Invariant predictions of epidemic patterns from radically different forms of seasonal forcing

David J.D. Earn; Irena Papst

REVIEW 2 major objections 4 minor 38 references

Invariant predictions of epidemic patterns from radically different forms of seasonal forcing

T0 review · 2 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A single matched bifurcation makes epidemic predictions indifferent to the shape of seasonal forcing.

desk verdict Solid numerical study of birth-fold invariance in seasonally forced SIR; the abstract oversells it, but the core result is real and worth refereeing. read the letter →

arxiv 1908.02843 v1 pith:BGSYMPIU submitted 2019-08-07 q-bio.PE math.DSnlin.CD

Irena Papst , David J.D. Earn This is my paper · ORCID

classification q-bio.PEmath.DSnlin.CD MSC 92D3037G1037N25

keywords seasonalforcingSIRepidemicmodelbifurcationinvarianceperioddoublingfoldstroboscopicmappredator-preymeaslesdynamics

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper asks whether predictions from seasonally forced epidemic models can be trusted when the true pattern of seasonal forcing is unknown. Its answer, for the standard SIR model, is yes: if the amplitude of forcing is chosen so that the principal period-doubling bifurcation occurs at the same value of the basic reproduction number $R_0$, then the positions of the fold bifurcations that give birth to 3- through 7-year epidemic cycles match across forcing shapes as different as school-term pulses, square waves, and sinusoids. The paper demonstrates this by numerical continuation across a continuous family of forcing functions and reports that the matched bifurcations agree to within about one percent. Because real forcing patterns are never known precisely, the result means idealized sinusoidal forcing can be used to predict qualitative changes in epidemic patterns, and the same invariance appears in a seasonally forced predator-prey model.

What carries the argument

The central machinery is the one-year stroboscopic map of the forced SIR model, which samples the continuous trajectory once per forcing period, together with numerical continuation of bifurcation curves. A one-parameter family of forcing functions $\mathrm{osc}_p(t)$ interpolates from term-time forcing ($p = -1$) through square wave forcing ($p = 0$) to sinusoidal forcing ($p = 1$) and beyond to impulse-like forcing ($p \to \infty$). The amplitude-matching function $\alpha(p)$ is obtained by continuing the stable period-doubling bifurcation in the $(p, \alpha)$ plane from the anchor point $p = -1$, $\alpha = 0.25$, $R_0 = 15.12$; this function fixes the amplitude that makes the principal period-doubling occur at the same $R_0$ for each shape. The claim is carried by the observation that along this curve, the birth folds of period-3 through period-7 cycles remain aligned in $R_0$, while death folds do not.

What would settle it

Compute the bifurcation diagram for forcing shapes outside the family, e.g., a realistic two-term school calendar with unequal term lengths, choosing amplitude by the same $\alpha(p)$ continuation rule; if the $R_0$ positions of the 3- to 7-year birth folds deviate by more than the roughly one percent spread in table 1, the invariance claim fails. Alternatively, repeat the continuation anchored at a different bifurcation, such as the period-3 birth fold, and check whether all the other folds still align.

Watch

Extended reading notes

Core claim

The central claim is that the key bifurcations of the standard seasonally forced SIR model are invariant to the shape of seasonal forcing after a single amplitude adjustment. Starting from term-time forcing with amplitude $\alpha = 0.25$, the authors continue the stable period-doubling bifurcation in the two-parameter (shape, amplitude) plane, obtaining an amplitude function $\alpha(p)$ that pins the period-doubling at $R_0 = 15.12$ for every forcing shape in the family. With no further adjustment, the fold bifurcations that create 3-, 4-, 5-, 6-, and 7-year epidemic cycles occur at essentially the same $R_0$ values for all shapes, with relative differences between $0.0005$ and $0.009$, at most about one percent. The invariance is quantitative for these birth folds and qualitative for the overall bifurcation structure; the corresponding death folds are not invariant. The paper additionally finds the same type of invariance in a seasonally forced predator-prey model and conjectures that the phenomenon may be general among periodically forced dynamical systems.

