Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

# Theoretical and semi-analytical simulation for a two-predator-one-prey model during the mating period

• Noufe H. Aljahdaly

Roles Conceptualization, Formal analysis, Investigation, Methodology, Software, Validation, Writing – original draft, Writing – review & editing

nhaljahdaly@kau.edu.sa

Affiliation Mathematics Department, Faculty of Sciences and Arts, King Abdulaziz University, Rabigh, Saudi Arabia

## Abstract

The article introduces a new application which is a system of equations of two predators and one prey with the term of interaction between male and female of predators and prey. Such term appears when male and female of predators feed on the same prey during their mating period. The mathematical model has been studied theoretically and semi-analytically. The positivity, boundedness, local and global stability are proved for the system. The logarithm of multistage differential transform method (MsDTM) is used to study this new application. The MsDTM is used because it globally converges to the solution, it is a highly accurate, fast and simple approach. The stability analysis as well as semi-analytical solutions of the system are obtained to understand the dynamic of the model. Moreover, the effects of several parameters in the system are presented. As a results, we obtain the periodic solution when when the growth rate of prey is larger than the growth rate of both type of predators.

## 1 Introduction

Mathematical models phrased in terms of differential equations are very useful tools to describe several problems and applications in science such as biology , physics , fluid dynamic , medical science [4, 5], etc. They are helpful to control the spread of infection, to understand the behavior of population, to observe the effect of several factors in population or to predict the solutions for different problems in the world. Some populations are described by a system of ordinary differential equations (ODEs) that introduces the interaction of different species .

The predator-prey model is a well known mathematical model and it was first introduced by Lotka  and Volterra . It has been used widely to study the dynamic of interaction of two species in population ecology and associated ecological interactions . In addition, researchers modified the model to study different biological and ecological scenarios. For example, they study the model among vertebrates and invertebrates in freshwater systems associated with fisheries  and with prey refuge and delay . The modified models have been utilized for studying the impulsive effects , the impact of prey herd dread from the two types of predators , effect of food sharing among predators , the effect of two different food chains , studying the stationary probability distributions of its population densities , studying with disease in prey  and so on.

One of the behaviors of some type of animals such as the discus fish is that the male stays with the female during mating period to protect her. During this period both male and female feed together in same prey. The model that describes the interaction between one predator and two preys has been studied in [18, 19], while the model of two predators and one prey has been studied in a few papers such as [20, 21]. Al Qudah modified the predator-prey model by considering the effect of the term XY2, where Y2 denotes the intersection between a male and a female of predators who feed on same prey (X) during mating period . Then, Aljahdaly and Al qudah found the analytical solutions and discussed the biological relevance of the solutions . The numerical solution was found by the exponential time differencing method . However, no contributions in the existing literature have studied the effect of the term XY2 on the two-predators-one-prey model.

Due to the difficulty of computing exact solutions for most ODEs that described real applications, semi-analytical solutions are found instead of exact solutions in explicit form . Our previous work  presented the predator-prey with the effect of the term XY2 and computed the semi-analytical solution by the differential transform method (DTM) , Adomian decomposition method (ADM) , 4th Runge-Kutta (4thRK)  and the multistage differential transform method (MsDTM). We proved that the MsDTM is more accurate than DTM and ADM and faster than the 4thRK. In fact, semi-analytical methods find the solution in the form of series and the other terms of series are obtained from given initial conditions. Some semi-analytical techniques are based on Taylor expansion about the initial point t0 such as Legendre Pseudospectral Scheme , residual power series method (RPSM) Laplace decomposition method (LDM) , DTM and ADM. These methods enable to find the solution in a very small time domain because the Taylor expansion is locally convergent. In other words, the DTM can find the solution only in small neighborhood about the initial point and can not obtain the solution in long domain. The DTM has been improved by using Padé approximation and by applying the multistage technique. Both improvements are able to find the solutions in long domain. However, the MsDTM is a reliable and optimal method to approximate solutions in a long domain for linear and nonlinear ODEs . The MsDTM is a modification of the DTM by dividing the original time domain into sub-domains and applying the DTM method at each subdomain. In addition, the nonlinearity is treated by transforming the nonlinear term to a special summation based on differential transformation form. The accuracy of the MsDTM is controlled by the iteration number and time step size. The accuracy for the 4thRK is controlled only by the time step size, therefore, the MsDTM can be faster than 4thRK .

The novelty of this paper is presenting a two-predator-one-prey mathematical model during the predator mating period. In addition the paper studies the model theoretically including a positivity, boundedness, as well as local and global stability. The paper introduces the MsDTM logarithm and interprets the results.

The paper is organized as follows: section (2) proposes the mathematical model, section (3) shows the positivity and boundedness, section (4) studies the local and global stability, section (4) describes the MsDTM for finding the semi-analytical solutions and these results are discussed in section (6), and the last section contains a conclusion of this work.

## 2 Mathematical model

In this section, we introduce the mathematical model that describes the interactions between two types of predators (i.e Y and Z) and one type of prey (X). We aim to study the behavior of the animal population during the mating period under the following assumptions:

• a male of the predators stays with the same female and they feed on the same prey during their mating period.
• the rate of growth for predators and prey are larger than the rate of decay by their natural death.

The modified diffusive predator-prey model in reference  studied the interaction of two predators (male and female from the same species) with one prey in the mating period. Herein, we extend this model to study two species of predators interacting with one prey during their mating period as follows (1) where X(t), Y(t) and Z(t) denote densities of prey, predator type (I) and predator type (II), respectively. The terms XY and XZ refer to the interaction between prey and predators of type (I) and type (II) respectively. The terms XY2 and XZ2 represent the interaction between the male and female of predators and prey. All parameters are nonnegative and have the following interpretations

• β1, β8, β11 are the growth rates of X, Y, Z respectively.
• β2, β9, β12 are the decay rates of X, Y, Z respectively due to natural death.
• β3, β10, β13 are the decay rates of X, Y, Z respectively due to competition of food on the same species.
• β4, β7 are the decay rates of prey due to the interaction between one predator and one prey.
• β5, β6 are the decay rates of prey due to the interaction between two predators and one prey.

## 3 Positivity and boundedness

Theorem 1 All solutions of system (1) are nonnegative and bounded in some region Ω subject to nonnegative initial conditions in Ω.

Proof

Let assume the initial conditions at t = 0 are nonnegative and rewrite the system (1) as where by integrating over [0, t], the following equalities are valid:  and Since X(0) ≥ 0,Y(0) ≥ 0, and Z(0) ≥ 0, we have X(t) ≥ 0, Y(t) ≥ 0 and Z(t) ≥ 0 for all t ≥ 0. Hence, the solution (X(t),Y(t), Z(t)) is nonnegative for nonnegative initial data. Next, we prove that all solutions will remain bounded. We start by rewriting the system (1) as follows Let β1 > β2, β3 > β9, β11 > β12, γ1 = max{(β1β2), (β3β9), (β11β12)}, and γ2 = min{β3, β10, β13}, thus, Hence, all the solutions of model (1) are nonnegative and bounded in the following region ## 4 Stability analysis

An equilibrium point of a dynamical system denotes a stationary condition for the dynamics. The local stability at equilibrium points indicates that the solutions remain near the equilibrium points with small perturbation, but can lead to a large change with large perturbation. On the other hand, the global stability means the solution remain nearby the equilibrium points with small or large perturbations. Thus, in this section, we will study the local and global stability of the equilibrium points.

### 4.1 Equilibrium points

To study the stability of the system (1), the equilibrium points are obtained by setting (2)

The solutions for the system (2) give the equilibrium points as follows: and . Note that the coexisting equilibrium point were excluded because it is negative points. Also, the trivial point was excluded because is not interesting in our investigation.

