Applications and future developments of Bayesian Networks in risk and impact assessments and environmental decision making

Open seminar and workshop

When: Wednesday March 29th 2017. From 10 to 14, followed by discussions

Where: The Blue Hall and the Red Room in the Ecology Building, Lund University, Sölvegatan 37, Lund, Sweden

Scroll down to view the list of talks and links to presentations

Background

Bayesian Networks is a type of probabilistic modelling with wide applications in science and decision making. BNs is a modelling framework that enable us to integrate evidence to inform decisions, based on causal relations between decision, states and impact variables. BNs allows integration of data and expert knowledge.

This workshop will demonstrate the application of BNs in risk and impact assessment and environmental management, and, discuss and critically reflect on the developments and applications of BNs in research and decision making, through the use of case studies.

Invited speakers

Wayne Landis, Institute of Environmental Toxicology, Huxley College of the Environment, Western Washington University, United States. Wayne is a Professor in Environmental Science and a Director of the Institute of Environmental Toxicology at Western Washington University. He has over 20 years of experience in ecological risk assessment research, using Bayesian networks to guide decision making. https://huxley.wwu.edu/people/landis

Sakari Kuikka, University of Helsinki, Finland. Sakari is a Professor in fisheries science and a head of the Fisheries and Environmental Management Group research group. http://www.helsinki.fi/science/fem/kuikka.html

Talks

Adaptive management using Bayesian tools to meet the challenges of uncertainty, climate change and the challenge of making decisions – Wayne Landis

Talk: BN Workshop Wayne Landis

Abstract: Environmental contamination, invasive species, changes in technology, and climate change are individually management challenges, but in reality, each intersects with the other.  This interaction is coupled with dramatic social changes in Europe, North America and across the world.  The question becomes how to effectively manage this mélange of factors within the cacophony of social norms.  Terms like “risk” and “uncertainty” also have multiple meanings, often with emotional undertones.  This talk will introduce the approach of using a Bayesian net based risk assessment framework coupled with an adaptive management framework.  Bayesian networks can be built to include multiple stressors affecting endpoints that represent ecological status and human well-being.  The process of adaptive management includes an initial risk assessment coupled to specific predictions regarding management options (hypotheses), followed by observation of the environment, and reconstruction of the Bayesian network to update the model.  The approach is closely coupled to decision making and is intended to be adaptable.  One of the goals in this approach is to accept uncertainty as a normal situation both in the understanding of the managed system, the efficacy of the management solutions, and in the societal norms.

Learning chains in oil spill risk analysis – Sakari Kuikka

Talk: BN Workshop Sakari Kuikka

Abstract: We review the experience obtained in developing integrative Bayesian models in interdisciplinary risk analysis focusing on oil spill in the Gulf of Finland. Moreover, we also discuss the future challenges in this demanding modeling task. We have applied Bayesian models to the oil spill risk analysis in interdisciplinary questions. Bayesian belief networks are flexible tools that can take into account the different research traditions and the various types of information sources. One of the advantages of using Bayesian decision analysis for management is that the uncertainty estimates are scientifically justified. Moreover, the Bayesian inference offers and important possibility to learn effectively from many sources of information, and the results of one integrative model can, and we argue that they should, be used as priors for next accidents so that the learning component from previous spills is as high as possible. Especially in cases where society is assumed to be risk averse, the uncertainty estimates have a crucial role.

Assessing multiple climate change impacts on water quality: a Bayesian Networks approach – Anna Sperotto

Talk: BN Workshop Anna Sperotto

Abstract: Bayesian Networks are employed for the implementation of a multi-risk model to assess cascading impacts induced by multiple stressors on water quality taking into account multiple climate and land use scenarios. Specifically, Bayesian Networks are applied as a meta-modelling tool for structuring and combining the information coming from existing hydrological models simulations, climate change and land use scenarios and to prioritize the contribute of different stressors on water quality status.

A Bayesian approach for safety barrier portfolio optimization – Alessandro Mancuso

Talk: BN Workshop Alessandro Mancuso

Abstract: In the framework of Probabilistic Risk Assessment (PRA), we develop a method to support the selection of cost-effective portfolios of safety measures. This method provides a systemic approach to determining the optimal portfolio of safety measures that minimizes the risk of the system and thus provides an alternative to using risk importance measures for guiding the selection of safety measures. We represent combinations of events leading to system failure with Bayesian Belief Networks (BBNs) which can be derived from traditional Fault Trees (FTs) and are capable of encoding event dependencies and multi-state failure behaviors. We also develop a computationally efficient enumeration algorithm to identify which combinations (portfolios) of safety measures minimize the risk of failure at different costs of implementing the safety measures. The method is illustrated by revisiting an earlier case study concerning the airlock system of a CANDU Nuclear Power Plant (NPP). The comparison of results with those of choosing safety measures based on risk importance measures shows that our approach can lead to considerably lower residuals risks at different cost levels.

