B4 · Publication Volume 9

Geomorphic Systems and Landscape Evolution

relief, base level, uplift, erosion and landscape history

Landscape mass balance, base level and competing histories
Landscape mass balance, base level and competing histories

Learning objectives

After this lesson, you should be able to define a geomorphic system by boundaries, stores and fluxes; distinguish absolute elevation, local relief and base level; explain how uplift, incision, weathering and deposition can produce similar landforms; use a simple stream-power relation without treating it as a universal law; and construct competing landscape histories from terraces, palaeosurfaces, weathering profiles and drainage geometry.

Start with a field problem

A broad upland surface is capped by iron-rich gravel. Deep valleys cut the surface, while isolated remnants occur at different elevations. Does the cap mark one once-continuous palaeosurface displaced by later deformation, several surfaces formed at different times, a transported gravel sheet draped across older relief, or convergent weathering on resistant lithologies?

Present elevation alone cannot decide. A terrace may record river incision, aggradation followed by re-incision, lateral migration or tectonic tilting. A low-relief summit may be an erosion surface, a structural bench, a lava or duricrust cap, or the surviving part of a buried surface. Begin with geometry and material continuity, then ask what sequence and boundary conditions are required by each hypothesis.

Core process model

A geomorphic system is a chosen volume with inputs, outputs, stores and internal transformations. Its boundary might be a catchment divide, coastline, fault block or mapping window. The choice controls the budget. Material crossing a catchment outlet is export; the same material is input to the downstream reach.

Local relief is the elevation difference over a declared neighbourhood. Base level is the limiting level toward which a process can lower a surface; it can be sea level, a lake, a resistant reach, a confluence or a local sediment-choked valley. Base level can rise, fall or migrate. A knickpoint may propagate upstream after a base-level fall, but it may also be fixed by resistant rock, a structure or a sediment transition.

For detachment-limited channel incision, a commonly used conceptual form is


E=K\left(\frac{A}{A_0}\right)^m S^n,

where E is incision or erosion rate, K groups erodibility and forcing, A is contributing area, A_0 is a reference area, S is channel slope, and m and n are model exponents. The relation expresses how discharge proxy and slope may combine; it does not account automatically for sediment cover, thresholds, width change, lithologic contrasts or transient floods.

Landscape state reflects inherited conditions as well as current forcing. Uplift can increase potential relief; weathering weakens substrate; transport evacuates material; deposition buries or protects a surface; cementation can reverse erodibility. Feedbacks matter: incision steepens adjacent slopes, slope supply may armour a channel, and vegetation or soil changes infiltration and runoff.

Evidence and measurement

Measure elevations against a declared vertical datum, and do not compare heights from incompatible reference surfaces. Derive relief at more than one neighbourhood size. Map terraces by tread, riser, sediment, strath and continuity rather than height alone. Log the material under a flat surface to distinguish bedrock bench, residual profile and deposited fill.

Drainage evidence includes network topology, abrupt bends, elbows of capture, barbed tributaries, wind gaps, underfit valleys, knickpoints and longitudinal profiles. None is unique. Test each against lithology, structure, mapped sediment and scale. Cosmogenic nuclides, luminescence, palaeomagnetism, geochronology of volcanic or authigenic minerals and biostratigraphy can constrain rates or ages, but every method dates a specific material and event rather than “the landscape” in general.

A robust palaeosurface correlation requires more than equal elevation. Compare substrate, regolith stratigraphy, cement generation, clast provenance, geomorphic position and independent age constraints. Express uncertain correlation as alternatives on the map.

Worked example

Use a synthetic channel cell for instruction. Let A=25 km², A_0=1 km², S=0.018, K=12 m/Ma, m=0.5 and n=1. Then


E=12(25)^{0.5}(0.018)=1.08\ \text{m/Ma}.

If the imposed uplift term is 2.00 m/Ma, the simplified local balance \partial z/\partial t=U-E predicts 0.92 m/Ma of relative surface rise. This is not a forecast for a real channel. It assumes steady area and slope, no sediment shielding, a compatible time-averaged K and no lateral migration. If the reach contains thick alluvium, a transport-limited model may be more appropriate.

Two histories can honour the same high surface. Model A correlates all cap remnants as one old surface later dissected. Model B treats the lowest remnants as younger valley fills that acquired similar cement. A testable prediction is that Model A should show comparable basal unconformity and weathering sequence across elevations, whereas Model B predicts different provenance or depositional fabric in the lower remnants.

Misinterpretations and uncertainty

Do not infer uplift rate by subtracting two present elevations and dividing by an assumed age. Rock uplift, surface uplift, exhumation and river incision are related but different. Do not treat planation as proof of long tectonic stability; low relief can form or persist under several combinations of erosion, deposition and resistant capping.

Power-law exponents fitted in one setting may absorb width, runoff, sediment and lithology effects. Extrapolation beyond the calibrated domain can create false precision. Mapping resolution changes apparent slope and drainage density. Correlative surfaces are especially vulnerable to circular reasoning when their assumed equivalence is also used to infer deformation.

Practical investigation

Map a synthetic or field landscape at three scales. At each scale, identify system boundary, relief measure, likely base levels, material stores and export paths. Construct two longitudinal river profiles and mark knickpoints, tributary junctions, lithologic boundaries and sediment reaches. Write at least three alternative explanations for each major break in slope.

Build two landscape-history panels with the minimum events needed to explain the observations. Use arrows only for explicitly named fluxes. Mark each correlation as observed, calculated or inferred and nominate one sample, exposure or survey that would discriminate the histories.

Mastery check

  1. Why does base level not always mean sea level?
  2. Which assumptions are hidden inside a stream-power calculation?
  3. What additional evidence is needed to correlate two flat surfaces?
  4. How can deposition preserve an old weathering profile?
  5. Which observation would distinguish a migrating knickpoint from a lithologically fixed step?

Sources and further reading