FORECAST JOURNAL DUE: Class before Thanksgiving Break Preface: You are to collect forecasts from local newspaper, radio, television, and/or internet sources for two cities in the country with a national weather service office nearby (for example Columbia and St. Louis). The object is to write down three sources of a forecast for the next 5 days, then verify that forecast with what actually happened including high and low temperature and whether or not it precipitated as recorded by the national weather service. You are then to determine which of your three sources did the best and why you think that is, including, was the meteorological setup easily predicted, for example was there a stretch of a few days that were dominated by high pressure making it easy or were there frontal passages or other turbulent weather events making it difficult for the forecaster. Who was the best at forecasting the next day, best at day 3, best at day 5? Does it look like your local media is just taking NWS or other mass internet weather service forecast? What does that say about the confidence of your local forecast? Also, how well was day 1 forecasted compared to day 5, what does this say about predictability in general? Instructions: You are to turn in a chart of your two locations and three forecast sources including the time of the initial forecast, forecast high, forecast low, forecast precip (yes or no), and the corresponding verification and deviation from verification. In addition you are to make comments on the forecast situation, and who actually did the best on the most sensitive days to forecast and who did the best for the five day period and why you think that might be. Make sure your second site is a national weather service site that you can get local climatological data archives. Columbia, MO archives can be found here: http://www.weather.gov/climate/index.php?wfo=lsx Click on preliminary climatology data (CF6), Columbia MO, and then select the month and GO. Some cities NWS page will be slightly different…always look for F6 data it is normally in a Local Climate link at the bottom right of the forecast page…email me if you are having trouble getting verification data and I will help out. Bonus Points: For bonus points make it as neat as possible using Excel or some other database program. Include some images in your write up to show the current meteorological set up, for example include an image of the surface map, satellite, or just give a good discussion of the situation. Also, maybe look at the average forecast for the three sources you select and calculate total errors combining penalties for high and low and precip in creative ways. Questions: Email Me ASAP at SALack@mizzou.edu
Example:
3-29-06 Forecast Columbia, MO KRCG Accuweather NWS Approx Time 7:53 AM 9:17 AM 7:54 AM Approx Time 8:53 AM 8:50 AM 8:54 AM High 68 69 67 Low 49 50 49 Precip yes yes yes High 72 72 72 Verification Low 53 53 53 Precip yes yes yes Diff High -4 -3 -5 Difference Diff Low -4 -3 -4 Precip Match yes yes yes Precip Match yes yes yes
Cleveland, OH WJW Accuweather NWS
High 66 66 65
Low 50 50 49
Precip no no no
High 64 64 64
Low 47 47 47
Precip no no no
Diff High 2 2 1
Diff Low 3 3 2 Difference
3-30-06 Forecast Columbia, MO KRCG Accuweather NWS High 65 62 62 Low 49 40 41 Precip yes yes yes High
3-30-06 Verification Low Precip Diff High
Diff Low
Precip Match
Cleveland, OH WJW Accuweather NWS
High 68 70 69
Low 45 50 45
Precip No Yes Yes
High
Low
Precip
Diff High
Diff Low
Precip Match