A review of Bayesian belief networks in ecosystem service modelling
TL;DR: This review discusses a number of BBN-based ESS models developed in the last decade and highlights the advantages and disadvantages of BBNs in ESS modelling and pinpoints remaining challenges for future research.
Abstract: A wide range of quantitative and qualitative modelling research on ecosystem services (ESS) has recently been conducted. The available models range between elementary, indicator-based models and complex process-based systems. A semi-quantitative modelling approach that has recently gained importance in ecological modelling is Bayesian belief networks (BBNs). Due to their high transparency, the possibility to combine empirical data with expert knowledge and their explicit treatment of uncertainties, BBNs can make a considerable contribution to the ESS modelling research. However, the number of applications of BBNs in ESS modelling is still limited. This review discusses a number of BBN-based ESS models developed in the last decade. A SWOT analysis highlights the advantages and disadvantages of BBNs in ESS modelling and pinpoints remaining challenges for future research. The existing BBN models are suited to describe, analyse, predict and value ESS. Nevertheless, some weaknesses have to be considered, including poor flexibility of frequently applied software packages, difficulties in eliciting expert knowledge and the inability to model feedback loops. BBNs are increasingly used to analyse, predict and value ecosystem services (ESS).Most BBN applications in ESS modelling target only a single service.Numerous advantages of BBNs in ESS modelling are demonstrated in current applications.Model drawbacks are absence of feedback loops and obligatory variable discretization.Spatially explicit modelling and modelling of ESS bundles are future opportunities.
Citations
More filters
TL;DR: In this article, the authors proposed to characterize three broad types of associations considering the ecological (supply side) and socio-economical (demand side) aspects of ecosystem services: supply-supply, supply-demand and demand-demand.
Abstract: Considering the increasing uptake of the concept of "ecosystem services" in landscape management and environmental policies, it is urgent to establish a consensual framework to assess the complex relationships among ecosystem services, considering both the supply- and the demand-sides. A diversity of approaches have been proposed to evaluate ecosystem services associations, but not all methods are equivalent and methodological choices need to be made depending on the scientific and policy questions at hand, as well as the type of data available. Based on previous classifications of ecosystem service associations, we propose to characterize three broad types of associations considering the ecological (supply side) and socio-economical (demand side) aspects of ecosystem services: supply-supply, supply-demand and demand-demand. We then review quantitative methods available and propose guidelines to assess those three categories of relationships among ecosystem services and identify their explanatory variables following three steps: (i) detecting ecosystem services associations, (ii) defining bundles and (iii) identifying the explanatory variables of ecosystem services associations. For each step, strengths and weaknesses of different statistical analysis and machine learning methods are described. The proposed interdisciplinary methodological approach takes one step toward embracing such complexity of socio-ecological systems as it considers ecosystem services delivery (supply-supply), stakeholders' needs (demand-demand), and on how stakeholders can benefit from the ecosystem services delivery (supply-demand). We illustrate how such a diverse spectrum of methods may apply for land management.
283 citations
TL;DR: In this paper, the authors argue that a prerequisite to progress in such public deliberations is that participants be very cognizant of the extreme relevance of soils to many aspects of their daily life, and that, as long as this prerequisite is satisfied, the combination of deliberative decision-making methods and of a sound scientific approach to quantify soil functions/services is a very promising avenue to manage effectively and ethically the priceless heritage that soils constitute.
Abstract: Over the last few years, considerable attention has been devoted in the scientific literature and in the media to the concept of "ecosystem" services of soils. The monetary valuation of these services, demanded by many governments and international agencies, is often depicted as a necessary condition for the preservation of the natural capital that soils represent. This focus on soil services is framed in the context of a general interest in ecosystem services that allegedly started in 1997, and took off in earnest after 2005. The careful analysis of the literature proposed in this article shows that, in fact, interest in the multifunctionality of soils emerged already in the mid-60s, at a time when hundreds of researchers worldwide were trying, and largely failing, to figure out how to put price tags meaningfully on "nature's services." Soil scientists, since, have tried to better understand various functions/services of soils, as well as their possible relation with key soil characteristics, like biodiversity. They have also tried to make progress on the challenging quantification of soil functions/services. However, researchers have shown very little interest in monetary valuation, undoubtedly in part because it is not clear what economic and financial markets might do with prices of soil functions/services, even if we could somehow come up with such numbers, and because there is no assurance at all, based on neoclassical economic theory, that markets would manage soil resources optimally. Instead of monetary valuation, focus in the literature has been put on decision-making methods, like Multi-Criteria Decision Analysis (MCDA) and Bayesian Belief Networks (BBN), which do not require the systematic monetization of soil functions/services and easily accommodate deliberative approaches involving a variety of stakeholders. A prerequisite to progress in such public deliberations is that participants be very cognizant of the extreme relevance of soils to many aspects of their daily life. We argue that, as long as this prerequisite is satisfied, the combination of deliberative decision-making methods and of a sound scientific approach to the quantification of soil functions/services is a very promising avenue to manage effectively and ethically the priceless heritage that soils constitute.
262 citations
Cites background from "A review of Bayesian belief network..."
...…(e.g., Sadoddin et al., 2009; Haines-Young, 2011; Farmani et al., 2012; Henriksen et al., 2012; Troldborg et al., 2013), in particular to quantify and map ecosystem services (Grêt-Regamey et al., 2013; Landuyt et al., 2013, 2015; Celio et al., 2014; Rositano and Ferraro, 2014; Taalab et al., 2015)....
[...]
TL;DR: It is argued that an extended matrix model could provide more than only scientifically sound and politically legitimate results and serve as a tool to improve cooperation between natural and social sciences, experts, stakeholders and decision makers.
254 citations
Cites background from "A review of Bayesian belief network..."
...The participative construction of such models could provide a go-between to bridge the gap between natural and social scientists, but also to realize inclusive and transparent ES-model development (see Landuyt et al., 2013 for an elaborated SWOT of Bayesian models for ES research)....
[...]
...Also, Bayesian modeling applications are of particular interest to integrate expert knowledge with multiple data sources in ES modeling (Haines-Young, 2011; Landuyt et al., 2013; Van der Biest et al., 2014; Gret-Regamey et al., 2013)....
[...]
TL;DR: In this paper, a new type of indicator called Benefit-relevant indicators (BRIs) is proposed, which explicitly reflects an ecosystem's capacity to provide benefits to society, ensuring that ecosystem services assessments measure outcomes that are demonstrably and directly relevant to human welfare.
158 citations
TL;DR: New methods by which BN model development and application are being joined with other tools and model frameworks are explored, including improving areas of Bayesian classifiers and machine-learning algorithms for model structuring and parameterization, and development of time-dynamic models.
Abstract: Bayesian network (BN) modeling is a rapidly advancing field. Here we explore new methods by which BN model development and application are being joined with other tools and model frameworks. Advances include improving areas of Bayesian classifiers and machine-learning algorithms for model structuring and parameterization, and development of time-dynamic models. Increasingly, BN models are being integrated with: management decision networks; structural equation modeling of causal networks; Bayesian neural networks; combined discrete and continuous variables; object-oriented and agent-based models; state-and-transition models; geographic information systems; quantum probability; and other fields. Integrated BNs (IBNs) are becoming useful tools in risk analysis, risk management, and decision science for resource planning and environmental management. In the near future, IBNs may become self-structuring, self-learning systems fed by real-time monitoring data. Such advances may make model validation difficult, and may question model credibility, particularly if based on uncertain sources of knowledge systems and big data.
158 citations
References
More filters
Book•
01 Jan 2001TL;DR: The book introduces probabilistic graphical models and decision graphs, including Bayesian networks and influence diagrams, and presents a thorough introduction to state-of-the-art solution and analysis algorithms.
Abstract: Probabilistic graphical models and decision graphs are powerful modeling tools for reasoning and decision making under uncertainty. As modeling languages they allow a natural specification of problem domains with inherent uncertainty, and from a computational perspective they support efficient algorithms for automatic construction and query answering. This includes belief updating, finding the most probable explanation for the observed evidence, detecting conflicts in the evidence entered into the network, determining optimal strategies, analyzing for relevance, and performing sensitivity analysis. The book introduces probabilistic graphical models and decision graphs, including Bayesian networks and influence diagrams. The reader is introduced to the two types of frameworks through examples and exercises, which also instruct the reader on how to build these models. The book is a new edition of Bayesian Networks and Decision Graphs by Finn V. Jensen. The new edition is structured into two parts. The first part focuses on probabilistic graphical models. Compared with the previous book, the new edition also includes a thorough description of recent extensions to the Bayesian network modeling language, advances in exact and approximate belief updating algorithms, and methods for learning both the structure and the parameters of a Bayesian network. The second part deals with decision graphs, and in addition to the frameworks described in the previous edition, it also introduces Markov decision processes and partially ordered decision problems. The authors also provide a well-founded practical introduction to Bayesian networks, object-oriented Bayesian networks, decision trees, influence diagrams (and variants hereof), and Markov decision processes. give practical advice on the construction of Bayesian networks, decision trees, and influence diagrams from domain knowledge. give several examples and exercises exploiting computer systems for dealing with Bayesian networks and decision graphs. present a thorough introduction to state-of-the-art solution and analysis algorithms. The book is intended as a textbook, but it can also be used for self-study and as a reference book.
4,566 citations
"A review of Bayesian belief network..." refers background or methods in this paper
...Moreover, software integration offers the potential to couple ESS BBNs with established ESS models....
[...]
...While the graphical representation of most BBN models is relatively simple compared to other modelling techniques, the use of numerous links and variables can considerably increase the complexity of ESS BBNs (Ordonez Galan et al., 2009; Getoor et al., 2004; Cain, 2001; Jensen, 2001)....
[...]
...Coupling of individual ESS BBNs offers the opportunity to expand existing BBNs towards models that assess bundles of ESS....
[...]
...This discretization often causes information loss (Aguilera et al., 2010; Jensen, 2001)....
[...]
...To obtain quantitative model outputs, Bayesian inference is used to propagate these probabilities through the network (Aguilera et al., 2011; Jensen, 2001)....
[...]
TL;DR: Nature's Services brings together world-renowned scientists from a variety of disciplines to examine the character and value of ecosystem services, the damage that has been done to them, and the consequent implications for human society.
Abstract: Life itself as well as the entire human economy depends on goods and services provided by earth's natural systems. The processes of cleansing, recycling, and renewal, along with goods such as seafood, forage, and timber, are worth many trillions of dollars annually, and nothing could live without them. Yet growing human impacts on the environment are profoundly disrupting the functioning of natural systems and imperiling the delivery of these services.Nature's Services brings together world-renowned scientists from a variety of disciplines to examine the character and value of ecosystem services, the damage that has been done to them, and the consequent implications for human society. Contributors including Paul R. Ehrlich, Donald Kennedy, Pamela A. Matson, Robert Costanza, Gary Paul Nabhan, Jane Lubchenco, Sandra Postel, and Norman Myers present a detailed synthesis of our current understanding of a suite of ecosystem services and a preliminary assessment of their economic value. Chapters consider: major services including climate regulation, soil fertility, pollination, and pest control philosophical and economic issues of valuation case studies of specific ecosystems and services implication of recent findings and steps that must be taken to address the most pressing concerns Nature's Services represents one of the first efforts by scientists to provide an overview of the many benefits and services that nature offers to people and the extent to which we are all vitally dependent on those services. The book enhances our understanding of the value of the natural systems that surround us and can play an essential role in encouraging greater efforts to protect the earth's basic life-support systems before it is too late. -- publisher's description
3,601 citations
"A review of Bayesian belief network..." refers background in this paper
...While scientists have studied this subject for decades (Daily, 1997), the term ESS was introduced to the general public in 2005 and led to the broad-scale public recognition of ecosystems and their value for humanwell-being (MEA, 2005)....
