How good are Bayesian belief networks for environmental management? A test with data from an agricultural river catchment

Abstract

SUMMARY1. The ecological health of rivers worldwide continues to decline despite increasing effort and investment in river science and management. Bayesian belief networks (BBNs) are increasingly being used as a mechanism for decision-making in river management because they provide a simple visual framework to explore different management scenarios for the multiple stressors that impact rivers. However, most applications of BBN modelling to resource management use expert knowledge and/or limited real data, and fail to accurately assess the ability of the model to make predictions.2. We developed a BBN to model ecological condition in a New Zealand river using field/GIS data (from multiple rivers), rather than expert opinion, and assessed its predictive ability on an independent dataset. The developed BBN performed moderately better than a number of other modelling techniques (e.g., artificial neural networks, classification trees, random forest, logistic regression), although model construction was more time-consuming. Thus the predictive ability of BBNs is (in this case at least) on a par with other modelling methods but the approach is distinctly better for its ability to visually present the data linkages, issues and potential outcomes of management options in real time. We have demonstrated this for a BBN of ecological condition in a New Zealand river, shown that model fit is better than that for other modelling techniques, and that improving habitat would be equally effective to reducing nutrients to improve ecological condition.

A review of macroinvertebrate‐ and fish‐based stream health modelling techniques

“…• Low predictive power • Model structure (distributions selection) must be defined a priori Low Low Damanik-Ambarita et al, 2016;Death et al, 2015;Donohue et al, 2006;Everaert et al, 2014;Gieswein et al, 2017;Holguin-Gonzalez, Everaert, et al, 2013;Jerves-Cobo et al, 2017;Kuemmerlen et al, 2014;Moya et al, 2011;Pont et al, 2009;Sauer, Domisch, Nowak, & Haase, 2011;Van Sickle et al, 2004Fukuda et al, 2013Gieswein et al, 2017;Grenouillet et al, 2011;Hermoso, Linke, Prenda, & Possingham, 2011;Kwon, Bae, Hwang, Kim, & Park, 2015;Leclere et al, 2011;Patrick & Yuan, 2017;Sui et al, 2014 Generalized additive models • Suitable for modelling nonlinear relationships • Uses nonparametric basis functions • Prone t...…”

“…Pressures on freshwater ecosystems comprise hydrologic and hydraulic stress, sediment intake, and substrate stability, among others, which can be represented by process-based models developed in Merriam et al, 2013Merriam et al, , 2015Pond et al, 2017;Waite et al, 2010, 2012Frimpong et al, 2005Van Sickle & Burch Johnson, 2008 Generalized linear models Death et al, 2015;Donohue et al, 2006;Everaert et al, 2014;Gieswein et al, 2017;Holguin-Gonzalez, Everaert, et al, 2013;Jerves-Cobo et al, 2017;Kuemmerlen et al, 2014;Moya et al, 2011;Pont et al, 2009;Sauer, Domisch, Nowak, & Haase, 2011;Van Sickle et al, 2004Fukuda et al, 2013Gieswein et al, 2017;Grenouillet et al, 2011;…”

“…Medium Maloney et al, 2012;Sauer et al, 2011Almeida et al, 2017Fukuda et al, 2013;Grenouillet et al, 2011;Maloney et al, 2012;Zhao et al, 2014 Shrinkage methods • Straightforward implementation and interpretation Ambelu et al, 2010;Death et al, 2015;Holguin-Gonzalez et al, 2014;Maloney et Grenouillet et al, 2011;He et al, 2010;Kwon et al, 2015; (Continues) Brown et al, 2012;Clapcott et al, 2012Clapcott et al, , 2017Clapcott et al, , 2014; J. T. May et al, 2015;Pilière et al, 2014;A. E. Steel et al, 2017;Tonkin et al, 2014;Wagenhoff et al, 2016;Waite et ...…”

“…Random forests Booker et al, 2015;Chinnayakanahalli et al, 2011;Clapcott et al, 2017;Death et al, 2015;Fox et al, 2017;Hill et al, 2017;Patrick & Yuan, 2017;Vander Laan et al, 2013;Waite et al, 2012Álvarez-Cabria et al, 2017Fukuda & De Baets, 2016;Fukuda et al, 2013;Grenouillet et al, 2011;He et al, 2010;Kwon et al, 2015 Chon, 2011;Gazendam et al, 2016;Goethals et al, 2007;Mathon et al, 2013; Ans M. Mouton et al, 2010;Sauer et al, 2011Chon, 2011Fukuda et al, 2013;…”

“…High Allan et al, 2012;Boets et al, 2015;Death et al, 2015;Forio et al, 2015;Xue Li et al, 2018;Mantyka-Pringle et al, 2014;McLaughlin & Reckhow, 2017Mantyka-Pringle et al, 2017Peterson et al, 2013;Turschwell et al, 2017;Vilizzi et al, 2013 (Continues) several areas such as ecohydrology and ecohydraulics. State refers to the outputs driven by the aforementioned pressures (e.g., extreme events, sediment, velocity, depth, substrate, and nutrients).…”