Exploring the roles of discharge and bedload composition in strath terrace formation through a naturally occurring experiment in the California Coast Ranges

Introduction

Pancho Rico Creek is a tributary of the Salinas River that flows across an area of modestly high topography known as the Gabilan Mesa. The Gabilan Mesa is in the central Coast Ranges of California, and is an exceptional locality to study stream terraces because variables that influence stream processes are well known and relatively simple (Dohrenwend, 1978). Stream capture (Figure 1) occurred when San Lorenzo Creek eroded laterally and breached its southwestern drainage divide (Garcia and Mahan, in review). The re-routing of water and sediment from San Lorenzo Creek into Pancho Rico Creek makes possible studying the effect of an abrupt change in the fundamental factors influencing stream-terrace formation in a natural setting. This is a significant "experiment" because, unlike experiments conducted in laboratory flumes or using numerical simulations, which can yield unrealistic results, the results of this experiment are produced by real geomorphic processes.

Definition and significance of strath terraces

Strath terraces are the type of stream terraces that form in tectonically active mountains, especially in drainage basins underlain by fine grained, marine sedimentary rock (Montgomery, 2004). Straths are planar, erosional surfaces that act as floodplains for bedrock streams, and they form adjacent to stream channels when rates of lateral stream erosion exceed rates of downward stream erosion (Montgomery, 2004; Figure 2). A strath terrace consists of a strath, as well as alluvium that overlies the strath, and strath terraces are created when a stream incises sufficiently below the strath so that it is no longer affected by fluvial processes (Figure 2). Strath terraces are significant because the straths at the bases of strath terraces represent a former stream channel altitude, and the numeric age of the overlying alluvium can be determined using techniques such as radiocarbon dating of organic matter (e.g. Wegmann and Pazzaglia, 2002), or optically stimulated luminescence dating (e.g. Garcia and Mahan, 2009). It follows that strath terraces can be used to calculate rates of downward stream erosion, which in some instances are equal to rates of rock uplift (e.g. Pazzaglia and Brandon, 2001). Accordingly, strath terraces have been intensely studied during the last two decades, and some controversies exist.

Factors that influence strath terrace formation

The seminal paper by Montgomery (2004) provides a new model explaining how and why strath terraces form. This model is based on the observation that straths form most commonly in drainage basins composed of fine grained, clay rich, marine sedimentary rock, because clay-rich sedimentary rocks are highly susceptible to mechanical weathering caused by shrinking and swelling associated with periodic wetting and drying. When clay-rich sedimentary rocks are exposed on a strath (Figure 1), they rapidly disintegrate, and the weathered debris is rapidly eroded. Moreover, the rock in the channel is not subject to wetting and drying cycles and associated weathering effects, so rock within the channel is relatively hard and resistant to stream erosion. The contrast in the competence of weathered floodplain rock and unweathered channel rock insures that lateral erosion occurs at a faster rate than downward erosion. Once a strath forms, it is preserved only if it is capped by a relatively resistant cover of alluvium, which in most instances is composed of the gravel that was being transported as bedload in the stream channel (Montgomery, 2004). This compelling model requires reevaluation of the factors previously accepted as influencing to strath terraces formation, such as bedload composition and stream discharge.

Merritts et al (1994) explored the relationship of stream discharge and uplift rate on development of strath terraces. They showed that straths will form only where there is discharge in excess of what is needed for a stream to erode downward at the same rate of rock uplift, and only a stream's capacity to erode beyond what is needed to keep pace with rock uplift is expended on lateral erosion. For example, if rock uplift rates are relatively high, and discharge is relatively low, straths may not form, or straths that form will be relatively narrow. If rock uplift rates and rock type are the same farther downstream in the same stream network, where catchment area and discharge are greater, straths are more likely to form, and will be wider than straths farther upstream (stream discharge is largely dependant on catchment area, which is the amount of terrain contributing water to a segment of a stream).

One way to test the hypotheses of Merritts et al (1994) is to conduct an experiment in which the catchment area and associated stream discharge of a stream increase abruptly. The history of strath terrace formation along Pancho Rico Creek is the result of such an experiment because its drainage basin expanded approximately 50% by stream capture sometime after approximately 11,800 to 15,500 years ago (Garcia and Mahan, 2009). Moreover, the bedrock geology of the Gabilan Mesa makes possible adding another facet to the experiment, which is the assessment of the influences of bedload composition on strath terrace formation.

