An extended Bayesian sediment fingerprinting mixing model for the full Bayes treatment of geochemical uncertainties

Recent advances in sediment fingerprinting research have seen Bayesian mixing models being increasingly employed as an effective method to coherently translate component uncertainties into source apportionment results. Here, we advance earlier work by presenting an extended Bayesian mixing model capable of providing a full Bayes treatment of geochemical uncertainties. The performance of the extended full Bayes model was assessed against the equivalent empirical Bayes model and traditional frequentist optimisation. The performance of models coded in different Bayesian software (‘JAGS’ and ‘Stan’) was also evaluated, alongside an assessment of model sensitivity to reduced source representativeness and non-conservative fingerprint behaviour. Results revealed comparable accuracy and precision for the full and empirical Bayes models across both synthetic and real sediment geochemistry datasets, demonstrating that the empirical treatment of source data here represents a close approximation of the full Bayes treatment. Contrasts in the performance of models coded in JAGS and Stan revealed that the choice of software employed can impact significantly upon source apportionment results. Bayesian models coded in Stan were the least sensitive to both reduced source representativeness and non-conservative fingerprint behaviour, indicating Stan as the preferred software for future Bayesian sediment fingerprinting studies. Whilst the frequentist optimisation generally yielded comparable accuracy to the Bayesian models, uncertainties around apportionment estimates were substantially greater and the frequentist model was less effective at dealing with non-conservative behaviour. Overall, the effective performance of the extended full Bayes mixing model coded in Stan represents a notable advancement in source apportionment modelling relative to previous approaches. Both the mixing model and the software comparisons presented here should provide useful guidelines for future sediment fingerprinting studies.

[1]  Daniel E. Schindler,et al.  Including source uncertainty and prior information in the analysis of stable isotope mixing models. , 2010, Environmental science & technology.

[2]  K. Hiscock,et al.  Apportioning sources of organic matter in streambed sediments: an integrated molecular and compound-specific stable isotope approach. , 2015, The Science of the total environment.

[3]  K. Hiscock,et al.  High‐temporal resolution fluvial sediment source fingerprinting with uncertainty: a Bayesian approach , 2015 .

[4]  M. Cowles,et al.  An enhanced Bayesian fingerprinting framework for studying sediment source dynamics in intensively managed landscapes , 2016 .

[5]  D. Walling.  The evolution of sediment source fingerprinting investigations in fluvial systems , 2013, Journal of Soils and Sediments.

[6]  Mark A. Nearing,et al.  Sediment tracers in water erosion studies: current approaches and challenges , 2013, Journal of Soils and Sediments.

[7]  A. Thompson,et al.  Apportionment of suspended sediment sources in an agricultural watershed using sediment fingerprinting , 2015 .

[8]  Stewart W. Franks,et al.  Sediment fingerprinting as an environmental forensics tool explaining cyanobacteria blooms in lakes , 2012 .

[9]  A. Massoudieh,et al.  Sediment source apportionment in Laurel Hill Creek, PA, using Bayesian chemical mass balance and isotope fingerprinting , 2015 .

[10]  K. Hiscock,et al.  Apportioning sources of organic matter in streambed sediments : 1 An integrated hydrogen and carbon stable isotope approach , 2015 .

[11]  P. Belmont,et al.  Toward generalizable sediment fingerprinting with tracers that are conservative and nonconservative over sediment routing timescales , 2014, Journal of Soils and Sediments.

[12]  Katrin Meusburger,et al.  Quantitative sediment source attribution with compound-specific isotope analysis in a C3 plant-dominated catchment (central Switzerland) , 2016 .

[13]  K. Hiscock,et al.  Combining two filter paper‐based analytical methods to monitor temporal variations in the geochemical properties of fluvial suspended particulate matter , 2014 .

[14]  S. Pulley,et al.  The impact of catchment source group classification on the accuracy of sediment fingerprinting outputs. , 2017, Journal of environmental management.

[15]  D. Walling,et al.  Source type ascription for fluvial suspended sediment based on a quantitative composite fingerprinting technique , 1997 .

[16]  J. Middelburg,et al.  A Bayesian compositional estimator for microbial taxonomy based on biomarkers , 2008 .

[17]  Jake M. Ferguson,et al.  Estimating the Diets of Animals Using Stable Isotopes and a Comprehensive Bayesian Mixing Model , 2012, PloS one.

[18]  G. Ollesch,et al.  Sediment fingerprinting in northern Jordan: element-specific correction factors in a carbonatic setting , 2015, Journal of Soils and Sediments.

[19]  Barry G Rawlins,et al.  Sensitivity of fluvial sediment source apportionment to mixing model assumptions: A Bayesian model comparison , 2014, Water resources research.

[20]  O. Evrard,et al.  A comparison of geological and statistical approaches to element selection for sediment fingerprinting , 2015, Journal of Soils and Sediments.

