The Subjectivity in Identification of Martian Channel Networks and Its Implication for Citizen Science Projects

Abstract

AbstractThe Martian surface is incised by numerous valley networks, which indicate the planet experienced sustained widespread flowing water in the past (e.g. Carr in Water on Mars, Oxford University Press, New York, 1996; Phil Trans R Soc A 370:2193–2212, 2012. 10.1098/rsta.2011.0500). Examining the distribution and geometries of these valley networks provides invaluable information about the Martian climate during the period of formation. The recent advancement in high resolution images has provided an opportunity to build upon past valley maps of Mars (Bahia et al. in LPSC 2018, 2018), however, the identification of these valley networks is extremely time-consuming. A citizen science project may aid in reducing this time-consuming process; this project conducts a valley mapping task with participants of varying expertise in valley mapping to determine whether a citizen science project of this kind should be worth pursuing. This was conducted in a region adjacent to Vogel Crater (36.1° S, 10.2° W). Repeated mapping of the area (a repeatability test) found that participants with low experience in valley mapping (22 a-level physics student’s representative of the public) were inconsistent when mapping valleys. Additionally, when comparing the results of participants within this group (a reproducibility test), the majority of reproduced valleys are false positives (i.e. incorrectly traced valleys). These results were consistent with those found for the medium experience group (45 2nd year geology undergraduates). The validated tracings of the low experience group improve upon the number and total length of valleys mapped by previous studies (Hynek et al. in J Geophys Res 115:1–14, 2010). To validate these valleys requires the input of an expert to remove false positives which is less time consuming than manually mapping the images; this may indicate that a citizen science project is worth pursuing. However, to effectively identify the maximum amount of valleys an expert is required.