Load-bearing premise

The invariance is established by matching one bifurcation at a single anchor point ($R_0 = 15.12$) for one measles-like parameter set; if a different anchor point or a different demographic regime breaks the alignment, the practical recommendation to use sinusoidal forcing would not generalize.

Editorial extensions

If this is right

Epidemiological transition analyses for measles and other childhood diseases can be conducted with sinusoidal forcing and still capture the $R_0$ values at which biennial, triennial, and longer epidemic cycles appear, provided the amplitude is adjusted to fix the principal period-doubling.
Rigorous mathematical results obtained under sinusoidal forcing, such as proofs of coexistence of multiple attractors or of chaos, are more likely to transfer to realistic forcing patterns than previously assumed.
The invariance extends beyond epidemics: the same matched-bifurcation behavior occurs in a seasonally forced predator-prey model, so ecosystem predictions may also be robust to forcing shape.
The practical scope of the invariance is the births of multi-year cycle branches: the folds that destroy those branches and some intermediate period doublings do shift with forcing shape, so quantitative predictions about those transitions still require accurate forcing.
The paper's conjecture suggests that any forced nonlinear oscillator with comparable resonance structure may show the same invariance, making idealized forcing a safer default in other fields.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

A natural testable extension is to apply the same amplitude-matching rule to forcing shapes outside the constructed family, such as real school calendars with unequal terms or climate-driven forcing in vector-borne diseases; if the birth folds still align, the invariance is more general than the paper's family.
The paper does not identify the conserved property of the forcing that produces the invariance; average spectral power is explicitly ruled out. If such a property exists, it might allow the correct amplitude for an arbitrary forcing shape to be computed without running a two-parameter continuation.
The anchor dependence is an open question: repeating the continuation from a different bifurcation, such as one of the birth folds themselves, would test whether the amplitude-matching rule is self-consistent across all key transitions.
If the invariance holds for other strongly nonlinear oscillators, it would give practical warrant to a general strategy: when only the period of forcing is certain, use the simplest periodic shape and calibrate amplitude to one measured transition.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

You should know two things about this paper. First, the core numerical result is real: after matching the principal period-doubling at R0=15.12, the five birth folds of the seasonally forced SIR model line up across forcing shapes with relative differences under 1%, which is genuinely surprising. Second, the title's "invariant predictions" is broader than what is shown: the invariance holds for birth folds, not for death folds and intermediate period-doublings, and it is demonstrated for one parameter set and one anchor point. The paper itself says the death folds don't line up (Sec 3.2), so the abstract overstates the result.

The systematic study across a continuous family of forcing functions is new. Earlier refs (6 and 17) noted qualitative similarity; this paper quantifies it via two-parameter continuation and shows exactly how the amplitude must trade off against shape. Table 1 and Figure 5 are clear. The honesty about which bifurcations do and do not line up is a real plus. The predator-prey result in the supplementary, though briefly presented, suggests the phenomenon may be more general than just measles.

The soft spots are three. First, the abstract says "key bifurcations" but the quantitative result covers only the birth folds. That should be fixed in revision. Second, the amplitude-adjustment function alpha(p) is anchored at a single point (R0=15.12 for measles parameters). Nothing in the paper tests whether the alignment survives with a different anchor or different host parameters (say, shorter/longer infectious period). The practical suggestion -- use a sinusoid with matched amplitude -- depends on that generalization. It may well be true, but it is currently an assumption. Third, the arXiv version is missing the electronic supplementary material and code. The text references it, but it isn't there. For a numerical paper, that's a reproducibility gap; the published version presumably includes it, but a reader of the arXiv version cannot verify table 1 without asking the authors.

The circularity worry turns out to be minor. The PD is fixed by construction, but the birth folds are not, and their alignment is the substantive finding.