Application of Bayesian Networks in integrated water resource management – Hazel Indrani Govender

Talk: BN Workshop Hazel Govender

Abstract: Water catchments are complex, with water resources receiving impacts from a vast range of land-use activities. Integrated Water Resource Management (IWRM) is a holistic approach that attempts to integrate the sustainable management of the water, land and related resources within the broader socio-economic and political context.  Risk assessment at a catchment scale, requires the consideration of multiple stressors and many causal relationships that result from the interactions between the components of ecosystems.  The Relative Risk Model, using Bayesian Networks (BNs), is used to assess the risks in a water stressed, economically critical water catchment in South Africa.  The focus on the study area by the authorities and government, has facilitated a number of research efforts and collaborations.  This is bringing together experts and a range of stakeholders in working towards protection of the water resources in the catchment.  This is beneficial to the application of Bayesian Networks as the information and data resulting from these research efforts can contribute to knowledge gaps and missing data.  This can facilitate updating of the BNs and contribute to an adaptive management approach to protecting water resources in the catchment.

What is needed to get Bayesian Networks robust to weaknesses in knowledge? – Ullrika Sahlin

Talk: BN Workshop Ullrika Sahlin

Abstract: Also the sun has its spots. Bayesian Networks are useful, but has its limitations. I will mention some problems with BNs coming from weaknesses in knowledge. Instead of leaving you in total misery, I will end with some suggestions on how to deal with these issues without totally abandoning Bayesian Networks.

This workshop is funded by the research school ClimBEco 

Local organisers were Indrani Hazel Govender and Ullrika Sahlin

 

meetings/risk research 0

Workshop: Introduction to Bayesian Networks for risk and impact assessment to support decision making

The aim of is workshop was to introduce Bayesian (Belief) Networks to students and researchers. We will provide a theoretical background together with hands on exercises from risk and impact assessments and decision making.

When: Tuesday March 28th at 10 to 16

Where: Heden (room bottom floor), Ecology Building, Lund University, Sölvegatan 37, Lund, Sweden

The workshop is free of charge. The workshop is supported by the research school ClimBEco.

Target audience:

Undergraduate and graduate students and researchers interested in learning how to build and use Bayesian (Belief) Networks to integrate evidence, assess risk or impacts, and evaluate decision alternatives. We will use examples from environmental risk and impact assessments, but the scope of the workshop is BNs with no particular type of application. No previous experience in BNs is needed. Participants are encouraged to study a literature list sent out prior to the workshop.

Motivation:

A Bayesian Network is a modelling framework that enable us to integrate evidence to inform decisions, based on causal relations between decision, state and impact variables. BNs allows integration of data and expert knowledge.

The workshop will be facilitated by:

Wayne Landis, Western Washington University, United States.

Wayne is a Professor in Environmental Science and a Director of the Institute of Environmental Toxicology at Western Washington University. He has over 20 years of experience in ecological risk assessment research, using Bayesian networks to guide decision making. https://huxley.wwu.edu/people/landis

Ullrika Sahlin, Center for Environmental and Climate Research, Lund, Sweden.

Ullrika is a researcher with focus on risk, uncertainty and decision making.

Schedule with links to presentations

Literature

In order to get more out of the workshop we encourage you to study the literature before the workshop.

  • Marcot, B. G., J. D. Steventon, G. D. Sutherland, and R. K. McCann. 2006. Guidelines for developing and updating Bayesian belief networks applied to ecological modeling and conservation. Canadian Journal of Forest Research 36:3063-3074. Link
  • J. Pearl. A short note on Bayesian networks prepared for the MIT Encyclopedia of the Cognitive sciences. Downloaded from http://ftp.cs.ucla.edu/pub/stat_ser/R246.pdf
  • The support documentation for using GENIE includes an introduction to decision analysis and Bayesian networks: http://support.bayesfusion.com/docs/genie/
  • Herring, C. E., J. Stinson, and W. G. Landis. 2015. Evaluating nonindigenous species management in a Bayesian networks derived relative risk framework for Padilla Bay, WA, USA. Integrated Environmental Assessment and Management 11:640-652. Link

We will run examples in R and GENIE (https://download.bayesfusion.com/). In order to do the hands-on exercises we recommend that you have these programs on you computer when you come to the workshop.

