Analysis of Vegetation-Water Interactions: Application and Comparison of Maximum-Likelihood Estimation and Bayesian Inference

Part of the Ecological Studies book series (ECOLSTUD, volume 240)


In environmental science, we often deal with complex systems, many processes and interactions with relatively little mechanistic understanding which is translated into our models. It is also not uncommon that there are not enough measurements to inform the parameters of these models directly. However, we have increasingly more data from observatory networks and remote sensing missions that can be mapped to the properties that we are trying to predict with our models. In that case, such data can be used to indirectly inform model parameters through likelihood-based estimation methods. This chapter contrasts two likelihood-based approaches, namely maximum-likelihood estimation and Bayesian inference. The simple examples used here reveal important and scientifically relevant dissimilarities between the two approaches. Bayesian approach allows us to treat all terms in models as probability distributions while producing readily interpretable results. The power of Bayesian approach becomes more apparent as we increase the complexity of our models to reflect the complexity of our study systems. This chapter demonstrates how Bayesian inference comes to the fore as not only a way to make environmental science more relevant but also as one of the best ways of progressing it.



I thank Michael Dietze, Florian Hartig, Martin De Kauwe, Jürgen Knauer and all members of the Boston University Ecological Forecasting Lab for their helpful comments and discussions.

Supplementary material (902 kb)
chapter9_MLEvsBayes (HTML 1278 kb)
464883_1_En_9_MOESM2_ESM.rmd (43 kb)
chapter9_MLEvsBayes (RMD 42 kb)
464883_1_En_9_MOESM3_ESM.rdata (16 kb)
DEopt_out (RDATA 15 kb)
464883_1_En_9_MOESM4_ESM.r (12 kb)
DETha_example (R 12 kb)
464883_1_En_9_MOESM5_ESM.rdata (37.5 mb)
US_Ha1_1991_2012 (RDATA 38422 kb)
464883_1_En_9_MOESM6_ESM.r (6 kb)
USHa1_example (R 6 kb)
464883_1_En_9_MOESM7_ESM.r (9 kb)
VSEM_example (R 8 kb)
464883_1_En_9_MOESM8_ESM.rdata (5.5 mb)
year_effect (RDATA 5626 kb)


