This graph shows the attenuation (as a percent of the incident wave height) as a function of incident wave height (in cm) for 1 Hz waves. As expected, the trend is towards higher attenuation at higher wave heights. This trend decreases as the number of stems decreases, perhaps due to the decreasing influence of the stems.
This graph is the same as the one above, except using 2 Hz data. Curiously, there is no apparent pattern here. Why this should be is unclear, however it suggests a need to repeat the study with significantly longer data sets, at least for this frequency. Note that despite identical wave generation schemes, the initial wave heights vary by as much as 12 percent, which is more than the change we are expecting to see. I strongly suspect that these results would not be repeatable, but could be improved through longer data sets.
This graph plots attenuation as a factor of incoming wave period. The results suggest that attenuation increases with increasing frequencies, as predicted. While more data would be required to determine the precise effect, we can predict that the spectrum of transmitted waves would be shifted towards longer periods compared to incoming waves.
Finally, we see the most surprising result: with the range of densities I explored, there seems to be no effect of stem density on wave transmission. What is clear from the graph, however, is that more study is necessary (the 3 Hz line, in particular, shows problems with the data). I remain confident that greater data accuracy, combined with a wider range of planting densities, would eventually yield some relationship.