Who should read this: anyone doing bifurcation analysis of seasonally forced epidemic models, and people interested in forced nonlinear oscillators more broadly. It's a good bit of work, worth a careful referee. My recommendation: send it to peer review. The authors should be asked to calibrate the abstract to the actual result, to say clearly that the invariance is for birth folds, and ideally to add a sensitivity check with a second anchor or parameter set. Without that, the paper is still a solid numerical contribution, but it doesn't quite deliver what the abstract promises.

Referee Report

2 major / 4 minor

Summary. This paper asks whether the detailed shape of seasonal forcing matters for the bifurcation structure of the standard seasonally forced SIR epidemic model. The authors construct a continuous family of forcing functions osc_p(t) that interpolates between term-time forcing (p = -1), square-wave (p = 0), sinusoidal (p = 1), and a delta-impulse-like limit (p large). Using numerical continuation with XPPAUT, they compute, for each shape p, the forcing amplitude alpha(p) that places the principal period-doubling (PD) bifurcation at the same value R0 = 15.12 as for term-time forcing. They then compare the positions of other bifurcations under this matched-amplitude condition. The main quantitative result, summarized in Table 1 and Figure 5, is that five 'birth fold' bifurcations, which create period-3 through period-7 attractors, occur at very similar R0 values across the six forcing shapes, with relative differences between 0.0004 and 0.009. The paper also reports that the corresponding 'death folds' and some intermediate period-doublings do not align, and that a similar qualitative invariance is observed in a seasonally forced predator-prey model. The authors conclude that key bifurcations are invariant to forcing shape when the amplitude is appropriately adjusted, and recommend sinusoidal forcing as a safe stand-in for real, poorly characterized seasonal forcing.

Significance. If the reported invariance holds beyond the specific calibrated case, this is a genuinely useful result for epidemiological modeling: it would validate the widespread use of sinusoidal forcing in studies of epidemic transitions and chaos, despite the fact that real seasonal forcing (e.g., school terms) is strongly non-sinusoidal. The numerical work is careful and reproducible: the continuation uses standard open-source software, the code is provided in supplementary material, and the relative-difference table permits quantitative scrutiny. A notable strength is that the birth-fold positions are not fitted; after fixing the principal PD by construction, the five folds are computed and found to align. The authors are also honest in Section 3.2 about the non-invariance of death folds and intermediate PDs. However, the abstract and conclusion substantially overstate the scope of the result: the quantitative invariance is demonstrated for only one anchor (R0 = 15.12), one parameter set (measles-like), and only the birth folds, not the entire bifurcation skeleton. The practical recommendation to use sinusoidal forcing 'with confidence' therefore goes beyond what is currently shown.

major comments (2)

[Abstract and Section 3.2] The abstract states that 'the key bifurcations of the standard epidemic model are invariant to the shape of seasonal forcing if the amplitude of forcing is appropriately adjusted,' and Section 4 repeats this claim in stronger language. Yet Section 3.2 explicitly states that the 'death folds' at the right edges of the fold branches and some intermediate period-doublings are not invariant, as shown by the open symbols in Figure 5. The concrete quantitative result in Table 1 is restricted to the principal PD (fixed by construction) and the five birth folds. This is more than a semantic nuance: the phrase 'key bifurcations' is precisely what a reader would use to decide whether sinusoidal forcing can reproduce the full bifurcation diagram. The abstract and conclusion should be revised to specify that the invariance is quantitative for the birth folds and only qualitative for the remainder of the bifurcation structure.
[Section 2.4] The amplitude-adjustment function alpha(p) is defined by continuing the stable period-doubling bifurcation from a single anchor point, R0 = 15.12, with fixed parameters mu = 0.02/year and 1/gamma = 13 days. The paper does not test whether the alignment of the five birth folds persists if the anchor is chosen differently, for example by continuing from another R0 value on the same PD branch or from a different bifurcation, or with different mu and gamma. Since the central claim and the practical recommendation depend on the matching being representative rather than accidental, the authors should either (a) report additional continuation runs for at least a few alternative anchors and parameter sets, or (b) explicitly limit the conclusion to the calibrated regime. Without this, the general statement that 'the key bifurcations ... are invariant' is not fully supported.

