104 lnternationalColloquiumon Vetum The enhanced uprange rim/wall collapse illustrated in Fig. 2b (and numerous other large oblique impacts on Venus) provides insight for why most craters exhibit acircular outline even though early-time energy transfer comprises a larger fraction of crater growth. Failure of the uprange rim/wall in response to the over,steepened wall and greater floor depth circularizes crater outlines. The rectilinear and conjugate scarp on the pattern uprange rim, however, indicates failure along prcimpact stresses. Hence, a corollary is that peak shock levels and particle motion may be reduced uprange during oblique impacts due to the downrange motion of the impactor, analogous to time dilation. References: [1] Gault D. E. and Wedekind J. A. (1978) Proc. LPSC9th, 3843-3875. [2] Moore H.J.(1979) U.S. Geol. Surv. Prof. Pap. 812-13, 47 pp. [3] Gauh D. E. and Wedekind J. A. (1977) In Impact and Explosion Cratering (D. Roddy etal., eds.), 123 I-1244, Pergamon, New York. [4] Holsapple K. A. and Sehmidt R. M, (1987) JGR, 92,6350-6376. [5] Schultz P. H, and Oault D. E.(1991) Meteoritics, 26. [6] Wichman R. W. and Schuhz P. I4.(1992) LPSC XXIII, 1521-1522. [7] Schultz P. H. (1988) In Mercury (F. Vilas et al., eds.), 274--335, Univ. of Arizona, Tucson. [8] Schultz P. H. (1992)JGR, inpress. /" i N93-14372 ii, 3/L) IMPACT-GENERATED WINDS ON VENUS: CAUSES AND EFFECTS. Peter H. Schultz, Brown University, Department of Geological Sciences, Box 1846, Providence RI 02912, USA. The pressure of the dense atmosphere of Venus significantly changes the appearance of ejecta deposits relative to craters on the Moon and Mercury. Conversely, specific styles and sequences of ejeeta emplacement can be inferred torepresent different intensities Fig, 2. (a) Crater(50 km diameter) exhibiting s partialcentral peak ring of atmospheric response winds acting over different timescales. offset uprange (from lower right) CI-60N 263. Radar look direction isfrom Three characteristic timescales can be inferred from the geologic upper left. (b) Larger enter (103 km diameter) exhibiting central peak ring record: surface scouring and impactor-controlled (angle and direc- offset downrange from present rim, opposite to occ_rmnr._ in(a)andFig. 1. tion) initiation of the long fluidized run-out flows; nonballistie Reversal in position is related to _ahanced tim/wall collapse upnmge that emplacement of inner, radar-bright ejecta facies and radar-dark widens and circularizes thecrateraround the deepest portion ofthe transient outer facies; and very late reworking of surface malerials. These crater cavity, which occurs uprange. Further crater widening follows pre- three timescales roughly correspond to processes observed in labo- existing structural grain. CI-30N 135. ratory experiments that can be scaled to conditions on Venus (with appropriate assumptions): coupling between the atmosphere and earlytime vapor/melt (target and impactor) that produces an intense IMPACT ANGLE shock that subsequently evolves into blast/response winds; less 20 ° 30 ° 50 ° 70 ° energetic dynamic response of the atmosphere to the outward- 1.0 moving ballistic ejecta curtain that generates nonthermal turbulent • lufbulent run-out flows eddies; and late recovery of the atmosphere to impact-generated xlaminar run-out flows thermal and pressure gradients expressed as low-energy but long- "2 lived winds. These different timescales and processes can be viewed D as the atmosphere equivalent ofshock reel ting, material motion, and 0.5 far-field seismic response in the target. O Early Processes (Direct EffecL_ of Blast and Fireball): Under _d vacuum conditions, the fate of the impactor is generally lost; even on the Earth, most impact melt sheets exhibit little trace of the impactor. The dense atmosphere of Venus, however, prevents 0,0 loL 05 oo escape of the impactor through rapid deceleration of ricochet debris and containment of the vapor cloud [1,2]. Figure la illustrates the Log (sin2e) time required for the atmospheric blast front to