Laserfiche WebLink
• <br /> <br />• <br />If the initial samples in each area did not provide a suitable estimate of the mean (i.e., the <br />inequality was false), additional samples would have been collected until the inequality (n„,;,, <_ n) <br />became true. This need for additional samples was manifested only once with regard to production <br />within the new Juniper Woodland reference area. In this instance, an extra ten samples were required to <br />meet statistical requirements. <br />Reference Area Comparability <br />Typically when reference areas are utilized to facilitate determinations of revegetation success for <br />vegetation cover and production, demonstrations are required prior to permit issuance to document that <br />the selected reference areas are statistically comparable (equal) to the vegetation types for which they <br />represent. The pre-mining demonstration of reference area comparability was made using the two-tailed <br />Student's t-test as indicated below. Reference area production and cover are compared at the 95% <br />confidence limit (alpha at 0.05 level) with the respective baseline area production and cover. The sample <br />°P' statistic (t*~ is calculated as follows: <br />t* _ (x, - x z) / (s~~ ln,) + (s~z l nz) <br />where <br />x, =mean of the baseline area <br />x Z =mean of the reference area <br />sz, =variance of the baseline area <br />s22 =variance of the reference area <br />n =sample size for the baseline area <br />n2 = sample size forthe reference area <br />The tabular "t" comparison statistic (ta) is calculated as follows: <br />to = ~(S12 /nl)tdl + (Sz2 /i72)ta2~/(Slz /n, + s22 /n?) <br />where <br />a = 0.051evel <br />Then if the inequality (t* <_ Ca) is true, the test is passed and the baseline and reference areas can be <br />considered to be from identical populations. <br />~ImA]P3 Q1P31EIEIIS Affiffi®CIIA7PIE~ INC. <br />Page A-6 Exhibit 5 - Appendix A -Vegetation Methods <br />