Icarus 145, 533–545 (2000) doi:10.1006/icar.2000.6346, available online at http://www.idealibrary.com on Two Different Supply Styles of Crater Outflow Materials on Venus Inferred from Numerical Simulations over DEMs Hideaki Miyamoto 1 and Sho Sasaki Geological Institute, School of Science, University of Tokyo, Tokyo 113-0033, Japan E-mail: miyamoto@geosys.t.u-tokyo.ac.jp Received January 4, 1999; revised November 15, 1999 The three-dimensional numerical simulation code for non- isothermal Bingham fluid was applied to fourtypical crateroutflows on Venus to estimate emplacement styles of outflow materials. We have succeeded in reproducing emplacement features qualitatively over digital elevation models (DEMs) obtained by the Magellan spacecraft, and we have shown that outflow materials can be formed by melts of basaltic crust. Therefore, the presence of crater outflows may not be evidence of the presence of previously suggested fluid materials, such as komatiite or carbonatite, although we failed to model the very low viscosity flow. Numerically estimated supply rates of flow materials are categorized into two types, a catastrophic style and a gentle style. Markham and Isabella crateroutflows are examples of catastrophic styles, and the flow materials required to form them were probably supplied at higherrates than 10 10 m 3 /s and the durations of material supply were less than 100 s. This very high supply rate is difficult to attribute to any volcanic process because typical eruption rates of terrestrial volcanic flows are less than 8 or 9 orders of magnitude lower than that. Such a style of emplacement is more plausibly considered as a result of the impact itself, for example, a catastrophic debris flow or a vapor cloud trig- gered by the impact. On the other hand, supply rates of Willard and Xantippe outflows, which are considered gentle-style flows, were estimated to be about 10 4 m 3 /s and the durations were more than 10 5 s. Such a gentle style of outflow emplacement is probably not a direct result of the impact but is ratherdue to a secondary pro- cess, possibly segregation and drainage of impact melt materials from within previously emplaced ejecta materials, or post-impact volcanism. c  2000 Academic Press Key Words: crater outflow; FEB; crater; ejecta; Venus; Venus, surface; impact processes; volcanism. INTRODUCTION On Venus, about one-half of the impact craters are accom- panied by radar bright lava-like deposits. They are called crater outflows (e.g., Asimow and Wood 1992, Chadwick and Schaber 1993), fluidized ejecta blankets (FEBs) (e.g., Komatsu et al. 1 Current address: Department of Geosystem Engineering, University of Tokyo, Tokyo 113-8656, Japan. 1991, Johnson and Baker 1994), or run-out flows (e.g., Sugita and Schultz 1996). These different terms are defined according to the different ideas about their origin. We use crater outflows here. They have thin, long, and sometimes lobate features that can be clearly distinguished from normal ballistic ejecta, not only by their length but also by their characteristic flow features. Al- though martian fluidized ejecta blankets (rampart craters) and lunar deposits near crater rims have morphologies similar to crater outflows, they are quite different in size and distribution from the crater outflows. The presence of such crater outflows only on Venus may imply environments or surface materials par- ticular to Venus. Kargel et al. (1994) suggested that extensive emplacement areas of outflows were due to their water-like rhe- ologies, similar to those of carbonatites. And they also advocated that the crater outflow was evidence for the presence of a carbon- atite aquifer. However, their suggestion is based on the idea that the laterally and longitudinally extensive features of outflows may be the result of their low viscosities. Vast emplacement areas may indicate low viscosity of emplaced materials, but a recent numerical study of lavas suggests that the emplacement area is not a direct indicator of flow rheology (Miyamoto and Sasaki 1998). There have been several attempts to estimate rheo- logical properties of crater outflows. Asimow and Wood (1992) tried to analyze some outflows using a simple model of isother- mal Bingham plastic flow on inclined planes. This model was first proposed by Hulme (1974) and was used for estimations of viscosity and yield strength of lava flows. They suggested that the depositional fluid may consist of fluidized solid debris, va- porized material, and/or melt. You and Raitala (1996) measured fractal dimensions of crater outflows to compare them with ter- restrial lava flows and tried to categorize types of crater outflows. Johnson and Gaddis (1996) applied the FLOW computer model of McEwen and Malin (1989) to crater outflows to demonstrate that relatively low viscosities, yield strength, and initial veloci- ties are required to duplicate the observed flow paths. But results of these works could not constrain their rheological properties well enough to determine the materials of outflow deposits. Several emplacement models have been proposed for crater outflows. They may result from segregation and drainage of melted or vaporized materials from within ejecta materials via 533 0019-1035/00 $35.00 Copyright c  2000 by Academic Press All rights of reproduction in any form reserved.