Journal of Sedimentary Research, 2006, v. 76, 889–902 Research Article DOI: 10.2110/jsr.2006.069 EXPERIMENTS ON SELF-CHANNELIZED SUBAQUEOUS FANS EMPLACED BY TURBIDITY CURRENTS AND DILUTE MUDFLOWS BIN YU, 1 ALESSANDRO CANTELLI, 2 JEFF MARR, 3 CARLOS PIRMEZ, 4 CIARA ´ N O’BYRNE, 4 AND GARY PARKER 2 1 Institute of Mountain Hazards and Environment, Chinese Academy of Science, P.O. Box 417, Chengdu 610041, China 2 Department of Civil and Environmental Engineering and Department of Geology, University of Illinois, Urbana, Illinois 61801, U.S.A. 3 St. Anthony Falls Laboratory, University of Minnesota, Minneapolis, Minnesota 55414, U.S.A. 4 Shell International Exploration and Production Inc., 3737 Bellaire Boulevard, Houston, Texas 77025, U.S.A. e-mail: parkerg@uiuc.edu ABSTRACT: The passage of turbidity currents over submarine fans often results in intense channelization. The channels are typically bounded by levees, and they vary from mildly to strongly meandering. The process of self-channelization remains, however, somewhat obscure. Preliminary experiments reported here demonstrate for the first time that self-channelization of subaqueous fans by turbidity currents can be reproduced at laboratory scale. The same experiments also illustrate self- channelization by dilute mudflows. The resulting weakly sinuous channels can be predominantly depositional, predominantly erosional, or some combination of the two. The channels can elongate to the length of the entire reach available for their formation. They can show both gradual shift and avulsion. A necessary condition for the formation of intricate channelization in the laboratory experiments appears to be a turbidity current that is insufficient to cover the entire area of the fan at any given time. Loose field analogs to the experimental channelized fans (i.e., analogs that are imprecise but contain many similarities) can often be found at the distal end of submarine fans. Two possible examples are cited, one from the Pochnoi system, Bering Sea, and one from the Gulf of Cadiz. INTRODUCTION Submarine fans are major deep-water sinks for terrestrial sediments brought across the continental shelf and slope. Many such fans are intensely channelized; the channels are often strongly meandering and bounded by well-developed levees. Examples of channelized submarine fans include the Amazon Submarine Fan (Pirmez 1994; Pirmez and Flood 1995), the Mississippi Submarine Fan (Twitchell et al. 1991) and the Bengal Fan (Schwenck et al. 2003). A levee-bounded meandering channel on the Amazon Submarine Fan is shown in Figure 1. It is generally believed that meandering, leveed channels on submarine fans are created by the passage of turbidity currents. Direct evidence for this is provided by the field measurements of Hay (1987a, 1987b) of turbidity currents in a meandering, leveed channel in Prince Rupert Inlet, British Columbia. The turbidity currents were created by the continuous disposal of mine tailings into the inlet; the meandering, leveed channel was constructed from deposited tailings. Imran et al. (1998) have proposed a mechanism for the self-channelization of turbidity currents through levee construction, and Imran et al. (1999) and Peakall et al. (2000) have also described how these channels might develop meander bends, Parsons et al. (2002) presented experimental results that demonstrate the role of lobe switching in submarine channel formation. The process by which a turbidity current flowing over an initially featureless surface self-channelizes so as to create the distributary network of sinuous leveed channels characteristic of submarine fans remains, however, incompletely understood (e.g., Imran et al. 1998). Here preliminary results demonstrate for the first time that turbidity currents at experimental scale can deposit subaqueous fans that are intricately channelized by low-relief, mildly sinuous channels. It is suggested that these experimental channelized fans not only bear superficial resemblance to weakly channelized distal regions of some submarine fans in the field but also may be created by a somewhat analogous mechanism. It should be pointed out that the flows reported here contain elements of both turbidity currents (i.e., dilute suspensions lacking any yield strength) and rather dilute mudflows (i.e., less dilute suspensions with some measurable yield strength). This issue is discussed and quantified in some detail below. The use of small-scale experiments to study submarine channelization has a precedent in the recent work of Me ´tivier et al. (2005). They introduced very small, laminar flows of saline (and thus dense) water over a submerged surface composed of lightweight plastic particles. The mode of sediment transport was essentially bedload, as could be seen through the transparent saline underflow. Under the right conditions the laminar flow incised to form a meandering channel of low sinuosity. The downstream end of the meandering channel consisted of a downstream- migrating depositional frontal lobe constructed from the material eroded from upstream. ‘‘Levees’’ could be formed as incision worked its way through an abandoned portion of a lobe. The preliminary experiments reported here are different from those of Me ´tivier et al. (2005) in several important ways: (1) the flows reported here are driven by suspended sediment, not dissolved salt, (2) the longest channels reported here are formed by deposition, not incision, and (3) the present experiments show distributary networks of channels rather than single channels. This notwithstanding, both sets of experiments show the Copyright E 2006, SEPM (Society for Sedimentary Geology) 1527-1404/06/076-889/$03.00