Douglas J Jerolmack

Department of Earth and Environmental Science

Douglas J Jerolmack

Research Interests - River Patterns

Research in this area focuses on the mechanics of sediment transport that give rise to the formation of patterns at various scales in rivers - from ripples ro river deltas. This work involves equal parts field, laboratory and modeling approaches.

Bed forms. Ripples and dunes are ubiquitous in alluvial rivers and represent a fundamental instablity at the sediment-fluid interface. Besides being a compelling pattern formation problem, bed forms are (1) the major form of flow resistance in many rivers and (2) the most common features used to reconstruct flow conditions from ancient river deposits preserved in rock. We study the formation and evolution of bed forms in the field (left) and laboratory in order to motivate and constrain dynamical mathematical models (right).

Collaborators: David Mohrig and Brandon McElroy (UT-Austin). Scale of image approximately 10 m across.

Braiding and branching. Rivers in non-cohesive sediments exhibit a braiding pattern, in which many small threads ceaselessly shift back and forth within a single well defined channel (the braid plain). This braiding pattern results from the deposition of bars, and as such is an instability of coupled fluid and sediment transport within the channel. Many depositional rivers also exhibit large-scale branching (anabranching or anastomosing) in which several well-defined channels are separated by stable floodplain. Ongoing research by our group indicates that branching results from the process of avulsion - the rapid abandonment by a river of its channel, in favor of a new path at lower elevation. We are actively researching the conditions that lead to avulsion, and how this large-scale instability generates various river patterns. This research has a large field component involving the Niobrara River in Nebraska, which is shown in pictures: top left - branching pattern caused by avulsions; top right - braiding pattern; bottom left - a bifurcation created by an avulsion. The bottom right shows snapshots of a cellular model for river avulsion that generates branching patterns.

Collaborators: Paul Heller (U. Wyoming), David Mohrig (UT-Austin) and Chris Paola (U. Minnesota)

River Deltas. River deltas develop at the land-sea interface under the competing influence of fluvial (river) and marine (ocean) processes. As such, deltas are a critical component in the source- to-sink system, as they sequester terrigenous (land-derived) sediment and modulate propagation of tectonic and climatic signals to the marine environment. The divergence of riverine flux that drives sedimentation on deltas is manifest in a network of distributary channels on the delta surface. Networks are highly variable in channel number and pattern. Few studies have addressed the morphodynamics of distributary networks, and no genetic model can explain observed network variability. This reflects the complicated nature of the problem: Network evolving processes are inherently stochastic and operate on ‘meso’ timescales (centuries – millennia) that are sufficiently long to complicate human observation but sufficiently short to escape reliable preservation in the stratigraphic record. One way to circumvent this limitation is carefully-controlled laboratory experiments developed in tandem with physically-based mathematical models. Our group is performing experiments on the evolution of deltas in non-cohesive and cohesive sediments to understand the formation of channel networks. We have constructed a simple mathematical model based in the idea that two fundamental processes act to create distributary channels. The first process is mouth-bar deposition at the shoreline and subsequent channel bifurcation; the second is avulsion. The former creates relatively small networks with power-law (fractal) distributions of channel length; the latter generates relatively few, large-scale distributaries.

Figure: Examples of river deltas. The deltas on the left have little or no wave influence, while the deltas in the right column have varying degrees of wave influence. The long channels were formed by avulsion, while the smaller channels formed by the repeated process of deposition of a mouth bar causing a bifurcation. Waves suppress the formation of smaller-scale channels

Figure: Left: Cartoon illustrating our mathematical model of the unit processes of distributary formation. (a) Expansion of a turbulent jet at the delta shoreline creates a mouth bar downstream of a channel tip. (b) Successive progradation of leveed channels and bifurcation due to mouth-bar deposition creates a branching (fractal) distributary network. (c) Aggradation leads to avulsion and the abandonment of the old channel network (dark gray) and creation of a large-scale avulsive distributary of length L. Each avulsive distributary spawns a new fractal distributary network at its tip. Abandoned channels on the delta plain are erased to floodplain deposition.
Right: Experimental river delta created in the EXperimental Earthscape Facility (XES) at Saint Anthony Falls Laboratory, University of Minnesota - photo courtesy of Chris Paola.

Collaborators: John Swenson (U. Minnesota-Duluth), Chris Paola (U. Minnesota), Wonsuck Kim (U. Illinois), John Martin (ExxonMobil)

Further Information

Research

Publications

Courses

Penn Sediment Dynamics Laboratory

Back to Profile

Department of Earth and Environmental Science
University of Pennsylvania, 254-b Hayden Hall, 240 South 33rd Street
Philadelphia, PA 19104-6316