Tuesday, April 1, 2008 : 10:40 a.m.

Scientific Integrity vs. Scientific Persuasion: Representing Model Uncertainty in the Courtroom

David R. Hargis, Phd and Leo S. Leonhart, PhD, Hargis + Associates Inc.

As technical practitioners, we share a responsibility to adhere to ethical standards in proffering our technical opinions, particularly when consumers of our services are the general, lay public.  Among these is the responsibility to make every effort to provide an honest and forthright characterization of limitations and uncertainties of our technical work.  Arguably the greatest opportunities to abuse such responsibilities arise in the courtroom.  In a courtroom venue, a technical expert may be tempted or lured across the threshold between scientific integrity and scientific persuasion.  One increasingly popular courtroom tool among hydrogeologists is the numerical model, which is capable of simulating real world processes and yielding predictions based on limited and uncertain data.  Accordingly, models are often perceived as sophisticated, perhaps unbiased instruments that yield accurate results when used by highly-credentialed experts. 

 

A recent case involved application of numerical models to evaluate specific questions posed by the court regarding persistence of agricultural pesticides in groundwater.  The questions were posed to project future treatment costs to meet drinking water standards at the wellhead.  The model results were presented simply as a year that the contaminant concentrations were predicted to decline below their respective MCLs.  No uncertainty or error was given to bound the model results. 

 

Essentially, the expert’s testimony was ipse dixit.  It was argued that the model calibration was “robust” based on the expert’s “X” years of modeling experience.  Therefore the predictions had to be valid.  In addition, other conceptual models that might have accounted for the available field data remained unconsidered.  This example typifies the dilemma before lay judges and juries when asked to weigh evidence manufactured by computer models, particularly when the responsible expert is remiss in representing the limitations and uncertainties in his opinion.  

David R. Hargis, Phd, Hargis + Associates Inc. David R. Hargis is President and founder of Hargis + Associates, Inc., Consultants in Hydrogeology/Engineering in San Diego, California. Dr. Hargis earned both BS and MS degrees in geology from the University of Hawaii-Manoa and his doctorate in hydrology and water resources from the University of Arizona. He is a Registered Geologist with over 30 years of experience as a consulting hydrogeologist and is on the adjunct faculty at the University of Arizona, where he teaches a seminar in forensic hydrology.

Leo S. Leonhart, PhD, Hargis + Associates Inc. Leo Leonhart received his bachelor's in geology from Youngstown State University, followed by a Master's in water resources from The Ohio State University. In 1978, he completed a doctorate in watershed hydrology from the University of Arizona. He is a registered geologist and certified hydrogeologist, practicing as a Principal Hydrogeologist at the Tucson office of Hargis + Associates, Inc., Consultants in Hydrogeology/Engineering. He is on the adjunct faculty in the Department of Hydrology & Water Resources at the University of Arizona and at the University of Phoenix.


2008 Ground Water Summit