The graph shows the range of swells directed at Sentinel through a typical October, based on 1736 NWW3 model predictions since 2006 (values every 3 hours). The wave model does not forecast wind and surf right at the shore so we have chosen the most applicable grid node based on what we know about Sentinel. In this particular case the best grid node is 26 km away (16 miles).
The rose diagram shows the distribution of swell sizes and swell direction, while the graph at the bottom shows the same thing without direction information. Five colours illustrate increasing wave sizes. Blue shows the smallest swells, less that 0.5m (1.5 feet) high. These were forecast only 0.5% of the time. Green and yellow show increasing swell sizes and red illustrates biggest swells greater than >3m (>10ft). In both graphs, the area of any colour is proportional to how frequently that size swell happens.
The diagram suggests that the dominant swell direction, shown by the longest spokes, was SSW, whereas the the prevailing wind blows from the SW. Because the wave model grid is offshore, sometimes a strong offshore wind blows largest waves away from Sentinel and away from the coast. We combine these with the no surf category of the bar chart. To keep it simple we don't show these in the rose plot. Because wind determines whether or not waves are surfable at Sentinel, you can view an alternative image that shows only the swells that were expected to coincide with glassy or offshore wind conditions. Over an average October, swells large enough to cause good for surfing waves at Sentinel run for about 24% of the time.
IMPORTANT: Beta version feature! Swell heights are open water values from NWW3. There is no attempt to model near-shore effects. Coastal wave heights will generally be less, especially if the break does not have unobstructed exposure to the open ocean.