1: \begin{abstract}

2:   The shallow water equations (SWE) are a widely used model for the

3:   propagation of surface waves on the oceans.  In particular, the SWE

4:   are used to model the propagation of tsunami waves in the open

5:   ocean.  We consider the associated data assimilation problem of

6:   optimally determining the initial conditions for the one-dimensional

7:   SWE in an unbounded domain from a small set of observations of the

8:   sea surface height and focus on how the structure of the observation

9:   operator affects the convergence of the gradient approach employed

10:   to solve the data assimilation problem computationally.  In the

11:   linear case we prove a theorem that gives sufficient conditions for

12:   convergence to the true initial conditions. It asserts that at least

13:   two observation points must be used and at least one pair of

14:   observation points must be spaced more closely than half the

15:   effective minimum wavelength of the energy spectrum of the initial

16:   conditions.  Our analysis is confirmed by numerical experiments for

17:   both the linear and nonlinear SWE data assimilation problems which

18:   reveal a relation between the convergence rate of gradient

19:   iterations and the number and spacing of the observation points.

20:   More specifically, these results show that convergence rates improve

21:   with increasing numbers of observation points and that at least

22:   three observation points are required for the practically useful

23:   results. Better results are obtained for the nonlinear equations

24:   provided more than two observation points are used. This paper is a

25:   first step in understanding the conditions for observability of the

26:   SWE for small numbers of observation points in more physically

27:   realistic settings.

28: \end{abstract}

29: