All the articles I've been editing were like this:
Colours, size, physical characteristics
Lookalike birds, possible mistakes
Subspecies, splits etc.
Type of habitat
Diet, typical behaviour, reproduction etc.
I think having both Description and Identification is not redundant. Just see the famous and excellent Channel-billed Toucan article by Rasmus. The Description part, as well as Distribution and Taxonomy parts split (as Niels mentioned), could be left to the author to decide.
In this Black Vulture
article I've made, I pointed charactersitics (later with information added by Rasmus) to tell Black and Turkey Vultures apart, while in Description I wrote about colours and overall features.