@spidergal,
spidergal wrote:
But I don't get the mathematical part at all. How do the factors interact with each other in that formula?
You're referring to Formula #6 on p. 386?
Q = (qrir * wwir) + (qrk * wwk) + (qrbd * wwbd) + (qroc * wwoc) + (qrpH * wwpH)
The small q's (which range in values from 0 to 1) tell you about how good the soil is based on a variety of criteria. For example q
rk describes soil erodability from worst (where q
rk = 0) to best (where q
rk = 1).
The formula gives us an overall measure of soil quality, Q, which is a
weighted average of the small q's. Why do they use a weighted average instead of a simple average? It's because some of the small q's are more important than others. For example, q
rk, the soil's score for soil erodibility, is important, so we set the value of w
wk to .25 in the formula for determining Q. The soil's score for bulk density, q
rbd, is less important, so we set the value of w
rdb to .10. The w's are all values that you have to multiply the q's by in order to determine the weighted average.
Then I think we use this Q (which will always have a value between 0 for bad soil and 1 for good soil) to estimate the T-value. The authors ran some kind of statistical test to see how well it did, compared to a tried-and-true method of estimating T-values, and apparently it passed.
Would you like to know how they came up with scores for soil erodibility, etc.? (Remember, a score of 1.0 for soil erodibility means that the soil is less erodible.)
[NOTE TO OTHERS READING: the "T-value" mentioned in the article should not be confused with the "t-values" you read about in stats courses.]