Lunar and Planetary Science XXXV (2004) 1544.pdf THE FORMATION OF AROMATUM CHAOS AND THE WATER DISCHARGE RATE AT RAVI VALLIS. Harald J. Leask, Lionel Wilson and Karl. L. Mitchell. Planetary Science Research Group, Environmental Science Dept., Lancaster University, Lancaster LA1 4YQ, UK. ([email protected]; [email protected]) Summary: The Aromatum Chaos depression- temperature near the equator is 210 K, this implies Ravi Vallis outflow channel system is sufficiently that the cryosphere thickness is ~[(273-210) K] / [15 simple that water flow rate and volume estimates can K km-1] = ~4 km, and if the cryosphere ice content is be made that throw light on processes operating to ~10% [1], the thickness of ice which can be removed form the Aromatum and Ravi features. Typical is only ~400 m. Even if one added the vertical extent discharge rates through Ravi Vallis are estimated at 3 of complete compaction in an underlying aquifer of x 106 m3 s-1. By assuming a high sediment load in the depth, say, 5 km and maximum pore space ~5% [1], water we find a minimum duration of ~2 months. the added vertical change would only be ~250 m. Too much water flowed in the channel to be The shape of Aromatum, especially the presence of explained by cryosphere melting alone, and drainage straight segments of the margin on the south side of a local aquifer system delineated by intrusions is (Fig. 1), strongly suggests that the formation of the clearly implicated. depression is related to the intrusion of one or more Aromatum Chaos: The main Aromatum Chaos dikes and the emplacement of a sill [2]. We therefore depression is a truncated triangle ~92 km long and an explore the possibility that the intrusions breached average of 30 km wide (Fig. 1). Its interior consists the cryosphere as well as melting ice within it, and mainly of a mass of blocky-chaotic terrain, with that some crustal material has been carried away by blocks generally becoming progressively smaller water seeping into the depression from an underlying towards its Eastern end. Some larger blocks at the aquifer system and overflowing into Ravi Vallis. Western end show some evidence of rotational Ravi Vallis Channels: The floor of Aromatum slumping and also appear to be less eroded, having Chaos rises to about 1 km below the rim at its Eastern flat tops showing a rather angular connection end as it connects into Ravi Vallis. This opening is between the flat top and the walls. The edges of the narrowest part of Aromatum Chaos and forms the Aromatum Chaos show strong evidence of local highest part of the floors of either the Chaos or the structural control and so, in an attempt to look for Vallis. There are multiple indicators of water flow in similar control of the interior, we examined all Ravi Vallis, especially teardrop shaped islands and MOLA profiles crossing the interior, and found 30 longitudinal floor grooving. There are also two small profiles which collectively crossed the tops of 20 areas of chaotic terrain, Iamuna Chaos and Oxia blocks. From these we measured the absolute heights Chaos, on the floor of the Vallis. Although there are (relative to Mars datum) of the tops of the blocks and indications that water overflowed the width of the their depths below the rim of the depression. The main channel for a while early in its formation, the tops of blocks lie between 1,008 m and 2,361 m presence of typically 6 deeply eroded and sometimes below the rim and show no systematic correlation braided sub-channels, each 50-100 m deep, within the with depth below rim or height above floor, main channel, suggests that the water was rarely suggesting piecemeal collapse rather than a structural more than 100 m deep in Ravi Vallis despite its control on their subsidence. typical ~700 m depth. A total volume of about 2000 We also integrated the topographic profiles to km3 has been eroded to form the channel. find the volume of material missing from the For a wide range of possible water depths from 50 depression, and extrapolated a smooth curve through m to bank-full, we used the average regional slope of the local minima of the profiles to find an the channel floor (0.0026) to calculate the water flow approximation to the shape of the depression speed in each sub-channel seen in each of 7 MOLA disregarding the blocks on the floor. The depression profiles. The Darcy-Weisbach bed friction is an asymmetric bowl-shape with a maximum depth coefficients (rather than the Manning coefficients, of 3.45 km below the rim level. It shallows more about which there are uncertainties that we shall slowly to the NE, where it connects with Ravi Vallis discuss elsewhere) were used for a range of channel (Fig. 1), than to the SW. The missing volume is roughnesses. Since the water depth is much greater ~4,100 km3 and the mean depth is ~1.49 km. than the roughness scale there is little spread in water It seems very unlikely that the volume of the speed values for a given water depth. Multiplying the depression represents compaction after removal of speed