minor comments (4)

[Section 4] The word 'sinuoidal' in the final paragraph should be 'sinusoidal'.
[Figure 3 caption] The caption says 'In the top two panels' and 'In the bottom two panels,' implying a 2x2 layout, but the body text in Section 2.4 refers to 'the term-time bifurcation diagram (figure 3, top panel)' and 'the bottom panel of figure 3.' Please clarify the layout of the figure panels and use consistent terminology (e.g., 'top-left,' 'bottom-right') throughout.
[Table 1] The table title says 'Invariance of fold bifurcations,' but the table includes a row for the period-doubling bifurcation. Either rename the table to indicate that the PD is shown for reference, or add a note that the PD is included to show the anchoring condition.
[Figure 5] The figure demonstrates the alignment of birth folds visually, but the numerical tolerance of the continuation is not stated in the caption. Adding a sentence about the continuation error (or stating that relative differences are given in Table 1) would help readers judge whether the observed scatter is within numerical precision.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the period-doubling alignment is an explicit calibration, and the reported fold-bifurcation invariance is computed independently.

full rationale

The paper's central claim is that, after matching the principal period-doubling (PD) bifurcation at R0 = 15.12, five birth folds of the seasonally forced SIR model are quantitatively invariant across forcing shapes. This is not circular because the matched PD is openly treated as a calibration, not as a prediction. Section 2.4 says the continuation 'can be thought of as a function, alpha(p), which specifies the forcing amplitude (alpha) that yields a PD at the same R0 value for any given shape of forcing pattern (p).' Figure 5's caption explicitly marks 'The PD that is fixed by construction' with solid black squares, and Section 3.4 repeats that the vertical alignment of the squares is because each occurs at R0 = 15.12. Thus the PD alignment is an input, not a derived outcome. The five fold bifurcations in Table 1 are computed separately by continuing one-parameter bifurcation diagrams at the calibrated amplitudes; their R0 values are not used to define alpha(p), and their close alignment (relative differences 0.0005 to 0.0090 for Fold 3 through Fold 7) is an independent numerical finding. The paper also discloses the limits of the invariance: Section 3.2 states that the folds at the right edges of the branches 'do not line up precisely' and that the quantitative invariance is restricted to the births, not deaths, of the branches. This candor further shows that the claim is not being stretched to cover anything forced by construction. The self-citations used for the anchor amplitude and measles parameters ([6], [17], [26]) supply starting values, but the invariance itself is established by the authors' own continuation calculations, not by assertoric citation. There is no self-definitional reduction, no fitted parameter renamed as a prediction, and no load-bearing uniqueness or ansatz imported from the authors' prior work. The derivation chain is self-contained for what it claims: a calibrated comparison across forcing shapes with the folds reported as computed, not as fitted.

Assumptions & free parameters 6 free parameters · 4 assumptions · 0 invented entities

The ledger lists the amplitude values matched to anchor the period-doubling as free parameters, and the modeling and numerical assumptions needed for the result. No new physical entities are introduced. The main unstated cost is the representativeness of the forcing family and the generalization of the invariance to untested parameter regimes.

free parameters (6)

alpha(p=-1) = 0.25
Term-time forcing amplitude estimated from measles data in refs 6,17,26; serves as the anchor for the continuation.
alpha(p=-0.5) = 0.1012
Chosen so the stable period-doubling bifurcation occurs at R0=15.12, matching the term-time case.
alpha(p=0) = 0.0782
Chosen so the stable period-doubling bifurcation occurs at R0=15.12.
alpha(p=0.25) = 0.0839
Chosen so the stable period-doubling bifurcation occurs at R0=15.12.
alpha(p=1) = 0.1
Chosen so the stable period-doubling bifurcation occurs at R0=15.12.
alpha(p=2) = 0.1182
Chosen so the stable period-doubling bifurcation occurs at R0=15.12.