decelerate to the speed of sound asafunction ofcrater size, where kis the fraction the Fig. 3. Effect of impact angle (from horizontal) onthe transverse diameter initial impactor energy (KEi) coupled to the atmosphere (EA). On of the central peak ring depreferenced tocrater diameter. As impact angle decreases (based on degree of eject_ symmetry), size of central peak ring Venus, the shock front dissipates before the crater finishes forming. becomes larger relative tocrater diameter. Such atrendisexpected ifcentral If the blast iscreated by deceleration and containment of early high- peak ring reflects the size of impactor and cratering efficiency decreases speed ejecta (downrange jetting and ricochet/vapor), then it will impact angle. precede ejecta emplacement and should exhibit asource area offset LPI Contribution No. 789 105 100 • - • • ,, i • i ..... k=E_KEi f k=E^/KEI i/erl_///l tu M,,,, _<_/llJ.Z ,-- 10 >t- 10 rrz tuO > I.tl_ (.3 tr 0 ,,rtr V--LL X/iS, ° Cl'liter ,orirllliiorl urider -'-TY-- _"I'<".°'atarcter_g"" _.,,-rU'IZ/,Z'_" A oZ-3 llz 0.1 0.1 10 100 I0 100 CRATER DIAMETER (kin) CRATER DIAMETER (kin) Fig. 1. (a)Recovery time foratmospheric blastto reducetothespeed ofsound onVenus scaled tocrater formation dine. The value of kreprer_nUthefractionof initial impactor kindle energ7(KE._coupled totheatmosphere tEA).OnVenustheblasteffec_ shouldprecede craterformation. Co)Recovery time foratmospheric pressure behindthe shock fronttoreturn toambient condition_ on Venus scaled to craterformation time,.Although atmospheric pressure hasmcovem,.A,high temperatures (diefireball) result in low denzltites. Thermal gradienu andmotion of fireball induce strong recovery winds that reworkejects it late times. ,mlhmal,_ (a) *tic,'kktilii nil (b) _: ' h_ty ._ "% Illi [I/11IIIIIIII iI i i i-/_jl Ii tl i i i _ui_ _por enltid__wC,_elmt_ ... =and-off _/ • =:;';_: m'lt_r pror=k I I I i /')'_i'l---- / I I I I I do_,nrange dbfing _ " Imli_, ..W-._ llli_ci_ /,.-,._7_/ _/'/i 7---- frz,_%om,elt r_i*s do,,m ntnge run d_p_s_o_ Fig. 2. (a) Scenario for sequenceand style of ejects emplacement atearlytimes based oninferences drawnfrom laboratory impact phenomena sealed to Venus andsurface features revealed byMagellan. Atearly times, kinetic energy andmomentum inthevapor cloud evolves into adownrange-moving fireball that creates strong winds downrange. (b) Scenario for lateprocesses. Winds and turlmlence created bythe outward-moving ejects curtain entrain coarse fractions to produce anavalanchelike flow of coarse inner ejecra deposits. Such deposits persist asradar-bright ejects deposits because of thetoughness and low ambient sudace winds.Finer fractions entrained insustained turbulence result inturbidity flows with potentially much greaterran-out distmces. Deposits from these flows will bemore suceptibleto subsequenterosion. downrange from the mater [2,3]. The downrange offset of the center (Fig. lb). If all craters were formed by vertical impacts, this would of origin of the shock is observed in laboratory impact experiments. mean that the craters form within a f'treball or behind the wake Features consistent with this interpretation can also be found around characterized by low density (high temperature, low pressure) as venusian craters and include [2] topographic barriers shadowing postulated in [4]. But most craters are formed by oblique impacts surface disruption from the blast; radar-dark/-bright striations con- (i.e., 75% formed at angles of 60°or less). Consequently, the early verging on the downrange rim rather than the crater; and diffuse fireball moves away from crater excavation initially at haloes at the base of small hills again focusing on adownrange rim hypervelocities. At low impact angles (<300), the energy coupled to "'source" (shock-dislodged debris drawn back into the rarefied, the atmosphere resembles arolling fireball containing vapor and rising fireball). Moreover, radar-dark parabola patterns commonly dispersed melt moving downrange until decelerated. Because the center on