What is 'depth' for compact height field?

Recast originally uses height for z-axis units, but A* uses depth.
What does it stand for?
What is the difference between them?
Why would A* choose to use depth?

In many places, A* uses wd = width * depth for looping

			for (int z = 0, pz = 0; z < wd; z += w, pz++) {
				for (int x = 0; x < w; x++) {

while Recast uses h

	for (int y = 0; y < h; ++y)
	{
		for (int x = 0; x < w; ++x)

They are essentially equivalent.
The A* Pathfinding Project has used the convention that the pathfinding plane is in the x,z plane, not the x,y plane (in hindsight, I really shouldn’t have done this, but it’s hard to change now unfortunately). So instead of height and the y coordinate, it uses depth and the z coordinate.

In some places in the source code, an optimization has however been made so that z is actually the z coordinate multiplied by width. This is done to save a few multiplications inside the loop. In those loops, the pz variable is the actual z coordinate. Sorry, I know that code is perhaps not the most self explanatory.