Geomorphic process domains in BC and Washington

Laurent Roberge



The current landscape of British Columbia has been created through three different processes. First, the general elevation and shape of the land was created by uplift from the subduction of the Juan de Fuca and Pacific tectonic plates as they collided with the North American plate during the Tertiary Period. Next, several periods of extensive glaciation carved out deep U-shaped valleys and rounded many mountains through the glacial buzzsaw process (Brardinoni and Hassan, 2006; Mitchell and Montgomery, 2006). These glacial processes occurred during the Quaternary Period, which is the geologic period spanning from around 2.5 million years ago to the present, although current glacier extent is minimal and only occurs at higher elevations. Finally, since the end of the last extensive glaciation (around 14,000 years ago), there have been small modifications due to erosion and deposition of sediment by diffusive, colluvial, and fluvial processes that continue today. Despite these latest processes, the glacial legacy still dominates the current morphology of British Columbia (Brardinoni and Hassan, 2006).

The extensive glaciation during the Quaternary Period caused a topographic anisotropy in British Columbia. Slopes parallel to previous glacier flow (i.e. along the longitudinal axis of glacially carved valleys) and slopes perpendicular to previous glacier flow (i.e. along the transverse axis, or valley walls) are governed by different geomorphic processes. Fluvial processes dominate along the longitudinal axis and colluvial processes dominate down the valley walls (Brardinoni and Hassan, 2006).

The maximum extent of glaciation covered the entire of British Columbia and the northern half of Washington State, USA. Except at high elevations, the climate further south was warm enough that no glaciation occurred. The mountains and hills of southern Washington were thus created by a different sequence of geomorphic processes than those in BC. Without the Quaternary glaciation, fluvial processes dominated throughout and no anisotropy exists (Mitchell and Montgomery, 2006).

We can better understand how these different geomorphic processes shape the landscapes described above by examining drainage basins. Drainage basins are made up of unchanneled hillslopes and stream channels, within which different geomorphic processes occur. A region in which a specific process dominates is called a geomorphic process domain (Brardinoni and Hassan, 2006).

Generally, diffusive hillslope processes occur in the steep area above the stream channel head, colluvial processes such as debris flows dominate in the upper stream channel areas where slope is still relatively steep, and fluvial processes dominate in the lower reaches, below the debris flow fans, where slope is relatively shallow. This shows an interesting phenomenon: the slope is constantly decreasing from the top of a drainage basin to the mouth of a stream. The contributing drainage area is the opposite; it increases lower down the stream channel. Using these observations, geomorphologists have developed models that approximate the slope of a stream channel at a given point as a function of its drainage area. Brardinoni and Hassan (2006) show this relationship using the equation:

                                                                                    S=KsA^(-θ)                      (1)

where S is slope at a given point, Ks is the channel steepness index and θ is the channel concavity index. This equation assumes a steady state landscape where uplift and erosion are equal.

Slope-area plots can be created to show the relationship in equation (1) and to better understand the processes that shape landscapes. These graphs plot the slope at each point along a line from the top of a drainage basin to the stream channel head and then down the stream channel (y-values) against the contributing drainage area for each point (x-values). The line created in a slope-area plot has several kinks in it (Figure 1), each occurring at the boundary of different geomorphic process domains, because of the different slope gradients associated with these processes. A logarithmic scale is used for both axes for two reasons. First, the slope-area relation is a power function, so the relationship is shown as a straight line in a logarithmic plot. Second, this scale is used because it increases the visual significance of smaller values and decreases that of large values. This is important as the most changes in geomorphic process domains occur in the upper reaches where the contributing area is very small, and there is almost no change in the lower reaches, where the contributing area becomes large. Both of these give rise to better graphical visualization of the different process domains.


Figure 1. Representation of slope-area plot emphasizing line kinks and geomorphic process domain boundaries (from Brardinoni and Hassan, 2006).

Assessing slope-area plots and identifying process domains from GIS data can be used for understanding landscape evolution and is also very useful for watershed management. Brardinoni and Hassan (2006) say it is “critical for predicting patterns of natural and anthropogenic disturbances, as well as in-stream habitat conditions.”

Other than the difference in Quaternary history, parts of British Columbia and Washington are very similar. There are similar patterns of precipitation and climate, as well as similar tectonic activity. In this study I create slope-area plots of previously glaciated areas in BC and non-glaciated areas in Washington. From these plots I identify the many geomorphic process domains and compare the differences between the two landscapes. I also analyze the topographic anisotropy found in British Columbia.


Background image from: <>