Laserfiche WebLink
' AOpendu A - Ve¢emlian Sampling for 1998 <br />Production was determined by hand clipping one-half square meter quadrants. The sampling point <br />' was determined by placing the quadrant to the rear and left of the random starting point for the <br />vegetative cover transect. The vegetation was clipped by forb or grass species and Separated by <br />annual or perennial life form, with the corresponding field "wet" weight recorded. The clipped <br />' vegetation was then oven dried and reweighed as a "dry" weight. The dry weights were converted <br />to pounds per acre for each respective sample area. <br />' Sample adequacy for vegetative cover and biomass production within each sampling area was <br />determined using the Snedecor-Cochran sample adequacy formula (Equation 3). <br />(3) <br />Where: <br />' Nm;n = the minimum number of transects or quadrants needed, <br />t = the a = .10 t-table value for asingle-tailed t-test with n-1 degrees of freedom, <br />1 s2 = the sample variance, <br />' d = .1 (level of precision for estimate of the mean to be within 10% of the actual mean), <br />and, <br />x = the sample mean. <br />Vegetative cover data from the transects were used to determine sample size adequacy. Sample <br />size adequacy was reached in the reference area with fifteen transects. Sample size adequacy was <br />reached in the reclaimed area with twenty-five transects. During the September reclaimed area <br />sampling event, sample size adequacy was reached in fifteen transects (Table A-1). <br />Biomass production data from the clipped quadrants were used to determine sample size adequacy. <br />' Sample size adequacy was reached in the reference area with fifteen quadrants. On the reclaimed <br />area, sample size adequacy was not achieved after fifty plots were clipped, the minimum number <br />needed to compare the data by nonparametric equation analysis. <br />The statistical analysis used for comparison of the production values sampled on the reference and <br />reclaimed areas was developed by the CDMG. A nonparametric procedure for estimating a <br />ranking was implemented due to the assumption that the data may or may not follow a normal <br />distribution. A lower confidence interval was developed for a ranking analogous to the median. <br />The data was ordered from smallest to largest with a mean and median calculated. The data was <br />' ordered using the ranking methodology in Quattro Pro V.6. A lower confidence interval of the <br />mean was calculated using Equation 4. The lower confidence interval was interpolated to test a <br />' BoM057 1999-053.wpd\February 26, 1999 A-2 <br />