2012-03-07_PERMIT FILE - C2010088 (14)

Home Browse Search

EXHIBIT 5 additional cover and shrub density samples were also collected. Phase III sampling corresponded to transects 31 through 50 and typically applied only to production transects, but in instances where sample adequacy was not achieved for plant cover, then additional cover and shrub density samples were collected. When sample adequacy was determined to be just a few more than the minimum of 15 cover of 30 production transects, additional transects were usually sampled sequentially until sample adequacy was achieved. For the Greasewood Reference Area, after 13 cover transects had been sampled, sample adequacy was found to equal 26.2 transects. After 19 transects had been sampled, sample adequacy was calculated to equal 50.8 transects. Since sample adequacy was increasing with each additional number of samples collected, evidence suggested that it would be necessary to sample to the maximum number of transects (50) and there was no need to periodically stop and reevaluate the data and determine whether sample adequacy had been achieved. However, after 35 transects had been sampled, from randomly located sample sites, and sample adequacy was calculated and found to equal 32.4 transects. Since sample adequacy was achieved, sampling stopped. On this site, the subsequent transects were sampled based upon which transect was located closest to the end of the transect being sampled. Given the fact that the transect location and bearing was randomized, the location of the closest transect to this point is considered to be random as well. Examination of Map 2.04.10 -3, Rhino Energy LLC, Fruita Unit Train Loadout - Greasewood Reference Area, clearly documents that the transect locations were randomly located and sampled on this site. Therefore, there was no bias due to sequence of how the sample transects were sampled. The DRMS regulations at Section 4.15.11 (1) states that "all aspects of the vegetation sampling program must be conducted to ensure a reasonable, unbiased estimate of the appropriate population parameter. Consistency in sampling shall be required in comparison between the reclaimed area and the undisturbed area. Both random and systematic sampling designs are acceptable." The Vegetation Guideline on page 5 states "it is important that unbiased samples be obtained. In general, the Division recommends that transect samples be randomly located." There is an abundance of documentation clearly showing that a "systematic sampling" approach yields data that is more biased and certainly not less biased as is being suggested by the DRMS and the most unbiased data is obtained from a completely randomized sampling approach. Similarity, if a "random" sampling effort is used to determine transect locations but then a "systematic sampling" approach is used to sample these coordinates, the end result would yield more not less sampling bias. By implementing a two -stage "random sampling approach," such as was used herein, by firstly selecting purely random coordinates and then secondly, by randomly sampling those coordinates or transects in a random manner, would yield less unbiased data than would a two -phase sampling approach that uses random sampling during Phase I and a systematic sampling approach during Phase II In a classic statistical treatise, written by D. R. Cox entitled "Planning of Experiments" published by John Wiley & Sons in 1958, "randomness" is defined as a "situation where there