin to pick up on patterns. Forecasting for an area for two years (forecast experience for each season twice) will give you significant background knowledge to pick up on forecast patterns that are common for your forecast area. Without keeping a forecast journal it can be more difficult to pick up on the mesoscale intricacies and weather patterns that are common during each season. Before starting your journal or when you first move to a new forecast area that you will be forecasting for it is a good idea to interview an experienced forecaster for that forecast area. A meteorologist at a nearby National Weather Service office is a good source of information. Ask questions such as, What are the mesoscale influences in the area? What is the weather typically like during each season? What unusual weather events do you see each year? Which weather forecast model seems to do best for this forecast region? Which MOS tends to do best in particular weather situations? What have you noticed that causes MOS to be way off? What are the forecast challenges for this region? What are common forecast mistakes you have seen forecasters make in this area? What should I do to learn more about how to be an experienced forecaster for this region? Take note of how the synoptic models look each day such as the jet stream pattern, position of troughs and ridges, air masses and surface winds and write down how they are influencing the local forecast area. With experience you will be able to see how these big scale influences impact the local forecast area. Weather patterns will often repeat, thus learning from a previous similar weather pattern will help when a similar pattern develops in the future. Take note of mesoscale variations in the forecast area such as with temperature and where precipitation develops. Often precipitation will develop first in favored areas. Some areas will tend to be cooler or warmer than other areas in the region. Write down the reasons for these mesoscale variations. This will help you get familiar with the local weather influences. EVENTS TO POST ANALYZE Events that are popular to post analyze are tornado events, severe weather events, hurricanes and winter storm events. Any unusual event to the forecast area is also popular to analyze. This is done by saving model images and analysis charts before and during the event. Radar, satellite, Skew-T data, MOS data, and other relevant data can also be saved. Once the event is over, these data can be referred back to. This is a good educational opportunity to learn reasons why a forecast went well or had challenges. It is the “big events” such as tornado events, hurricane events, severe storm events and winter storm events when the general public is going to take an exponentially higher interest in the weather. Thus it is a good idea to focus on forecasting these events well. These are the forecasts and verifications that are most likely to be remembered by the general public. Successfully forecasting these events can launch a forecast career especially if a track record a several successful forecasts of big events can be established. Much of forecasting though is day to day type forecasting. Getting good with forecasting the high temperature, low temperature, temperature trend and precipitation threat are important also to establishing a successful forecast career. Performing a post analysis on typical weather days is also important to learning about the intricacies of the forecast region. THE PAYOFF In this conclusion on post analyses we look at several reasons why they are important to do. Strong post analysis skills require keeping a daily forecast journal which gives an overview of the synoptic and mesoscale influences on the forecast area, the MOS predictions, your predictions and the actual conditions that occurred. Take notes of the differences between the computer forecasts, your forecasts and what actually occurs and hypothesize reasons for the differences. Several reasons why performing post analyses and keeping a forecast journal can increase forecasting skill are given below: 1. Make it easier to pick up on a similar weather pattern from the past when making a forecast 2. Helps the forecaster pick up on the typical synoptic and mesoscale influences on the forecast area 3. Helps the forecaster learn from mistakes that may have not been noticed otherwise 4. Increases use of weather data and forecasting skills instead of relying solely on the NWS/SPC forecasts 5. Gives the forecaster a specific method and record to learn from mistakes 6. Helps you learn which model MOS does best in certain situations 7. Helps the forecaster to be more cognizant of the various factors influencing the forecast 8. Reduces the frequency of repeating forecast mistakes 9. Gives you good reference material (models, Skew-T, MOS, radar, satellite, etc.) of big weather events 10. Leads to a more rapid increase in forecasting skill |