  |
|
Towards a synthesis of qualitative and quantitative uncertainty assessment
|
|  |
|
| Posted by Jeroen on Monday, August 04 2003 @ 23:15:20 CEST
|
|
|
|
|
 A new short paper is posted to the NUSAP net:
Jeroen van der Sluijs, Penny Kloprogge, James Risbey, and Jerry Ravetz,
TOWARDS A SYNTHESIS OF QUALITATIVE AND QUANTITATIVE UNCERTAINTY ASSESSMENT: APPLICATIONS OF THE NUMERAL, UNIT, SPREAD, ASSESSMENT, PEDIGREE (NUSAP) SYSTEM
The paper has been prepared on invitation of the US Environmental Protection Agency for the International Workshop on Uncertainty, Sensitivity, and Parameter Estimation for Multimedia Environmental Modeling, August 19-21, 2003, Rockville, MD, US, organised by the Federal Working Group on Uncertainty and Parameter Estimation under the Federal Interagency Steering Committee on Multimedia Environmental Modeling (ISCMEM).
A novel approach to uncertainty assessment, known as the NUSAP method (Numeral Unit Spread Assessment Pedigree) has been applied to assess qualitative and quantitative uncertainties in three case studies with increasing complexity: (1) the monitoring of VOC emissions from paint in the Netherlands, (2) the TIMER energy model, and (3) two environmental indicators from the Netherlands 5th Environmental Outlook. The VOC monitoring involves a simple calculation scheme with 14 parameters. The TIMER model is a complex non-linear dynamic system model, which consists of over 300 parameters. The indicators in the Environmental Outlook result from calculations with a whole chain of soft-linked model calculations, involving both simple and complex models. We show that the NUSAP method is applicable not only to simple but also to complex models in a meaningful way and that it is useful to assess not only parameter uncertainty but also (model) assumptions. The method provides a means to prioritize uncertainties and focus research efforts on the potentially most problematic parameters and assumptions, identifying at the same time specific weaknesses in the knowledge base. With NUSAP, nuances of meaning about quantities can be conveyed concisely and clearly, to a degree that is quite impossible with statistic methods only.
Click here to download the short paper.
|
|
|
|
| |
|
