 The basic idea of model output statistics or MOS forecasts is that they use predicted variables from dynamic models along with actual observations to produce a forecast for a variable that dynamic models may not predict well. As an example, let's look at how MOS could predict daily high temperatures as a function of dynamic model 18-hour forecast for 1,850 mbar of thickness. First, a series of high-temperature observations is collected and matched with the corresponding dynamic model 18-hour predictions of 1,850 mbar of thickness. That's represented by these red dots on the graph, which plots observed high temperature along the axis on the left and model forecast for 1,850 mbar of thickness along the bottom. This data set happens to be taken from observations and model forecasts for Columbus, Ohio during the month of January. After a sufficient number of observations has been collected, a linear regression is performed to find a simple equation that best characterizes the observed data. The line of best fit here is marked in green. With the regression equation, when the dynamic model predicts a value for 1,850 mbar of thickness, the linear equation is used to convert this value directly to a daily maximum temperature forecast. In this case, let's say the dynamic model predicted a thickness of 1,220 m. Using the line on our graph or plugging that thickness into our equation, we can find that the corresponding MOS high temperature forecast would be 15 degrees Fahrenheit. I should note that this particular sample of one month of data is too short to actually compute a statistically significant linear regression equation. You need a much larger sample of several years of data. Of course, real MOS equations also use more than one predictor from a dynamic model output in order to forecast high temperature or any other variable. But I hope you now have a basic idea of how MOS takes output from dynamic models and converts it to helpful forecasts for variables that dynamic models sometimes don't predict very well. Thank you.