In anticipation of having to submit my Masters Thesis for review by Canada’s 151st birthday, I’ve been working to finish up the Thesis itself as well as the two results figures my work has been geared to create over the past year-and-a-half.
First is rim crest low (RCL) location vs. melt deposit location, either both locations “coincide” (same direction from crater), “within 45” (locations within 45* of each other), “within 90” (locations between 45* and 90*), or “90 or greater” (melts located anywhere on the other side of the crater from the RCL):
Second is the above figure vs. a ratio “R” = (rim crest high – rim crest low)/(depth of crater) that represents topographic variation about the crater rim — now complete with error bars:
Special thanks to Dr. Neish for providing me the data for the Moon and Venus, given in grey and orange, respectively, in the above figures (plus errors). Thanks also to her for a statistical analysis of the data (in the interest of time), and for concluding that Mercury’s craters are statistically similar to Venus’ craters (p = 0.49) and very different from the Moon’s craters (p = 0.00023). Here, larger values of p imply more similar pairs of datasets — implying further that there’s a higher probability that the two datasets tested came from the same, original set of data (the “null hypothesis”). This statistical method, “Anderson-Darling (AD),” also mentioned briefly in Josh’s blog here, was done in Neish et al. (2017) for the Moon and Venus as well (where 0.01 < p > 0.09, in comparison)….
Back to work for me!