Uplift may approximate spatially uniform landscape erosion rates in a long-term, orogenic-scale sense, but approximate flux steady state is not useful for geomorphologists interested in evolution of mountain ranges consisting of drainage basins. Valley floors attain steady-state longitudinal profiles quickly where unit stream power is large. Hillslopes do not attain constant slope length and relief because huge mass and diffused erosive power cause endless response time.

Disparity between stream and hillslope erosion rates results in two types of non-steady state mountains. Relief increases in the tectonically active Southern Alps of New Zealand as ridgecrests rise above downcutting stream channels with equilibrium characteristics. Relief decreases in less tectonically active watersheds, such as the Appalachians and Sierra Nevada of California during the Cenozoic. Erosion rates are greater on ridgecrests than on valley floors graded to a constant base level.

Ridgecrests and stream channels are hillslope boundaries in non-steady-state ~100 and ~1,000 ky California Coast Range watersheds. Ridgecrest erosion lags far behind midslopes and footslopes of tectonically active basins. Convex slopes and high denudation rates prevail. Mean slope increases to about 0.99 (45°) near the basin mouth. Most sediment yield from tectonically inactive basins comes from midslopes, because footslopes are surfaces of transportation graded to valley floor base levels. Mean slope decreases to only 0.15 (8°) near the basin mouth.

Crustal-dynamics equilibrium is not topographic equilibrium. Steady-state may be a convenient assumption for models of landscape evolution, but we will learn more by studying constant change in response to tectonic and climatic controls.