Abstract
Climate change impacts ecosystems variably in space and time. Landscape features may confer resistance against environmental stressors, whose intensity and frequency also depend on local weather patterns. Characterizing spatio-temporal variation in population responses to these stressors improves our understanding of what constitutes climate change refugia. We developed a Bayesian hierarchical framework that allowed us to differentiate population responses to seasonal weather patterns depending on their “sensitive” or “resilient” states. The framework inferred these sensitivity states based on latent trajectories delineating dynamic state probabilities. The latent trajectories are composed of linear initial conditions, functional regression models, and additive random effects representing ecological mechanisms such as topological buffering and effects of legacy weather conditions. Further, we developed a Bayesian regularization strategy that promoted temporal coherence in the inferred states. We demonstrated our hierarchical framework and regularization strategy using simulated examples and a case study of native brook trout (Salvelinus fontinalis) count data from the Great Smoky Mountains National Park, southeastern USA. Our study provided insights into ecological processes influencing brook trout sensitivity. Our framework can also be applied to other species and ecosystems to facilitate management and conservation.
Similar content being viewed by others
References
Ashcroft MB (2010) Identifying refugia from climate change. J Biogeogr 37:1407–1413
Berliner LM (1996) Hierarchical Bayesian time series models. In: Maximum entropy and Bayesian methods: Santa Fe, New Mexico, USA, 1995 Proceedings of the fifteenth international workshop on maximum entropy and bayesian methods, pp 15–22. Springer
Bradshaw WE, Holzapfel CM (2006) Evolutionary response to rapid climate change. Science 312:1477–1478
Burden F, Winkler D (2009) Bayesian regularization of neural networks. Methods Appl Artif Neural Netw 23–42
Carline RF, McCullough BJ (2003) Effects of floods on brook trout populations in the Monongahela National Forest, West Virginia. Trans Am Fish Soc 132:1014–1020
Chakraborty A, Gelfand AE, Wilson AM, Latimer AM, Silander JA (2011) Point pattern modelling for degraded presence-only data over large regions. J R Stat Soc Ser C Appl Stat 60:757–776
Dakos V, Matthews B, Hendry AP, Levine J, Loeuille N, Norberg J, Nosil P, Scheffer M, De Meester L (2019) Ecosystem tipping points in an evolving world. Nat Ecol Evol 3:355–362
Dunham JB, Young MK, Gresswell RE, Rieman BE (2003) Effects of fire on fish populations: landscape perspectives on persistence of native fishes and nonnative fish invasions. For Ecol Manag 178:183–196
Elwood JW, Waters TF (1969) Effects of floods on food consumption and production rates of a stream brook trout population. Trans Am Fish Soc 98:253–262
Farr MT, Green DS, Holekamp KE, Zipkin EF (2021) Integrating distance sampling and presence-only data to estimate species abundance. Ecology 102:e03204
Felus YA, Saalfeld A, Schaffrin B (2005) Delaunay triangulation structured kriging for surface interpolation. Surv Land Inf Sci 65:27
Fletcher RJ Jr, Hefley TJ, Robertson EP, Zuckerberg B, McCleery RA, Dorazio RM (2019) A practical guide for combining data to model species distributions. Ecology 100:e02710
Girosi F, Jones M, Poggio T (1995) Regularization theory and neural networks architectures. Neural Comput 7:219–269
Glauber RJ (1963) Time-dependent statistics of the Ising model. J Math Phys 4:294–307
Griffith DA (2020) Some guidelines for specifying the geographic weights matrix contained in spatial statistical models 1. In: Practical handbook of spatial statistics, pp 65–82. CRC press
Gunderson LH (2000) Ecological resilience: in theory and application. Annu Rev Ecol Syst 31:425–439
Habera JW, Kulp MA, Moore SE, Henry TB (2010) Three-pass depletion sampling accuracy of two electric fields for estimating trout abundance in a low-conductivity stream with limited habitat complexity. N Am J Fish Manag 30:757–766
Hakala JP, Hartman KJ (2004) Drought effect on stream morphology and brook trout (Salvelinus fontinalis) populations in forested headwater streams. Hydrobiologia 515:203–213
Hamilton JD (1994) State-space models. Handb Econ 4:3039–3080
Hans C (2009) Bayesian lasso regression. Biometrika 96:835–845
Hare DK, Benz SA, Kurylyk BL, Johnson ZC, Terry NC, Helton AM (2023) Paired air and stream temperature analysis (PASTA) to evaluate groundwater influence on streams. Water Resources Research page e2022WR033912
Hastie T, Mallows C (1993) A statistical view of some chemometrics regression tools: discussion. Technometrics 35:140–143
Hazzard A (1932) Some phases of the life history of the eastern brook trout, Salvelinus fontinalis Mitchell. Trans Am Fish Soc 62:344–350
Hefley TJ, Broms KM, Brost BM, Buderman FE, Kay SL, Scharf HR, Tipton JR, Williams PJ, Hooten MB (2017) The basis function approach for modeling autocorrelation in ecological data. Ecology 98:632–646
Hooten MB, Hobbs NT (2015) A guide to Bayesian model selection for ecologists. Ecol Monogr 85:3–28
Hudy M, Thieling TM, Gillespie N, Smith EP (2008) Distribution, status, and land use characteristics of subwatersheds within the native range of brook trout in the eastern United States. N Am J Fish Manag 28:1069–1085
Hughes JP, Guttorp P, Charles SP (1999) A non-homogeneous hidden Markov model for precipitation occurrence. J R Stat Soc Ser C Appl Stat 48:15–30
Isaac JL, De Gabriel JL, Goodman BA (2008) Microclimate of daytime den sites in a tropical possum: implications for the conservation of tropical arboreal marsupials. Anim Conserv 11:281–287
Jia X, Willard J, Karpatne A, Read JS, Zwart JA, Steinbach M, Kumar V (2021) Physics-guided machine learning for scientific discovery: an application in simulating lake temperature profiles. ACM/IMS Trans Data Sci 2:1–26
Johnson DS, Laake JL, Ver Hoef JM (2010) A model-based approach for making ecological inference from distance sampling data. Biometrics 66:310–318
Johnstone JF, Allen CD, Franklin JF, Frelich LE, Harvey BJ, Higuera PE, Mack MC, Meentemeyer RK, Metz MR, Perry GL, Schoennagel T (2016) Changing disturbance regimes, ecological memory, and forest resilience. Front Ecol Environ 14:369–378
Kanno Y, Letcher BH, Hitt NP, Boughton DA, Wofford JE, Zipkin EF (2015) Seasonal weather patterns drive population vital rates and persistence in a stream fish. Glob Change Biol 21:1856–1870
Kanno Y, Kulp MA, Moore SE (2016) Recovery of native Brook Trout populations following the eradication of nonnative Rainbow Trout in southern Appalachian Mountains streams. N Am J Fish Manag 36:1325–1335
Kanno Y, Kulp MA, Moore SE, Grossman GD (2017) Native brook trout and invasive rainbow trout respond differently to seasonal weather variation: spawning timing matters. Freshw Biol 62:868–879
Keppel G, Van Niel KP, Wardell-Johnson GW, Yates CJ, Byrne M, Mucina L, Schut AG, Hopper SD, Franklin SE (2012) Refugia: identifying and understanding safe havens for biodiversity under climate change. Glob Ecol Biogeogr 21:393–404
Kleinhesselink AR, Adler PB (2018) The response of big sagebrush (Artemisia tridentata) to interannual climate variation changes across its range. Ecology 99:1139–1149
Loarie SR, Duffy PB, Hamilton H, Asner GP, Field CB, Ackerly DD (2009) The velocity of climate change. Nature 462:1052–1055
Lu X, Hooten MB, Raiho AM, Swanson DK, Roland CA, Stehn SE (2023) Latent trajectory models for spatio-temporal dynamics in Alaskan ecosystems. Biometrics
Merriam ER, Fernandez R, Petty JT, Zegre N (2017) Can brook trout survive climate change in large rivers? If it rains. Sci Total Environ 607:1225–1236
Mohankumar NM, Hefley TJ, Silber KM, Boyle WA (2023) Data fusion of distance sampling and capture–recapture data. Spat Stat 55:100756
Mork D, Wilson A (2022) Treed distributed lag nonlinear models. Biostatistics 23:754–771
Morris JS (2015) Functional regression. Annu Rev Stat Appl 2:321–359
Ogle K, Barber JJ, Barron-Gafford GA, Bentley LP, Young JM, Huxman TE, Loik ME, Tissue DT (2015) Quantifying ecological memory in plant and ecosystem processes. Ecol Lett 18:221–235
Park T, Casella G (2008) The Bayesian lasso. J Am Stat Assoc 103:681–686
Peltier DM, Barber JJ, Ogle K (2018) Quantifying antecedent climatic drivers of tree growth in the Southwestern US. J Ecol 106:613–624
Polson NG, Scott JG (2010) Shrink globally, act locally: sparse Bayesian regularization and prediction. Bayesian Stat 9:105
Raiho AM, Scharf HR, Roland CA, Swanson DK, Stehn SE, Hooten MB (2022) Searching for refuge: a framework for identifying site factors conferring resistance to climate-driven vegetation change. Divers Distrib 28:793–809
Ramsay JO, Dalzell CJ (1991) Some tools for functional data analysis. J R Stat Soc Ser B (Methodol) 53:539–561
Renner IW, Elith J, Baddeley A, Fithian W, Hastie T, Phillips SJ, Popovic G, Warton DI (2015) Point process models for presence-only analysis. Methods Ecol Evol 6:366–379
Roghair CN, Dolloff CA, Underwood MK (2002) Response of a brook trout population and instream habitat to a catastrophic flood and debris flow. Trans Am Fish Soc 131:718–730
Royle JA (2004) N-mixture models for estimating population size from spatially replicated counts. Biometrics 60:108–115
Scharf HR, Raiho AM, Pugh S, Roland CA, Swanson DK, Stehn SE, Hooten MB (2022) Multivariate Bayesian clustering using covariate-informed components with application to boreal vegetation sensitivity. Biometrics 78:1427–1440
Skelly DK, Joseph LN, Possingham HP, Freidenburg LK, Farrugia TJ, Kinnison MT, Hendry AP (2007) Evolutionary responses to climate change. Conserv Biol 21:1353–1355
Smith AC, Brown EN (2003) Estimating a state-space model from point process observations. Neural Comput 15:965–991
Thornton M, Shrestha R, Wei Y, Thornton P, Kao S, Wilson B (2022) Daymet: daily surface weather data on a 1-km grid for North America, Version 4 R1. ORNL DAAC, Oak Ridge
Tibshirani R (1996) Regression shrinkage and selection via the lasso. J R Stat Soc Ser B (Methodol) 58:267–288
U.S. Geological Survey (2016). NHDPlus Version 2
Wadsworth C, Vera F, Piech C (2018) Achieving fairness through adversarial learning: an application to recidivism prediction. arXiv:1807.00199
Warren DR, Robinson JM, Josephson DC, Sheldon DR, Kraft CE (2012a) Elevated summer temperatures delay spawning and reduce redd construction for resident brook trout (Salvelinus fontinalis). Glob Change Biol 18:1804–1811
Warren J, Fuentes M, Herring A, Langlois P (2012b) Spatial–temporal modeling of the association between air pollution exposure and preterm birth: identifying critical windows of exposure. Biometrics 68:1157–1167
Wikle CK, Berliner LM, Cressie N (1998) Hierarchical Bayesian space–time models. Environ Ecol Stat 5:117–154
Williams PM (1995) Bayesian regularization and pruning using a Laplace prior. Neural Comput 7:117–143
Williams SE, Shoo LP, Isaac JL, Hoffmann AA, Langham G (2008) Towards an integrated framework for assessing the vulnerability of species to climate change. PLoS Biol 6:e325
Wood DM, Welsh AB, Todd Petty J (2018) Genetic assignment of brook trout reveals rapid success of culvert restoration in headwater streams. N Am J Fish Manag 38:991–1003
Wu F-Y (1982) The Potts model. Rev Mod Phys 54:235
Xu C, Letcher B, Nislow K (2010a) Size-dependent survival of brook trout Salvelinus fontinalis in summer: effects of water temperature and stream flow. J Fish Biol 76:2342–2369
Xu C, Letcher BH, Nislow KH (2010b) Context-specific influence of water temperature on brook trout growth rates in the field. Freshw Biol 55:2253–2264
Zucchini W, MacDonald IL, Langrock R (2017) Hidden Markov models for time series: an introduction using R. CRC Press, Cambridge
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors have no relevant financial or non-financial interests to disclose.
Data Availability
The datasets generated and analysed during the current study will be available on ScienceBase: https://www.sciencebase.gov/catalog/item/5f62407d82ce38aaa236148b.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix A: Simulation Study
Appendix A: Simulation Study
We generated count data from \(n = 20\) sites for \(T = 10\) years. For \(i = 1, \dots , 20\) and \(t = 1, \dots , 10\), we simulated covariates \(\varvec{x}_i = \left( 1, x_{i, 1}, x_{i, 2}\right) '\) and \(\varvec{h}_{i, t} = \left( h_{i, t, 1}, h_{i, t, 2}\right) '\) from independent standard normal distributions. To specify the latent trajectory, we let \(\varvec{m}_i = \varvec{x}_i\) for the initial conditions. We simulated a continuous driver covariate, \(w_{i, 1}(\tau )\), by sampling from independent standard normal distributions, and an indicator driver covariate, \(w_{i, 2}(\tau )\), by sampling from independent Bernoulli distributions with probability 0.1. Further, we generated spatially correlated random effects, \(\varvec{\epsilon }_{t, q} = \left( \epsilon _{1, t, q}, \dots , \epsilon _{10, t, q}\right) ', q = 1, 2\), using Equation 6. We calculated state probabilities, \(\rho _{i, t, q}\), using Equation 4, and generated \(z_{i, t, q} \sim \text {Bern}\left( \rho _{i, t, q}\right) \). We calculated population densities, \(\lambda _{i, t}\), using Equation 3, and generated true abundance, \(N_{i, t}\), using Equation 2, where we let \(A_i = 1\) for all sites. Finally, we generated observed counts, \(y_{i, t, j}, j = 1, 2, 3\), using Equation 1.
In terms of model fitting, we first selected the optimal tuning parameter from \(\varvec{c} = (0, 0.5, 1, 1.5, 2)'\). The array was concluded from the iterative selection process described in Section 2.2. We conducted a three-fold cross-validation, where at each non-overlapping fold, we randomly designated two-thirds of the simulated counts as the training set and the remaining one-third as the test set. We fit the models under different penalties to the training sets and evaluated their predictive performance on the test sets using the procedure described in Section 3.2. We then fit the best predictive model to all simulated data and summarized marginal posterior distributions. We repeated the above data simulation and model fitting processes twenty times to derive the empirical coverage rate of the inferred 95% credible intervals (Table 2).
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Lu, X., Kanno, Y., Valentine, G.P. et al. Regularized Latent Trajectory Models for Spatio-temporal Population Dynamics. JABES (2024). https://doi.org/10.1007/s13253-024-00616-y
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1007/s13253-024-00616-y