by the depth and average channel width for that cryosphere ice: If the geothermal gradient on Mars is depth we obtain a water flux. These were summed ~15 K km-1 and the annual average mean surface for each MOLA profile and the values averaged. For Lunar and Planetary Science XXXV (2004) 1544.pdf bank-full flow the flux would have been ~2.6 x 107 For example at 20% load the aquifer water volume m3 s-1 but for water depths in the 50-100 m range, would have to be 21350 km3, the water flux 2.4 x 106 which we consider much more likely, the flux would m3 s-1, and the duration 103 days. have been 2.3-4.6 x 106 m3 s-1, with formal errors of Assuming aquifer porosities of ~5% and vertical ~30%. We adopt 3 x 106 m3 s-1 as a best estimate. extents of ~5 km, the surface areas of the aquifer Flood duration and total water volume: There systems needed to contain water volumes of 7625 is no unambiguous way of estimating the duration of and 21350 km3 are 30500 and 85400 km2, the water flow in Ravi Vallis and hence the total respectively. The surface of the region immediately water volume released. However, a lower limit can to the West of Aromatum Chaos shows clear be estimated by assuming that all of the missing evidence of fracturing, pitting and collapse, and crustal volume not attributable to cryosphere ice extrapolation of the regional topography implies that removal was carried away as sediment at the it was originally at a higher elevation than the rim of maximum possible water loading, of order 40% [3, Aromatum Chaos. The surface area affected is 4]. If ice formed 10% of the cryosphere [1], the ~37000 km2, which would therefore correspond to missing rock volume is [0.9 x (4100 + 2000) =] 5490 the area of drained aquifer needed to supply the Ravi km3. If this represented 40% of the flood volume, flood if the sediment load had been about 30%. It is then the remaining water volume was [(60/40) x 5490 very tempting to conclude that the scenario proposed =] 8235 km3. This water volume would have here, of volcanic intrusions disrupting the cryosphere consisted of the ~610 km3 derived from melted and allowing drainage of a section of part of the cryosphere ice and an additional [8235 - 610 =] 7625 regional aquifer system confined by dikes acting as km3 derived from the underlying aquifer system. aquicludes, is the explanation for the Aromatum-Ravi Using our best (~50-100 m water depth) estimate system. of typical flood discharge, 3 x 106 m3 s-1, and References: [1] Hanna, J.C. and Phillips, R.J. (2003) allowing for the fact that only 60% of this is water, LPS XXXIV, abstract #2027. [2] Head, J.W. & Wilson, L. i.e. ~1.8 x 106 m3 s-1, the implied duration of flow is (2002) pp. 27-57 in Geol. Soc. Lond. Spec. Publ. 202. [3] [(7625 x 109 m3) / (1.8 x 106 m3 s-1)] = ~4.2 x 106 s, Komar, P.D. (1980) Icarus, 42, 317-329. [4] Carr, M.H. i.e. ~49 days. Of course, if the flood were not (1987) Nature, 326, 30-35. capable of transporting as much as 40% by volume Acknowledgement: This work was partially supported solid debris, the durations would be increased and the by PPARC grant PPA/G/S/2000/00521. aquifer water volume estimate would also increase. 316° 317° 318° 319° 320° 321° 0 -750 -1250 -2250 0 5 -250 250 -250 -750 -37 0 -250 000 -3500 -1250 -750 -2 -3250 25 -1000 0 -3000 -750 -250 -2 500 -1500 -1500 -2750 -500 -1500 -1250 -1000 0 -1000 0 -1750 -750 -750 -1250 -1250 -1750 -375 -150 0 -1500 -1500 250 -1500 -32500 -3500 250 -300 0° -1500 -2750 0° -1250 -2500 0 -250 250 0 -1 250 -1250 -1500 -425 -250 -750 -4000 -2250 0 -250 -1250 -2000 250 250 250 0 500 -1000 -1250 -400 -1000 500 -1000-500 -1250 -3750 0 -750 250 250 -1000 -750 -500 -1500 -1250 -1750 -4250 -1250 250 -250 -1000 -1250 -50 -4500 250 -250 0 -3500 250 250 0 -1250 -1250 -1250 -75 -1000 0 -750 0 -400 0 -1250 -1500 -1250 50 -3 -4000 0 -1000 -7 75 0 -3750 250 -750 -350 -4250 -1000 -4500 -250 -100 -500 250 -3250 0 -1000 0 -1000 -325 0 -3000 -1000 -3500 -250 -2750 -100 -750 -2500 -500 0 0 -100 250 -1250 -750 -500 -4 0 -2250 50 0 -1000 -1500 -4500 -500 -750 0 0 -750 -250 -1250 0 -1500 -1750 -1250-200 0 -2000 0 -750 -500 -1750 -50 0 0 -750 250 -500 -1250 -400 -2500 -2 -1000 -4250 -2000 -3750 -1000 -750 -1500 -2500 -750 -1500 -1250 -500 0 -1° -1750 -1° -1000 -1500 00 -250 -500 -10 -250 -2750 0 -3500 -2250 -2000 -2250 -1000 -1500 -1750 -2 -500 -750 -1250 -250 50 -1750 -250 0 -750 -2500 0 -1500 -50 -1500 0 -2500 -1250 -750 0 -500 -1000 50 -500 0 -1250 -750 75 -1 -1 -1750 0 25 -250 -1250 -2000 -100 -2250 -750250 -750 0 0 -1250-1500 0 -1000 -500 -500 -1000 -250 0 0 -750 -250 00 -1250 0 -5 -500 -250 -1000 -1500 -1250 -250 250 -1500 0 -500 250 -250 250 -500 -250 -500 -500 -750 250 -750 -1000 0 -1250 250 -500 -1000 -750 -500 -750 -250 -500 -1 -5 -500 500 500 -500 -500 00 0 500 0 0 750 250-250 0 -2° 250 -250 -2° 0 -1250 -500 -1000 -250 1000 250 25 0 -250 500 0 -500 25 -1250 750 -1000 -500 250 316° 317° 318° 319° 320° 321° m -3000 -2750 -2500 -2250 -2000 -1750 -1500 -1250 -1000 -750 -500 -250 0 250 500 750 1000 1250 1500 1750 2000 Elevation Fig.