Originally Posted by Deetz and Adams (1934, p 91)
In general, for equal distribution of scale error, the standard parallels are placed within the area represented at distances from its northern and southern limits each equal to one-sixth of the total meridional distance of the map. It may be advisable in some localities, or for special reasons, to bring them closer together in order to have greater accuracy in the center of the map at the expense of the upper and lower border areas.
As stated, It makes perfect sense to me that if needed, greater accuracy can be achieved by selecting two SPs at a range closer to 16º. But why choose two parallels with a change of
28º especially when the chart's ( E (LO) 2 ) latitude coverage ranges only from 52º 20' to 56º 20'? I see no point in selecting those SP and having the chart lie between the Parallel of Origin and the northerly SP where scale contraction error would be almost at its peak and surely above 1%.
I also must admit the math level of 'Savric B, Jenny B. 2016. Automating the selection of standard parallels for conic map projections: Computers and Geosciences, 90, 202–212.' is beyond my current knowledge. I would've loved to post the result to the distortion measure ( 204 Eq(2) ) of the chart though.