 

students/teaching 0

Bayes@Lund2017 – note the date and call for abstracts

Save the date! The fourth edition of Bayes@Lund will be held on 20 April 2017 in Lund, Sweden.

The purpose of this conference is to bring together researchers, practitioners and students working with or interested in Bayesian methods. Bayes@Lund aims at being accessible to those with little experience of Bayesian methods while still being relevant to experienced practitioners.

It will feature keynote lectures from Prof. Darren Wilkinson and Prof. Richard McElreath, in addition to a number of contributed talks. On the 19th April there will be a free introductory tutorial to Bayesian analysis.

The conference is free to attend and registration will open in February. For more information see the conference web page: http://www.maths.lu.se/bayeslund2017/

*** Call for Contributions ***

We’re looking for contributed talks and any topic related to Bayesian analysis is relevant including, but not restricted to:

* Case studies. Have you used Bayesian methods in your research? Describe what you did, and how it worked out.
* Methodology development. Are you developing novel Bayesian methods that you want to share?
* Tutorial style talks. Want to share your favorite tool or modeling technique? This does not have to be novel research, just a useful tool for research!
* Teaching Bayes. Do you have experiences teaching Bayesian methods. What were the challenges, and do you have any useful tips?

To contribute a talk please send a plain text email (no pdf attachments) with the title of your talk, authors names with affiliations, email addresses, and an abstract (not more than 150 words) to bayeslund2017@gmail.com.

As the audience will be highly heterogeneous, the talk is expected to be accessible and engaging for a multidisciplinary audience.

Deadline for abstracts submission is the 10th of February, 2017  (notification of acceptance sent by 15 February).

Each accepted speaker is given 20 minutes (plus 5 minutes for questions).

We’re looking forward to see you in Lund!

 

The organizers

Umberto Picchini and Ullrika Sahlin

Uncategorized 0

Intensive and fun -2016’s graduate course on risk, uncertainty and decision making is now completed

This was a course funded by the research schools LUCID and ClimBeco. 26 students participated. We had two physical meetings. One in Lund and one in Gothenburg. The one in Gothenburg included a 1.5 day risk conference – the 2nd Nordic Chapter Risk meeting. We were a bit afraid that it would be too much, but on the contrary – we felt that the students got exposed to high quality keynotes and contributed talks, saw the width of risk research and actually made new connections. The small format of the risk conference (about 100 people) and friendly atmosphere made them feel welcomed.

Me and Åsa Knaggård decided to have this course since we had a similar one when we were PhD-students. Now, ten years later – it was time to pay back. We learnt a lot and hope that the students did too. It was excellent to have Daniel Slunge on the team to help us out in Gothenburg and using his network to get good lecturers.

~Ullrika

lundpict

 

gbgpictstudents16

Uncategorized 0

Great talks and networking at the 2nd Nordic chapter risk meeting

Last week risk researchers and other interested in the risk field came to Gothenburg to take part in the 2nd annual meeting of the Nordic chapter of the Society for Risk analysis. Last year in Lund was successful so our expectations were high. I must say that this year became even more successful.

The keynote speakers were excellent – all of them. Read more on the web page for the conference

We had fun, the atmosphere was friendly and the topics highly relevant and interesting.

Check out the flow on twitter #srae16 to see what was going on.

A more in depth summary is available nordic_sra_meeting_gbg2016. Thanks to Randa Kachef for the reporting and the high quality photos.

walkommen

chair

renn

Ullrika and Åsa Boholm (chair) opens the meeting. Åsa talks to the president of SRA James Lambert. Ortwin Renn in his keynote.

mingel3

 

folk

ata

mingel

 

 

 

 

 

 

 

 

 

 

 

At the Nordic chapter SRA business meeting Ullrika (president) and Marja (president elect) informed about the next meeting which will be in Finland 2017

3rd

The European SRA conference will be in Sweden 2018.

nextsrae

 

 

 

 

More keynotes: Nick Pidgeon, Sven-Ove Hansson, Barbara Czarniawska, Jaana Huso-Kallio

pidgeonsvenovekeynot

efsa

The Nordic chapter handed out two awards: Caroline Fredriksson for the best student presentation and Christina Mauléon & Maria Spante for the best presentation.

awardstudentaward

meetings/risk research 0

The ESREL risk image competition over

This year the ESREL conference in Glasgow decided to do a risk image competition. What a wonderful idea. Time to go creative.