assumptions (4)

standard math The numerical continuation software (XPPAUT) correctly computes bifurcation curves for the stroboscopic map.
The central numerical results depend on the reliability of the continuation algorithm; the paper does not provide an analytic proof of the bifurcation positions.
domain assumption The standard seasonally forced SIR model (equation 2.1) adequately represents childhood disease transmission dynamics.
The practical conclusions about real disease dynamics rely on the model being a faithful representation of diseases like measles.
ad hoc to paper The constructed family of forcing functions osc_p(t) (supplementary S1.1) spans 'radically different' realistic seasonal patterns.
The family is purpose-built for this paper; its representativeness of real forcing patterns is assumed, not established from data.
domain assumption The invariance observed at p=-1,-0.5,0,0.25,1,2 holds for all shapes in the continuous family.
Only six discrete values of p are tested; the claim of invariance across the family assumes no exceptional behavior between these points.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Invariant predictions of epidemic patterns from radically different forms of seasonal forcing." pith.science (2026). https://pith.science/paper/BGSYMPIU

@misc{pith2026190802843,
  author       = {Pith},
  title        = {Pith review of: Invariant predictions of epidemic patterns from radically different forms of seasonal forcing},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BGSYMPIU}},
  note         = {Machine review of arXiv:1908.02843}
}

read the original abstract

Seasonal variation in environmental variables, and in rates of contact among individuals, are fundamental drivers of infectious disease dynamics. Unlike most periodically-forced physical systems, for which the precise pattern of forcing is typically known, underlying patterns of seasonal variation in transmission rates can be estimated approximately at best, and only the period of forcing is accurately known. Yet solutions of epidemic models depend strongly on the forcing function, so dynamical predictions---such as changes in epidemic patterns that can be induced by demographic transitions or mass vaccination---are always subject to the objection that the underlying patterns of seasonality are poorly specified. Here, we demonstrate that the key bifurcations of the standard epidemic model are invariant to the shape of seasonal forcing if the amplitude of forcing is appropriately adjusted. Consequently, analyses applicable to real disease dynamics can be conducted with a smooth, idealized sinusoidal forcing function, and qualitative changes in epidemic patterns can be predicted without precise knowledge of the underlying forcing pattern. We find similar invariance in a seasonally forced predator-prey model, and conjecture that this phenomenon---and the associated robustness of predictions---might be a feature of many other periodically forced dynamical systems.

Figures

Figures reproduced from arXiv: 1908.02843 by the authors.

**Figure 2.** Several members of the family of forcing functions, oscp(t), described in 2.2. Panels, from top to bottom: p = −1 (term-time forcing), p = −0.5, p = 0 (square wave forcing), p = 0.25, p = 1 (sinusoidal forcing), p = 2. These shape parameter values correspond to those used in figure 5. Details of the construction are given in electronic supplementary material, S1.1. 14 [PITH_FULL_IMAGE:figures/full_fig_p014_2.png] view at source ↗

**Figure 3.** R0 bifurcation diagrams for the annual stroboscopic map of the seasonally forced SIR model (equation 2.1) with different patterns and amplitudes of forcing. In the top two panels, the forcing pattern is different but the associated amplitudes are the same. In the bottom two panels, the forcing pattern is the same but the amplitudes are different. The fixed parameter values are µ = 0.02/year and 1/γ = 13 days (corres… view at source ↗

Figures from the paper (2 more)

**Figure 4.** Figure 4: Continuation of the stable period doubling (PD) bifurcation in the two-dimensional (p, α) parameter plane (see 2 and electronic supplementary material, S1.2.2). The continuation was initiated at term-time forcing (p = −1) with the amplitude estimated from data [6,17,26…

**Figure 5.** Figure 5: Graphical representation of bifurcation invariance in the seasonally forced SIR model (equation 2.1). For six forcing patterns (p, left vertical axis) and amplitudes determined by the function shown in figure 4 (α(p), right vertical axis), the values of R0 (horizonta…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

38 extracted references · 38 canonical work pages

[1]

Absolute humidity modulates inﬂuenza survival, transmis- sion, and seasonality

Shaman J, Kohn M. Absolute humidity modulates inﬂuenza survival, transmis- sion, and seasonality. Proc. Natl. Acad. Sci. USA. 2009;106(9):3243–3248

work page 2009
[2]

Patterns of spread of inﬂuenza A in Canada

He D, Dushoﬀ J, Eftimie R, Earn DJD. Patterns of spread of inﬂuenza A in Canada. Proc. R. Soc. Lond. B. 2013;280(1770):20131174

work page 2013
[3]

Epidemiological eﬀects of seasonal oscillations in birth rates

He D, Earn DJD. Epidemiological eﬀects of seasonal oscillations in birth rates. Theor Popul Biol. 2007;72:274–291

work page 2007
[4]

Human birth seasonality: latitudinal gradient and interplay with childhood disease dynamics

Martinez-Bakker M, Bakker KM, King AA, Rohani P. Human birth seasonality: latitudinal gradient and interplay with childhood disease dynamics. Proc. R. Soc. Lond. B. 2014;281(1783)

work page 2014
[5]

Recurrent outbreaks of measles, chickenpox and mumps

London W, Yorke JA. Recurrent outbreaks of measles, chickenpox and mumps. I. Seasonal variation in contact rates. Am J Epidemiol. 1973;98(6):453–468. 8

work page 1973
[6]

A simple model for complex dynamical transitions in epidemics

Earn DJD, Rohani P, Bolker BM, Grenfell BT. A simple model for complex dynamical transitions in epidemics. Science. 2000;287(5453):667–670

work page 2000
[7]

Seasonal dynamics of recurrent epidemics

Stone L, Olinky R, Huppert A. Seasonal dynamics of recurrent epidemics. Na- ture. 2007;446(7135):533–536

work page 2007
[8]

Eﬀects of school closure on incidence of pandemic inﬂuenza in Alberta, Canada

Earn DJD, He D, Loeb MB, Fonseca K, Lee BE, Dushoﬀ J. Eﬀects of school closure on incidence of pandemic inﬂuenza in Alberta, Canada. Ann Intern Med. 2012;156(3):173–181

work page 2012

Show all 38 references

[9]

An age-structured model of pre- and post-vaccination measles trans- mission

Schenzle D. An age-structured model of pre- and post-vaccination measles trans- mission. IMA J Math Appl Med Biol. 1984;1:169–191

1984
[10]

Infectious Diseases of Humans: Dynamics and Control

Anderson RM, May RM. Infectious Diseases of Humans: Dynamics and Control. Oxford: Oxford University Press; 1991

1991
[11]

Mathematical models in population biology and epidemiology

Brauer F, Castillo-Chavez C. Mathematical models in population biology and epidemiology. vol. 40 of Texts in Applied Mathematics. New York: Springer- Verlag; 2001

2001
[12]

Periodicity in epidemiological models

Hethcote H, Levin SA. Periodicity in epidemiological models. In: Levin SA, Hallam TG, Gross LJ, editors. Applied Mathematical Ecology. Biomathmatics
[13]

Chaos versus noisy periodicity: alternative hypotheses for childhood epidemics

Olsen LF, Schaﬀer WM. Chaos versus noisy periodicity: alternative hypotheses for childhood epidemics. Science. 1990;249:499–504

1990
[14]

The mathematics of infectious diseases

Hethcote HW. The mathematics of infectious diseases. SIAM Rev. 2000;42(4):599–653

2000
[15]

Dynamical resonance can ac- count for seasonality of inﬂuenza epidemics

Dushoﬀ J, Plotkin JB, Levin SA, Earn DJD. Dynamical resonance can ac- count for seasonality of inﬂuenza epidemics. Proc. Natl. Acad. Sci. USA. 2004;101(48):16915–16916

2004
[16]

Chaos and biological complexity in measles dynamics

Bolker BM, Grenfell BT. Chaos and biological complexity in measles dynamics. Proc. R. Soc. Lond. B. 1993;251:75–81

