Accounting for environmental change in continuous-time stochastic population models

Abstract

The demographic rates (e.g., birth, death, migration) of many organisms have been shown to respond strongly to short- and long-term environmental change, including variation in temperature and precipitation. While ecologists have long accounted for such nonhomogeneous demography in deterministic population models, nonhomogeneous stochastic population models are largely absent from the literature. This is especially the case for models that use exact stochastic methods, such as Gillespie’s stochastic simulation algorithm (SSA), which commonly assumes that demographic rates do not respond to external environmental change (i.e., assumes homogeneous demography). In other words, ecologists are currently accounting for the effects of demographic stochasticity or environmental variability, but not both. In this paper, we describe an extension of Gillespie’s SSA (SSA + ) that allows for nonhomogeneous demography and examine how its predictions differ from a method that is partly naive to environmental change (SSAn) for two fundamental ecological models (exponential and logistic growth). We find important differences in the predicted population sizes of SSA + versus SSAn simulations, particularly when demography responds to fluctuating and irregularly changing environments. Further, we outline a computationally inexpensive approach for estimating when and under what circumstances it can be important to fully account for nonhomogeneous demography for any class of model.

This is a preview of subscription content, log in to check access.

Fig. 1
Fig. 2
Fig. 3
Fig. 4

References

  1. Adolph SC, Porter WP (1993) Temperature, activity, and lizard life histories. Am Nat 142(2):273–295

    CAS  PubMed  Article  Google Scholar 

  2. Allen E (2016) Environmental variability and mean-reverting processes. Discrete & Continuous Dynamical Systems - B 21:2073–2089

    Article  Google Scholar 

  3. Allen EJ, Allen LJ, Schurz H (2005) A comparison of persistence-time estimation for discrete and continuous stochastic population models that include demographic and environmental variability. Math Biosci 196(1):14–38

    PubMed  Article  Google Scholar 

  4. Angert AL, Huxman TE, Barron-Gafford GA, Gerst KL, Venable DL (2007) Linking growth strategies to long-term population dynamics in a guild of desert annuals. J Ecol 95(2):321–331

    Article  Google Scholar 

  5. Angilletta M (2009) Thermal adaptation: a theoretical and empirical synthesis. Oxford University Press, London

    Google Scholar 

  6. Bartlett MS (1955) An introduction to stochastic processes: with special reference to methods and applications. Cambridge University Press, Cambridge

    Google Scholar 

  7. Bartlett MS (1957) Measles periodicity and community size. J R Stat Soc Ser A (General) 120(1):48–70

    Article  Google Scholar 

  8. Black AJ, McKane AJ (2012) Stochastic formulation of ecological models and their applications. Trends Ecol Evol 27(6):337–345

    PubMed  Article  Google Scholar 

  9. Boguñá M, Lafuerza LF, Toral R, Serrano MA (2014) Simulating non-Markovian stochastic processes. Phys Rev E 90:042,108

    Article  CAS  Google Scholar 

  10. Buffon GLL (1774) Histoire naturelle, générale et particulière servant de suite à la théorie de la terre, et d’introduction à l’histoire des miné raux. De l’Imprimerie royale á Paris

  11. Collins M, Knutti R, Arblaster J, Dufresne JL, Fichefet T, Friedlingstein P, Gao X, Gutowski W, Johns T, Krinner G, Shongwe M, Tebaldi C, Weaver A, Wehner M (2013) Long-term climate change: projections, commitments and irreversibility. In: Stocker T, Qin D, Plattner G K, Tignor M, Allen S, Boschung J, Nauels A, Xia Y, Bex V, Midgley P (eds) Climate change 2013: the physical science basis. Contribution of working Group I to the fifth assessment report of the intergovernmental panel on climate change. Cambridge University Press, pp 1029–1136, Chap 12