apoint downrange and not the crater [2]. In contrast, the vapor/meh is higher in density than conditions in the fireball, it time for recovery of the atmosphere to ambient pressure (or density collapses within the fireball to form long run-out density flows and temperature) is much longer than the time for crater formation controlled by local topography, well in advance of ejecta emplace- 106 lntemationalColloquiumon Venus merit as observed on Venus [5,6]. Such aprocess accounts for the results in late-stsge reworking, if not self-destruction, of ejects long run-out flows eonsistendy originating downrange in oblique facies emplaced earlier. Surface expression should include bedforms impacts (i.e., opposite the missing ¢jecta sector) even if'uphill from (e.g., meter-scale dunes and deeicentirneter-scale ripples) reflect- thecrater rim. A_nosphcric turbulence and recovery winds decoupled ing eddies created in the boundary layer at the surface. Because from the gradient-controlled basal run-out flow continues down- radar imaging indicates small-scale surface roughness (as well as range and produces wind streaks in the lee of topographic highs. resolved surface features), regions affected by such long-lived low- Turbulence accompanying the basal density flows may also produce energy processes can extend to enormous dislances. Such areas are wind streak patterns. Uprange the atmosphere isdrawn inbehind the not directly related to ejecta emplacement but reflect the atmo- fireball (and enhanced by the impinging impactor wake), resulting spheric equivalent to distant seismic waves inthe target. Late-stage in strong winds that will last atleast as long as the time for crater atmospheric processes also include interactions with upper-level formation (i.e., minutes). Such winds can entrain and saltate surface winds. Deflection of the winds around the advancing/expanding materials as observed in laboratory experiments [2,3] and inferred FLrebali creates aparabolic-shaped interface aloft. This ispreserved from large transverse dunes uprange on Venus [2]. in the fall-out of finer debris forimpacts directed into the winds aloft Atmospheric Effects on Ballistic EJeeta: Even on Venus, (from the west) but self-destructs if the impact is directed with the target debris will be ballistically ejected and form aconical ejects wind. Exception to this rule occurs for larger crater (>60 kin) curtain until its outward advance is decelerated by the atmosphere. sufficient to interrupt the flow pattern not only by the fireball but The well-defined, radial ejecta delineating the uprange missing also by the ejects curtain. ejecta sector of craters formed by oblique impacts demonstrate References: [1] Schultz P. H. and Gauh D. E. (1990) In GSA ballistic control of ejection. As the inclined ejects curtain advances Spec. Pap. 247 (V. L. Sharpton and P. D. Ward, eds.), 239-261. outward, however, itcreates turbulent vortices, which have been [2] Schultz P. H.(1992) JGR, inpress. [3]Schultz P. H. (1992) JGR, observed in the laboratory experiments [2] and modeled theoreti- inpress. 1.4]Ivanov B. A. eta]. (1986)Proc.LPSC16th, inJGR, 91, cally [7]. The ejects curtain gradually becomes more vertical ia D413--IM30. [5] Schultz P. H. (1991) Eos, 73, 288. [6] Phillips response to atmospheric resistance. The atmospheric density is R. J. et al. (1991) Science, 252, 288-296. [71 Barnouin O. and sufficient todecelerate meter-sized ejecta to terminal velocities [8] Schultz P. H. (1992)LPSCXXIII, 65--66. [8] Schultz P. H. and Gault that will be entrained in and driven byresponse winds induced bythe D. E. (1979)JGR, 84, 7669-7687. [9] Schultz P. H. et al. (1981) outward-moving curtain. While larger ejecta are deposited, smaller In Multi-Ring Basins, Proc. LPS 12A (P. H. Schultz et al., eds.), size fractions become entrained in an outward ejects flow. Based on 181-195, Pergamon, New York. [10] Barnouin O. and Schultz P. H. diversion of such flows by low-relief barriers near the rims of (1992) LPSC XXIII, 65-66. [11] Schultz