[21]  D. Walling,et al.  A reconnaissance survey of the source of interstitial fine sediment recovered from salmonid spawning gravels in England and Wales , 2003, Hydrobiologia.

[22]  Desmond E. Walling,et al.  Fingerprinting suspended sediment sources in the catchment of the River Ouse, Yorkshire, UK , 1999 .

[23]  G. Verstraeten,et al.  Unravelling changing sediment sources in a Mediterranean mountain catchment: a Bayesian fingerprinting approach , 2013 .

[24]  R. Clarke.  A bootstrap calculation of confidence regions for proportions of sediment contributed by different source areas in a ‘fingerprinting’ model , 2015 .

[25]  William H. Blake,et al.  Tracing crop-specific sediment sources in agricultural catchments , 2012 .

[26]  Andrew Gelman,et al.  The No-U-turn sampler: adaptively setting path lengths in Hamiltonian Monte Carlo , 2011, J. Mach. Learn. Res..

[27]  A. Papanicolaou,et al.  An un-mixing model to study watershed erosion processes , 2008 .

[28]  Rutger van Haasteren,et al.  Gibbs Sampling , 2010, Encyclopedia of Machine Learning.

[29]  M. Wieczorek,et al.  Suspended sediment source apportionment in Chesapeake Bay watershed using Bayesian chemical mass balance receptor modeling , 2013 .

[30]  D. Lobb,et al.  The role of gravel channel beds on the particle size and organic matter selectivity of transported fine-grained sediment: implications for sediment fingerprinting and biogeochemical flux research , 2015, Journal of Soils and Sediments.

[31]  D. Phillips,et al.  Bayesian stable isotope mixing models , 2012, 1209.6457.

[32]  S. Pulley,et al.  The uncertainties associated with sediment fingerprinting suspended and recently deposited fluvial sediment in the Nene river basin , 2015 .

[33]  Gerard Govers,et al.  A mixing model to incorporate uncertainty in sediment fingerprinting , 2014 .

[34]  S. Franks,et al.  Uncertainty-based assessment of tracer selection, tracer non-conservativeness and multiple solutions in sediment fingerprinting using synthetic and field data , 2015, Journal of Soils and Sediments.

[35]  Thomas Udelhoven,et al.  A rapid spectral-reflectance-based fingerprinting approach for documenting suspended sediment sources during storm runoff events , 2010 .

[36]  S. Franks,et al.  Quantitative sediment fingerprinting using a Bayesian uncertainty estimation framework , 2002 .

[37]  Hugh G. Smith,et al.  Sediment fingerprinting in agricultural catchments: A critical re-examination of source discrimination and data corrections , 2014 .

[38]  D. Lobb,et al.  The behavioural characteristics of sediment properties and their implications for sediment fingerprinting as an approach for identifying sediment sources in river basins , 2013 .

[39]  O. Evrard,et al.  Preface—Addressing challenges to advance sediment fingerprinting research , 2015, Journal of Soils and Sediments.

[40]  D. Walling,et al.  Quantifying fine‐grained sediment sources in the River Axe catchment, southwest England: application of a Monte Carlo numerical modelling framework incorporating local and genetic algorithm optimisation , 2012 .

[41]  J. Olley,et al.  An examination of geochemical modelling approaches to tracing sediment sources incorporating distribution mixing and elemental correlations , 2015 .

[42]  M. Plummer,et al.  CODA: convergence diagnosis and output analysis for MCMC , 2006 .

[43]  G. Mateu-Figueras,et al.  Isometric Logratio Transformations for Compositional Data Analysis , 2003 .

[44]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[45]  A. L. Collins,et al.  Contemporary fine‐grained bed sediment sources across the River Wensum Demonstration Test Catchment, UK , 2013 .

[46]  Jim Freer,et al.  Uncertainties in data and models to describe event dynamics of agricultural sediment and phosphorus transfer. , 2009, Journal of environmental quality.

[47]  J. Olley,et al.  Accuracy of mixing models in predicting sediment source contributions. , 2014, The Science of the total environment.

[48]  I. Fuller,et al.  Characterization and quantification of suspended sediment sources to the Manawatu River, New Zealand. , 2016, The Science of the total environment.

[49]  D. Lobb,et al.  Sources of variability in fatty acid (FA) biomarkers in the application of compound-specific stable isotopes (CSSIs) to soil and sediment fingerprinting and tracing: A review. , 2016, The Science of the total environment.

[50]  Hugh G. Smith,et al.  Comparing catchment sediment fingerprinting procedures using an auto-evaluation approach with virtual sample mixtures. , 2015, The Science of the total environment.

[51]  Martyn Plummer,et al.  JAGS: A program for analysis of Bayesian graphical models using Gibbs sampling , 2003 .