This 10 part series will detail forecasting tricks that can be used to try to outforecast MOS. Outforecasting MOS is an important skill for a forecaster. MOS stands for Model Output Statistics and they are used as a guide for temperature prediction and precipitation prediction by forecasters. Model consensus is the average of the high temperatures, low temperatures or precipitation amount predicted by several forecast models.


When putting together a forecast make sure to check the upstream conditions before finalizing the forecast. The upstream direction is the direction the wind is blowing from. Typically the weather moves from west to east or a close variation of that. Often the weather will also approach from south to north or north to south. The weather can approach from east to west in certain weather situations also. When putting the forecast together check the analysis charts and surface charts and see what direction the wind is coming from. Then note the temperature and dewpoints upstream from your location. Also note the characteristics of precipitation if precipitation is occurring upstream on radar. Check immediately upstream to 200 or so miles upstream. Also note how quickly the upstream air is advecting toward the forecast area.

One reason to check upstream temperature and dewpoint is to see if the MOS forecast temperatures and dewpoints for your location over the next day are reasonable. The models may not have handled upstream conditions as well as they should have. For example, suppose the MOS predicted low temperature is 25 F. It is in the late afternoon and you notice 150 miles upstream are temperatures around 35 F with dewpoints in the teens. Once this upstream air advects in overnight and radiational cooling occurs you may come to the conclusion that the temperature will fall below 25 F for the low. From experience you will get a feel for what weather to expect from certain upstream conditions. Ask yourself does the MOS make sense given the air that will advect into the forecast area. Often it will seem reasonable but at other times you will conclude the MOS is either underestimating or overestimating temperatures. One model may handle the upstream conditions better than another. Suppose three models have low temperatures of 18 F, 25 F and 24 F. You notice the upstream conditions are advecting in cold and dry air overnight. The 24 F and 25 F in your opinion do not seem reasonable lows given the upstream conditions. In that situation you may want to go with the 18 F temperature or colder than model consensus.

Here is a situation in which upstream conditions can cause the low temperature to be much colder than model consensus. Suppose there is snow on the ground and clear skies is expected overnight. It is the afternoon and the temperature is 30 F. The forecast models have predicted lows of 15 F, 13 F and 10 F. You notice that 150 miles upstream the temperatures are in the low 20s with dewpoints near 0 F. In this situation the actual low will likely be well below model consensus. The radiational cooling from snow, clear skies, a long night, very low dewpoint and Cold Air Advection will combine for a very cold night. In the lower single digits or colder most likely.

The upstream conditions could also indicate to you that very dry or moist air is approaching. This will have a significant influence on temperature. Drier air with clear skies tends to warm the highs more than expected and cool the lows more than expected. Very moist air tends to reduce warming during the day and does not allow as much cooling at night. The models may pick up on this but if they are not modeling upstream conditions very well then MOS could be off once the dry air or moist air moves in.

There will be some situations in which the wind is very light. If this is the case there will not be much advection and the upstream conditions are not as important. With light wind and clear skies, the high will tend to go above MOS consensus and the low will tend to go below MOS consensus. When winds are light and the skies are clear then a cool pool of air will develop right at the surface due to radiational cooling at night (boundary layer decoupling) and a warm pool of air will develop right at the surface due to solar heating during the day. Since these pools of air are more pronounced when the skies are clear and winds very light, MOS will underestimate their influence on the temperature.

With MOS precipitation forecasts you will want to check the radar before finalizing your short term forecast. The upstream characteristics of precipitation will often be the characteristics of the precipitation once it reaches your forecast region. Note the areal coverage of precipitation, intensity and other characteristics. Look over the MOS percent chance, forecasted precipitation amount and other precipitation characteristics and ask yourself if it is reasonable.

Combining the information about the weather that is approaching with the model data and your forecast experience will benefit your forecasting method and accuracy.