Errata for Wave Curves: Simulating Lagrangian water waves on dynamically deforming surfaces

E1 EFFECTIVE GRAVITY IN “PADDLE” SCENE The simulations of the “paddle” scene in [Skrivan et al. 2020] have a parameter set incorrectly. Instead of using our proposed effective gravity g∗, these examples only use the fixed gravity g. To rectify this mistake, the authors re-ran the simulations with the effective gravity enabled. We provide corrected figures here, and we provide corrected animations in the supplementary files accompanying this document. Figure E1 in this paper should replace Figure 8 in the original paper. It shows the difference between a low-resolution simulation with only 100k wave curve points, compared to a detailed simulation with 800k curve points. These images have fewer visible waves compared to the original paper, because the modified effective gravity causes different wave seeding and decay rates in the simulation. Figure E2 in this paper should replace Figure 9 in the original paper. It shows a simulation with effective gravity enabled and how it looks with only a single wavelength simulated. We do not see a significant qualitative difference between these simulations and those of the original paper. Figure E3 in this paper should replace Figure 12 in the original paper. It shows the comparison between our seeding and random seeding, with effective gravity enabled. The differences between the two simulations are still very clear — our method produces more waves that align with the flow, while random seeding produces random wave orientations before the geometric stretching of the flow naturally destroys poorly-aligned waves. The only difference between the seeding in the original paper and this one is the effective gravity term in the growth equation: