sign in sign up
<<
Home , Gale (crater)

Constraints on 's recent equatorial wind regimes from interior layered deposits and comparison with general circulation model results E. Sefton-Nash1 , N. A. Teanby 2, C. Newman3 , R. A. Clancy 2, M. I. Richardson3 1. Department of and Space Sciences, University of California Los Angeles, 595 Charles Young Drive East, Los Angeles, CA 90095, USA. [email protected] 2. School of Earth Sciences, University of Bristol, Queen’s Road, Bristol, UK, BS8 1RJ. 3. Ashima Research, 600 S. Ave., Pasadena, CA 91106, USA.

Introduction Aeolis Crater (west) Candor In some sites yardangs showed a 1 1 1 10 10 10 A B distinctive teardrop shape (e.g. Aeolian features on Mars are transient on a range of timescales. 2 2 2 Figure 3B), allowing inference of a 0 0 0 100 Ma 10 100 Ma 10 100 Ma 10 unique wind direction. In other The orientations of features such as and yardangs are controlled by the prevailing −1 −1 −1 sites, yardangs with more elongate 10 10 Ma 10 10 10 Ma

wind regime over thier respective intervals of formation. No. Craters/km No. Craters/km 10 Ma No. Craters/km and parallel topography allowed 10 ka 10 ka 10 ka −2 −2 −2 10 10 10 inference only of a 180° ambigu- 1 Ma 1 Ma 1 Ma Statistical analysis of the orientations of young features allows probing of Mars's recent 100 ka 100 ka 100 ka 5 10 20 40 60 100 150 5 10 20 40 60 100 150 1 2 5 10 20 40 60 100150 ous trend line (e.g. Figure 3A). wind regime and comparison with wind elds predicted by general circulation models Diameter (m) Diameter (m) Diameter (m) (GCMs). Gordii Dorsum Crater orientation was determined

1 1 1 by measurement of the trend line We collect statistical distributions of transverse dune trends and yardang azimuths at nine 10 10 10 parallel to the mean azimuth of 2 2 2 sites on Mars (Figure 1), and compare measured feature orientations to those predicted 0 0 0 10 10 10 each dune crest (Figure 3C). Dune from interpretation of time-integrated wind vectors from the MarsWRF GCM [1, 7]. 100 Ma 100 Ma 100 Ma slip-face normals were perpen- −1 −1 −1 10 10 Ma 10 10 Ma 10 10 Ma

No. Craters/km No. Craters/km No. Craters/km dicular to this, but for transverse We focus on features atop interior layered deposits because their young surface ages [2] 10 ka 10 ka 10 ka −2 −2 −2 dunes wind direction remains 180° 10 10 10 and erodible nature makes them applicable to determination of Mars's modern wind 1 Ma 1 Ma 1 Ma C D 100 ka 100 ka 100 ka ambiguous. regime. 5 10 20 40 60 100 150 5 10 20 40 60 100 150 5 10 20 40 60 100 150 Diameter (m) Diameter (m) Diameter (m) 90 20000 Figure 2: Crater size-frequency distributions on ILDs compared to model surface-age isochrons of [5].

60 15000 We determine eective resurfacing age, and therefore maximum aeolian feature age, for ILDs at each site using small crater populations observed in HiRISE images. 30 Gordii Dorsum 10000 (North & ) Danielson Crater ArabiaArabia TerraTerra Gordii Dorsum (North(South)) Six sites harboured sucient crater populations (e.g. Figure 3E) to enable dating: Aeolis, Candor E Gale Crater Iani Chaos tion (m) 0 (East & West) GaleGaleAeolis Aeolis CraterCrater (West(East)) Candor Chasma a Chasma, Danielson Crater, Gale Crater (east), Gordii Dorsum (south) and Iani Chaos. 5000 Iani Chaos

−30 MOLA ele v Isochrons in Figure 2 are for model surface ages based on [5], which represents a renement and 0 synthesis of crater-dating techniques. This includes a correction applied to account for the loss of small ( D < 20 m ) bolides in Mars's atmosphere [6]. Error bars represent a factor of two either side of −60 Figure 3: A) Yardang with 180° ambiguous inferred wind −5000 the measured crater density, as suggested by [4]. Uncertainties are greatest if only small (D < 100m) direction B) Yardang with unique inferred wind direction craters are used [5] and we therefore interpret our crater statistics to give order of magnitude resur- (blue bars in Figure 4, except for Danielson crater, in −90 0 60 120 180 240 300 360 facing age estimates, rather than constrained model ages. which all yardang orientations are duplicated in the op- posite sense.). C) Transverse dune crests (red bars in Figure 1: Location of sites selected for this study. Base map is MOLA 32ppd topography [3]. In general, aeolian features on ILDs at these sites are younger than 0.1-10Ma. Figure 4). D) Aeolian feature population in one HiRISE image. E) Small crater population.

Results (Figure 4) Aeolis Candor Chasma Danielson Crater Gale Crater (east) Gale Crater (west) Gordii Dorsum (south) Gordii Dorsum (north) Iani Chaos 0 0 0 0 0 0 0 Dunes 330 30 330 30 330 30 330 30 330 30 330 30 330 30

Dune trend predicted by maximizing gross-bedform normal transport (GBNT) 300 60 300 60 300 60 300 60 300 60 300 60 300 60 Primary maximum: -2 Weighted by Fryberger (1979) ux with threshold = 0 Nm 30% 15 20% 15% 30% 30% 20% 15% -2 10 20 10 10 20 20 10 15 10 Weighted by Fryberger (1979) ux with threshold = 0.008 Nm No dunes counted. 5 5 10 10 5 5 270 90 270 90 270 90 270 90 270 90 270 90 No dunes counted. 270 90 -2 Weighted by Fryberger (1979) ux with threshold = 0.016 Nm Secondary maximum of signicant magnitude (Iani Chaos only): -2 Weighted by Fryberger (1979) ux with threshold = 0.008 Nm 240 120 240 120 240 120 240 120 240 120 240 120 240 120 -2 Weighted by Fryberger (1979) ux with threshold = 0.016 Nm

210 150 210 150 210 150 210 150 210 150 210 150 210 150 180 180 180 180 180 180 180

0 0 0 1 0 0 0 0 0 0 330 30 330 30 330 30 330 30 330 1 30 330 1 30 330 30 330 30 330 30 1 1 300 60 300 60 300 60 300 60 300 60 300 60 300 60 300 60 300 60 Yardangs 1 1 1 40% 30% 15% 20 30 20 10 6 8 10% 30 40% 30 40 50% 20 30% 30 40% 8 10% 1 Maximum MarsWRF vector frequency 10 10 5 2 4 10 20 10 20 10 10 20 2 4 6 270 90 270 90 270 90 270 90 270 90 270 90 270 90 270 90 270 90 2 Secondary vector frequency maximum 1

240 2 120 240 120 240 120 240 120 240 120 240 120 240 120 240 120 2402 120

2 210 150 210 150 210 150 210 150 210 2 150 210 150 210 150 210 150 210 150 180 180 180 180 180 180 180 180 180

0 0 0 0 0 0 0 0 0 330 30 330 30 330 30 330 30 330 30 330 30 330 30 330 30 330 30 Normalized frequency of MarsWRF wind vector orientation: 300 60 300 60 300 60 300 60 300 60 300 60 300 60 300 60 300 60 Unweighted

Weighted by vector mean stress, u* 1 1 1 1 1 1 1 1 1 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 Weighted by Fryberger (1979) ux 270 90 270 90 270 90 270 90 270 90 270 90 270 90 270 90 270 90 -2 Weighted by Fryberger (1979) ux with 0.008 Nm threshold Weighted by Fryberger (1979) ux with 0.016 Nm-2 threshold 240 120 240 120 240 120 240 120 240 120 240 120 240 120 240 120 240 120 5 Weighted by u* ux with 0.008 Nm-2 threshold

210 150 210 150 210 150 210 150 210 150 210 150 210 150 210 150 210 150 180 180 180 180 180 180 180 180 180

10 6 8 8 2 4 4 8 6 Gross bedform-normal transport (GBNT) 2 4 6 6 2 −2 GBNT with a stress threshold of 0.008 Nm 4 2 2 4 4 2 2 2 2

Relative Gross Bedform- Normal Transport 0 0 0 0 0 0 0 0 0 0 30 60 90 120 150 180 0 30 60 90 120 150 180 0 30 60 90 120 150 180 0 30 60 90 120 150 180 0 30 60 90 120 150 180 0 30 60 90 120 150 180 0 30 60 90 120 150 180 0 30 60 90 120 150 180 0 30 60 90 120 150 180 Bedform Azimuth ( °) Bedform Azimuth ( °) Bedform Azimuth ( °) Bedform Azimuth ( °) Bedform Azimuth ( °) Bedform Azimuth ( °) Bedform Azimuth ( °) Bedform Azimuth ( °) Bedform Azimuth ( °)

GCM sampling & prediction of feature orientation Conclusions References In 4 out of 7 sites, dune orientations are predicted within ~10° by the MarsWRF GCM [1] Richardson, M. I. et al. (2007) J. Geophys. Res. 112 (E9). At each site we sampled the MarsWRF GCM [1,7] for each minute of a Mars year, by maximizing gross bedform-normal transport. Locations that do not match often [2] Warner, N. H. et al. (2011) J. Geophys. Res. Plan. 116 (E06003), 1-29. giving a total of 963360 wind vectors, spread uniformly throughout the year. show non-unidirectional wind elds or topographic variation. [3] Zuber, M. T., et al. (1992) J. Geophys. Res. 97 (E5), 7781-7797. [4] Hartmann, W. K. (1999) Met. Plan. Sci. 34, 167-177. For each point we extracted zonal (u) and meridional (v) wind velocities, the friction We calculate the angle clockwise from north where T is maximized Use of higher stress thresholds (0.016Nm), produces better ts for dune populations [5] Hartmann, W. K. (2005) Icarus 174, 294-320. velocity, u*, and the atmospheric density, ρ (required to construct wind stress, σ, (Figure 4, lower row), as well as for a secondary maximum (for Iani at a number of locations, constraining the activation energy required for saltation. [6] Popova, O. et al. (2003) Met. Plan. Sci. 38 (6), 905-925. using σ=ρu*). Chaos, lines, top row). [7] Toigo, A. D. et al. (2012) Icarus 221 (1), 276-288. Areas where features do not match GCM output are often more topographically vari- [8] Rubin, D. M. & Hunter, R. E. (1987) Science 237, 276–278. u and v were interpolated to an altitude of 1.5m using similarity theory (the lowest Predicted dune crest orientations for maximum GBNT and maximum able (Candor Chasma, Iani Chaos) suggesting that the MarsWRF GCM is unable to re- model altitude was 10m). Wind speeds at 1.5m were deemed representative of GBNT with various wind stress thresholds (dashed and dotted lines) solve localized topographic wind forcing at these small scales. those that control transport via saltation. are shown as red lines in Figure 4, upper row. Yardang populations approximately match the maximum GCM vector frequency at GCM vector frequencies are plotted in angular bins (Figure 4, third row down) and most sites, but at some they match secondary and tertiary directions, indicating the weighted by various formulations of wind stress. We also consider various values of need for better understanding of the relationship between yardang erosion and wind saltation threshold friction velocity u* t, the minimum value required to keep grains in saltation. elds.

Transverse dune crests developed by a temporally dynamic wind eld tend not to Slight osets between observed and predicted yardang orientations, coupled with align normal or parallel to the direction of transport, but to trend such matches between observed and predicted dune orientations (e.g. Gordii Dorsum), that the maximum gross bedform-normal transport, GBNT (T) is achieved [8]: may indicate that yardangs are less in equilibrium with present day dune and wind elds (if dunes are active).