Pith. sign in

REVIEW 3 major objections 4 minor 49 references

A numerical study of a PDE-ODE system with a stochastic dynamical boundary condition: a nonlinear model for sulphation phenomena

T0 review · 3 major / 4 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read The paper constructs a fully discrete, positivity-preserving scheme for the stochastic sulphation model and proves pathwise boundedness under explicit mesh and initial-data conditions.

desk verdict Useful applied numerical study with a checkable stability result, but the main theorem's hypotheses are not yet connected to the implemented scheme; worth refereeing. read the letter →

arxiv 2412.16307 v1 pith:JT6X26NN submitted 2024-12-20 math.NA cs.NAmath.PR

Francesca Arceci , Daniela Morale , Stefania Ugolini This is my paper · ORCID
classification math.NAcs.NAmath.PR MSC 65M0665C3035K5760H1565M12
keywords sulphationstochasticdynamicalboundaryconditionPearsondiffusionLampertitransformFTCSfinitedifferenceschemepositivitypreservationmovingfrontculturalheritagedegradation
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 sets out to show that the randomly forced sulphation system—sulphur dioxide eating into calcium carbonate with a stochastic boundary concentration—can be simulated by a fully discrete scheme that respects the physics of the model. The construction splits the solution into a heat component that carries the random boundary and a nonlinear component with a deterministic boundary, then samples the boundary with a Lamperti-transformed Pearson diffusion and advances both parts with explicit finite differences. The authors prove pathwise positivity, boundedness, and stability under explicit conditions on the spatial grid, the time step, and the initial calcite density. If these claims hold, the scheme is a working laboratory for qualitative questions: how boundary noise spreads into the material, how porosity changes penetration, and how fast reactions sharpen the degradation front. The wider value is a template for random-boundary reaction-diffusion problems where concentrations must never leave their physical range.

What carries the argument

The load-bearing mechanism is the splitting $s=u+v$, in which $u$ solves the heat equation with the stochastic boundary condition and $v$ solves a nonlinear, nonlocal equation with zero boundary data; diffusion smooths the irregular boundary noise before it reaches the reaction part. Boundary sampling is carried by a Lamperti transform $Y=2\arcsin(\sqrt{\Psi/\eta})$ that converts the non-Lipschitz Pearson diffusion into additive noise, followed by a sloping smooth truncation of the drift so that the explicit Euler\,--\,Maruyama update is stable and monotone. The PDE discretization is a forward-time, centred-space (FTCS) scheme, and the proof of Proposition 4.3 runs on coefficient positivity: under (47), (56), and (57), every coefficient multiplying the old values in the $s$-update is nonnegative, so the interval $[0,\tilde{\eta})\times[0,\bar c_0]$ is invariant pathwise.

What would settle it

Run the implemented scheme with boundary data built from a concrete quadrature for (41), starting with $\Psi_0$ close to $\eta$ and a coarse $\Delta t$; if any computed $\tilde{\psi}^n$ leaves $[0,\tilde{\eta})$ or any $s$-value turns negative, Proposition 4.3's conclusion fails for that quadrature and the proof's hypothesis is violated.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that the splitting $s=u+v$ used for the well-posedness of the continuous problem can be transplanted to a numerical scheme with a rigorous positivity statement. The discrete heat component $u$ inherits the stochastic boundary values $\tilde{\psi}^n$, produced by the Lamperti\,--\,sloping-smooth-truncation sampler, while the nonlinear component $v$ and the calcite density $c$ are advanced by an explicit scheme. Proposition 4.3 proves by induction that every pathwise solution of the $s$-update (54) and $c$-update (55) stays in $[0,\tilde{\eta})\times[0,\bar c_0]$ whenever $\bar\Delta\le 1/2$, the initial data satisfy (56), and the time step satisfies (57); the induction rests on checking that each coefficient of the update is nonnegative under these conditions. Corollary 4.4 then bounds $v$ between $-\tilde{\eta}$ and $\tilde{\eta}$, and Proposition 4.5 converts those bounds into $L^2$ and maximum-norm stability, pathwise and in mean. The numerical sections use the scheme to describe slow and fast reaction regimes, including the formation of a moving front with variance concentrated around it.