  1. Amrhein A, Greenland S, McShane B et al (2019) Scientists rise up against statistical significance. Nature 567:305–307. CrossRefGoogle Scholar
  2. Andrieu C, de Freitas N, Doucet A, Jordan MI (2003) An introduction to MCMC for machine learning. Mach Learn 50:5–43. CrossRefGoogle Scholar
  3. Arora RK (2015) Optimization: algorithms and applications. Chapman and Hall, LondonCrossRefGoogle Scholar
  4. Bolker BM (2007) Ecological models and data in R. Princeton University Press, PrincetonGoogle Scholar
  5. Brookshire ENJ, Gerber S, Webster JR, Vose JM, Swank WT (2011) Direct effects of temperature on forest nitrogen cycling revealed through analysis of long-term watershed records. Glob Chang Biol 17:297–308. CrossRefGoogle Scholar
  6. Carpenter B, Gelman A, Hoffman MD, Lee D, Goodrich B, Betancourt M et al (2017) Stan: a probabilistic programming language. J Stat Softw 76:1–32. CrossRefGoogle Scholar
  7. Clark JS (2005) Why environmental scientists are becoming Bayesians. Ecol Lett 8:2–14. CrossRefGoogle Scholar
  8. Clark JS (2007) Models for ecological data. Princeton University Press, PrincetonCrossRefGoogle Scholar
  9. Clark JS, Carpenter SR, Barber M, Collins S, Dobson A, Foley JA et al (2001) Ecological forecasts: an emerging imperative. Science 293:657–660. CrossRefGoogle Scholar
  10. De Kauwe MG, Medlyn BE, Zaehle S, Walker AP, Dietze MC, Hickler T et al (2013) Forest water use and water use efficiency at elevated CO2: a model-data intercomparison at two contrasting temperate forest FACE sites. Glob Chang Biol 19:1759–1779. CrossRefGoogle Scholar
  11. De Kauwe MG, Medlyn BE, Knauer J, Williams CA (2017) Ideas and perspectives: how coupled is the vegetation to the boundary layer? Biogeosciences 14:4435–4453. CrossRefGoogle Scholar
  12. de Valpine P, Turek D, Paciorek CJ, Anderson-Bergman C, Temple Lang D, Bodik R (2017) Programming with models: writing statistical algorithms for general model structures with NIMBLE. J Comput Graph Stat 26:403–413. CrossRefGoogle Scholar
  13. Dietze MC (2017) Ecological forecasting. Princeton University Press, PrincetonCrossRefGoogle Scholar
  14. Dietze MC, Fox A, Beck-Johnson LM, Betancourt JL, Hooten MB, Jarnevich CS et al (2018) Iterative near-term ecological forecasting. Proc Natl Acad Sci U S A 115:1424–1432. CrossRefGoogle Scholar
  15. Dietze M, Lynch H (2019) Forecasting a bright future for ecology. Front Ecol Environ 17:3. CrossRefGoogle Scholar
  16. Dormann CF, Calabrese JM, Guillera-Arroita G, Matechou E, Bahn V, Bartoń K, Beale CM, Ciuti S, Elith J, Gerstner K, Guelat J, Keil P, Lahoz-Monfort JJ, Pollock LJ, Reineking B, Roberts DR, Schröder B, Thuiller W, Warton DI, Wintle BA, Wood SN, Wüest RO, Hartig F (2018) Model averaging in ecology: a review of Bayesian, information-theoretic, and tactical approaches for predictive inference. Ecol Monogr 88:485–504. CrossRefGoogle Scholar
  17. Ellison AM (2004) Bayesian inference in ecology. Ecol Lett 7:509–520. CrossRefGoogle Scholar
  18. Fer I, Kelly R, Moorcroft PR, Richardson AD, Cowdery EM, Dietze MC (2018) Linking big models to big data: efficient ecosystem model calibration through Bayesian model emulation. Biogeosciences 15:5801–5830. CrossRefGoogle Scholar
  19. Forum – P values and model selection (2014) Ecology 95:609–653. CrossRefGoogle Scholar
  20. Gelman A, Carlin JB, Stern HS, Rubin DB (2003) Bayesian data analysis. Chapman and Hall, LondonCrossRefGoogle Scholar
  21. Golding N (2018) Greta: simple and scalable statistical modelling in R. R package version 0.3.0.
  22. Hartig F, Calabrese JM, Reineking B, Wiegand T, Huth A (2011) Statistical inference for stochastic simulation models – theory and application. Ecol Lett 14:816–827. CrossRefGoogle Scholar
  23. Hartig F, Dyke J, Hickler T, Higgins SI, O’Hara RB, Scheiter S et al (2012) Connecting dynamic vegetation models to data – an inverse perspective. J Biogeogr 39:2240–2252. CrossRefGoogle Scholar
  24. Hartig F, Minunno F, Paul S (2018) BayesianTools: general-purpose MCMC and SMC samplers and yools for Bayesian statistics. R package version 0.1.5.
  25. Hooten MB, Hobbs NT (2015) A guide to Bayesian model selection for ecologists. Ecol Monogr 85:3–28. CrossRefGoogle Scholar
  26. Jaynes ET (2003) Probability theory: the logic of science. Cambridge University Press, CambridgeCrossRefGoogle Scholar
  27. Keenan TF, Carbone MS, Reichstein M, Richardson AD (2011) The model-data fusion pitfall: assuming certainty in an uncertain world. Oecologia 167:587–597. CrossRefGoogle Scholar
  28. Kim Y, Knox RG, Longo M, Medvigy D, Hutyra LR, Pyle EH et al (2012) Seasonal carbon dynamics and water fluxes in an Amazon rainforest. Glob Chang Biol 18:1322–1334. CrossRefGoogle Scholar
  29. Knauer J, Zaehle S, Medlyn BE et al (2018a) Towards physiologically meaningful water-use efficiency estimates from eddy covariance data. Glob Change Biol 24:694–710. CrossRefGoogle Scholar
  30. Knauer J, El-Madany TS, Zaehle S, Migliavacca M (2018b) Bigleaf – an R package for the calculation of physical and physiological ecosystem properties from eddy covariance data. PLoS One 13:e0201114. CrossRefGoogle Scholar
