Issue in the bottom drag formulation of the kdvimex class. Current code has self.H^2 in the denominator, where self.H is the greatest depth input in the background density profile i.e. spatially constant and potentially arbitrary (kdv classes will extrapolate past this max depth id need be). In Holloway 97 the vertical scale (h) was a spatially variable function of Beta which is much smaller than the depth.
Issue in the bottom drag formulation of the kdvimex class. Current code has self.H^2 in the denominator, where self.H is the greatest depth input in the background density profile i.e. spatially constant and potentially arbitrary (kdv classes will extrapolate past this max depth id need be). In Holloway 97 the vertical scale (h) was a spatially variable function of Beta which is much smaller than the depth.