Paper 4, Section II, 17C

Methods | Part IB, 2015

Describe the method of characteristics to construct solutions for 1st-order, homogeneous, linear partial differential equations

α(x,y)ux+β(x,y)uy=0\alpha(x, y) \frac{\partial u}{\partial x}+\beta(x, y) \frac{\partial u}{\partial y}=0

with initial data prescribed on a curve x0(σ),y0(σ):u(x0(σ),y0(σ))=h(σ)x_{0}(\sigma), y_{0}(\sigma): u\left(x_{0}(\sigma), y_{0}(\sigma)\right)=h(\sigma).

Consider the partial differential equation (here the two independent variables are time tt and spatial direction xx )

ut+uux=0\frac{\partial u}{\partial t}+u \frac{\partial u}{\partial x}=0

with initial data u(t=0,x)=ex2u(t=0, x)=e^{-x^{2}}.

(i) Calculate the characteristic curves of this equation and show that uu remains constant along these curves. Qualitatively sketch the characteristics in the (x,t)(x, t) diagram, i.e. the xx axis is the horizontal and the tt axis is the vertical axis.

(ii) Let x~0\tilde{x}_{0} denote the xx value of a characteristic at time t=0t=0 and thus label the characteristic curves. Let x~\tilde{x} denote the xx value at time tt of a characteristic with given x~0\tilde{x}_{0}. Show that x~/x~0\partial \tilde{x} / \partial \tilde{x}_{0} becomes a non-monotonic function of x~0\tilde{x}_{0} (at fixed tt ) at times t>e/2t>\sqrt{e / 2}, i.e. x~(x~0)\tilde{x}\left(\tilde{x}_{0}\right) has a local minimum or maximum. Qualitatively sketch snapshots of the solution u(t,x)u(t, x) for a few fixed values of t[0,e/2]t \in[0, \sqrt{e / 2}] and briefly interpret the onset of the non-monotonic behaviour of x~(x~0)\tilde{x}\left(\tilde{x}_{0}\right) at t=e/2t=\sqrt{e / 2}.

Typos? Please submit corrections to this page on GitHub.