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.


In the absence of cold thermal advection, radiational cooling is the primary reason temperatures cool during the overnight hours. The earth is constantly emitting longwave radiation. At night there is no solar radiation coming in thus the longwave emission represents a constant net energy loss. Temperatures tend to cool during the overnight hours with the low temperature occurring around sunrise.

The following help contribute to longwave energy loss at night and thus cooling. Each is listed below along with the reason:

Clear skies- When skies are clear the longwave radiation is better able to escape into space. When clouds are present the clouds will reflect and emit longwave radiation back to the surface. With clouds there is less net longwave energy escape. It is ideal to have clear skies all night.

Long night- There are months in which night is longer. When night is longer there is more time for longwave energy to escape. It is ideal to have as long of a night as possible.

Dry air- Even when skies are clear, water vapor in the air will absorb and emit longwave radiation. Humid air can act like cloud cover when it comes to longwave energy trying to escape. It is ideal to have low dewpoints throughout the atmosphere.

Snow cover- Snow is a strong emitter and reflector of radiation. At night, especially when it is clear, snow cover will enhance the cooling that takes place. Fresh snow is better at this than old snow or snow that is not covering all the ground. It is ideal to have a solid layer of fresh snow cover.

Light wind- When the wind is light, the air near the surface will not mix as much. Since the ground surface is the source of the cooling, air temperature will be colder close to the surface. When the wind is very light there will be less mixing with warmer air aloft. This is especially true if the winds aloft are light also. When winds are very light a shallow layer of cold air will develop just above the ground surface (boundary layer decoupling). It is ideal to have a calm wind all night.

Dry soil- There will not be much evaporation into the air when soils are dry. Dry soil reduces the chance of fog developing. Fog will greatly limit the amount of cooling that could occur. Even without fog, the evaporating moisture will not let as much longwave energy escape. It is ideal for it not to have rained for the past several days and for the soil to have a lack of moisture emitting vegetation.

Rural area- Rural areas cool off more at night than urban areas. Urban concrete does not cool as quickly.

When all of the above occur at the same time then it is likely the low temperature will fall below MOS model consensus. This is true with or without snow cover but if there is snow cover it is not uncommon for the low to be 10 F or less below MOS model consensus. Several of the above factors in varying degrees can join together to cause the low temperature to be below model consensus. Study the MOS low temperatures and ask yourself if it could be a situation where a great amount of radiational cooling occurs. If you think the potential is there for more cooling than the MOS low temperatures indicate then adjust your low downwards. With practice you will begin to outforecast MOS in these significant overnight cooling situations.