[...]
Book•
01 Jan 1990TL;DR: In this paper, the authors present a software tool for uncertainty analysis, called Analytica, for quantitative policy analysis, which can be used to perform probability assessment and propagation and analysis of uncertainty.
Abstract: Preface 1. Introduction 2. Recent milestones 3. An overview of quantitative policy analysis 4. The nature and sources of uncertainty 5. Probability distributions and statistical estimation 6. Human judgement about and with uncertainty 7. Performing probability assessment 8. The propagation and analysis of uncertainty 9. The graphic communication of uncertainty 10. Analytica: a software tool for uncertainty analysis 11. Large and complex models 12. The value of knowing how little you know Index.
2,666 citations
TL;DR: It is shown that if the network is singly connected (e.g. tree-structured), then probabilities can be updated by local propagation in an isomorphic network of parallel and autonomous processors and that the impact of new information can be imparted to all propositions in time proportional to the longest path in the network.
2,266 citations
TL;DR: The Models-3 CMAQ system as mentioned in this paper is a community multiscale air quality modeling system that includes a meteorological modeling system for the description of atmospheric states and motions, emission models for man-made and natural emissions that are injected into the atmosphere, and a chemistry-transport modelling system for simulation of the chemical transformation and fate.
Abstract: This article describes the governing equations, computational algorithms, and other components entering into the Community Multiscale Air Quality (CMAQ) modeling system. This system has been designed to approach air quality as a whole by including state-ofthe-science capabilities for modeling multiple air quality issues, including tropospheric ozone, fine particles, acid deposition, and visibility degradation. CMAQ was also designed to have multiscale capabilities so that separate models were not needed for urban and regional scale air quality modeling. By making CMAQ a modeling system that addresses multiple pollutants and different spatial scales, it has a “one-atmosphere” perspective that combines the efforts of the scientific community. To implement multiscale capabilities in CMAQ, several issues (such as scalable atmospheric dynamics and generalized coordinates), which depend on the desired model resolution, are addressed. A set of governing equations for compressible nonhydrostatic atmospheres is available to better resolve atmospheric dynamics at smaller scales. Because CMAQ is designed to handle scale-dependent meteorological formulations and a large amount of flexibility, its governing equations are expressed in a generalized coordinate system. This approach ensures consistency between CMAQ and the meteorological modeling system. The generalized coordinate system determines the necessary grid and coordinate transformations, and it can accommodate various vertical coordinates and map projections. The CMAQ modeling system simulates various chemical and physical processes that are thought to be important for understanding atmospheric trace gas transformations and distributions. The modeling system contains three types of modeling components (Models-3): a meteorological modeling system for the description of atmospheric states and motions, emission models for man-made and natural emissions that are injected into the atmosphere, and a chemistry-transport modeling system for simulation of the chemical transformation and fate. The chemical transport model includes the following process modules: horizontal advection, vertical advection, mass conservation adjustments for advection processes, horizontal diffusion, vertical diffusion, gas-phase chemical reactions and solvers, photolytic rate computation, aqueous-phase reactions and cloud mixing, aerosol dynamics, size distributions and chemistry, plume chemistry effects, and gas and aerosol deposition velocity estimation. This paper describes the Models-3 CMAQ system, its governing equations, important science algorithms, and a few application examples. This review article cites 114 references. DOI: 10.1115/1.2128636
1,993 citations
"A review of Bayesian belief network..." refers background in this paper
...Sufficient process knowledge to develop mechanistic models to quantify ESS like improved air quality, carbon sequestration in soils and erosion prevention is one possible reason for the lack of BBN applications (Skjemstad et al., 2004; Byun and Schere, 2006; Merritt et al., 2003)....
[...]