Author
![]() |
Harmel, Daren |
![]() |
SMITH, PATTI - Texas A&M University |
![]() |
MIGLIACCIO, KATI - University Of Florida |
Submitted to: Book Chapter
Publication Type: Book / Chapter Publication Acceptance Date: 5/27/2010 Publication Date: 10/22/2010 Citation: Harmel, R.D., Smith, P., Migliaccio, K. 2010. Chapter 12: Uncertainty in measured water quality data. In: Li, Y., Migliaccio, K., editors. Water Quality Concepts, Sampling, and Analyses. CRC Press. p. 227-239. Interpretive Summary: Water quality assessment, management, and regulation continue to rely on measured water quality data, in spite of advanced modeling capabilities. However, very little information is available on one very important component of the measured data - the inherent measurement uncertainty. Although all measurements are in fact uncertain to some degree, the uncertainty in measured data is rarely estimated and thus typically is ignored. While some would argue that uncertainty is too difficult a concept for stakeholders and decision-makers to understand and that presentation of uncertainty (especially substantial uncertainty) will diminish usefulness of the results, these common philosophical justifications for ignoring data uncertainty are tenuous at best. Support for uncertainty estimation is not a condemnation of previous data collection efforts that have ignored measurement uncertainty. Instead, it is a call for future water quality projects to estimate measurement uncertainty and clearly explain the uncertainty to both scientific and non-scientific users of those data. The environmental and socio-economic ramifications of decisions based on water quality data are too important for the inherent uncertainty to continue to be ignored. To facilitate uncertainty estimation and enhanced understanding, this chapter presents recent scientific advances related to uncertainty in measured water quality data. It nullifies previously-used technical justifications for ignoring measurement uncertainty by summarizing current scientific understanding, describing a user-friendly uncertainty estimation method, and presenting uncertainty estimates for measured data. Technical Abstract: Water quality assessment, management, and regulation continue to rely on measured water quality data, in spite of advanced modeling capabilities. However, very little information is available on one very important component of the measured data - the inherent measurement uncertainty. Although all measurements are in fact uncertain to some degree, the uncertainty in measured data is rarely estimated and thus typically is ignored. While some would argue that uncertainty is too difficult a concept for stakeholders and decision-makers to understand and that presentation of uncertainty (especially substantial uncertainty) will diminish usefulness of the results, these common philosophical justifications for ignoring data uncertainty are tenuous at best. Support for uncertainty estimation is not a condemnation of previous data collection efforts that have ignored measurement uncertainty. Instead, it is a call for future water quality projects to estimate measurement uncertainty and clearly explain the uncertainty to both scientific and non-scientific users of those data. The environmental and socio-economic ramifications of decisions based on water quality data are too important for the inherent uncertainty to continue to be ignored. To facilitate uncertainty estimation and enhanced understanding, this chapter presents recent scientific advances related to uncertainty in measured water quality data. It nullifies previously-used technical justifications for ignoring measurement uncertainty by summarizing current scientific understanding, describing a user-friendly uncertainty estimation method, and presenting uncertainty estimates for measured data. |