1993
[17]

Transients and attractors in epidemics

Bauch CT, Earn DJD. Transients and attractors in epidemics. Proc. R. Soc. Lond. B. 2003;270(1524):1573–1578

2003
[18]

Heidelberg: Springer-Verlag; 1989. p. 193–211

1989
[19]

Interepidemic intervals in forced and unforced SEIR models

Bauch CT, Earn DJD. Interepidemic intervals in forced and unforced SEIR models. In: Ruan S, Wolkowicz G, Wu J, editors. Dynamical Systems and Their Applications in Biology. vol. 36 of Fields Institute Communications. Toronto: American Mathematical Society; 2003. p. 33–44

2003
[20]

A Century of Transitions in New York City’s Measles Dynamics

Hempel K, Earn DJD. A Century of Transitions in New York City’s Measles Dynamics. J. R. Soc. Interface. 2015;12(106):20150024. 9

2015
[21]

Epidemic threshold conditions for seasonally forced SEIR models

Ma J, Ma Z. Epidemic threshold conditions for seasonally forced SEIR models. Math Biosci Eng. 2006;3(1):161–172

2006
[22]

Nonlinear Dynamics and Chaos: With Applications to Physics, Biology, Chemistry, and Engineering

Strogatz SH. Nonlinear Dynamics and Chaos: With Applications to Physics, Biology, Chemistry, and Engineering. CRC Press; 2018

2018
[23]

Elements of applied bifurcation theory

Kuznetsov YA. Elements of applied bifurcation theory. vol. 112 of Applied Math- ematical Sciences. 3rd ed. New York: Springer-Verlag; 2004

2004
[24]

Simulating, analyzing, and animating dynamical systems: a guide to XPPAUT for researchers and students

Ermentrout B. Simulating, analyzing, and animating dynamical systems: a guide to XPPAUT for researchers and students. Software, Environments, and Tools. Philadelphia: Society for Industrial and Applied Mathematics; 2002

2002
[25]

Inﬁnite subharmonic bifurcation in an SEIR model

Schwartz IB, Smith HL. Inﬁnite subharmonic bifurcation in an SEIR model. J Math Biol. 1983;18:233–253

1983
[26]

Melnikov analysis of chaos in a simple epidemiological model

Glendinning P, Perry LP. Melnikov analysis of chaos in a simple epidemiological model. J Math Biol. 1997;35(3):359–373

1997
[27]

Eﬀects of the infectious period distribution on predicted transitions in childhood disease dynamics

Krylova O, Earn DJD. Eﬀects of the infectious period distribution on predicted transitions in childhood disease dynamics. J. R. Soc. Interface. 2013;10:20130098

2013
[28]

Measles in England and Wales — I: An Analysis of Factors Underlying Seasonal Patterns

Fine PEM, Clarkson JA. Measles in England and Wales — I: An Analysis of Factors Underlying Seasonal Patterns. Int J Epidemiol. 1982;11(1):5–14

1982
[29]

Parameterizing state- space models for infectious disease dynamics by generalized proﬁling: measles in Ontario

Hooker G, Ellner SP, De Vargas Roditi L, Earn DJD. Parameterizing state- space models for infectious disease dynamics by generalized proﬁling: measles in Ontario. J. R. Soc. Interface. 2011;8(60):961–974

2011
[30]

Predicting epidemiological transitions in infectious disease dynamics: Smallpox in historic London (1664-1930) [PhD]

Krylova O. Predicting epidemiological transitions in infectious disease dynamics: Smallpox in historic London (1664-1930) [PhD]. McMaster University, Canada; 2011

1930
[31]

Fast estimation of time-varying transmission rates for infectious diseases [MSc]

deJonge M. Fast estimation of time-varying transmission rates for infectious diseases [MSc]. McMaster University, Canada; 2014

2014
[32]

Understanding Nonlinear Dynamics

Kaplan DT, Glass L. Understanding Nonlinear Dynamics. Textbooks in math- ematical sciences. New York: Springer-Verlag; 1995

