Hi Chaitanya, Any of these timestepping schemes can be implemented in Firedrake. As for which is best, this should be determined from the properties of those schemes (perhaps it is your homework?). All the best Cjc On 22 Oct 2015 01:15, "Chaitanya Raj Goyal" <chaitanyaraj108@gmail.com> wrote:

Hi all,

I am new to Fenics and recently came across Firedrake. I am trying to solve a 2D seismic wave equation for multiple layers of soil with varying properties with absorbing boundary conditions.

1. I was wondering that between the Sympletic Euler, Newmark Beta and Crank Nicolson, (or any other) which is the best method to discretise the wave equation {u_tt = c^2 \Delta u + f} for implementation in firedrake. Could you please share your opinion.

2. If you happen to be aware of a specific good document that derives such a time stepping scheme for wave equation, could you kindly point towards it.

3. Also, would firedrake be a better platform to implement the problem instead of fenics?

I apologize if the questions may seem trivial, but I just started my Masters in Civil Engineering, and it would be awesome to get first hand opinion from math modelling experts.

I' ll be very grateful if you could guide me on the above issues!

Thank you,

Sincerely yours, Chaitanya Goyal