  31. Kruschke JK (2015) Doing Bayesian data analysis: a tutorial with R, JAGS, and Stan, 2nd edn. Academic Press, Cambridge, MAGoogle Scholar
  32. Kurt W (2019) Bayesian statistics the fun way. No Starch Press, San FranciscoGoogle Scholar
  33. LeBauer DS, Dietze MC, Bolker BM (2013) Translating probability density functions: from R to BUGS and back again. R Journal 5:207–209. CrossRefGoogle Scholar
  34. Link WA, Barker RJ (2006) Model weights and the foundations of multimodel inference. Ecology 87:2626–2635.[2626:MWATFO]2.0.CO;2 CrossRefGoogle Scholar
  35. Luo Y, Ogle K, Tucker C, Fei S, Gao C, LaDeau S, Clark JS, Schimel DS (2011) Ecological forecasting and data assimilation in a data-rich era. Ecol Appl 21:1429–1442. CrossRefGoogle Scholar
  36. Lunn D, Jackson C, Best N, Thomas A, Spiegelhalter DJ (2012) The BUGS book – a practical introduction to Bayesian analysis. Chapman and Hall, LondonCrossRefGoogle Scholar
  37. MacBean N, Maignan F, Bacour C, Lewis P, Peylin P, Guanter L et al (2018) Strong constraint on modelled global carbon uptake using solar-induced chlorophyll fluorescence data. Sci Rep 8:2045–2322. CrossRefGoogle Scholar
  38. Martin O (2018) Bayesian analysis with Python, 2nd edn. Packt Books, BirminghamGoogle Scholar
  39. McElreath R (2015) Statistical rethinking: a Bayesian course with examples in R and Stan. Chapman and Hall, LondonGoogle Scholar
  40. McNaughton K, Jarvis P (1991) Effects of spatial scale on stomatal control of transpiration. Agric For Meteorol 54:279–302. CrossRefGoogle Scholar
  41. Medlyn BE, Duursma RA, Eamus D, Ellsworth DS, Prentice IC, Bartons CVM et al (2011) Reconciling the optimal and empirical approaches to modelling stomatal conductance. Glob Chang Biol 17:2134–2144. CrossRefGoogle Scholar
  42. Medlyn BE, De Kauwe MG, Lin Y, Knauer J, Duursma RA, Williams CA et al (2017) How do leaf and ecosystem measures of water-use efficiency compare? New Phytol 216:758–770. CrossRefGoogle Scholar
  43. Novick KA, Ficklin DL, Stoy PC, Williams CA, Bohrer G, Oishi AC et al (2016) The increasing importance of atmospheric demand for ecosystem water and carbon fluxes. Nat Clim Chang 6:1023–1027. CrossRefGoogle Scholar
  44. O’Hagan A (2019) Expert knowledge elicitation: subjective but scientific. Am Stat 73:69–81. CrossRefGoogle Scholar
  45. Plummer M (2003) JAGS: a program for analysis of Bayesian graphical models using Gibbs sampling. In: Hornik K, Leisch F, Zeileis A (eds) Proceedings of the 3rd international workshop on distributed statistical computing (DSC 2003). Technische Universität Wien, Vienna. Google Scholar
  46. Plummer M (2016) rjags: Bayesian graphical models using MCMC. R package version 4-6.
  47. Richardson AD, Hollinger DY (2005) Statistical modeling of ecosystem respiration using eddy covariance data: maximum likelihood parameter estimation, and Monte Carlo simulation of model and parameter uncertainty, applied to three simple models. Agric For Meteorol 131:191–208. CrossRefGoogle Scholar
  48. Richardson AD, Braswell BH, Hollinger DY, Burman P, Davidson EA et al (2006) Comparing simple respiration models for eddy flux and dynamic chamber data. Agric For Meteorol 141:219–234. CrossRefGoogle Scholar
  49. Rue H, Martino S, Chopin N (2009) Approximate Bayesian inference for latent Gaussian models using integrated nested Laplace approximations (with discussion). J Roy Stat Soc B 71:319–392. CrossRefGoogle Scholar
  50. Smith BJ, Deonovic B et al (2014) Mamba: Markov chain Monte Carlo for Bayesian analysis in julia. 2014. julia software package.
  51. Smith WK, Reed SC, Cleveland CC, Ballantyne AP, Anderegg WRL, William WR et al (2016) Large divergence of satellite and Earth system model estimates of global terrestrial CO2 fertilization. Nat Clim Chang 6:306–310. CrossRefGoogle Scholar
  52. Stoy PC, Mauder M, Foken T, Marcolla B, Boegh E, Ibrom A et al (2013) A data-driven analysis of energy balance closure across FLUXNET research sites: the role of landscape scale heterogeneity. Agric For Meteorol 171-172:137–152. CrossRefGoogle Scholar
  53. Twine TE, Kustas WP, Norman JM, Cook DR, Houser PR, Meyers TP, Prueger JH, Starks PJ, Wesely ML (2000) Correcting eddy-covariance flux underestimates over a grassland. Agric For Meteorol 103(3):279–300. CrossRefGoogle Scholar
  54. van Oijen M, Rougier J, Smith R (2005) Bayesian calibration of process-based forest models: bridging the gap between models and data. Tree Physiol 25:915–927. CrossRefGoogle Scholar
  55. Varadhan R, Borchers HW, ABB Corporate Research (2018) dfoptim: derivative-free optimization. R package version 2018.2-1.
  56. Williams M, Richardson AD, Reichstein M, Stoy PC, Peylin P, Verbeeck H et al (2009) Improving land surface models with FLUXNET data. Biogeosciences 6:1341–1359. CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  1. 1.Department of Earth and EnvironmentBoston UniversityBostonUSA

Personalised recommendations