P. H., this volume. craters, the transition from ballistic to nonballistic emplacement [12] Jones E. M. and San ford M. T. I1(1982) In GSA Spec. Pap. 190 occurs within about 0.5 crater radii of the rim. This observation (L. Silver and P. Schultz, eds.), 175-186. [13] Sehultz P. H. (1992) underscores the fact that dynamic atmospheric pressure signifi- JGR, in press. [14] Schultz P. H. and Gault D. E. (1982) In GSA cantly restricts outward advance of the ejecta curtain. The scaled Spec. Pap. 190 (L. Silver and P. Schultz, eds.), 153-174. [15] Post run-out distance (distance from the crater rim scaled to crater R. L. (1974)AtzWL-TR-74-51. _ ^ diameter. D) of the ejecta flow should decrease on Venus as D-°.s. ,;N93-14373 unless consumed by crater rim collapse. Because of the high atmospheric density, collapse of near-rim ejecta into aflow crudely MAGELLAN PROJECT PROGRESS REPORT. J.F. Scott, rescmbles an avalanche comprised of coarsc debris and blocks. But D. G. Griffith, J. M. Gunn, R. G. Piereson, J. M. Stewart, A. M. high winds and turbulence created by the outward-moving curtain Tavormina, and T. W. Thompson, Jet Propulsion Laboratory, separate during terminal emplacement of the inner flow, thereby California Institute of Technology, Pasadena CA 91109, USA. winnowing the freer fractions and creating anoverrunning turbidity flow that continues outward. The Magellan spacecraft was placed into orbit around Venus on Turbidity flows containing freer fractions can extend to much August 10, 1990 and started radar data acquisition on September 15, larger distances until turbulence supporting entrained debris no 1990. Since then, Magellan has completed mapping over 2.75 longer can support thc load. Because turbulent wind velocitics rotations of the planet (as of mid-July 1992). Synthetic aperture gready exceed ambient surface winds, such vortices are also capable radar (SAR), altimetry, and radiometry observations have covered of mobilizing surface materials. It is suggcsted that the radar-dark 84% of the surface during the first mission cycle from mid- lobes cxtcnding beyond the inner radar-bright ejecta [2,6] reflect September 1990 through mid-May 1991. this process. In addition, many craters arc surrounded by a very Operations in the second mission cycle from mid-May 1991 diffuse boundary that masks low-relief ridges and fractures; this through mid-January 1992 emphasized filling the larger gaps (the boundary may indicate thc limits of athird stage of flow scparafion south polar region and asuperior conjunction) from that f'wstcycle. and deposition. The observed radar-dark signature requires such An Orbit Trim Maneuver (OTM) was performed atthe beginning of ejecta to be less than afew centimeters. In contrast with the coarse, cycle 2 in order to interleave altimeter footprints atperiapsis. This radar-bright inner facies, the outer radar-dark facies will be more yielded better altimetric sampling of the equatorial regions of susceptible to later erosion by ambient or other impact-generated Venus. Some 94% of the planet was mapped at the end of mission winds because the size fractions were sorted by a similar process. cycle 2. This is consistent with observed removal or reworking of craters Observations in the third mission cycle from mid-January to believed to beold, based on superposed tectonic features. mid-September 1992 emphasized reimaging of areas covered in Late Recovery Winds (Secondary Effects of Atmospheric cycle 1and cycle 2 such that digital stereo and digital terrain data Turbulence): On planets without atmospheres, the effects of products can be produced. Atransponder anomaly in January 1992 early, high-speed ejecta and impactor are typically lost. On Venus, (just before mission cycle 3started) forced the project touse aradar however, the dense atmosphere not only contains this energy data downlink of 115 Kbs instead of 268 Kbs. Although data fraction, but the long recovery time of the atmosphere (Fig. 1b) acquisition iscurtailed, some 30--40% of the planet will be mapped