Dutch Experience of Assuring the Quality of Information from Government Exercises Using Models

1
Dutch Experience of Assuring the Quality of
Information from Government Exercises Using Models

• Arthur Petersen, Netherlands Environmental
  Assessment Agency - RIVM

2
Will we meet our national Kyoto target?

• Message expectations are that we make it

3
Environmental Outlooks 1- 4 (1988 - 1997)

• Uncertainties play a central role
• But
• extremely limited set of scenarios used
• dealing with uncertainty in an intuitive manner
• no systematic attention paid to uncertainty
• lack of transparency (van Asselt et al. 2001)

4
Example QA in EO5 (2000)

• RIVM/MNP-internal software EO5 Explorer
• Focus group and workshop on uncertainties
• Descriptions of uncertainties in main text and
  text boxes
• ? RIVM/MNP still learning (van Asselt et
  al. 2001)

5
Dataflow visualisation in EO5
Model Y
Dataset 1 (e.g. scenario)
Dataset 2
Model X
indicator
6
(No Transcript)
7
(No Transcript)
8
Directly to (meta)data
9
...and to model metadata
10
But how to propagate uncertaintiesthrough the
chain?

• As exercise in NUSAP, Penny Kloprogge (Utrecht
  University) studied the influence and pedigree of
  assumptions in two chains (2003)

11
Importance of uncertainty communication (van der
Vlist, Director-General for Environmental
Protection)

• Politicians must decide under uncertainty
• The Netherlands Environmental Assessment Agency
  (RIVM/MNP) must bring all relevant scientific
  arguments into the political debate through their
  assessments
• RIVM/MNP must inspire the Ministry in openness
  and dealing with uncertainties
• ? Communication about uncertainty is crucial

12
Origin of RIVM/MNP Guidance for Uncertainty
Assessment and Communication

• Scientific Audit (2000)RIVM should consider
  making a project to systematically address
  terminology, methodology, interpretation and
  communication of incertitude.
• RIVM report van Asselt et al. (2001)RIVM needs
  a protocol or at least a guideline that describes
  which uncertainty management steps need to be
  taken in which phase of the assessment process.

13
Purpose of the Guidance

• Offer assistance to analysts
• Focussed on the following parts of assessments
• problem framing
• involvement of stakeholders
• selection of indicators
• appraisal of knowledge base
• mapping and assessment of relevant uncertainties
• reporting of uncertainty information

14
Structure of the Guidance
Reminder listInvokes ReflectionPortal to QS
Advice on Quantitative Qualitative tools for UA
Downloads www.nusap.net - keyword rivm/mnp
15
Web-application
16
Mini-Checklist
17
QS-Questionnaire
18
QS-HintsActions
19
Typology of uncertainties dimensions

• Location
• Level of uncertainty
• statistical uncertainty, scenario uncertainty,
  recognised ignorance
• Nature of uncertainty
• knowledge-related uncertainty,
  variability-related uncertainty
• Qualification of knowledge base (backing)
• weak, fair, strong
• Value-ladenness of choices
• small, medium, large

20
Typology of uncertainties locations

• Context
• ecological, technological, economic, social and
  political representation
• Expert judgement
• narratives, storylines, advices
• Model
• model structure, technical model, model
  parameters, model inputs
• Data
• measurements, monitoring data, survey data
• Outputs
• indicators, statements

21
(No Transcript)
22
(No Transcript)