Visualising and Communicating Uncertain Flood Inundation Maps David Leedal 1 , Jeff Neal 2 , Keith Beven 1,3 and Paul Bates 2 . (1)Lancaster Environment Centre, Lancaster University, Lancaster, UK; (2)School of Geographical Sciences, University of Bristol, Bristol, UK; (3)Geocentrum, Uppsala University, Uppsala, Sweden
12
Embed
Visualising and Communicating Uncertain Flood Inundation Maps David Leedal 1, Jeff Neal 2, Keith Beven 1,3 and Paul Bates 2. (1)Lancaster Environment Centre,
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Visualising and Communicating Uncertain Flood Inundation Maps
David Leedal1, Jeff Neal2, Keith Beven1,3 and Paul Bates2.
(1) Lancaster Environment Centre, Lancaster University, Lancaster, UK;(2) School of Geographical Sciences, University of Bristol, Bristol, UK;
(3) Geocentrum, Uppsala University, Uppsala, Sweden
Guidelines for flood risk mapping
Guidelines and framework for best practice in uncertain flood risk mapping (FRMRC2 WP1.7) provides:• A comprehensive background in state-of-the-art
thinking and methods for uncertainty analysis
• A breakdown of the flood risk modelling procedure into 7 key processes
• A series of decision trees for each process
• A set of case studies showing examples of the guidelines in action
Types of uncertainty
The Guidelines and Framework emphasises methods for aleatory and epistemic uncertainty.
• Aleatory: arising from the natural variability of the process
• Epistemic: shortcoming in knowledge about the process
Addressing epistemic uncertainty
Objective is to elicit and record expert opinion in a reflexive way and to document the thoughts, decisions and processes of those involved.
The ‘Guidelines and Framework’ suggests the modelling process should be documented in sufficient detail to provide a record of the decisions and methods used during the modelling exercise
What are the benefits of documenting a modelling exercise?
• Transparency – providing a record of which processes were carried out and why
• Which model was used and why?
• Which parameters were adjusted and within what range? Why?
• What topography was used?
• How where bridges treated?
• How many MC realisation were performed?
• etc…
What are the benefits of documenting a modelling exercise?
• Improve work practice – standardisation etc
• Method of communication with others
• Transfer skills and experience
• Receive support (and criticism)
These methods address epistemic uncertainty by:
• Explicitly communicating the degree to which a factor is understood
• Describing how a factor was addressed
• Making the process open so that others can:
• appreciate the degree of understanding
• contribute to better understanding if possible
• Over time produce a catalogue of cases that can be studied
In the mean time...
These methods address aleatory uncertainty:
• Monte Carlo
• Event generators
• GLUE
• Bayesian methods
• …many more (applied separately and in combination)
Carlisle uncertain flood inundation study
Carried out by Jeff Neal (Bristol) and Caroline Keef (JBA)
• Boundary condition upstream input event generator produced 47000 multivariate input scenarios (with model identified from observed level + rating curve record)
• LISFLOOD-FP 2D hydrodynamic model simulated flood spreading over 5m grid for each scenario (using HPC)
• 40GB data generated
• Frequency of depth exceedence for each model cell can be calculated from data set
Data visualisation
The ‘Guidelines and Framework’ outlines the need for a modelling study to provide a clear method to visualise the complex data sets produced by uncertainty analysis. This method should:
• Allow non-experts to gain an insight into the identified uncertainty in the study
• Provide a means to support decision making if necessary
The Google maps uncertain flood inundation visualisation tool