  12. Cox DR (1955) Some statistical methods connected with series of events. J R Stat Soc Ser B Methodol 17 (2):129–164

    Google Scholar 

  13. Cox DR, Isham V (1980) Point processes. Chapman and Hall, London

    Google Scholar 

  14. Crow LH (1974) Reliability analysis for complex repairable systems. Reliability and Biometry 13:379–410

    Google Scholar 

  15. Davidson J, Andrewartha HG (1948) The influence of rainfall, evaporation and atmospheric temperature on fluctuations in the size of a natural population of Thrips imaginis (Thysanoptera). J Anim Ecol 17(2):200–222

    Article  Google Scholar 

  16. Davidson AC, Hinkley DV (1997) Bootstrap methods and their application. Cambridge University Press, Cambridge

    Google Scholar 

  17. Dell AI, Pawar S, Savage VM (2011) Systematic variation in the temperature dependence of physiological and ecological traits. Proc Natl Acad Sci U S A 108(26):10,591–10,596

    CAS  Article  Google Scholar 

  18. Deutsch CA, Tewksbury JJ, Huey RB, Sheldon KS, Ghalambor CK, Haak DC, Martin PR (2008) Impacts of climate warming on terrestrial ectotherms across latitude. Proc Natl Acad Sci U S A 105(18):6668–6672

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  19. Devroye L (1986) Non-uniform random variate generation. Springer, Berlin

    Google Scholar 

  20. Doering CR, Sargsyan KV, Sander LM (2005) Extinction times for birth-death processes: exact results, continuum asymptotics, and the failure of the Fokker–Planck approximation. Multiscale Model Simul 3(2):283–299

    Article  Google Scholar 

  21. Donat MG, Alexander LV (2012) The shifting probability distribution of global daytime and night-time temperatures. Geophys Res Lett 39(14):L14707

    Article  Google Scholar 

  22. Duan Q, Liu J (2015) A first step to implement Gillespie’s algorithm with rejection sampling. Statistical Methods & Applications 24(1):85–95

    Article  Google Scholar 

  23. Estay SA, Clavijo-Baquet S, Lima M, Bozinovic F (2011) Beyond average: an experimental test of temperature variability on the population dynamics of Tribolium confusum. Popul Ecol 53(1):53–58

    Article  Google Scholar 

  24. Estay SA, Lima M, Bozinovic F (2014) The role of temperature variability on insect performance and population dynamics in a warming world. Oikos 123(2):131–140

    Article  Google Scholar 

  25. Fay PA, Carlisle JD, Knapp AK, Blair JM, Collins SL (2003) Productivity responses to altered rainfall patterns in a c4-dominated grassland. Oecologia 137(2):245–251

    PubMed  Article  Google Scholar 

  26. Finkelstein JM (1976) Confidence bounds on the parameters of the weibull process. Technometrics 18(1):115–117

    Article  Google Scholar 

  27. Gillespie DT (1977) Exact stochastic simulation of coupled chemical reactions. J Phys Chem 81(25):2340–2361

    CAS  Article  Google Scholar 

  28. Gillespie DT (2001) Approximate accelerated stochastic simulation of chemically reacting systems. J Chem Phys 115(4):1716– 1733

    CAS  Article  Google Scholar 

  29. Gokhale CS, Papkou A, Traulsen A, Schulenburg H (2013) Lotka–Volterra dynamics kills the Red Queen: population size fluctuations and associated stochasticity dramatically change host-parasite coevolution. BMC Evol Biol 13(1):254

    PubMed  PubMed Central  Article  Google Scholar 

  30. Grandell J (1976) Doubly stochastic Poisson processes. Springer, Berlin

    Google Scholar 

  31. Hart SP, Schreiber SJ, Levine JM (2016) How variation between individuals affects species coexistence. Ecol Lett 19(8):825– 838

    PubMed  Article  Google Scholar 

  32. Hartmann D, Klein Tank A, Rusticucci M, Alexander L, Bronnimann S, Charabi Y, Dentener F, Dlugokencky E, Easterling D, Kaplan A, Soden B, Thorne P, Wild M, Zhai P (2013) Observations: atmosphere and surface. In: Stocker T, Qin D, Plattner GK, Tignor M, Allen S, Boschung J, Nauels A, Xia Y, Bex V, Midgley P (eds) Climate change 2013: the physical science basis. Contribution of working Group I to the fifth assessment report of the intergovernmental panel on climate change, vol 2. Cambridge University Press, Cambridge. Book Section, pp 159– 254