Load-bearing premise

The load-bearing premise is that the discrete boundary values fed into the scheme stay inside the allowed interval as the proof assumes, and the paper does not verify this for the quadrature it uses to compute them; a second fragile point is that the well-posedness theorem silently assumes the boundary starts at zero, which is not true for general positive initial data.

Editorial extensions

If this is right

  • If the stability bounds hold, the scheme offers a practical recipe for choosing time steps when simulating the stochastic sulphation model: condition (57) ties $\Delta t$ to the spatial mesh, the reaction rate $\lambda$, and the initial calcite bound.
  • The simulations imply that boundary noise does not stay at the boundary: for slow reactions the variance of both $\rho$ and $c$ spreads gradually over the whole domain, while for fast reactions it concentrates in a thin region around the moving front.
  • In the fast-reaction regime the discrete solutions display the same qualitative front formation known from the deterministic fast-reaction limit, now with stochastic boundary data; this supports using the scheme to study black-crust formation under random pollution histories.
  • The numerically estimated spatial accuracy order is about one for both sulphur dioxide and calcite, which is consistent with the explicit first-order finite-difference structure and gives an error scale for the reported statistics.

Reading between the lines

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

  • Editorial extension: no quadrature rule for the integral in (41) is specified, so the implemented boundary sequence could leave the interval $[0,\tilde{\eta})$; proving a discrete analogue of Proposition 3.1 for that quadrature is the missing step that would make Proposition 4.3 airtight.
  • Editorial extension: the same Lamperti\,--\,LSST-plus-FTCS construction should transfer to other bounded Pearson boundary noises, such as Wright\,--\,Fisher processes with $\eta=1$, where positivity of concentrations or frequencies is a hard modelling constraint.
  • Editorial extension: the concentration of sample variance near the moving front suggests a stochastic front-speed law; one could test whether the front position scales like $\sqrt{\lambda t}$ with noise-induced corrections by estimating it from many realizations.
  • Editorial extension: if the scheme is correct, it gives a direct way to quantify uncertainty in cultural heritage risk assessments, since it returns full space-time distributions of pollutant concentration and material loss for given environmental noise parameters.
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.

Referee Report

3 major / 4 minor

Summary. This paper presents a numerical study of a half-line reaction-diffusion/ODE system modelling marble sulphation, with a stochastic Dirichlet boundary condition given by a Pearson diffusion. The authors recall well-posedness and boundedness of the boundary SDE, simulate it through a Lamperti transform combined with a sloping smooth truncation (LSST), and then discretize the PDE by a splitting strategy: an FTCS heat equation carrying the stochastic boundary data, coupled with an explicit finite-difference scheme for the nonlinear reaction part with deterministic boundary condition. The main theoretical contributions are the pathwise stability of the heat component (Proposition 4.2) and the positivity/boundedness result for the full discrete system (Proposition 4.3) under explicit conditions (47), (56), and (57). The paper closes with extensive numerical experiments on noise propagation, porosity effects, slow and fast reaction regimes, moving fronts, and statistical summaries over 500 sample paths.

Significance. If the stated results are fully connected to the implemented algorithm, the paper provides a useful, positivity-preserving and stable numerical method for a stochastic boundary-value problem that is relatively unexplored numerically. The strengths are the explicit and checkable coefficient conditions in Proposition 4.3, the pathwise character of the stability analysis, the use of a Lamperti-type transform to preserve the boundary domain, and the careful qualitative experiments that identify the moving-front regime. The honest deferral of full convergence analysis to the companion paper [5] is a clear limitation but is not by itself disqualifying for a numerical study. However, two load-bearing gaps currently separate the theorems from the code: the inverse Lamperti transform in the displayed schemes misses the factor eta, and the boundary sequence used in the PDE scheme is defined through an unspecified continuous-time integral with no proof that its discretization satisfies the hypothesis of Proposition 4.3.

