1: \begin{abstract}

2: The use of proper orthogonal decomposition (POD) to explore the complex 

3: fluid flows that are common in engineering applications is increasing and has yielded new physical 

4: insights.  However, for most engineering systems the dimension of the dynamics is expected to 

5: be very large yet the flow field data is available only for a finite time.  In this context, it is important 

6: to establish the convergence of the POD in order to accurately estimate such quantities as the 

7: Karhunen-Lo\`{e}ve dimension. Using direct numerical simulations of Rayleigh-B\'{e}nard 

8: convection in a finite cylindrical geometry we explore a regime exhibiting extensive chaos and 

9: demonstrate the consequences of performing a POD with a finite amount of data. In 

10: particular, we show that the convergence in time of the eigenvalue spectrum, the eigenfunctions, 

11: and the dimension are very slow in comparison with the time scale of the convection rolls and that 

12: the errors incurred by not using the asymptotic values can be significant.   We compute the 

13: dimension using two approaches, the method of snapshots and a Fourier method 

14: that exploits the azimuthal symmetry. We find that the convergence rate of the Fourier method 

15: is vastly improved over the method of snapshots. The dimension is found to be extensive 

16: as the system size is increased and for a dimension measurement that captures 90\% of 

17: the variance in the data the Karhunen-Lo\`{e}ve dimension is about 20 times larger 

18: than the Lyapunov dimension.

19: \end{abstract}

20: