The avalanche danger maps illustrate the potential danger when attempting to ski these three descents. Whilst they can provide important information in regard to the most dangerous locations, they are limited by the amount and type of data available. Thus, when considering a ski-mountaineering descent like these three, it is essential to use information from multiple sources and understand that avalanches are dynamic processes which can be difficult to predict. 

Firstly, it must be recognized that the data used in this study was obtained from multiple sources which is likely to have introduced uncertainty and reduced the accuracy of the results. This, in turn, affects the quality of the decisions that can be made. 

The wind data had to be converted into a raster and resampled in order to fit the DEM, which could have reduced the accuracy. Further, we were unable to find prevailing wind data, which would have been much more useful than just using the wind speed. Knowledge of the prevailing wind is necessary to more accurately determine the avalanche risk, as the wind direction can influence how the snow forms and collects on the slope. For example, if the wind packs snow on the lee side of the slope, it is more likely for a slab avalanche to form (Montgomery, 1954). 

Additionally, the usefulness of the data was also limited by the spatial and temporal resolution as well as the information it provided. As avalanches can often begin from a single point, having a high spatial resolution is extremely important in order to accurately depict the magnitude of the danger associated with a location. Terrain and weather patterns can vary significantly within space, particularly in mountainous regions. The DEM data we obtained was of 20 m resolution, thus limiting our ability to illustrate areas of highest risk. Temporally, the average annual wind speed was used, which will, of course, vary significantly at different times of the day and the year. 

Due to the low spatial resolution of our data, we chose not to include curvature in our analysis as it would not be accurate enough to provide useful information. However, it is important to acknowledge the uncertainty that arises without curvature data. Convex slopes are inherently less stable and often wind load more than concave slopes, resulting in a greater risk (Avalanche.org, 2020). 

Moreover, whilst snowpack is a significant factor to consider when evaluating avalanche risk, it is difficult to acquire and use for a study of this nature. Snowpack can vary considerably throughout the winter, thus avalanche risk will vary on a daily basis depending on snowfall. For example, a sudden large increase in snowfall will result in substantial fresh snow which can increase avalanche risk. However, if the snowfall is large and consistent, it can actually build a stronger snowpack, resulting in a lower avalanche risk (Mountain Academy, n.d.). As the snowpack changes every year and is likely to become more variable with climate change, we decided not to include it as it would be too difficult to accurately represent unless we were able to produce daily maps with accurate snowpack data for each day.  

Overall, the complex and dynamic nature of avalanches makes it difficult to analyze and understand the level of risk associated with them. The increasing variability in weather patterns due to climate change will likely increase this uncertainty and could result in a greater area of high-risk terrain. Thus, examining daily avalanche forecasts and having knowledge in regard to avalanche safety is essential when skiing in dangerous terrain.