I do not have a Mathematics degree. I did graduate in Economics (B.A. 1974) from the University of Waterloo. Yes I did take a third year Statistics course and passed it with a lot of hard work. That said I can't remember the last time I actually refreshed my memory with Log scales, Logarithmics or for gosh sakes Calculus which I genuinely despised. It's been well over half a century ago approximately.
Why did I suggest that the vertical scale ( i.e. y axis I believe) was a Logarithmic scale? First of all bear with me if my terminology is incorrect here. Most linear scales for example might be scaled such as 0, 2,4,6,8,10,12 etc. up the y axis. Or they could be 0, 10, 20, 30, 40, 50, 60, 70, 80, etc.
I believe that I got fooled by the vertical scales in Jen Lyndall's (Integral) report because they were scaled as follows: .01, .1, 1, 10, 100 Note that each higher number up the vertical scale is TEN times the number preceding it. Hence the actual numerical difference between spaces of similar length isn't the same as it is with the two linear scales in the preceding paragraph. In this scale presented in this paragraph the numerical difference between marked points is .1-.01= .09 The next numerical difference between marked points is 10-1= 9 The next numerical difference between marked points is 100-10=90 .
Ms. Lyndall's graphs actually have quite a variety of vertical scales on them including .1, 10, 1000 as well as .1, 10, 1000, 100,000 and some that I find rather bizarre such as .03, .1, .3 followed by .03, .1, .3, 1, 3 As weird as these last two are in that they are neither equal difference in values between them like paragraph 2, NOR is each higher up the scale number a similar multiple as in paragraph 3 (i.e. 10 times).
What is my conclusion? Both Log scales (paragraph 3 at first blush) and I hesitate what to call the scales in paragraphs 3 and 4 with either the exact same multiples (i.e. 10) used moving up the vertical scale or as in paragraph 4 the .03, .1, .3 and .03, .1, .3, 1, 3 with different multiples between marked points ( 3.3, 3, ) and (3.3, 3, 3.3, 3) now appear to me to be an aberration.
To be clear I have been referring to Jen Lyndall's Figures 4-24 up to 4-57. which plot concentrations of DDT and Dioxin/Furans on the vertical y axis and either location as in Reaches 4, 3, 2, 1 downstream or a number of graphs have dates (years) on the horizontal x axis. Now all the graphs have concentrations on the vertical y axis either mg/kg (parts per million) or ng/kg (parts per trillion) depending on which of two contaminants (DDT or Dioxins/Furans) they are looking at. Regardless these concentrations are marked whereas Logarithmic numerics are not, which has me now believing that these are all linear scales albeit some of them I still believe are bizarrely scaled and labelled.
Is there a purpose to this. Of course there is. All Log scales AND all linear scales using multiples of the previous marked value as they rise up the vertical scale reduce the visual aspect of exactly how high the highest values actually are. If they went up the vertical scale in a linear fashion such as 2,4,6,8,10,12,14,16,18 can you imagine how large the page would have to be to register say a 114 value? Therefore by using a scale such as 2, 10, 50, 250 (multiples of 5) you can readily fit in the 114 value PLUS visually the 114 value sits much lower on the page and doesn't so drastically appear out of place with all the lower values.
The other advantage for unscrupulous polluters is simple confusion. You can pretend that you are transparent and putting all data on graphs when in fact you are intentionally making the graphs as confusing as possible.
P.S. Speaking of NOT putting all data in reports shouldn't Table 3-3 of this report under Maximum Detected Concentration for Background (Creeks/Ditches) have a value of 24.4 ng/kg (ppt.) for Dioxins/Furans rather than the 9.0 ng/kg listed? This just happens to be the Stroh Drain which coincidentally I'm sure is one of the major coverups Lanxess are involved in.