major comments (3)
  1. [Section 2.3, Eqs. (25)-(26)] The inverse Lamperti transform is correctly stated in Eq. (17) as Psi = eta sin^2(Y/2), but the LSST sampling scheme (25) and the time-continuous version (26) define psi^n = sin^2(y^n/2) and psi_t = sin^2(y_t/2) without the factor eta. Since all later bounds and experiments use eta = 1.5, the boundary process generated by the scheme as written lies in [0,1) rather than in [0,eta), which would either scale the boundary data by 1/eta relative to the model or indicate a typo in the displayed equations. This factor must be restored, because the boundary data enter every PDE simulation and the bound (36) depends on eta.
  2. [Section 4.1, Eq. (41) and Proposition 4.3] The discrete boundary sequence feeding the scheme is defined in (41) through the continuous-time integral exp(-lambda integral_0^{t_n} Psi_u du), but no quadrature rule for that integral is specified, and no proof is given that the computed values tilde-psi^n remain in [0,tilde-eta). Proposition 4.3's proof assumes exactly that interval property for every n, and its statement covers only m = 1,...,M, although the boundary value m = 0 is needed in the heat scheme (42). The gap is easy to close: with a nonnegative quadrature of a nonnegative integrand, exp(-lambda Q) in (0,1], so phi(c0 exp(-lambda Q)) >= phi(c0) and tilde-psi^n <= eta/phi(c0) = tilde-eta. This argument should be stated and the quadrature chosen explicitly so that the theorem formally covers the implemented algorithm.
  3. [Theorem 3.2] The statement of Theorem 3.2 assumes tilde-Psi(0)=0 as part of the hypotheses, but this is not a consequence of condition (10) and is inconsistent with (35) when Psi_0>0. In that case the splitting u(0,x)=0, v(t,0)=0 creates a boundary discontinuity unless s_0=0. The theorem should either state the compatibility condition explicitly and restrict the admissible data, or the splitting should be modified to absorb a nonzero initial boundary value. The numerical experiments appear to use Psi_0=0, so this may be a presentation issue in practice, but the theorem as written is formally incomplete for general data.
minor comments (4)
  1. [Section 4.1, Eq. (45)] The summation in the variation-of-constants formula appears off by one: for the recurrence U^{n+1}=A U^n + bar-Delta tilde-U_0^n, the solution should sum j=0,...,n-1 with A^{n-1-j}, not j=0,...,n with A^{n-j}. As written, U^n depends on tilde-U^n, which is not yet available at time t_n.
  2. [Figures 2 and 9] Some captions and inline references use inconsistent parameter names: Figure 2 refers to sigma2 = 1 where Table 3 and the surrounding text use sigma3 = 1, and Figure 9(a) is labelled lambda1 = 0.1 while Table 3 sets lambda1 = 1. These should be harmonized.
  3. [Section 5.2] The text says the accelerated regime considers lambda_i, i=2,3, but the discussion also refers to lambda1; please clarify whether the slow case is lambda1=1 or the lambda1=0.1 shown in Figure 9(a).
  4. [Section 4.2.2] The sentence introducing the numerical spatial accuracy estimate should state more explicitly that only the spatial error is measured for a fixed time step, and that the full convergence analysis is deferred to [5]; this would avoid the impression that the scheme's complete error is being estimated.

Circularity Check

0 steps flagged · score 1.0 of 10

No circular reduction is exhibited: Proposition 4.3 proves boundedness directly from the scheme coefficients under stated CFL and initial-data conditions, LSST convergence is imported from an external source, and the reader-identified gaps are rigor gaps rather than circular derivations.

full rationale

The central numerical claim, Proposition 4.3, is a discrete maximum-principle argument for (54)-(55): under (47), (56), (57) and the boundary bound ~ψ^n∈[0,~η), it proves s_m^{n+1}∈[0,~η) and c_m^{n+1}∈[0,¯c_0] by checking the signs of the coefficients of the explicit scheme. This conclusion is not a fit and does not assume the qualitative results it is later used to support. The boundary sequence is an input of the theorem; its bound follows from ψ^n=η sin²(y_n/2)∈[0,η] and monotonicity of φ, so the unspecified quadrature in (41) is a completeness gap, not a circular reduction. LSST strong-convergence statements are cited from the independent external source [15] (Chen and Gan), and the error tables compare coarser LSST solutions against a finer-step LSST solution, an explicitly acknowledged numerical benchmark rather than a fitted parameter renamed as a prediction. The self-citations [3,35] supply the continuous model and the u/v splitting ansatz, but the discrete stability proof is self-contained and does not reduce to them. The flagged issues, such as Theorem 3.2 silently assuming ~Ψ(0)=0 and the omitted convergence-rate analysis, are correctness or presentation concerns and do not make any derivation equivalent to its inputs by construction.

