skip to main content

DOE PAGESDOE PAGES

This content will become publicly available on March 14, 2019

Title: An overview of methods to identify and manage uncertainty for modelling problems in the water-environment-agriculture cross-sector

Uncertainty pervades the representation of systems in the water–environment–agriculture cross-sector. Successful methods to address uncertainties have largely focused on standard mathematical formulations of biophysical processes in a single sector, such as partial or ordinary differential equations. More attention to integrated models of such systems is warranted. Model components representing the different sectors of an integrated model can have less standard, and different, formulations to one another, as well as different levels of epistemic knowledge and data informativeness. Thus, uncertainty is not only pervasive but also crosses boundaries and propagates between system components. Uncertainty assessment (UA) cries out for more eclectic treatment in these circumstances, some of it being more qualitative and empirical. Here in this paper, we discuss the various sources of uncertainty in such a cross-sectoral setting and ways to assess and manage them. We have outlined a fast-growing set of methodologies, particularly in the computational mathematics literature on uncertainty quantification (UQ), that seem highly pertinent for uncertainty assessment. There appears to be considerable scope for advancing UA by integrating relevant UQ techniques into cross-sectoral problem applications. Of course this will entail considerable collaboration between domain specialists who often take first ownership of the problem and computational methods experts.
Authors:
 [1] ;  [2]
  1. Australian National Univ., Canberra, ACT (Australia)
  2. Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
Publication Date:
Report Number(s):
SAND-2017-9715J
Journal ID: ISSN 2198-350X; 656893
Grant/Contract Number:
AC04-94AL85000; NA0003525
Type:
Accepted Manuscript
Journal Name:
Mathematics for Industry
Additional Journal Information:
Journal Volume: 28; Related Information: Book Series. FMfi2016: Agriculture as a metaphor for creativity in all human endeavours; Journal ID: ISSN 2198-350X
Publisher:
Springer
Research Org:
Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
Sponsoring Org:
USDOE National Nuclear Security Administration (NNSA)
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICS AND COMPUTING; Uncertainty quantification; Water resources Uncertainty assessment; Mathematics-for-Industry
OSTI Identifier:
1429841

Jakeman, Anthony J., and Jakeman, John Davis. An overview of methods to identify and manage uncertainty for modelling problems in the water-environment-agriculture cross-sector. United States: N. p., Web. doi:10.1007/978-981-10-7811-8_15.
Jakeman, Anthony J., & Jakeman, John Davis. An overview of methods to identify and manage uncertainty for modelling problems in the water-environment-agriculture cross-sector. United States. doi:10.1007/978-981-10-7811-8_15.
Jakeman, Anthony J., and Jakeman, John Davis. 2018. "An overview of methods to identify and manage uncertainty for modelling problems in the water-environment-agriculture cross-sector". United States. doi:10.1007/978-981-10-7811-8_15.
@article{osti_1429841,
title = {An overview of methods to identify and manage uncertainty for modelling problems in the water-environment-agriculture cross-sector},
author = {Jakeman, Anthony J. and Jakeman, John Davis},
abstractNote = {Uncertainty pervades the representation of systems in the water–environment–agriculture cross-sector. Successful methods to address uncertainties have largely focused on standard mathematical formulations of biophysical processes in a single sector, such as partial or ordinary differential equations. More attention to integrated models of such systems is warranted. Model components representing the different sectors of an integrated model can have less standard, and different, formulations to one another, as well as different levels of epistemic knowledge and data informativeness. Thus, uncertainty is not only pervasive but also crosses boundaries and propagates between system components. Uncertainty assessment (UA) cries out for more eclectic treatment in these circumstances, some of it being more qualitative and empirical. Here in this paper, we discuss the various sources of uncertainty in such a cross-sectoral setting and ways to assess and manage them. We have outlined a fast-growing set of methodologies, particularly in the computational mathematics literature on uncertainty quantification (UQ), that seem highly pertinent for uncertainty assessment. There appears to be considerable scope for advancing UA by integrating relevant UQ techniques into cross-sectoral problem applications. Of course this will entail considerable collaboration between domain specialists who often take first ownership of the problem and computational methods experts.},
doi = {10.1007/978-981-10-7811-8_15},
journal = {Mathematics for Industry},
number = ,
volume = 28,
place = {United States},
year = {2018},
month = {3}
}