2D Gauge wave tests

For computational limitations, the 2D wave runs were stopped at $t=100$. The ${\hat W}$ algorithm has smaller error, presumably as a result of the extra averaging operators.

This Figure indicates the error in $g_{xx}$ for 2D wave with A=.5, the ${\hat W}$ algorithm:

Using the $\ell_\infty$ norm and the $N=100$ and $N=200$ runs with the $\hat W$ algorithm, wei measured the following convergence rates at t=10 and t=100:

$${\tilde c_r^{(10)}} = 2.084, {\tilde c_r^{(100)}} = 1.568$$ (10)