4.2.2 Improving HWR Predictions
Our results demonstrate that an OT below 4 m and a consistent exposure of MI within the headwall were the most important factors in producing the fastest rates of HWR at Peninsula Point. A simple index was created to test if accounting for these two in-situ controls can improve the prediction of HWR spatial variability across the site. Along the 20 transects that covered the MI surface model, we measured inland from the 2017 headwall position a distance equal to the 2016 to 2017 HWR. The average ice elevation and the average OT along this distance were then given a value based on the criteria in Table 2. These two values were averaged and divided into the 2016 to 2017 HWR distance along each transect to produce a predicted 2017 to 2018 HWR (green line labelled MI). For example, consider a transect that recorded 5 m of retreat between 2016 and 2017, but had no exposed ice and an overburden > 4 m in both 2016 and 2017. If this transect is modelled to have an exposure of ice and an OT < 4 m after 2017, using Table 2, the predicted HWT from 2017 to 2018 would be 6.7 m (5 m divided by 0.75).
These predictions were compared with similar metrics based on the assumption that the headwall properties present in 2017 extended inland (blue line labelled OBS), an extrapolation of the 2016 to 2017 HWR (red line labelled EXT) and with the HWR of 7 m a-1 on active retreating headwalls on Peninsula Point as described by Mackay (1986) (pink line labelled MAC), and plotted together in Figure 10.
Table 2: Criteria and values used to create the 2017 to 2018 HWR prediction