The Annals of Applied Statistics

Gaussian process framework for temporal dependence and discrepancy functions in Ricker-type population growth models

Marcelo Hartmann, Geoffrey R. Hosack, Richard M. Hillary, and Jarno Vanhatalo

Full-text: Open access


Density dependent population growth functions are of central importance to population dynamics modelling because they describe the theoretical rate of recruitment of new individuals to a natural population. Traditionally, these functions are described with a fixed functional form with temporally constant parameters and without species interactions. The Ricker stock-recruitment model is one such function that is commonly used in fisheries stock assessment. In recent years, there has been increasing interest in semiparametric and temporally varying population growth models. The former are related to the general statistical approach of using semiparametric discrepancy functions, such as Gaussian processes (GP), to model deviations around the expected parametric function. In the latter, the reproductive rate, which is a key parameter describing the population growth rate, is assumed to vary in time. In this work, we introduce how these existing Ricker population growth models can be formulated under the same statistical approach of hierarchical GP models. We also show how the time invariant semiparametric approach can be extended and combined with the time varying reproductive rate using a GP model. Then we extend these models to the multispecies setting by incorporating cross-covariances among species with a continuous time covariance structure using the linear model of coregionalization. As a case study, we examine the productivity of three Pacific salmon populations. We compare the alternative Ricker population growth functions using model posterior probabilities and leave-one-out cross validation predictive densities. Our results show substantial temporal variation in maximum reproductive rates and reveal temporal dependence among the species, which have direct management implications. However, our results do not support inclusion of semiparametric discrepancy function and they suggest that the semiparametric discrepancy functions may lead to challenges in parameter identifiability more generally.

Article information

Ann. Appl. Stat., Volume 11, Number 3 (2017), 1375-1402.

Received: August 2016
Revised: December 2016
First available in Project Euclid: 5 October 2017

Permanent link to this document

Digital Object Identifier

Mathematical Reviews number (MathSciNet)

Zentralblatt MATH identifier

Model evidence marginal likelihood population growth fisheries density dependence temporal dependence interspecific dependence


Hartmann, Marcelo; Hosack, Geoffrey R.; Hillary, Richard M.; Vanhatalo, Jarno. Gaussian process framework for temporal dependence and discrepancy functions in Ricker-type population growth models. Ann. Appl. Stat. 11 (2017), no. 3, 1375--1402. doi:10.1214/17-AOAS1029.

Export citation


  • Barnard, J., McCulloch, R. and Meng, X.-L. (2000). Modeling covariance matrices in terms of standard deviations and correlations, with application to shrinkage. Statist. Sinica 10 1281–1311.
  • Bradford, M. J. (1995). Comparative review of Pacific salmon survival rates. Can. J. Fish. Aquat. Sci. 52 1327–1338.
  • Bradford, M. J., Lovy, J. and Patterson, D. A. (2010). Infection of gill and kidney of Fraser River sockeye salmon, Oncorhynchus nerka (Walbaum), by Parvicapsula minibicornis and its effect on host physiology. J. Fish Dis. 33 769–779.
  • Brännstron, Å. and Sumpter, D. J. T. (2005). The role of competition and clustering in population dynamics. Proc. R. Soc. Lond., B Biol. Sci. 272 2065–2072.
  • Brynjarsdóttir, J. and O’Hagan, A. (2014). Learning about physical parameters: The importance of model discrepancy. Inverse Probl. 30 114007, 24.
  • Buckland, S. T., Newman, K. B., Fernández, C., Thomas, L. and Harwood, J. (2007). Embedding population dynamics models in inference. Statist. Sci. 22 44–58.
  • Burgner, R. L. (1991). Life history of sockeye salmon (Oncorhynchus nerka). In Pacific Salmon Life Histories 3–117. UBC Press, Vancouver, BC.
  • Cadigan, N. G. (2013). Fitting a non-parametric stock-recruitment model in R that is useful for deriving MSY reference points and accounting for model uncertainty. ICES J. Mar. Sci. 70 56–67.
  • Chen, D. G. and Irvine, J. R. (2001). A semiparametric model to examine stock-recruitment relationships incorporating environmental data. Can. J. Fish. Aquat. Sci. 58 1178.
  • Chen, D. G. and Ware, D. M. (1999). A neural network model for forecasting fish stock recruitment. Can. J. Fish. Aquat. Sci. 56 2385–2396.
  • Cooke, S. J., Hinch, S. G., Anthony, P. F., Lapointe, M. F., Jones, S. R. M., Macdonald, J. S., Patterson, D. A., Healey, M. C. and Kraak, G. V. D. (2004). Abnormal migration timing and high en route mortality of sockeye salmon in the Fraser River, British Columbia. Fisheries 29 22–33.
  • DFO (2016). Department of Fisheries and Oceans. Salmonid Enhancement Program, Weaver Creek Spawning Channel. Unpublished data.
  • Dorner, B., Peterman, R. M. and Haeseker, S. L. (2008). Historical trends in productivity of 120 Pacific pink, chum, and sockeye salmon stocks reconstructed by using a Kalman filter. Can. J. Fish. Aquat. Sci. 65 1842–1866.
  • Essington, T. E., Quinn, T. P. and Ewert, V. E. (2000). Intra- and inter-specific competition and the reproductive success of sympatric Pacific salmon. Can. J. Fish. Aquat. Sci. 57 205–213.
  • Gelfand, A. E., Kim, H.-J., Sirmans, C. F. and Banerjee, S. (2003). Spatial modeling with spatially varying coefficient processes. J. Amer. Statist. Assoc. 98 387–396.
  • Geweke, J. (1992). Evaluating the accuracy of sampling-based approaches to the calculation of posterior moments. In Bayesian Statistics, 4 (PeñíScola, 1991) 169–193. Oxford Univ. Press, New York.
  • Healey, M. C. (1991). Life history of chinook salmon (Oncorhynchus tshawytscha). In Pacific Salmon Life Histories 311–394. UBC Press, Vancouver, BC.
  • Heard, W. R. (1991). Life history of pink salmon (Oncorhynchus gorbuscha). In Pacific Salmon Life Histories 119–230. UBC Press, Vancouver, BC.
  • Hilborn, R. and Walters, C. J. (1992). Quantitative Fisheries Stock Assessment: Choice, Dynamics and Uncertainty. Chapman & Hall, New York.
  • Hillary, R. M. (2012). Practical uses of non-parametric methods in fisheries assessment modelling. Mar. Freshw. Res. 63 606.
  • Hinch, S. G., Cooke, S. J., Farrell, A. P., Miller, K. M., Lapointe, M. and Patterson, D. A. (2012). Dead fish swimming: A review of research on the early migration and high premature mortality in adult Fraser River sockeye salmon Oncorhynchus nerka. J. Fish Biol. 81 576–599.
  • Jost, C. and Ellner, S. P. (2000). Testing for predator dependence in predator–prey dynamics: A non-parametric approach. Proc. R. Soc. Lond., B Biol. Sci. 267 1611–1620.
  • Kass, R. E. and Raftery, A. E. (1995). Bayes factors. J. Amer. Statist. Assoc. 90 773–795.
  • Kennedy, M. C. and O’Hagan, A. (2001). Bayesian calibration of computer models. J. R. Stat. Soc. Ser. B. Stat. Methodol. 63 425–464.
  • Kuikka, S., Vanhatalo, J., Pulkkinen, H., Mäntyniemi, S. and Corander, J. (2014). Experiences in Bayesian inference in Baltic salmon management. Statist. Sci. 29 42–49.
  • Lewis, S. M. and Raftery, A. E. (1997). Estimating Bayes factors via posterior simulation with the Laplace-Metropolis estimator. J. Amer. Statist. Assoc. 92 648–655.
  • Mäntyniemi, S., Uusitalo, L., Peltonen, H., Haapasaari, P. and Kuikka, S. (2013). Integrated, age-structured, length-based stock assessment model with uncertain process variances, structural uncertainty, and environmental covariates: Case of central Baltic herring. Can. J. Fish. Aquat. Sci. 1326 1317–1326.
  • Mäntyniemi, S., Whitlock, R. E., Perälä, T. A., Blomstedt, P. A., Vanhatalo, J. P., Ricon, M. M., Kuparinen, A. K., Pulkkinen, H. P. and Kuikka, S. O. (2015). General state-space population dynamics model for Bayesian stock assessment. ICES J. Mar. Sci. 72 2209–2222.
  • Mardia, K. V. and Goodall, C. R. (1993). Spatial–temporal analysis of multivariate environmental monitoring data. In Multivariate Environmental Statistics. North-Holland Ser. Statist. Probab. 6 347–386. North-Holland, Amsterdam.
  • Maunder, M. N. and Deriso, R. B. (2011). A state–space multistage life cycle model to evaluate population impacts in the presence of density dependence: Illustrated with application to delta smelt (Hyposmesus transpacificus). Can. J. Fish. Aquat. Sci. 68 1285–1306.
  • Minto, C., Flemming, J. M., Britten, G. L. and Worm, B. (2014). Productivity dynamics of Atlantic cod. Can. J. Fish. Aquat. Sci. 71 203–216.
  • Morita, K., Morita, S. H. and Fukuwaka, M.-a. (2006). Population dynamics of Japanese pink salmon (Oncorhynchus gorbuscha): Are recent increases explained by hatchery programs or climatic variations? Can. J. Fish. Aquat. Sci. 63 55–62.
  • Munch, S. B. and Kottas, A. (2009). A Bayesian modeling approach for determining productivity regimes and their characteristics. Ecol. Appl. 19 527–537.
  • Munch, S. B., Kottas, A. and Mangel, M. (2005). Bayesian nonparametric analysis of stock–recruitment relationships. Can. J. Fish. Aquat. Sci. 62 1808–1821.
  • Myers, R. A., Mertz, G. and Bridson, J. (1997). Spatial scales of interannual recruitment variations of marine, anadromous, and freshwater fish. Can. J. Fish. Aquat. Sci. 54 1400–1407.
  • Neal, R. M. (2003). Slice sampling. Ann. Statist. 31 705–767. With discussions and a rejoinder by the author.
  • Newman, K. B., Fernández, C., Thomas, L. and Buckland, S. T. (2009). Monte Carlo inference for state-space models of wild animal populations. Biometrics 65 572–583.
  • O’Hagan, A. (2006). Bayesian analysis of computer code outputs: A tutorial. Reliab. Eng. Syst. Saf. 91 1290–1300.
  • Perälä, T. and Kuparinen, A. (2015). Detecting regime shifts in fish stock dynamics. Can. J. Fish. Aquat. Sci. 72 1619–1628.
  • Peterman, R. M., Pyper, B. J. and Grout, J. A. (2000). Comparison of parameter estimation methods for detecting climate-induced changes in productivity of Pacific salmon (Oncorhynchus spp). Can. J. Fish. Aquat. Sci. 57 181–191.
  • Peterman, R. M., Pyper, B. J. and MacGregor, B. W. (2003). Use of the Kalman filter to reconstruct historical trends in productivity of Bristol Bay sockeye salmon (Oncorhynchus nerka). Can. J. Fish. Aquat. Sci. 60 809–824.
  • Pulkkinen, H. and Mäntyniemi, S. (2013). Maximum survival of eggs as the key parameter of stock-recruit meta-analysis: Accounting for parameter and structural uncertainty. Can. J. Fish. Aquat. Sci. 70 527–533.
  • Quinn, T. J. and Deriso, R. B. (1999). Quantitative Fish Dynamics. Oxford Univ. Press.
  • Rasmussen, C. E. and Williams, C. K. I. (2006). Gaussian Processes for Machine Learning. MIT Press, Cambridge, MA.
  • Ricker, W. E. (1954). Stock and recruitment. J. Fish. Res. Board Can. 5 559–623.
  • Riihimäki, J. and Vehtari, A. (2010). Gaussian processes with monotonicity information. J. Mach. Learn. Res. Workshop Conf. Proc. 9 645–652. (AISTATS 2010 Proceedings).
  • Rosberg, G. E., Scott, K. J. and Rithaler, R. (1986). Review of the International Pacific Salmon Fisheries Commission’s Sockeye and Pink salmon enhancement facilities on the Fraser River. Technical Report, Department of Fisheries and Oceans, Vancouver, Canada.
  • Rose, K. A., Junior, J. A. C., Winemiller, K. O., Myers, R. A. and Hilborn, R. (2001). Compensatory density dependence in fish populations: Importance, controversy, understanding and prognosis. Fish Fish. 2 293–327.
  • Simpson, D., Rue, H., Riebler, A., Martins, T. G. and Sørbye, S. H. (2017). Penalising model component complexity: A principled, practical approach to constructing priors. Statist. Sci. 32 1–28.
  • Sugeno, M. and Munch, S. B. (2013). A semiparametric Bayesian method for detecting Allee effects. Ecology 94 1196–1204.
  • Sundararajan, S. and Keerthi, S. S. (2001). Predictive approaches for choosing hyperparameters in Gaussian processes. Neural Comput. 13 1103–1118.
  • Thorson, J. T., Jensen, O. P. and Zipkin, E. F. (2014). How variable is recruitment for exploited marine fishes? A hierarchical model for testing life history theory. Can. J. Fish. Aquat. Sci. 983 973–983.
  • Thorson, J. T., Ono, K. and Munch, S. B. (2014). A Bayesian approach to identifying and compensating for model misspecification in population models. Ecology 95 329–341.
  • Tokuda, T., Goodrich, B., Van Mechelen, I., Gelman, A. and Tuerlinckx, F. (2012). Visualizing distributions of covariance matrices. Technical report, University of Leuwen, Belgium and Columbia University, USA. Available at
  • Traxler, G. and Rankin, J. (1989). An infectious hematopoietic necrosis epizootic in sockeye salmon Oncorhynchus nerka in Weaver Creek spawning channel, Fraser River system, BC, Canada. Dis. Aquat. Org. 6 221–226.
  • Vanhatalo, J., Riihimäki, J., Hartikainen, J., Jylänki, P., Tolvanen, V. and Vehtari, A. (2013). GPstuff: Bayesian modeling with Gaussian processes. J. Mach. Learn. Res. 14 1175–1179.
  • Vehtari, A. and Ojanen, J. (2012). A survey of Bayesian predictive methods for model assessment, selection and comparison. Stat. Surv. 6 142–228.
  • Zeng, Z., Nowierski, R. M., Taper, M. L., Dennis, B. and Kemp, W. P. (2010). Complex population dynamics in the real world: Modeling the influence of time-varying parameters and time lags. Ecology 79 2193–2209.