Assumptions & free parameters 5 free parameters · 6 assumptions · 0 invented entities

The central results rest on standard SDE theory, the external LSST convergence theorem [15], and the authors' own well-posedness theorem [35]. The numerical stability analysis is derived from the discrete coefficients, so the ledger is dominated by domain assumptions rather than fitted parameters.

free parameters (5)
  • Pearson boundary parameters alpha, gamma, eta = alpha=7, gamma=1, eta=1.5
    Chosen from the preliminary Milano SO2 study in [3]; not fitted in this paper, but they determine the strength and range of the boundary noise in all simulations.
  • Noise intensities sigma = 0, 0.25, 0.7, 1
    Hand-picked to illustrate small and large boundary noise; the qualitative conclusions are specific to this range.
  • Reaction rates lambda = 1, 10, 100
    Hand-picked to represent slow, intermediate, and fast regimes; the moving-front result depends on these values.
  • Material parameters phi1, phi2, c0 = phi1=0.2, phi2=-0.01, c0=10
    Chosen by hand; the positivity condition (56) explicitly depends on them.
  • LSST truncation exponent k = k=0.22
    A numerical parameter in the slope-smooth truncation (23); the convergence rates imported from [15] depend on k.
assumptions (6)
  • standard math Existence and pathwise uniqueness of the Pearson SDE (5) on [0,eta] under condition (10), via Skorokhod and Ikeda-Watanabe theorems.
    Invoked in Proposition 2.1 to justify the boundary process Psi used throughout.
  • standard math Boundary classification of 0 and eta as entrance boundaries, with the inverse-moment bound (11) taken from [44].
    Used in Proposition 2.4 to ensure Psi stays in (0,eta) and to define the Lamperti transform (16).
  • domain assumption Strong convergence of the LSST scheme with the rates stated in Proposition 2.8, transferred from the Wright-Fisher setting of [15].
    The boundary sampling scheme (25) relies on this external convergence result; the transfer to the Pearson parameters (19) is assumed.
  • domain assumption Well-posedness of the PDE-ODE system (32)-(34) with the stochastic boundary condition, Theorem 3.2 from [35].
    The splitting s=u+v and the whole numerical strategy inherit this unproved-in-this-paper existence and uniqueness result.
  • domain assumption The sulphation reaction is described by system (1) with porosity linear in calcite, phi(c)=phi1+phi2 c.
    This is the physical model from [6]; the numerical study does not validate it against experiments.
  • ad hoc to paper Theorem 3.2 requires ~Psi(0)=0, restricting the theory to zero initial boundary data.
    Eq. (35) gives ~Psi(0)=Psi_0/phi(c0) unless Psi_0=0; the statement silently assumes the zero case, which is unflagged and limits the generality of the existence result used.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A numerical study of a PDE-ODE system with a stochastic dynamical boundary condition: a nonlinear model for sulphation phenomena." pith.science (2026). https://pith.science/paper/JT6X26NN

@misc{pith2026241216307,
  author       = {Pith},
  title        = {Pith review of: A numerical study of a PDE-ODE system with a stochastic dynamical boundary condition: a nonlinear model for sulphation phenomena},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JT6X26NN}},
  note         = {Machine review of arXiv:2412.16307}
}
read the original abstract

We investigate the qualitative behaviour of the solutions of a stochastic boundary value problem on the half-line for a nonlinear system of parabolic reaction-diffusion equations, from a numerical point of view. The model describes the chemical aggression of calcium carbonate stones under the attack of sulphur dioxide. The dynamical boundary condition is given by a Pearson diffusion, which is original in the context of the degradation of cultural heritage. We first discuss a scheme based on the Lamperti transformation for the stochastic differential equation to preserve the boundary and a splitting strategy for the partial differential equation based on recent theoretical results. Positiveness, boundedness, and stability are stated. The impact of boundary noise on the solution and its qualitative behaviour both in the slow and fast regimes is discussed in several numerical experiments.

Figures

Figures reproduced from arXiv: 2412.16307 by the authors.