The task was to illustrate risk. After several months of scratching my head I saw the deadline approaching too fast.

My children came to the rescue. My oldest son had built something in lego that with some slight imagination looked like the Eiffel tower. At the same time the middle son played the board game risk, which we all know has a lot of dices in it. I was cleaning my room and found a dusty mirror in a corner.

Bam – there it was. The image illustrating what it means to assess the probability of unique event – such as a terrorist attack or something which we just have not experienced before.

 

litenbildriskHow do we assess the probability of a unique event? It is not repeatable. Sometimes we just have to believe – somehow. 

Did I win? Well, almost. I got in the final with a nice certificate to put on the wall.

 

Thanks Lesley Walls for arranging this stimulating competition

imagecompsahlin

~Ullrika

 

 

Uncategorized 0

What is Robust analysis?

Does this question appeals to you?

Do you want to apply Bayesian modelling for the purpose to quantify uncertainty and support decision making? Do you feel inspired reading texts about uncertainty? Can you tell a bee from a wasp? Do you want to do a PhD in an interdisciplinary environment?

Then perhaps our new PhD project on Robust analysis is something for you. Last day to apply is October 3rd 2016.

 

Uncategorized 0

Graduate course in Risk, Uncertainty and Decision Making

The course is over. Check this post for a summary.

Registration is now closed for the PhD course in risk, uncertainty and decision making.

risk_measurement_400_clr_5483

We make a number of decisions every day. These decisions are based on conscious or unconscious analysis, as well as assessments of risks and uncertainties. We are often unaware of how we make these decisions, and how we compare alternatives. The decisions can be about everyday situations, like if to bring an umbrella, or about complex and urgent issues that face society, like issues connected with the environment or sustainability. The course focuses on a number of issues: How can risk be assessed? Can risk be quantified? How do risk assessments affect decision and policy making? What role does risk communication and media play? How do individuals perceive risks and risk management? How does the concept of risk influence the way we understand the world and try to control it? Risk, uncertainty, and decision analysis implies systematic efforts to understand the context and consequences of decisions.

The aim of the course is to give a deeper understanding of the theoretical perspectives and methods in risk research within different disciplines, and tools for interdisciplinary risk research. The course contains lectures and seminars. The lectures will provide rich accounts of different disciplinary perspectives.  The participants will in seminars get a chance to collaborate on questions requiring an interdisciplinary approach.

Target group: Graduate students from different disciplines

Extent: 7.5 credits

The course has two physical meetings:

  • October 13-14th in Lund (hosted by LUCRAM, Lund University), starts 9.00 the 13th, ends 15.00 the 14th
  • November 13-16th in Gothenburg (hosted by the School of Business, Economics and Law, Gothenburg University), starts 14.00 the 13th, ends 12.00 16th

Participants are expected to prepare in advance to the meetings by reading the assigned literature. The course is examined through attendance of the physical meetings and a completed paper where at least two themes of the course is discussed in relation to the participants PhD subject or field of interest.

The second meeting includes the second Nordic risk conference held in November 14-15th 2016. As course participant you have two options: (1)  To send in an abstract to the conference for an oral presentation. If accepted the participant should pay the conference fee to the conference organisers. The conference fee is 1000 SEK for PhD and master students (and 1500 SEK otherwise). (2) To participate in the conference without giving any oral presentation. If so, the course will compensate the conference fee after a completed course.

This course is funded by the research program LUCID and the research school ClimBECO and is given as part of the Lund University Centre for Risk Assessment and Managment (LUCRAM). It is open for any PhD student to attend. The number of places is limited.

Course directors: Åsa Knaggård (LU), Ullrika Sahlin (LU) och Daniel Slunge (GU).

