-
Views
-
Cite
Cite
A. John Bailer, Useless Arithmetic: Why Environmental Scientists Can't Predict the Future by PILKEY, O. H. and PILKEY-JARVIS, L., Biometrics, Volume 64, Issue 2, June 2008, Pages 651–652, https://doi.org/10.1111/j.1541-0420.2008.01026_1.x
Close - Share Icon Share
Extract
Useless Arithmetic describes a number of mathematical models for physical systems that have been misused or naively applied to the detriment of societal decisions that were supported by the application of these models. This very accessible book begins with an illustration of how fisheries modeling efforts did not prevent the collapse of the cod populations in the Grand Banks. This first chapter sets the tone for the remainder of the book. Models used to predict the cod stock are blamed, in part, for the catastrophic failure of this population. Critiques of these models include being overly simplistic, failing to address uncertainty in the model predictions, and making assumptions to bridge data gaps. These critiques are all familiar to a statistical reader.
The second chapter introduces the authors' vocabulary for describing models and examples of failures of models including predictions of the age of the earth and prediction of AIDS cases in Africa. Modelers might find the definitions incomplete, and statisticians would long for a greater discussion of uncertainty and variability here. Chapter 3 provides history of the Yucca Mountain nuclear waste storage facility along with a discussion of modeling questions such as the infiltration of water into the waste storage facilities. The authors express concern about unreasonable, refutable assumptions that are used to extrapolate the performance of a storage system 100,000 years into the future.