  33. Heisler-White JL, Knapp AK, Kelly EF (2008) Increasing precipitation event size increases aboveground net primary productivity in a semi-arid grassland. Oecologia 158(1):129–140

    PubMed  Article  Google Scholar 

  34. Henson SM, Costantino RF, Cushing JM, Desharnais RA, Dennis B, King AA (2001) Lattice effects observed in chaotic dynamics of experimental populations. Science 294(5542):602–605

    CAS  PubMed  Article  Google Scholar 

  35. Huang W, Hauert C, Traulsen A (2015) Stochastic game dynamics under demographic fluctuations. Proc Natl Acad Sci U S A 112(29):9064–9069

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  36. Huey RB, Kingsolver JG (1989) Evolution of thermal sensitivity of ectotherm performance. Trends Ecol Evol 4(5):131–135

    CAS  PubMed  Article  Google Scholar 

  37. Huey RB, Stevenson R (1979) Integrating thermal physiology and ecology of ectotherms: a discussion of approaches. Am Zool 19(1):357–366

    Article  Google Scholar 

  38. Huntingford C, Jones PD, Livina VN, Lenton TM, Cox PM (2013) No increase in global temperature variability despite changing regional patterns. Nature 500(7462):327–330

    CAS  PubMed  Article  Google Scholar 

  39. Huxman TE, Snyder KA, Tissue D, Leffler AJ, Ogle K, Pockman WT, Sandquist DR, Potts DL, Schwinning S (2004) Precipitation pulses and carbon fluxes in semiarid and arid ecosystems. Oecologia 141 (2):254–268

    PubMed  Article  Google Scholar 

  40. Kaplan N (1973) A continuous time Markov branching model with random environments. Adv Appl Probab 5(1):37–54

    Article  Google Scholar 

  41. Keeling M, Ross J (2008) On methods for studying stochastic disease dynamics. J R Soc Interface 5 (19):171–181

    CAS  PubMed  Article  Google Scholar 

  42. Kendall DG (1948) On the generalized “birth-and-death” process. Ann Math Stat 19(1):1–15

    Article  Google Scholar 

  43. Kessler DA, Shnerb NM (2007) Extinction rates for fluctuation-induced metastabilities: a real-space WKB approach. J Stat Phys 127(5):861–886

    Article  Google Scholar 

  44. Kingsolver JG, Diamond SE, Buckley LB (2013) Heat stress and the fitness consequences of climate change for terrestrial ectotherms. Funct Ecol 27(6):1415–1423

    Article  Google Scholar 

  45. Kingsolver JG, Higgins JK, Augustine KE (2015) Fluctuating temperatures and ectotherm growth: distinguishing non-linear and time-dependent effects. J Exp Biol 218(14):2218– 2225

    PubMed  Article  Google Scholar 

  46. Knapp AK, Smith MD (2001) Variation among biomes in temporal dynamics of aboveground primary production. Science 291(5503):481–484

    CAS  PubMed  Article  Google Scholar 

  47. Kolmogoroff A (1931) ÜBer die analytischen methoden in der wahrscheinlichkeitsrechnung. Math Ann 104:415–458

    Article  Google Scholar 

  48. Kolpas A, Nisbet RM (2010) Effects of demographic stochasticity on population persistence in advective media. Bull Math Biol 72(5):1254–1270

    PubMed  Article  Google Scholar 

  49. Kramer AM, Drake JM (2010) Experimental demonstration of population extinction due to a predator-driven Allee effect. J Anim Ecol 79(3):633–639

    PubMed  Article  Google Scholar 

  50. Kramer AM, Drake JM (2014) Time to competitive exclusion. Ecosphere 5(5):1–16

    Article  Google Scholar 

  51. Lande R (1993) Risks of population extinction from demographic and environmental stochasticity and random catastrophes. Am Nat 142(6):911–927

