Superfund and Mining Megasites Lessons from the Coeur d'Alene River Basin (2005) / Chapter Skim
Currently Skimming:
Appendix F. Assessment of the Probabilistic Model for Estimating Metal Loading and Effectiveness of Remedial Action
Appendix F. Assessment of the Probabilistic Model for Estimating Metal Loading and Effectiveness of Remedial Action
Pages 459-484
The Chapter Skim interface presents what we've algorithmically identified as the most significant single chunk of text within every page in the chapter.
Select key terms on the right to highlight them within pages of the chapter.
From page 459... ...
(URS Greiner, Inc.
|
From page 460... ...
2-18 to 2-19) , for the upper basin, conceptual site model (CSM)
|
From page 461... ...
, the dissolved metal loading F(t) at the same specific location is written as F(t)
|
From page 463... ...
APPENDIX F 463 their variability due to their unpredictable fluctuations with time, or of the correlations between such uncertainties or variabilities. The PTM attempts to account probabilistically for the uncertainties and variabilities.
|
From page 464... ...
464 APPENDIX F Unfortunately, however, simply stating such claims does not make them true; in this case, they are not true. The analysis presented in the PTM lacks any scientific basis.
|
From page 465... ...
APPENDIX F 465 the basin, the lack of concentration measurements in groundwater, and the very limited information on groundwater flow deny the information needed to evaluate the hypothesis or propose any more correct one on which to build a plausible analysis. THE EVALUATION OF TIME IS INCORRECT Even if the principal hypothesis used in the PTM was correct and the calculation of the immediate postremediation situation was adequately approximated, the treatment of time variation following remediation is incorrect.
|
From page 466... ...
C-6, gives an estimate value of 0.001 to 0.005 for upland waste rock, compared with 1 for the reference source, tailingsaffected sediments) , but that tells us nothing about the relative mass per unit volume ultimately available for leaching in these two source types.
|
From page 467... ...
APPENDIX F 467 available for leaching or erosion. Only one of the potential remedial actions (chemical fixation)
|
From page 468... ...
468 APPENDIX F A Belief in the Generality of Equations 1.12 It is stated that "By varying exponent n, the relationship F(t)
|
From page 469... ...
APPENDIX F 469 depends on the arbitrary time periods tp1 through tpX", PTM, p.
|
From page 470... ...
470 APPENDIX F decay afterheat in a nuclear reactor; this is the weighted sum of many exponentials with different decay constants and behaves roughly as a power law, at least over suitably defined intervals) , but prediction of such behavior requires evaluation of all the sources separately, and no a priori guess about the behavior is likely to be adequate.
|
From page 471... ...
APPENDIX F 471 of variation and assuming that the uncertainty in estimates followed a lognormal distribution. However, no data on mobility, exposure to leaching, or erosion are presented or summarized.
|
From page 472... ...
472 APPENDIX F The PTM gives no indication of how the RLPj and Rij values were obtained in any way that would allow reproduction or challenge of their values; indeed, it is unclear how any reader could determine a preference for the sets of values given in the PTM over almost any other set of plausible but arbitrary values. It is claimed that (PTM, p.
|
From page 473... ...
APPENDIX F 473 Including item 1 in this list as natural variablity is incorrect; item 1 describes uncertainty, not natural variability. Only item 2 corresponds to natural variability, and nothing involving our knowledge of it will change it.14 In the case of loading or stream flow, for example, the natural variability is represented in the PTM by a probability distribution representing the fluctuations that occur from time to time in load or flow.
|
From page 474... ...
474 APPENDIX F will also be lognormally distributed. An important implication of both pre- and post-remediation loading being lognormal is that the effects of remedial action should also be lognormal (because products and quotients of lognormal distributions are also lognormal.
|
From page 475... ...
/ CL, (1.14) and the metal loading F(t)
|
From page 476... ...
476 APPENDIX F Associated with R(t) is an uncertainty distribution but no unpredictable time variability (R(t)
|
From page 477... ...
is 0.643, and that of the logarithm of their ratio, ln(L/CL) , is 0.22517 (obtained from the joint measurements of concentration, hardness, and flow rate; [EPA 2000, for hardness and flow measurements; URS Greiner Inc.
|
From page 478... ...
478 APPENDIX F TABLE F-1 Effect of the Two Erroneous Correlation Calculations Introduced in the PTM Standard Deviation of ln(Lr(t)
|
From page 479... ...
APPENDIX F 479 definitions, it is likely to be untrue in general. Indeed, it is quite likely that a converse theorem holds -- for any lognormal approximation to a given PDF, there exist statistics of that PDF that are not conservatively estimated by the lognormal approximation.
|
From page 480... ...
480 APPENDIX F no correlation possible between true values of L and TEM,19 which are single values. · "The BHSS data do not represent the true values of CDS and CS, which are uncertain" (PTM, p.
|
From page 482... ...
482 APPENDIX F only approximations of the expected values of the normal order statistics, so the technique is approximate at best (better approximations of normal order statistics are available (Royston 1993, 1995)
|
From page 483... ...
December 2000 (as cited in URS Greiner, Inc., and CH2M Hill 2001a)
|
From page 484... ...
Environmental Protection Agency, Region 10, Seattle, WA, by URS Greiner, Inc., Seattle, WA, and CH2M Hill, Bellevue, WA. September 20, 2001.
|
Key Terms
This material may be derived from roughly machine-read images, and so is provided only to facilitate research.
More
information on Chapter Skim is available.