Extensive testing of the kinematic wave routing method determined that an error exists in the method as implemented in HEC-1 whereas the calculations are performed correctly in HMS. For each time step, the cross sectional flow area must be calculated at each node in the reach. In HEC-1 this was done incorrectly for the first node, where it used the inflow from the previous time step to compute area instead of the inflow for the current time step. This seemingly small difference in the routing algorithms can lead to differences in the routed peak flow. So far differences of up to 5% have been observed.