MOS guidance forecasts have been produced for stations and provided to NWS forecasters and private entities for over three decades. As the numerical weather prediction models became more accurate, MOS followed that trend. Up until a few years ago, the MOS forecasts produced for observation locations met the basic need for guidance. With the advent of the Interactive Forecast Preparation System and the National Digital Forecast Database, gridded MOS forecasts became needed as guidance for forecasters. One method of providing such grids is to objectively analyze the MOS forecasts for points.

A basic successive correction method has been extended to analyze MOS forecasts and surface weather variables. This method is being applied to station MOS forecasts to provide guidance to NWS forecasters for producing grids of sensible weather elements such as temperature, clouds, and snow amount. These forecasts have been implemented for the conterminous United States for most weather elements contained in routine weather forecasts. Of prime importance is how the terrain is used to provide a variation with elevation where appropriate.

While it might be expected that some accuracy of the point MOS forecasts would be lost by gridding them, withheld data tests have shown this to not be true. However, careful quality control has shown some deficiencies in the point forecasts due primarily to low quality observations as predictand data, and we have been able to discard some stations with bad forecasts. In addition, putting the data onto a relatively fine grid has presented a few problems due to: (1) some MOS equations were developed regionally and the regional borders were apparent in the analysis, (2) the sets of stations for which we have forecasts are not the same for different model run cycles, and (3) the equations were produced on samples not covering the same period and with different predictors. In order to address these difficulties, as well as to account for the possibility the GFS model on which the MOS is based does not perform exactly the same way at different cycles, we use two cycles of MOS forecasts to make one analysis, adjusting appropriately for projection. It has been shown that the quality does not suffer by this juxtaposition of two cycles.

This paper will describe the analysis technique and provide verification and examples of the operational product. We will also describe our extension of the method to Alaska.