Having some experience with avalanche danger modeling in GIS, I propose that the map-based idea being discussed here, though a cool and fun exercise, would ultimately produce a product that is at best inaccurate and at worst misleading.
Jeremy makes this point above:
So anyway, it is possible... the real question is whether NWAC wants to be publishing computer generated maps that say a particular slope is or isn't safe. A danger rose is much more generic, and probably safer from a liability standpoint.
Allow me to elaborate: our current ability to model and predict avalanche formation conditions is quite limited at spatial scales relevant to individual slopes. Current implementations of the danger rose concept are necessarily vague - when it comes to assessing the stability of an individual slope, it is up to the b.c. traveler to combine the larger-scale information from the forecast, danger rose, weather, etc. with more location-specific information from snow pits, wind drift observations, and instability cues.
A forecast product that attempts to show a danger rating on a specific slope due to specific snowpack conditions will likely be inaccurate, due to our inability to accurately represent regional and local spatial variability. The current state of avalanche science just does not support this application.
Further, consider how this proposed product would be used. In the hands of an experienced backcountry traveler with knowledge of the product's limitations, it is not likely to add a significant benefit to their decision-making process. They will understand that significant spatial variation is likely and will assess each slope individually just as if they only had information from the standard danger rose and not the higher-resolution GIS product.
However, in the hands of a less-experienced traveler, this product could either provide a false sense of security or misrepresent actual conditions. Think of the surface hoar pockets that are not predicted by the GIS model, e.g. ...
Slope-scale GIS models of avalanche danger are certainly useful as research tools, but an accurate slope-scale model is still in 'holy grail' territory.
Coarser-resolution products are in operational use in Europe. With the data streams available to NWAC, perhaps one of the Euro-style products would be a better role model - basically a weather data/forecast map with a little snowpack info added in, showing broad-scale variation by range/region or 'massif'... I agree with Gary that a map product at the region scale would be quite useful. Check out Colorado's forecast product as an example...
Anyways, just my $0.02