1995
[33]

The Geometry of Biological Time

Winfree AT. The Geometry of Biological Time. vol. 12 of Interdisciplinary Ap- plied Mathematics. 2nd ed. New York: Springer-Verlag; 2001

2001
[34]

Multiple attractors, catastrophes and chaos in seasonally perturbed predator-prey communities

Rinaldi S, Muratori S, Kuznetsov Y. Multiple attractors, catastrophes and chaos in seasonally perturbed predator-prey communities. Bull Math Biol. 1993;55(1):15–35. 10

1993
[35]

A universal bifurcation diagram for seasonally perturbed predator-prey models

Gragnani A, Rinaldi S. A universal bifurcation diagram for seasonally perturbed predator-prey models. Bull Math Biol. 1995;57(5):701–712

1995
[36]

Seasonal dynamics of Daphnia and algae explained as a periodically forced predator-prey system

Scheﬀer M, Rinaldi S, Kuznetsov YA, van Nes EH. Seasonal dynamics of Daphnia and algae explained as a periodically forced predator-prey system. Oikos. 1997;p. 519–532

1997
[37]

Bifurcations and chaos in a periodic predator-prey model

Kuznetsov YA, Muratori S, Rinaldi S. Bifurcations and chaos in a periodic predator-prey model. Int J Bifurc Chaos. 1992;2(01):117–128

1992
[38]

birth folds

Doedel EJ, Oldeman BE. AUTO-07P: Continuation and bifurcation software for ordinary diﬀerential equations. Montreal, Canada: Concordia University; 2011. 11 Table 1: Invariance of fold bifurcations [22] at diﬀerent R0 values when the principal period doubling (PD) bifurcation a...

2011

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [1]

Absolute humidity modulates inﬂuenza survival, transmis- sion, and seasonality

Shaman J, Kohn M. Absolute humidity modulates inﬂuenza survival, transmis- sion, and seasonality. Proc. Natl. Acad. Sci. USA. 2009;106(9):3243–3248

work page 2009

[2] [2]

Patterns of spread of inﬂuenza A in Canada

He D, Dushoﬀ J, Eftimie R, Earn DJD. Patterns of spread of inﬂuenza A in Canada. Proc. R. Soc. Lond. B. 2013;280(1770):20131174

work page 2013

[3] [3]

Epidemiological eﬀects of seasonal oscillations in birth rates

He D, Earn DJD. Epidemiological eﬀects of seasonal oscillations in birth rates. Theor Popul Biol. 2007;72:274–291

work page 2007

[4] [4]

Human birth seasonality: latitudinal gradient and interplay with childhood disease dynamics

Martinez-Bakker M, Bakker KM, King AA, Rohani P. Human birth seasonality: latitudinal gradient and interplay with childhood disease dynamics. Proc. R. Soc. Lond. B. 2014;281(1783)

work page 2014

[5] [5]

Recurrent outbreaks of measles, chickenpox and mumps

London W, Yorke JA. Recurrent outbreaks of measles, chickenpox and mumps. I. Seasonal variation in contact rates. Am J Epidemiol. 1973;98(6):453–468. 8

work page 1973

[6] [6]

A simple model for complex dynamical transitions in epidemics

Earn DJD, Rohani P, Bolker BM, Grenfell BT. A simple model for complex dynamical transitions in epidemics. Science. 2000;287(5453):667–670

work page 2000

[7] [7]

Seasonal dynamics of recurrent epidemics

Stone L, Olinky R, Huppert A. Seasonal dynamics of recurrent epidemics. Na- ture. 2007;446(7135):533–536

work page 2007

[8] [8]

Eﬀects of school closure on incidence of pandemic inﬂuenza in Alberta, Canada

Earn DJD, He D, Loeb MB, Fonseca K, Lee BE, Dushoﬀ J. Eﬀects of school closure on incidence of pandemic inﬂuenza in Alberta, Canada. Ann Intern Med. 2012;156(3):173–181

work page 2012