    PubMed  Article  Google Scholar 

  52. Loik ME, Breshears DD, Lauenroth WK, Belnap J (2004) A multi-scale perspective of water pulses in dryland ecosystems: climatology and ecohydrology of the Western USA. Oecologia 141(2):269–281

    PubMed  Article  Google Scholar 

  53. Mangel M, Tier C (1993) A simple direct method for finding persistence times of populations and application to conservation problems. Proc Natl Acad Sci U S A 90(3):1083–1086

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  54. Marion G, Renshaw E, Gibson G (2000) Stochastic modelling of environmental variation for biological populations. Theor Popul Biol 57(3):197–217

    CAS  PubMed  Article  Google Scholar 

  55. Meisner MH, Harmon JP, Ives AR (2014) Temperature effects on long-term population dynamics in a parasitoid–host system. Ecol Monogr 84(3):457–476

    Article  Google Scholar 

  56. Miquel J, Lundgren PR, Bensch KG, Atlan H (1976) Effects of temperature on the life span, vitality and fine structure of drosophila melanogaster. Mech Ageing Dev 5:347–370

    CAS  PubMed  Article  Google Scholar 

  57. Nisbet RM, Martin BT, de Roos AM (2016) Integrating ecological insight derived from individual-based simulations and physiologically structured population models. Ecol Model 326:101– 112

    Article  Google Scholar 

  58. Novoplansky A, Goldberg DE (2001) Effects of water pulsing on individual performance and competitive hierarchies in plants. J Veg Sci 12(2):199–208

    Article  Google Scholar 

  59. Okuyama T (2015) Demographic stochasticity alters the outcome of exploitation competition. J Theor Biol 365:347–351

    PubMed  Article  Google Scholar 

  60. Orrock JL, Fletcher RJ (2005) Changes in community size affect the outcome of competition. Am Nat 166 (1):107–111

    PubMed  Article  Google Scholar 

  61. Orrock JL, Watling JI (2010) Local community size mediates ecological drift and competition in metacommunities. Proc R Soc Lond B Biol Sci 277(1691):2185–2191

    Article  Google Scholar 

  62. Ovaskainen O, Meerson B (2010) Stochastic models of population extinction. Trends Ecol Evol 25(11):643–652

    PubMed  Article  Google Scholar 

  63. Paaijmans KP, Heinig RL, Seliga RA, Blanford JI, Blanford S, Murdock CC, Thomas MB (2013) Temperature variation makes ectotherms more sensitive to climate change. Glob Chang Biol 19(8):2373–2380

    PubMed  PubMed Central  Article  Google Scholar 

  64. Palamara GM, Carrara F, Smith MJ, Petchey OL (2016) The effects of demographic stochasticity and parameter uncertainty on predicting the establishment of introduced species. Ecol Evol 6(23):8440–8451

    PubMed  PubMed Central  Article  Google Scholar 

  65. Parmesan C (2006) Ecological and evolutionary responses to recent climate change. Annu Rev Ecol Evol Syst 37(1):637–669

    Article  Google Scholar 

  66. Pearl R, Reed LJ (1920) On the rate of growth of the population of the United States since 1790 and its mathematical representation. Proc Natl Acad Sci U S A 6(6):275–288

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  67. Pedruski MT, Fussmann GF, Gonzalez A (2015) Predicting the outcome of competition when fitness inequality is variable. Royal Society Open Science 2(8):150,274

    Article  Google Scholar 

  68. R Core Team (2017) R: a language and environment for statistical computing r foundation for statistical computing, Vienna, Austria

  69. Ross S (2014) Introduction to probability models, 11th edn. Academic, New York

    Google Scholar 

  70. Shaffer ML (1981) Minimum population sizes for species conservation. Bioscience 31(2):131–134

    Article  Google Scholar 

  71. Simonis JL (2012) Demographic stochasticity reduces the synchronizing effect of dispersal in predator–prey metapopulations. Ecol 93(7):1517–1524

    Article  Google Scholar 

  72. Soetaert K, Petzoldt T, Setzer RW (2010) Solving differential equations in R: package desolve. J Stat Softw 33(9):1–25