Figure 1
Figure 1. shows the effect of the smoothing truncation on the drift f given by (21) for a specific set of parameters and different diffusion coefficients. - 0 2 -100 -50 0 50 100 150 - 0 2 -100 -50 0 50 100 150 (a) (b) [PITH_FULL_IMAGE:figures/full_fig_p010_1.png] view at source ↗
Figure 2
Figure 2. Log-log of the estimations of the L 2 -errors at time T (∆, eˆ2,T,∆)∆ (solid line-first raw), and the uniform one (∆, ǫˆ2,∆)∆ (solid line-second row). Parameters are: α = 7, γ = 1, η = 1.5 and different diffusion coefficients: (a)-(c) σ1 = 0.25; (b)-(d) σ3 = 1. Dashed black line: slope reference 1; dashed blue line: slope reference 0.8; dashed green line: slop reference 0.6. σ1 = 0.25 σ3 = 1 ∆/δ eˆ2,T,∆ eˆ2,T,∆ 2 8 … view at source ↗
Figure 3
Figure 3. Log-log representation of the estimations of the [PITH_FULL_IMAGE:figures/full_fig_p014_3.png] view at source ↗
Figures from the paper (9 more)
Figure 4
Figure 4. Figure 4: Two random trajectories (filled line) kept within t [PITH_FULL_IMAGE:figures/full_fig_p014_4.png]
Figure 5
Figure 5. Figure 5: Random heat solution in the case ϕ1 = 0.2, for different values of σ. Top-left: σ0 = 0; top-right: σ1 = 0.25; bottom-left : σ2 = 0.7; bottom-right : σ3 = 1 [PITH_FULL_IMAGE:figures/full_fig_p025_5.png]
Figure 6
Figure 6. Figure 6: Pathwise different contributions to the complete PD [PITH_FULL_IMAGE:figures/full_fig_p025_6.png]
Figure 7
Figure 7. Figure 7: The role of the diffusion σ. A single path of the SO2 concentration ρ = ϕs for different levels of boundary noise. Top-left: σ0 = 0; top-right: σ1 = 0.25; bottom-left: σ2 = 0.7; bottom￾right: σ3 = 1. Gypsum porosity may reach a level of 70%. The higher the porosity, th…
Figure 8
Figure 8. Figure 8: The role of the gypsum porosity via ϕ1. Evolution of a realization of SO2 (first row) and the corresponding marble concentration (second row) for ϕ1 = 0.02 (left) and a higher porosity ϕe1 = 0.7 (right). Other parameters: λ = 1, σ2 = 0.7. dioxide and the consequent deg…
Figure 9
Figure 9. Figure 9: The role of the activation energy λ. Evolution of a single realization of SO2 con￾centration (first column) and the corresponding marble density (second column) for increasing reaction rate: (a) λ1 = 0.1; (b) λ2 = 10; (c) λ3 = 100. Other parameters: ϕ1 = 0.2, ϕ2 = −0.0…
Figure 10
Figure 10. Figure 10: Estimation of the distribution : 25% (green-blue [PITH_FULL_IMAGE:figures/full_fig_p029_10.png]
Figure 11
Figure 11. Figure 11: Estimation of the distribution: standard deviat [PITH_FULL_IMAGE:figures/full_fig_p030_11.png]
Figure 12
Figure 12. Figure 12: Estimation of the impact of the noise at the bounda [PITH_FULL_IMAGE:figures/full_fig_p031_12.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

49 extracted references · 48 canonical work pages

  1. [35]

    Maurelli, D.Morale, S.Ugolini, Well-posedness of a reaction-diffusion model with stochas- tic dynamical boundary conditions arXiv preprint, arXiv:2 308.06847 (2023)

    M. Maurelli, D.Morale, S.Ugolini, Well-posedness of a reaction-diffusion model with stochas- tic dynamical boundary conditions arXiv preprint, arXiv:2 308.06847 (2023)

    work page 2023

  2. [15]

    L. Chen, S. Gan, Strong convergence and stationary dist ribution of an explicit scheme for the Wright–Fisher model, Journal of Computational and Appl ied Mathematics, Vol. 424, (2023)

    work page 2023

  3. [5]