### 4.2 Local stability

Theorem 2 The points, and are locally asymptotically stable if the following conditions are satisfied

• (1): for the equilibrium point: Local stability conditions: β8β9, β11β12, • (2): for the equilibrium point: Local stability conditions: • (3): for the equilibrium point: Local stability conditions: • (4): for the equilibrium point: Local stability conditions: Proof

The Jacobian matrix for the system (1) is defined as  where A1 = β1β2 − 2β3Xβ5Y2β4Yβ6Z2β7Z, A2 = β8β9 + β4X + 2β5XY − 2β10Y, A3 = β11β12 + β7X + 2β6XZ − 2β13Z. To study the stability of each equilibrium point, we substitute the point Pi, i = 1, 2, 3, 4 into the matrix J and the result is JP. Then, looking for negative eigenvalues of JP implies the stability conditions for each equilibrium point.

1. (1): the equilibrium point: is non negative if β8β9, β11β12, and the eigenvalue of is negative if 2. (2): is non-negative if β8β9, β11β12 and the eigenvalues of are negative if 3. (3): is non-negative if β8β9, β11β12, and the eigenvalues of are negative if 4. (4): is non- negative if β1β2, and the eigenvalues of are negative if Thus, the points, and are asymptotically stable associated to the presented stability conditions. The relevant ecological interpretation for stability analysis that the system has stable equilibrium point (1) if the growth rate of predators is greater than the natural death rate for both types of predators. The equilibrium point (2) is stable if the growth rate of the predator type (I) is less than the death rate and the growth rate of predator type (II) is greater than its death rate. In case of the opposite of this scenario, the equilibrium point (3) is stable. The equilibrium point (4) is stable if the growth rate of the prey is greater than the death rate which is a result of decreasing growth rate of predators type (I) and (II).

### 4.3 Global stability

Theorem 3 The equilibrium points are asymptotically global stable under the following stability conditions

1. (1) for point , the conditions are β8β9, β11β12, β10 ≠ 0, β13 ≠ 0 and β1β2,
2. (2) for point , the conditions are β11β12, β13 ≠ 0, β1β2, β8β9,
3. (3) for point , the conditions are β8β9, β10 ≠ 0, β1β2, β11β12,
4. (4) for point , the conditions are β1β1, β3 ≠ 0, β8β9, β11β12, and β5 = β4 = β6 = β7 = 0.

Proof

The global stability can be studied by employing the Lyapunov equation. Consider where These functions are defined on and (X*, Y*, Z*) is an equilibrium point. The function V(X, Y, Z) is zero at all equilibrium points and positive for other values. An equilibrium point p is globally stable if the derivative evaluated at p is non-positive, that is if .

• for point , the point is positive defined under the conditions β8β9, β11β12, β10 ≠ 0, β13 ≠ 0 and from the equilibrium equations we have   , if β1β2.
• for point , the point is positive defined under the conditions β11β12, β13 ≠ 0 and from the equilibrium equations we have   , if β1β2, β8β9.
• for point , the point is positive defined under the conditions β8β9, β10 ≠ 0 and from the equilibrium equations we have   , if β1β2, β11β12.
• for point , the point is positive defined under the conditions β1β1, β3 ≠ 0 and from the equilibrium equations we have   , if β8β9, β11β12, and β5 = β4 = β6 = β7 = 0.

## 5 The multistage differential transform method

The idea of the multistage differential transform method is based on dividing the domain of time into subdomains [ti ti+1]. In each subdomain, we apply the differential transform method (DTM) where the initial condition at ti. The DT is defined as follows (3) Let assume an ODE with initial condition at t0 = 0 and a solution u(t). The operators in the ODE are transferred to DT operators as in Table 1. Then, we obtain the scheme in terms of U(k) and U(k + 1). The U(0) is the initial condition and U(k + 1) is obtained by applying U(k). The solution u(t) is , where N is the arbitrary iteration number. The readers are referred to the references  for more details.

Table 1. Differential transformations.

https://doi.org/10.1371/journal.pone.0289410.t001

## 6 Semi-analytical simulation

In this section, the proposed model will be computed by MsDTM. First step, we divide the domain [0, T] into subdomain [ti, ti+1] where h = [ti+1ti]/100 is the step size. Next, we assume , where X0(ti), Y0(ti), Z0(ti) are the initial values and the Xk+1, Yk+1, Zk+1 are obtained by applying DTM method in each subinterval as follows for k = 0, 1, …, α − 1. The solution of each case in Table 2 is plotted in Fig 1. The solutions clarify that if there are two types of predator such that male and female feed together on same prey during mating period, the predator who has low growth rate decays fast. However, the optimal ecological scenario is case (4) because it gives a periodic solution. The case (4) denotes that case when the growth rate of prey is larger than the growth rate of predators (I) and (II).