    Article  Google Scholar 

  73. Stroustrup N, Anthony WE, Nash ZM, Gowda V, Gomez A, López-Moyado IF, Apfeld J, Fontana W (2016) The temporal scaling of Caenorhabditis elegans ageing. Nature 530:103– 107

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  74. Uhlenbeck GE, Ornstein LS (1930) On the theory of the Brownian motion. Phys Rev 36:823–841

    CAS  Article  Google Scholar 

  75. van den Broek J, Heesterbeek H (2007) Nonhomogeneous birth and death models for epidemic outbreak data. Biostatistics 8(2):453–467

    Article  Google Scholar 

  76. van Kampen NG (1992) Stochastic processes in physics and chemistry. Elsevier, Amsterdam

    Google Scholar 

  77. Varughese M, Fatti L (2008) Incorporating environmental stochasticity within a biological population model. Theor Popul Biol 74(1):115–129

    CAS  PubMed  Article  Google Scholar 

  78. Verhulst P (1845) Recherches mathématiques sur la loi d’accroissement de la population. Nouveaux mémoires de l’Académie Royale des Sciences et Belles-Lettres de Bruxelles 18:14–54

    Google Scholar 

  79. Vestergaard CL, Génois M (2015) Temporal gillespie algorithm: fast simulation of contagion processes on time-varying networks. PLoS Comput Biol 11(10):1–28

    Article  CAS  Google Scholar 

  80. Von Neumann J (1951) Various techniques used in connection with random digits. Appl Math Ser 12:36–38

    Google Scholar 

  81. Wilcox C, Possingham H (2002) Do life history traits affect the accuracy of diffusion approximations for mean time to extinction? Ecol Appl 12(4):1163–1179

    Article  Google Scholar 

  82. Yaari G, Ben-Zion Y, Shnerb NM, Vasseur DA (2012) Consistent scaling of persistence time in metapopulations. Ecol 93(5):1214–1227

    Article  Google Scholar 

  83. Yule GU (1925) A mathematical theory of evolution, based on the conclusions of Dr. J. C. Willis, F.R.S. Philos Trans R Soc Lond B: Biol. Sci. 213(402–410):21–87

    Article  Google Scholar 

Download references

Acknowledgements

GL conceived of the study and ran the simulations. Authors GL and BAM did the analysis of the simulations and wrote the paper. The Janus supercomputer is a joint effort of the University of Colorado Boulder, the University of Colorado Denver and the National Center for Atmospheric Research. We thank the ESA Theoretical Ecology section for a poster award to GL, which provided early support for the project.

Funding

In this study, preliminary simulation work used the Janus supercomputer, which is supported by the National Science Foundation (DEB, award number 1457660) and the University of Colorado Boulder.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Geoffrey Legault.

Appendices

Appendix A: Diagnostic tests of SSA + method

To ensure we were generating appropriate inter-event times, \(\tau \), using our implementation of the SSA + method, we performed two diagnostic tests:

  1. 1.

    We simulated a non-stationary exponential growth model using our SSA + implementation with a rate function of \(F(t)=\alpha t^{-\beta }\) (a power law function) and compared our results to taking samples from a Weibull distribution with parameters \(\alpha \) and \(\beta \), as discussed in the main text. These results should align nearly exactly if our implementation is correct, which appears to be the case (Fig. 5).

  2. 2.

    We set the environment function to a constant value (i.e., demography was stationary) and compared the results of 50,000 simulations of the SSA + method to an independent SSA implementation (since demography is stationary in this case, we refer to it simply as the SSA rather than the SSAn, though both use the same methods). With a constant environment, the SSA + method should match the SSA method, which appears to be the case (Fig. 6).