    Arceci, F

    F. Arceci, F. De Vecchi, D.Morale, and S.Ugolini, Semi-d iscretization of a reaction-diffusion system with stochastic dynamical boundary conditions: con vergence results. arXiv preprint, (2024-to appear)

    work page 2024

  4. [1]

    Al ` ı, V

    G. Al ` ı, V. Furuholt, R. Natalini, and I. Torcicollo, A ma thematical model of sulphite chemical aggression of limestones with high permeability. Part I. Modeling and qualitative analysis, Transport in Porous Media, Volume 69, Number 1 (20 07), 109-122

    work page

  5. [2]

    Al ` ı, V

    G. Al ` ı, V. Furuholt, R. Natalini, and I. Torcicollo, A ma thematical model of sulphite chemical aggression of limestones with high permeability. Part II. Numerical approximation, Transport in Porous Media, Volume 69, Number 1 (2007), 175-1 88

    work page 2007

  6. [3]

    Arceci, L.M

    F. Arceci, L.M. Giordano, M. Maurelli, D.Morale, and S.U golini, Randomness in a non- linear model of sulphation phenomena, in Mathematcal Model ing in Cultural Heritage, MACH2021, (G.Bret, C.Cavaterra, M.Solci, M.Spagnuolo Eds ), Springer Nature, 2023

    work page 2023

  7. [4]

    A react ion diffusion model with a stochas- tic boundary condition

    F. Arceci, M. Maurelli, D.Morale, and S.Ugolini “A react ion diffusion model with a stochas- tic boundary condition”. In: MACH2023. Mathematcal Modeli ng in Cultural Heritage, Springer Nature. To appear.(2024)

    work page 2024

  8. [6]

    Aregbda-Driollet, F

    D. Aregbda-Driollet, F. Diele, and R. Natalini A mathema tical model for the sulphur dioxide aggression to calcium carbonate stones: numerical approxi mation and asymptotic analysis, SIAM J. APPL. MATH. (2004) Vol. 64, No. 5, pp. 1936-1667

    work page 2004

Show all 49 references
  1. [7]

    Barbu, S

    V. Barbu, S. Bonaccorsi, L. Tubaro, Stochastic paraboli c equations with nonlinear dynam- ical boundary conditions, J. Math. Anal. Appl. (2015) 427, 4 84-–498

    2015

  2. [8]

    Bonazza, P

    A. Bonazza, P. Messina, C. Sabbioni, C. M. Grossib, P. Bri mblecombe, Mapping the impact of climate change on surface recession of carbonate buildin gs in Europe, Science od the Total Encironment, 407, (2009), 2039-2050

    2009

  3. [9]

    Bonetti, C

    E. Bonetti, C. Cavaterra, F. Freddi, M. Grasselli, R. Nat alini, A nonlinear model for marble sulphation including surface rugosity: theoretical and nu merical results, Communications on Pure & Applied Analysis,18(2): 977-998,(2019), 32

    2019

  4. [10]

    Bonetti, C

    E. Bonetti, C. Cavaterra, F. Freddi, M. Grasselli, R. Na talini, A nonlinear model for marble sulphation including surface rugosity and mechanical dama ge, Nonlinear Analysis: Real World Applications, 73,(2023)

    2023

  5. [11]

    Bonaccorsi, G

    S. Bonaccorsi, G. Ziglio, A semigroup approach to stoch astic dynamical boundary value problems. In: Systems, control, modeling and optimization , 55–65. IFIP Int. Fed. Inf. Process., 202, Springer, New York, 2006

    2006

  6. [12]

    Brze´ zniak, B

    Z. Brze´ zniak, B. Goldys, S. Peszat and F. Russo, Second order PDEs with Dirichlet white noise boundary conditions, J. Evol. Equ. 15 (2015), 1–26

    2015

  7. [13]

    Capasso, D

    V. Capasso, D. Bakstein, An Introduction to Continuous -Time Stochastic Processes: The- ory, Models, and Applications to Finance, Biology, and Medi cine. Springer (2021)

    2021

  8. [14]

    L. Chen, S. Gan, X. Wang, First order strong convergence of an explicit scheme for the stochastic SIS epidemic model, Journal of Computational an d Applied Mathematics, Vol- ume 392, 2021, 113482

    2021

  9. [16]