The experiment

When the catchment area of Pancho Rico Creek expanded through stream capture (Figure 1), the total catchment area of Pancho Rico Creek increased from 99.5 km² to 156.6 km² (Garcia, 2006). Additionally, prior to stream capture, the entire Pancho Rico Creek drainage basin was formed in fine-grained, marine sedimentary rock, and gravel bedload consisted exclusively of relatively weak, clay rich, fine grained marine sedimentary rocks. Since stream capture occurred, Pancho Rico Creek's catchment includes areas underlain by the Franciscan Complex, and gravel bedload has been dominated by relatively hard, crystalline rocks of the Franciscan Complex (chert, as well as igneous and metamorphic rocks; Garcia, 2006).

The effect of steam capture profoundly influenced the development of strath terraces along Pancho Rico Creek, which is readily apparent because pre- and post-capture strath terrace sediments have distinctly different compositions. Pre-capture terrace sediment was deposited during the Pleistocene to Holocene climatic transition, when the Pancho Rico Creek channel and Pancho Rico Creek floodplains were buried by extensive debris flows (Garcia and Mahan, 2009). The surfaces of pre-capture terraces are the remnants of the sedimentary deposit that filled the Pancho Rico Creek stream valley during the Pleistocene to Holocene climatic transition, therefore, pre-capture terraces are paired terraces (Figure 3; Garcia, 2006; Garcia and Mahan, 2009). On the other hand, post-capture strath terraces consist of relatively thin (5 m or less) Pancho Rico Creek channel gravel and floodplain deposits, are at many different heights above the Pancho Rico Creek channel, and are unpaired terraces that constitute an uninterrupted record of alternating lateral erosion and downward erosion by Pancho Rico Creek since stream capture occurred (Garcia and Mahan, 2009). A complete record of alternating lateral and vertical erosion by Pancho Rico Creek, which consists of straths at many different levels above the Pancho Rico Creek channel, is preserved because the alluvial covers protecting the straths are composed of relatively hard, post-capture, Franciscan-Complex derived bedload (Garcia and Mahan, 2009). The relatively strong gravelly covers protect the straths from weathering and erosion, insuring their preservation in the alluvial stratigraphic record. In pre-capture time, straths cut by Pancho Rico Creek were preserved only when buried under substantially thick (up to 20 m) alluvial covers associated with a significant climatic event (the Pleistocene to Holocene climatic transition), because the pre-capture bedload covering the straths was relatively weak (Garcia and Mahan, 2009). Lastly, in both pre- and post-capture time, straths formed only where Pancho Rico Creek's catchment area exceeds/exceeded approximately 60 km² (Figure 4), because that is the size of catchment area needed to provide Pancho Rico Creek with sufficient discharge to erode downward at the same rate as rock uplift and also erode laterally (Garcia, 2006).

• Garcia, A. F., 2006. Thresholds of strath genesis deduced from landscape response to stream piracy by Pancho Rico Creek in the Coast Ranges of central California. American Journal of Science 306, 655-681.
• Garcia, A. F., and Mahan, S. A., 2009. Sediment storage and transport in Pancho Rico Valley during and after the Pleistocene-Holocene transition, Coast Ranges of central California (Monterey County). Earth Surface Processes and Landforms 34, 1136-1150.
• Garcia, A. F., and Mahan, S. A., in review. Geomorphic evidence for latest-Quaternary time tilting of the Gabilan Mesa, central Coast Ranges of California. Submitted to Earth Surface Processes and Landforms.
• Merritts, D. J., Vincent, K. R., Wohl, E. E., 1994. Long river profiles, tectonism, and eustasy: a guide to interpreting fluvial terraces. Journal of Geophysical Research 99 (B7), 14,031-14,050.
• Montgomery, D. R., 2004. Observations on the role of lithology in strath terrace formation and bedrock channel width. American Journal of Science 304, 454-476.
• Pazzaglia, F. J., Brandon, M. T., 2001. A fluvial record of long-term steady-state uplift and erosion across the Cascadia forearc high, western Washington State. American Journal of Science 301, 385-481.