The person shall follow the procedures of this subsection to determine if a response action under this chapter is necessary to protect human health and the environment, and if a response action is necessary, then to determine if the remedy standard is attained. If the person satisfactorily demonstrates that all reasonably available analytical technology (e.g., selected ion monitoring) has been used to show that the COC cannot be measured to the method quantitation limit due to sample specific interferences, then the person shall be allowed to determine attainment based on the sample detection limit. The person shall make these determinations using the procedures described in either paragraph (1) or (2) of this subsection.
  (1) The person may make a direct comparison between individual measurements of COC concentrations within environmental media and the critical PCLs. If the concentrations of a COC within an environmental medium exceeds a critical PCL, then a response action is required.
  (2) The person may determine if a response action is required by using appropriate statistical methods provided in subparagraphs (A) or (B) of this paragraph.
    (A) In order to determine if the concentrations of the COC at an affected property exceed a critical PCL the person shall conduct a statistical test of the following set of hypotheses:
      (i) the null hypothesis (Ho ) is that the mean of the COC concentrations in the affected property is equal to or greater than the critical PCL;
      (ii) the alternative hypothesis (Ha ) is that the mean COC concentration is less than the critical PCL;
      (iii) the test is performed at a Type I error rate of 5%; and
      (iv) any statistical model used for testing this hypothesis set must be demonstrated to meet these performance standards.
    (B) In order to determine if the concentration of a COC in an environmental medium at the affected property is greater than the COC concentration for background areas, the person will use a statistical test meeting the following performance standards:
      (i) the null hypothesis (Ho ), in conjunction with any supporting assumptions, is equivalent to the statement that the mean of the COC concentrations in the two areas are identical;
      (ii) the alternative hypothesis (Ha ), is equivalent to the statement that the mean of the COC concentrations at the affected property exceeds that population of background concentrations; and
      (iii) the test is performed at a Type I error rate of 20% and the test must have a demonstrable power of 80% for an alternative hypothesis equivalent to a 100% difference in population means in the Student's ''t'' test. Alternative statistical methods for comparing affected property COC concentrations to background COC concentrations may be approved by the executive director.

Source Note: The provisions of this §350.79 adopted to be effective September 23, 1999, 24 TexReg 7436; amended to be effective March 19, 2007, 32 TexReg 1526