Fig 1. Plot the solutions of X(t), Y(t) and Z(t) where h = 0.01 using MsDTM for X0 = 5, Y0 = 3 and Z0 = 2.

(a) Case (1), (b) Case (2), (c) Case (3), (d) Case (4).

https://doi.org/10.1371/journal.pone.0289410.g001

Table 2. Different value of parameters.

https://doi.org/10.1371/journal.pone.0289410.t002

## 7 Conclusion

The article is a first contribution in the literature that produces two-predator-one-prey system with the effect of XY2 term. It also solved the aforementioned mathematical model by MsDTM. The used method is fast, reliable and global convergent. The stability of the dynamic of the system was investigated. We found that the equilibrium points are asymptotically stable. The advantages of the used method have been proved in our previous work . The method is more accurate than ADM, DTM, LADM and is faster than 4thRK. In fact that the 4thRK method discretizes the domain and its speed based on the step size of the discretization, but the speed of MsADM based on two factors: the length of subdomain and number of taken Taylor expansion terms. Therefore we can control the speed and accuracy of MsDTM by two factors.

The results proved that when two types of predators live in a same spot with the assumptions that male and female feed on the same prey during mating period, the ecological balance is achieved when the growth rate of prey is grater than both types of predators. Otherwise, the population of prey will decay fast and then the predator populations will decay as results of lack of food.

In future work, MsDTM can be used to solve different ODEs in the science and the presented model can be studied with fractional derivative.

## Supporting information

### S1 File.

https://doi.org/10.1371/journal.pone.0289410.s001

(BST)

## Acknowledgments

The Deanship of Scientific Research (DSR), King Abdulaziz University (KAU), Jeddah, Saudi Arabia has funded this Project under grant No.(G: 164-665-1443). The authors, therefore, acknowledge with thanks DSR technical and financial support.