    Chueshov and B

    I. Chueshov and B. Schmalfuss, Parabolic stochastic pa rtial differential equations with dynamical boundary conditions, Differential Integral Equat ions 17 (2004), 751–780

    2004

  10. [17]

    S., CardellL, C., Rivas, T ., Randazzo, L., La Russa, M

    Comite, V., Pozo-Antonio, J. S., CardellL, C., Rivas, T ., Randazzo, L., La Russa, M. F., & Fermo, P. Environmental I International Journal of Conser vation Science, (2020), 11

    2020

  11. [18]

    Comite, A

    V. Comite, A. Bergomi, M. Formenti, C.A. Lombardi, M. Bo relli, D. Morale, S. Ugolini, P. Fermo, C. Cavaterra, C. Castellano, C. Della Pina, Experi mental study on the catalytic action of heavy metals in the sulphation process and formati on of black crusts by accelerated agei...

  12. [19]

    G. Da Prato and J, Zabczyk, Evolution equations with whi tenoise boundary conditions, Stochastics: An International Journal of Probability and S tochastic Processes 42.3-4 (1993), 167–182

    1993

  13. [20]

    Debussche, M

    A. Debussche, M. Fuhrman, G. Tessitore , Optimal contro l of a stochastic heat equation with boundary-noise and boundary-control ESAIM Control Op timisation and Calculus of Variations 13(1) (2007), pp. 178–205

    2007

  14. [21]

    Delbaen and H

    F. Delbaen and H. Shirakawa. An interest rate model with upper and lower bounds. Asia- Pacific Financial Markets, 9(3):191–209, (2002)

    2002

  15. [22]

    D’Onofrio, P

    G. D’Onofrio, P. Patie, and L. Sacerdote, Jacobi proces ses with jumps as neuronal models : A first passage time analysis. SIAM Journal on Applied Mathe matics. (2023)

    2023

  16. [23]

    D’Onofrio (Ed.), Bounded Noises in Physics, Biology , and Engineering

    A. D’Onofrio (Ed.), Bounded Noises in Physics, Biology , and Engineering. Birkh¨ auser (2013)

    2013

  17. [24]

    Domingo, A

    D. Domingo, A. D’Onofrio, F. Flandoli, Boundedness vs u nboundedness of a noise linked to Tsallis q-statistics. The role of the overdamped approxi mation, J. Mathematical Physics 58(3), (2017) 33

    2017

  18. [25]

    R. C. Griffiths and D. Span´ o. Diffusion processes and coale scent trees. In Probability and mathematical genetics, volume 378 of London Math. Soc. Lect ure Note Ser., pages 358–379. Cambridge Univ. Press, Cambridge, 2010

    2010

  19. [26]

    R. C. Griffiths, P. A. Jenkins, and D. Span` o. Wright–Fish er diffusion bridges. Theoretical Population Biology, 122:67 – 77, 2018. Paul Joyce

    2018

  20. [27]

    Giordano, D

    L.M. Giordano, D. Morale, A Fractional Brownian–Hawke s Model for the Italian Electricity Spot Market: Estimation and Forecasting, Journal of Energy Markets, Vol. 14, No. 3, 2021

    2021

  21. [28]

    F. R. Guarguaglini, R. Natalini, Global existence of so lutions to a nonlinear model of sulpha- tion phenomena in calcium carbonate stones,Nonlinear Anal ysis: RealWorld Applications 6 (2005) 477 – 494

    2005

  22. [29]

    F. R. Guarguaglini, R. Natalini. Global existence and u niqueness of solutions for multidi- mensional weakly parabolic systems arising in chemistry an d biology. Communications on Pure & Applied Analysis, 2007, 6 (1) : 287-309. doi: 10.3934/ cpaa.2007.6.287

    2007

  23. [30]

    F. R. Guarguaglini, R. Natalini, Fast Reaction Limit an d Large Time Behavior of Solutions to a Nonlinear Model of Sulphation Phenomena. Communicatio ns in Partial Differential Equations (2007), 32:2, 163-189

    2007

  24. [31]

    Hutzenthaler, A

    M. Hutzenthaler, A. Jentzen, and P, E. Kloeden, Strong C onvergence of an explicit nu- merical method for SDE with non globally Lipschitz continuo us coefficients. The Annals of Applied Probability 22, no. 4 (2012): 1611–41