Fig. 5
figure5

The probability density of 1,000,000 arrival times, obtained by sampling from a Weibull distribution with scale parameter α and shape parameter β (black), or using our SSA + method for a non-stationary stochastic process with rate function F(t) = αtβ (blue). In both cases, α = 1 and β = 1.5

Fig. 6
figure6

The frequency distribution of population sizes at t = 100 for 50,000 simulations of exponential growth (a) and logistic growth (b), simulated using the SSA (red) or SSA + (blue) methods with a constant environment. Colors are transparent, so the color purple indicates overlap between the SSA and SSA + methods. Also displayed are the (overlapping) expected values, \(\bar {N}\) (rounded) for each simulation method

Finally, we also examined the extent to which 10,000 simulations was sufficient to capture the ensemble mean and variance of the stochastic simulations. To do so, we examined how the means and variances of the exponential growth models were affected by the number of simulations used. We found only very small differences in the means and variances from 1000 to 10,000 simulations used (see Fig. 7 for this comparison).

Fig. 7
figure7

The expected population sizes at t = 100 of the SSA + method (exponential growth) for different numbers of simulations (x-axis). Displayed are the population sizes for the increasing (a), fluctuating (b), and irregularly changing (c) environment functions. Points represent the mean population size (across simulations), while bars are the standard deviations of the simulations

Appendix B: R code for SSA + method

This section contains sample code for simulating 1 run of the Gillespie algorithm (SSA or SSAn in the main text) and 1 run of the extended, nonhomogeneous (i.e., non-stationary) version of the algorithm (SSA + ) for an exponential growth model. The code is specifically designed to allow for environment- or time-dependent demography, but in this example the environment is constant. Additional code and more detailed examples are available at https://github.com/legault/SSAplus.

figurea
figureb
figurec

Appendix C: Comparison of SSAn and SSA + for slower environment functions

In Fig. 8, we compare SSAn and SSA + predictions for the slower environment functions, specifically “increasing 1”, “fluctuating 1”, and “irregular 1”. The qualitative results are similar to those in Fig. 3 (in main text), albeit with more overlap between the distributions.

Fig. 8
figure8

The frequency distribution of population sizes at t = 100 for the exponential growth (ac) and logistic growth (df) models, simulated using either the SSAn (red) or SSA + (blue) method. Displayed are the population size distributions for environment functions: increasing 1 (a, d), fluctuating 1 (b, e), and irregular 1 (c, f). Colors are transparent, so purple indicates overlap between the SSAn and SSA + methods. Also displayed are the expected values, \(\bar {N}\) (rounded to nearest integer), for each simulation method (same coloration as above). Note the different scales on both axes for each panel

Appendix D: CDFs beginning at different time points

In Fig. 9, we compare the CDFs for the exponential growth model assuming the simulations began at \(t = 3\) (Fig. 9a–c) or \(t = 5\) (Fig. 9d–f), with respect to the environment functions. For the fluctuating and irregular environments, these CDFs differ from those displayed in Fig. 2 of the main text. For example, at \(t = 3\) environment function “fluctuating 1” is decreasing rapidly thereby lowering the birth rate and consequently the CDF in Fig. 9b. In contrast, the function “fluctuating 2” is increasing rapidly at \(t = 3\) and the resulting CDF is elevated compared to the SSAn CDF.

Fig. 9
figure9

Cumulative probabilities over time (starting at t = 3 for (ac); at t = 5 for (df)) that the next demographic event will occur for the exponential growth model under increasing (a, d), fluctuating (b, e), and irregular (c, f) environments. The solid lines represent the probabilities for the SSAn method, which assumes that demographic processes occur at constant rates over time (in this case, the rate at t = 3 or t = 5). Dashed and dotted lines represent the probabilities obtained from the SSA + method, which accurately accounts for nonhomogeneous demographic rates (compare to Fig. 1). As in Fig. 2, the CDFs assume an “initial” (i.e., beginning at t = 3 or t = 5) population size of 100

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Legault, G., Melbourne, B.A. Accounting for environmental change in continuous-time stochastic population models. Theor Ecol 12, 31–48 (2019). https://doi.org/10.1007/s12080-018-0386-z

Download citation

Keywords

  • Demographic stochasticity
  • Environmental change
  • Stationary
  • Non-stationary
  • Homogeneous
  • Nonhomogeneous
  • Environment-dependent demography
  • Stochastic simulation algorithm
  • Poisson process