The difference is most likely caused by the following
1) The initial surface was most likely 3D Faces (maybe maybe not?) - when you bring in 3D Faces, the lines around the edges are all 3D Breaklines. 3D Breaklines break the model up into areas for rendering purposes, so in this case every face is shaded individually, because each is bounded by 3D breaklines.
2) When you made a second surface and added the first surface to it, and then you added a boundary to the surface to limit it, new nodes are formed around the boundary so that it can cut out the piece it needs, the second surface now only has triangles and no 3D Breaklines to render from, so what happens is the rendering engine calculates approximate shading which smooths the surface graphically. If you want to sharpen it up, you have to have 3D breaklines to do that. So I would explode it using the Explode surface command to create 3D Breaklines around each Triangle, and then reform the surface from those 3D breaklines and you will end up with a sharp faced model again.
Rendering each triangle individually is more graphics intensive, that is why we use breakline bounded areas and render each area (collection of triangles) as one area and then as you hit the boundary of the area (a Breakline) you will get a sharp shadow transition across the boundary.
It makes no difference to the volumes or the color coded elevations, only to the shadows cast by the model.
Hope that this helps
Alan