Comparison of Different Crater Counting Methods Applicated to Parana

Introduction: Martian valley networks are good indicators of the past climate. These features are located primarily in heavily cratered terrain of Noachian age (>3.7 Ga) [1] but the exact timing of these features is not well understood. Recent global studies using superposed craters along different fluvial valleys [2, 3] determined that most formed during the Late Noachian to Early Hesperian. However, valley networks are small, linear features that make them difficult to accurately date using traditional cratercounting techniques. In this study, we set out to test the reliability of the counting methods used in [2] and [3] while comparing the results to other methods using fine scale age determination from small diameter craters. We applied these different methods to Parana Valles using a large dataset of high-resolution images from different instruments that allow us to count smaller craters while providing a good sampling of crater populations that is necessary to reduce uncertainties in age determination. Dataset: We used three different types of images to cover the Parana Valles study area, including CTX (6m/pixel) and THEMIS visible images (17 to 35m/pixel), which provided optimal resolution for part of the study area. Where there were no available high resolution images, we used THEMIS IR night time images (100 m/pixel). The mosaic of all these images was used to map the Parana Valles network and to count craters. Additionally, we used digital topography from MOLA at a resolution of 460 m/pixel to delineate drainage basin and subbasin divides using automatic techniques established for digital elevation data of terrestrial watersheds [4]. Parana Valles Basin: We studied a large subbasin of the Parana Valles drainage system that is located upstream from the Parana/Loire basin and situated between Margarifer and Noachis Terra. The study covers an area equal to 26 452 km2 and is drained by 3586 km of valleys organized dendritically (Fig. 1). This results in a drainage density of 0.13 km, and we determined that the highest order stream is a 6 order using the Strahler system [5]. The primary network is divided into two main branches where the source areas are located between 400 and 700 m elevation and terminate at an elevation of -1000 m. Two different morphologies are observed in the main tributaries. Between -1000 and 200 m elevation the valley is rectangular with a width of ~2 km. We observed that this valley is partially filled by eolian deposits. At an elevation of ~200 m, the main valley in the southern branch, becomes wider (between 3 and 8 km), exhibits a wide V-shape profile and incises interfluves. We suggest that the differences in morphology are the result of variation of lithology.