Discretization of continuous predictor variables in Bayesian networks: An ecological threshold approach

Lucena-Moya, Paloma; Brawata, Renee L.; Kath, Jarrod; Harrison, Evan; Elsawah, Sondoss; Dyer, Fiona

doi:10.1016/j.envsoft.2014.12.019

Cited by 18 publications

(11 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is likely that the sequencing of deceased organisms, an artifact of eDNA sampling, was a major contributor to the variability in the benthic eDNA data (Chariton et al ). These error rates are consistent with other BN models that predicted benthic taxa by using case learning to parameterize the CPTs (Lucena‐Moya et al ).…”

Section: Resultssupporting

confidence: 89%

“…A single file was used because the eDNA data set for individual estuaries was relatively small. Using benthic samples from different estuaries to parameterize CPTs is consistent with Lucena‐Moya et al (), who also used BNs and case learning to predict ecological assemblages.…”

Section: Methodssupporting

confidence: 75%

“…The CPTs in the SEQ BN‐RRMs were parameterized with one method: case files and the expectation‐maximization (EM‐learning) algorithm, an automated function available in Netica (Norsys ). We chose this method based on the study by Lucena‐Moya et al (), who also used case learning to parameterize predictor nodes for benthic endpoints. A case file is a compilation of data sets that go together to provide information about the variables (nodes) in the model.…”

Section: Methodsmentioning

confidence: 99%

“…The EM‐learning algorithm iteratively calculates the maximum likelihood estimates for the node states in the model given the case file data and the model structure. Expectation‐maximum learning has been routinely used for other environmental BN modeling (Pollino et al ; Lucena‐Moya et al ) and was selected over other algorithms because it deals well with missing data.…”

Section: Methodsmentioning

confidence: 99%

See 3 more Smart Citations

Using Bayesian networks to predict risk to estuary water quality and patterns of benthic environmental DNA in Queensland

Graham

Chariton

Landis

2018

Integr Envir Assess & Manag

View full text Add to dashboard Cite

Predictive modeling can inform natural resource management by representing stressor-response pathways in a logical way and quantifying the effects on selected endpoints. This study demonstrates a risk assessment model using the Bayesian network relative risk model (BN-RRM) approach to predict water quality and, for the first time, eukaryote environmental DNA (eDNA) data as a measure of benthic community structure. Environmental DNA sampling is a technique for biodiversity measurements that involves extracting DNA from environmental samples, amplicon sequencing a targeted gene, in this case the 18s rDNA gene (which targets eukaryotes), and matching the sequences to organisms. Using a network of probability distributions, the BN-RRM model predicts risk to water quality objectives and the relative richness of benthic taxa groups in the Noosa, Pine, and Logan estuaries in Southeast Queensland (SEQ), Australia. The model predicts Dissloved Oxygen more accurately than the chlorophyll a water quality endpoint and photosynthesizing benthos more accurately than heterotrophs. Results of BN-RRM modeling given current inputs indicate that the water quality and benthic assemblages of the Noosa are relatively homogenous across all sub risk regions, and that the Noosa has a 73%-92% probability of achieving water quality objectives, indicating a low relative risk. Conversely, the Middle Logan, Middle Pine, and Lower Pine regions are much less likely to meet objectives (15%-55% probability), indicating a relatively higher risk to water quality in those regions. The benthic community richness patterns associated with risk in the Noosa are high Diatom relative richness and low Green Algae relative richness. The only benthic pattern consistently associated with the relatively higher risk to water quality is high richness of fungi species. The BN-RRM model provides a basis for future predictions and adaptive management at the direction of resource managers. Integr Environ Assess Manag 2018;00:000-000. © 2018 SETAC.

show abstract

Section: Resultssupporting

confidence: 89%

Section: Methodssupporting

confidence: 75%

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

See 2 more Smart Citations

Using Bayesian networks to predict risk to estuary water quality and patterns of benthic environmental DNA in Queensland

Graham

Chariton

Landis

2018

Integr Envir Assess & Manag

View full text Add to dashboard Cite

show abstract

“…To avoid inconsistency in parameterization and to handle high sensitivities in the ecological domain, we used a weighting of expert inputs [46]. We decided not to lower or limit the number of parent nodes to keep the flexibility as high as possible.…”

Section: Limitations Of Bbn Developmentmentioning

confidence: 99%

A Bayesian Belief Network Approach to Predict Damages Caused by Disturbance Agents

Radl¹,

Lexer²,

Vacik³

2017

Forests

View full text Add to dashboard Cite

Abstract:In mountain forests of Central Europe, storm and snow breakage as well as bark beetles are the prevailing major disturbances. The complex interrelatedness between climate, disturbance agents, and forest management increases the need for an integrative approach explicitly addressing the multiple interactions between environmental changes, forest management, and disturbance agents to support forest resource managers in adaptive management. Empirical data with a comprehensive coverage for modelling the susceptibility of forests and the impact of disturbance agents are rare, thus making probabilistic models, based on expert knowledge, one of the few modelling approaches that are able to handle uncertainties due to the available information. Bayesian belief networks (BBNs) are a kind of probabilistic graphical model that has become very popular to practitioners and scientists mainly due to considerations of risk and uncertainties. In this contribution, we present a development methodology to define and parameterize BBNs based on expert elicitation and approximation. We modelled storm and bark beetle disturbances agents, analyzed effects of the development methodology on model structure, and evaluated behavior with stand data from Norway spruce (Picea abies (L.) Karst.) forests in southern Austria. The high vulnerability of the case study area according to different disturbance agents makes it particularly suitable for testing the BBN model.

show abstract

Dynamic Bayesian Networks to Assess Anthropogenic and Climatic Drivers of Saltwater Intrusion: A Decision Support Tool Toward Improved Management

Rachid

Alameddine

Najm

et al. 2020

Integr Envir Assess & Manag

View full text Add to dashboard Cite

show abstract

Discretization of continuous predictor variables in Bayesian networks: An ecological threshold approach

Cited by 18 publications

References 33 publications

Using Bayesian networks to predict risk to estuary water quality and patterns of benthic environmental DNA in Queensland

Using Bayesian networks to predict risk to estuary water quality and patterns of benthic environmental DNA in Queensland

A Bayesian Belief Network Approach to Predict Damages Caused by Disturbance Agents

Dynamic Bayesian Networks to Assess Anthropogenic and Climatic Drivers of Saltwater Intrusion: A Decision Support Tool Toward Improved Management

Contact Info

Product

Resources

About