|
Midwest | I think it's easier to understand the 0-100% version so that's what I'll give an example for. This takes each year's yield data and first looks at each year's map individually. Year 1 - Corn - Yields are from 120-220 in that field that year. So, 120=0% (the bottom) and 220=100% (the max) Everything else is given it's percentage value depending on where it is between those. Your one square had 160 that year so that's 40. Year 2 you had beans, they yielded from 10-80 bu/ac so 10=0% and 80=100%. Your same square that year had 45 bu beans so that's 50%. (50) Corn in year 3, yields were 100-250 and that square got 200 like you said so that gets a 67. Beans in year 4, yielded from 5-60 and that square got 60, the max, so that's 100 for that year. We'll stop there but you could keep going. So........your square had a 40, 50, 67, and 100. The average of those normalized numbers is around 64 which is the normalized average number you see in the results for just that square. Each other square gets the same treatment for its own yield values.
Say you had a square that dominated each year for some reason. It had the max for each year, so 220 corn in year 1, 80 bu beans in year 2, 250 bu corn in year 3, and 60 bu beans in year 4. That's 100, 100, 100, 100 so your average is 100. (the max possible) If you had a square that needed tile, deer ate everything else, and didn't get treated seed there or something that one might end up with normalized numbers for each year of 5, 8, 0, and 10 which would give that square a final value of about 6. (not good) | |
|
normalization of data