## References

1. 1. Dubey B, Das B, Hussain J. A predator-prey interaction model with self and cross-diffusion. Ecological Modelling. 2001;141(1-3):67–76
2. 2. Salas AH, El-Tantawy S, Aljahdaly NH. An exact solution to the quadratic damping strong nonlinearity Duffing oscillator. Mathematical Problems in Engineering. 2021;2021
3. 3. Behera S, Aljahdaly N, Virdi J. On the modified (G/G2)-expansion method for finding some analytical solutions of the traveling waves. Journal of Ocean Engineering and Science. 2022;7(4):313–320
4. 4. Alharbey R, Aljahdaly NH. On fractional numerical simulation of HIV infection for CD8+ T-cells and its treatment. Plos one. 2022;17(3):e0265627 pmid:35324957
5. 5. Aljahdaly NH, Alharbey R. Fractional numerical simulation of mathematical model of HIV-1 infection with stem cell therapy. AIMS Mathematics. 2021;6(7):6715–6726
6. 6. Saha S, Samanta G. Modeling of insect-pathogen dynamics with biological control. Mathematical Biology and Bioinformatics. 2020;15(2):268–294
7. 7. Lotka AJ. Elements of physical biology. Science Progress in the Twentieth Century (1919-1933). 1926;21(82):341–343
8. 8. Volterra V. Variations and fluctuations of the number of individuals in animal species living together. Animal ecology. 1926; p. 409–448
9. 9. Nadjah K, Abdelouahab MS. Stability and Hopf bifurcation of the coexistence equilibrium for a differential-algebraic biological economic system with predator harvesting. Electronic Research Archive. 2021;29(1):1641–1660
10. 10. Heithaus MR. Predator-prey interactions. Biology of sharks and their relatives. 2004;17:487–521
11. 11. Mondal S, Samanta G. Dynamics of a delayed predator-prey interaction incorporating nonlinear prey refuge under the influence of fear effect and additional food. Journal of Physics A: Mathematical and Theoretical. 2020;53(29):295601
12. 12. Sugie J, Ishihara Y. Attraction Region for the Classical Lotka- Volterra Predator- Prey model Caused by impulsive Effects. Qualitative Theory of Dynamical Systems. 2021;20(2):1–26
13. 13. Bentout S, Djilali S, Kumar S. Mathematical analysis of the influence of prey escaping from prey herd on three species fractional predator-prey interaction model. Physica A: Statistical Mechanics and its Applications. 2021;572:125840
14. 14. Mondal S, Samanta G. Dynamics of an additional food provided predator?prey system with prey refuge dependent on both species and constant harvest in predator. Physica A: Statistical Mechanics and its Applications. 2019;534:122301
15. 15. Sahoo D, Samanta G. Comparison between two tritrophic food chain models with multiple delays and anti-predation effect. International Journal of Biomathematics. 2021;14(03):2150010
16. 16. Wang S, Wang Z, Xu C, Jiao G. Sensitivity analysis and stationary probability distributions of a stochastic two-prey one-predator model. Applied Mathematics Letters. 2021;116:106996
17. 17. Saha S, Samanta G. A prey-predator system with disease in prey and cooperative hunting strategy in predator. Journal of Physics A: Mathematical and Theoretical. 2020;53(48):485601
18. 18. Sahoo D, Samanta G. Impact of Fear Effect in a Two Prey-One Predator System with Switching Behaviour in Predation. Differential Equations and Dynamical Systems. 2021; p. 1–23
19. 19. Saha S, Samanta G. Modelling of a two prey and one predator system with switching effect. Computational and Mathematical Biophysics. 2021;9(1):90–113
20. 20. Long Y, Wang L, Li J. Uniform persistence and multistability in a two-predator?one-prey system with inter-specific and intra-specific competition. Journal of Applied Mathematics and Computing. 2021; p. 1–28
21. 21. Singh J, Kilicman A, Kumar D, Swroop R. Numerical study for fractional model of nonlinear predator-prey biological population dynamic system. Preprints. 2018
22. 22. Al Qudah M. Existence, uniqueness solution of a modified predator-prey model. Nonlinear Analysis and Differential Equations. 2016;4:669–677
23. 23. Aljahdaly NH, Alqudah MA. Analytical solutions of a modified predator-prey model through a new ecological interaction. Computational and mathematical methods in medicine. 2019;2019 pmid:31737084
24. 24. Aljahdaly NH, Ashi H. Exponential time differencing method for studying prey-predator dynamic during mating period. Computational and Mathematical Methods in Medicine. 2021;2021 pmid:34594395
25. 25. Aljahdaly NH, El-Tantawy S, Wazwaz AM, Ashi H. Adomian decomposition method for modelling the dissipative higher-order rogue waves in a superthermal collisional plasma. Journal of Taibah University for Science. 2021;15(1):971–983
26. 26. Aljahdaly NH; AIP Publishing LLC. New application through multistage differential transform method. AIP Conference Proceedings. 2020;2293(1):420025.
27. 27. Ayaz F. On the two-dimensional differential transform method. Applied Mathematics and computation. 2003;143(2-3):361–374
28. 28. Wazwaz AM, El-Sayed SM. A new modification of the Adomian decomposition method for linear and nonlinear operators. Applied Mathematics and computation. 2001;122(3):393–405
29. 29. Butcher J. Runge-kutta methods. Scholarpedia. 2007;2(9):3147
30. 30. Tohidi E, Navid Samadi O, Shateyi S. Convergence analysis of Legendre Pseudospectral scheme for solving nonlinear systems of Volterra integral equations. Advances in Mathematical Physics. 2014;2014
31. 31. Hussain M, Khan M. Modified Laplace decomposition method. Applied Mathematical Sciences. 2010;4(36):1769–1783
32. 32. Aljahdaly NH, El-Tantawy S. On the multistage differential transformation method for analyzing damping Duffing oscillator and its applications to plasma physics. Mathematics. 2021;9(4):432