    2012

  25. [32]

    Ikeda and S

    N. Ikeda and S. Watanabe, Stochastic Differential Equati ons and Diffusion Processes, North-Holland mathematical library,North-Holland Publi shing Company (1981)

    1981

  26. [33]

    P. E. Kloeden and E. Platen, Numerical solutions of stoc hastic differential equations, Ap- plications of Mathematics, 23, Springer (1992)

    1992

  27. [34]

    R. Lord, R. Koekkoek, D.V. Dijk, A comparison of biased s imulation schemes for stochastic volatility models, Quant. Finance 10(2) (2010), 177-194

    2010

  28. [36]

    A stochastic interac ting particle model for the marble sulphation process

    D. Morale, G. Rui, and S. Ugolini. “A stochastic interac ting particle model for the marble sulphation process”. In: MACH2023. Mathematcal Modeling i n Cultural Heritage, Springer Nature. To appear.(2024)

    2024

  29. [37]

    Morale, L

    D. Morale, L. Tarquini, and S. Ugolini. A probabilistic interpretation of a non-conservative and path-dependent nonlinear reaction-diffusion system for the marble sulphation in Cul- tural Heritage. arXiv: 2407.19301 [math.PR] (2024)

    2024 arXiv

  30. [38]

    S. Pal. Wright–Fisher diffusion with negative mutation r ates. Ann. Probab., 41(2):503–526, 2013 (2013)

    2013

  31. [39]

    Saba, E.E

    M. Saba, E.E. Qui˜ nones-Bola˜ nosa and A. L. Barbosa L´ opez, A review of the mathemat- ical models used for simulation of calcareous stone deterio ration in historical buildings, Atmospheric Environmen, 180, 156-166, 2018

    2018

  32. [40]

    Skorohod, Studies in the theory of random processe s, Addison-Wisley 1965 (Originally published in Kiev)

    A.V. Skorohod, Studies in the theory of random processe s, Addison-Wisley 1965 (Originally published in Kiev). 34

    1965

  33. [41]

    and Taylor H.M., A Second Course in Stochastic Processes, (1981) Academic Press

    Karlin S. and Taylor H.M., A Second Course in Stochastic Processes, (1981) Academic Press

    1981

  34. [42]

    Dangerfield, D

    C.E. Dangerfield, D. Kay, S. MacNamara, K. Burrage, A bou ndary preserving numerical algorithm for the Wright-Fisher model with mutation, BIT Nu mer. Math. 2012, 52, 283–304

    2012

  35. [43]

    Ghayebi, S.M

    B. Ghayebi, S.M. Hosseini and Bl¨ omker, Dirk, Numerica l Solution of the Burgers Equation With Neumann Boundary Noise, Journal of Computational and A pplied Mathematics,311 (2016)

    2016

  36. [44]

    Sorensen, J.L

    M. Sorensen, J.L. Forman, The Pearson diffusions: A class of statistically tractable diffusion processes[J], Scand. J. Stat. 35(3) (2008) 438-465

    2008

  37. [45]

    Revuz and M

    D. Revuz and M. Yor. Continuous martingales and Brownia n motion, 293, Springer (1999)

    1999

  38. [46]

    L. C. G. Rogers and D. Williams. Diffusions, Markov Proces ses, and Martingales: Volume 2, It ¨ 0 Calculus. Cambridge university press (2000)

    2000

  39. [47]

    Strikwerda, Finite Difference Schemes and Partial D ifferential Equations, Second Edi- tion, SIMA, 2004

    J.C. Strikwerda, Finite Difference Schemes and Partial D ifferential Equations, Second Edi- tion, SIMA, 2004

    2004

  40. [48]

    and G¨ unther, M

    Teng, L., Ehrhardt, M. and G¨ unther, M. Modelling stoch astic correlation. J.Math.Industry 6, 2 (2016)

    2016

  41. [49]

    Thomas, Numerical Partial Differential Equations: Finite Difference Methods, Springer New York, NY (1995) 35

    J.W. Thomas, Numerical Partial Differential Equations: Finite Difference Methods, Springer New York, NY (1995) 35

    1995

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.