Along with the vertically binned ice shelf thickness distribution, Fig. 7(b) also shows the mean melt rates within each

depth bin (right axis) for nine different experiments, corresponding to the strongest (130), weakest (30), and intermediate wind forcing (100) for each of the three different hydrographic scenarios, temporally averaged over the respective last model year. The results generally reflect the spatial pattern of Fig. 7(a), with high melt rates above 10 m year−1 only occurring at deep ice below 400 m, and melt rates of less than 1 m year−1 at ice depths between 200 m and 400 m for all experiments. Somewhat higher melt rates of up to 3 m year−1 also occur at locations of very shallow ice above 5-Fluoracil solubility dmso 200 m depth, corresponding to enhanced melting Alectinib mouse near the ice front. The contribution to the total basal mass balance within a given depth bin, obtained by multiplying the vertically binned mean melt rates by the ice shelf area distribution, is shown in Fig. 7(c), with three main features being evident

from the graph.3 Firstly, the deep and shallow melting respond in opposite ways to winds. Melting of shallow ice above 400 m increases with the strength of the wind forcing, whereas melt rates below 400 m are largest for the weakest winds for all hydrographic scenarios. Secondly, melting of both deep ice and shallow ice, are strongest in the constant summer scenario and weakest in the constant

winter scenario for equal wind forcings. Thirdly and perhaps most noticeably, the melting response is strongly modulated by the uneven distribution of ice shelf area. In most experiments, the basal mass loss is Quinapyramine dominated by weak melting of large areas of shallow ice, while substantial changes of the mass loss at very deep ice only occur for the extremely large deep melt rates in the ANN-30 and SUM-30 experiments shown in Fig. 7(c). The characteristic depth-dependent melting response to varying forcing is summarized in Fig. 9(a) and (b). The colored curves are identical in both panels, showing the total amount of melting for the entire ice shelf as function of the wind forcing. The colored patches show the contribution of melting only from ice deeper than 300 m (Fig. 9(a)), or from melting at ice shallower than 300 m (Fig. 9(b)), respectively. For an applied surface stress above 60% of the climatological average (indicated by the vertical lines in Fig. 9), the melting response in all hydrographic scenarios is dominated by changes of the shallow melting contribution, which correlates roughly linearly with the applied surface stress (Fig. 9(b)).