Quite recently a paper on dust devils on Mars was published: https://arxiv.org/abs/1708.00484
Three quotes that could be of special interest to you (Brian Jackson et al. 2017):
"Instead, we can adapt their formulation for our estimated typical, actual diameter of 13 m (Figure 9), which gives an occurrence rate ten times higher, 1 event per sol per km2. [...] This estimate includes detection biases and a population average should be used instead of typical values, but these results are roughly consistent with terrestrial field studies that estimate 40% of dust devils lift visible amounts of dust (Lorenz and Jackson, 2015)."
"Figure 8 (b) shows the resulting Γobs distribution. Consistent with estimates of the distributions of dust devil widths (e.g. Reiss and Lorenz, 2016), wider devils are less common than narrow devils, and Figure 8 (b) shows a power-law model for the differential distribution with index -1.6 provides a reasonable fit.
The distribution also declines for dust devils narrower than [...]18 m, seemingly at odds with the results of Reiss and Lorenz (2016), which conducted a visual survey of dust devil tracks and reported a population dominated by devils with a diameters < 10 m. However, the apparent decline on the narrow-end of the distribution is likely the result of selection bias in Ellehoj et al. (2010). Given the typical wind velocity of 5 m/s measured at the Phoenix site, a devil with a width ∼ 10 m would produce a signal with only a ∼2-s duration."
"For example, the devils with the deepest pressure profiles seem to occur preferentially around mid-day local time both on Mars Ellehoj et al. (2010) and the Earth Jackson and Lorenz (2015)."
Which means that for the 25 sq mile map (about time you Americans finally switched to metres now that even Myanmar and Liberia are converting) there should be around 64*0.4=25.6 visible dust devils per sol, skewed towards midday. They should be generally small at around 10m diameter or less, although a rare one might reach a width of around 100m. They should move at a speed of some 5 m/s on average.
Customer support service by UserEcho