Influence of Target Yield Stress on Crater Dimensions: a Numerical Approach Based on Chicxulub

Introduction: AUTODYN [1] is used to consider the effect of altering the yield stress of geological materials on the resultant crater dimensions. This investigation is part of a series of fundamental inquiries into the sensitivities of AUTODYN, in order to fully validate the code as an appropriate tool to study large planetary impact crater formation. The majority of current AUTODYN applications are for weaponry, defence and civil engineering problems. AUTODYN is extremely competent in analysing rapid events typical of this genre, and indeed is well suited to resolving peak shock pressure decay in the initial stages of large planetary impacts [2]. However, its potential for analysing the longer timescales required for the full impact process (i.e. to the end of the modification stage), which can require timescales on the order of hours, has been less widely explored. Motivation for using Chicxulub: An imperative part of validating a numerical code is to assess its reliability at reproducing “real-life” data. A well-studied terrestrial crater was therefore selected for this study. An additional motive for investigating Chicxulub is its association with the impact event implicated in the demise of the dinosaurs, 65Mya. However, despite the plethora of data resulting from numerous intensive seismic, gravity and numerical studies, different authors have arrived at a wide range of dimensions for the crater (Table. 1). Furthermore, Chicxulub is classified as both a peak ring [3,7] or multi-ringed basin [4-5, 7,8] complex crater by different authors. Despite the agreement that there are three possible rings, the first of which is located at a diameter of 80km e.g. [5, 7, 8], estimates for the location of subsequent rings are varied. Rings two and three are inferred by [5] to be at 130km and 195km respectively; whereas [8] considers them to be at 200km and 250km. Modelling this crater with AUTODYN therefore also has the potential to constrain the parameters of this impact event. Model initialisation: Grid setup. A 600x100km domain was defined, using axial symmetry, by an SPH grid (18945 particles) coupled to a Lagrange grid (42600 cells) and initialised with gravity (9.81m/s). Material model. The success of a simulation relies strongly on the material data that are available. Any material model (i.e. equation of state, strength, failure) that is applied has many parameters, (e.g. yield stress, shear modulus, hydrodynamic tensile limit) and changing just one parameter could significantly affect the final results. In addition, yield stresses of typical geological materials can cover a range of several orders of magnitude [9].Therefore, to see whether crater dimensions are strongly affected by yield stress, a series of test simulations were executed. Preliminary tests configure the materials with identical yield stresses, to simply illustrate the affect of altering one parameter at a time. Subsequent investigation will examine other significant parameters. For results presented here, the linear DruckerPrager strength model and hydro-dynamic tensile limit failure model was applied [10]. Materials. Materials chosen for the simulations imitate the idealised stratigraphy at the Chicxulub site: 100m water overlying 3km sediments, and a 27km thick crust overlying the mantle e.g. [11,12] Due to the resolution of the model, the water layer is considered negligible. Subsequent layers are represented as limestone, granite and dunite e.g. [11, 12]. Tillotson and Shock equations of state were used for limestone and granite, and dunite, respectively.