The course will be given in English

Cost: The conference fee to attend the Nordic Chapter SRA conference

Lund – October

Where: Kärret bottom floor in the Ecology Building, Sölvegatan 37 (close to LTH student union house)
Thursday 13
9.30 to 12.00 Henrik Tehler (Division of Risk Management and Social Safety, LTH): Introduction to risk assessment
13.00 to 14.30 Annika Wallin (Department of Philosophy, Lund University): Individual decision making
14.45 to 15.45 Henrik Hassel (Division of Risk Management and Social Safety, LTH): Quantitative descriptions of risk
16.00 to 17.00 Ullrika Sahlin (Centre for Environmental and Climate Research, Lund University): Probability

Friday 14
9 to 10.30 Johannes Persson, (Department of Philosophy, Lund University): Risk philosophy
11 to 12 Frederik Lundtofte (School of Economics and Management, Lund University): Risk in an economist’s perspective
13 to 14 Literature discussion
14 to 15 Ullrika Sahlin: Mind the risk – a case-study on planning under risk and uncertainty

Göteborg – November
Sunday 13
14 to 15.15 Åsa Knaggård (Department of Political Science, Lund University): Policy-making under risk and uncertainty
15.30 to 16.45 Ragnar Löfstedt (King’s Centre for Risk Management, King’s College London): Risk communication in the 21st century
17 to 18.15 Nick Pidgeon (School of Psychology, Cardiff University): Advances in risk communication
18.30 to 19 Joint discussion with students

Monday 14
9.30 to 18 Risk conference and group discussions

Tuesday 15
9 to 13 Risk conference
14.30 to 15.30 Tomas Bergström (Institute of Biomedicine, Gothenburg University): Ticks – case-study on risk perception and risk communication
16 to 17 Daniel Slunge (Centre for Environment and Sustainability, Gothenburg University): Ticks – case-study on risk perception and risk communication

Wednesday 16
9 to 10 Daniel Slunge: Ticks – case-study on risk perception and risk communication
10.30 to 12 Ullrika Sahlin: Uncertainty

 

meetings/risk research/students/teaching 2

Join the SRA-E: NORDIC meeting during the SRA conference in Bath on June 21st

The Nordic chapter of SRA Europe invites members and non-members to a chapter meeting during the SRA conference in Bath, UK.

The meeting will take place in room CB 5.6 in the Chancellors Building on Tuesday 21st June at 12.30 – 1.30pm.

It will be possible to attend the meeting via skype. Send an email to Ullrika.Sahlin [at] cec.lu.se if you want to join.

Invitation as a pdf

Info about SRA-E:NORDIC

Agenda

  1. Information about the SRA Nordic chapter

1.1. The purpose of SRA-E:NORDIC

1.2. Past activities

  1. Information of planned activities

2.1. Invitation to submit abstract and session proposals to the 2nd Nordic Chapter conference in Gothenburg Nov 2016

2.2. Planned courses in the Nordic/Baltic region (contributions are welcome)

2.3. Planned workshops in the Nordic/Baltic region (contributions are welcome)

  1. Discussions and future activities

Shared risk issues, Suggestions for policy impact, research needs etc. (contributions are welcome)


Board members:

Ullrika Sahlin, Lund University, Sweden (President for two years  – starting from June 2015)

Marja Ylönen, Technical Research Centre of Finland, Finland (President elect – to take over in June 2017 and then be president for two years)

Stian Antonsen, SINTEF, Norway

Susanne Öhman, Mid Sweden University, Sweden

Igor Kozine, Technical University of Denmark, Denmark

Aistė Balžekienė, Kaunas University of Technology, Lithuania

Student representatives:

Ingibjörg Lilja, University of Iceland, Iceland

Björn Arvidsson, Lund University, Sweden

 

Uncategorized 0

EFSAs treatment of uncertainty under trial

The way uncertainty is treated in risk assessment is under scrutiny. There is a need for this since we today face more complex problems and need to take into account possible risks for which knowledge is weak.

One example is the ongoing work in the EU on how we treat uncertainty in environmental and health risk assessments.

In March 2016 the European Food Safety Authority (EFSA) published the draft scientific Guidance document on Uncertainty in EFSA Scientific Assessment. The Guidance has been exposed to a public consultation and stakeholders provided many comments, which now have been addressed.

According to EFSA, “the Guidance is applicable to all areas of EFSA, all types of scientific assessment and all types of uncertainty affecting scientific assessment. It does not prescribe specific methods for uncertainty analysis but rather provides a harmonised and flexible framework within which different methods may be selected, according to the needs of each assessment.”

EFSA’s scientific expert panels will use the Guidance for a trial period for about two years. In 2017 the Guidance will be revised according to the problems identified. This is a great opportunity to discuss the treatment of uncertainty, not only in EFSA’s risk assessments. We wrote an oped in Sydsvenskan about this.

EFSA is the agency responsible for the risk assessments of threats posed to living organisms for the EU commission. EFSA is situated in Parma, Italy.

 

info/risk concept/risk research/Uncategorized 0