Synoptic II Journal Entry #1 --> Forecast period: 0Z March 1st- 0z March 3rd
Tonight (0z March 1st - 12Z March 1st): Few clouds in the early evening hours with a 30% chance of some snowfall. No accumulation expected. Skies gradually clearing overnight. Low temperature of 25° with winds out of the West at 5-10mph
Tomorrow: Clear skies with a high temperature of 36° expected. Light Northwesterly winds throughout the day eventually shifting Southwesterly into the evening. No precipitation expected.
Tomorrow Night: Low temperature of 20°. Southwesterly winds around 5-7mph expected, with some increasing cloud cover overnight. No precipitation expected.
Wednesday: Partly cloudy skies early becoming cloudy by the afternoon. High temperature of 36° with a 20% chance of some snowfall occurring after a cold front moves through the area in the mid-afternoon hours. No accumulation expected. Winds will be out of the Northwest up to 15mph.
Weather Observations (that lead to public forecast)
My forecasting method for this particular forecasting period stayed consistent throughout the entire period of my forecast. I started by observing the current conditions of Plymouth, NH on Monday, Feb 28th.
From there, I kept the current conditions in mind and used the "850 Method" for forecasting the overnight low temperature. When looking at the 850mb analysis, I look to two different areas. One is the Vortex website in order to determine the minimum 850 temperature and for the other, I used the PSU EWALL to look at 850mb temperature as well as probability and precipitation amounts. When taking the coldest 850mb temperature and converting it to °F, I had to adjust the overnight minimum temperature that I had originally calculated. I then looked at other factors such as the 10m winds from the Texastornado website and saw some light and variable winds could be expected. I then took into account any cloud cover, by looking at 700mb Relative Humidity on the same website as well as the Low-Upper Level Cloud forecast given by the PSU E-WALL. I noticed that some cloud cover was expected to increase into the early morning hours, so I deducted 5° from my coldest 850mb temperature. Afterward, taking into account the current observations of Plymouth that Monday afternoon I had to deduct a few more degrees for some fresh snow cover that our area had. I also used a persistence method, to give myself a 20% chance of precipitation occurring overnight. Finally, my overnight public forecast was created (above).
In terms of the Tuesday forecast. I used the same "850 Method" but this time for the daytime maximum. In terms of the models and what to look at for, I went to the same areas: 700mb Relative Humidity and the low-upper level cloud forecast to determine cloud cover and 10m winds to forecast the wind. However, I also looked more closely to the 850mb upper air analysis and noticed some indication of cold air advection in the Plymouth area. Taking all of this into account and the fact that I noticed 10mph winds out of the Northwest with clear skies, I was able to adjust my daytime maximum temperature by subtracting about 5°, and create the rest of my public forecast.
For the Tuesday night low, again I looked to the same method. I noticed some increasing cloud cover overnight, as compared to the clear sky forecast I had made during the day. I also looked to the 850 method for the overnight low temperature. Once again, I discovered some warm air advection moving into the area and was able to adjust my overnight minimum temperature. I found Southwesterly winds on the 10m wind charts, which corresponded to the warm air advection that I found. And finally, I saw no indication of precipitation, so I was able to reassure that I could forecast a probability of 0%.
On Wednesday, I looked at the same charts and used the same methods when forecasting the daytime maximum temperature, winds, clouds, etc. However, when looking at the PSU EWALL, I noticed a probability of some precipitation occurring within that day. I then looked to the Short Range Surface forecast on the National Weather Service's website and saw a cold front. I took into account the cold front and cloud cover, as well as a vorticity analysis for that day, and forecasted a 20% chance of snow.
Forecast Verification
For the Feb 28th-March 1st overnight low: I had forecasted 25° with a 30% probability of precipitation. The actual temperature was 23° with 0 precipitation. However, there was a slight trace.
Tuesday: I had forecasted a temperature of 36° with no precipitation. The verification was a high temperature of 32° and no precipitation.
Tuesday Night: My forecast was a low temperature of 20° with no precipitation. The verification was a low temperature that got down to 5° with no precipitation.
Wednesday: I forecasted a high temperature of 36° with a 20% chance of precipitation. The verification was 38° with only a trace of precipitation.
In terms of my forecast, I was very pleased with the results. The only outlier was the Tuesday night low temperature. The winds went calm that night and the cloud cover did not exactly increase (like I had forecasted) until the daytime hours. I believe that overall for this particular forecast period, my methods were about where they should be. However, since there is always room for improvement, looking back, I realized that I did not exactly check my forecasts for validity with the Model Output Statistics. If I had, I could have possibly been able to adjust my forecasts accordingly, not only to help me with Tuesday night's low temperature, but for my entire forecast in general. I might have been able to get a little "closer" to the actual verification. I definitely plan on taking advantage of the models as well as experimenting with new techniques to enhance my forecasting skills.
