Hindawi

Mathematical Problems in Engineering

Volume 2020, Article ID 5309814, 18 pages
https://doi.org/10.1155/2020/5309814

Research Article
Bifurcation and Chaos of a Discrete Predator-Prey Model with
Crowley–Martin Functional Response Incorporating
Proportional Prey Refuge

P. K. Santra,1 G. S. Mahapatra,2 and G. R. Phaijoo 3

1
Maulana Abul Kalam Azad University of Technology, Kolkata 700064, India
2
Department of Mathematics, National Institute of Technology Puducherry, Karaikal 609609, India
3
Department of Mathematics, Kathmandu University, Dhulikhel 45200, Nepal

Correspondence should be addressed to G. R. Phaijoo; [email protected]

Received 12 September 2019; Revised 29 November 2019; Accepted 6 December 2019; Published 22 January 2020

Academic Editor: Elisa Francomano

Copyright © 2020 P. K. Santra et al. This is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

The paper investigates the dynamical behaviors of a two-species discrete predator-prey system with Crowley–Martin functional
response incorporating prey refuge proportional to prey density. The existence of equilibrium points, stability of three fixed
points, period-doubling bifurcation, Neimark–Sacker bifurcation, Marottos chaos, and Control Chaos are analyzed for the
discrete-time domain. The time graphs, phase portraits, and bifurcation diagrams are obtained for different parameters of the
model. Numerical simulations and graphics show that the discrete model exhibits rich dynamics, which also present that the
system is a chaotic and complex one. This paper attempts to present a feedback control method which can stabilize chaotic orbits
at an unstable equilibrium point.

1. Introduction time domain. Upadhyay and Naji [46] presented three-

species food chain model with Crowley–Martin type func-
The dynamical behavior of the prey-predator system in tional response for the dynamical study. Zhou [47] pre-
ecology has been the subject of study for many researchers sented qualitative analysis of the predator-prey model for
and hence they have provided a substantial contribution to Crowley–Martin functional responses. Tripathi et al. [48]
the growth of the population models [1–8] in the contin- applied Crowley–Martin functional response on a delayed
uous-time domain. In case of populations that have over- density dependent predator-prey model. Li et al. [49] pre-
lapping generations and the birth processes occuring sented a predator-prey model with Crowley–Martin func-
continuously, the ordinary differential equations are used for tional response for positive periodic solutions.
modeling of predator-prey interaction [9–30]. Many species, A certain degree of protection can be offered by nature to
such as monocarpic plants, and semelparous animals have a set number of prey populations by providing refuges. The
discrete non overlapping generations, and their births occur effects of prey refuges on the population dynamics are in-
in regular breeding seasons. Their interactions are charac- tricate and complex in nature; however, it can be considered
terized by difference equations or formulated as discrete- to be constituted by two components for modeling purposes.
time mappings. Discrete-time predator-prey models are The first component of refuge effects, which governs positive
capable of exhibiting much more complicated dynamics in growth of prey and negatively that of predators, consists of
comparison to the corresponding continuous-time models a reduction in prey mortality owing to decreasing predation
[31–45]. success. The second one may arise as the trade-offs and
The dynamical study of the prey-predator model with byproducts arising from the prey’s hiding behavior, which
different functional responses is an effective approach for could cause advantages or even be detrimental for all the
2 Mathematical Problems in Engineering

interacting populations. González-Olivares and Ramos- functional response is modified to incorporate prey refuges
Jiliberto [8] presented prey refuges in a simple prey-predator in the above model, and the term φ(x, y) becomes the
system. Huang et al. [9] presented stability analysis of a prey- functional response in the form of φ(x − xr , y). Based on
predator system incorporating a prey refuge. Ma et al. [12] a laboratory study of dragonfly, Crowley and Martin present
discussed the effect of prey refuges on a prey-predator functional response φ(x, y) � x/(1 + αx)(1 + βy), where
system. Chen et al. [13] studied the prey refuge on α > 0 describe the handling time and β > 0 is the magnitude
a Leslie–Gower predator-prey model. of interference among predators. In the proposed model, we
Extensive studies have been carried out by many re- consider refuge xr � bx, and to derive a discrete-time model
searchers on refuge concept [9–30], and most of these papers let (dx/dt) � ((xt+h − xt )/h) and (dy/dt) � ((yt+h − yt )/h),
consider refuge as a constant number of prey or proportional where xt and yt are the densities of the prey and predator
to prey. These earlier works have been taken into account populations in discrete time t. Let h ⟶ 1 and f � 1, then
that the use of refuges by a fraction of prey, or a constant the (n + 1)th generation of the prey and predator pop-
number prey, is influential in existing a stabilizing effect in ulations by replacing t by n is obtained as follows:
the dynamics of the interacting populations. This paper will r c(1 − b)xn yn
test the above statement assuming that the quantity of prey xn+1 � (r + 1)xn 􏼠1 − x 􏼡− ,
k(r + 1) n 􏼈1 + α(1 − b)xn 􏼉 1 + βyn 􏼁
in refuges is proportional to the prey. We also conduct an
analysis of the dynamic properties of such a system via d(1 − b)xn yn
modification of the well-known Lotka–Volterra predator- yn+1 � .
􏼈1 + α(1 − b)xn 􏼉 1 + βyn 􏼁
prey model with prey self-limitation incorporating Crowl-
ey–Martin functional response [46, 47]. In this paper, we (2)
consider the refuge term proportional to prey density. Let us consider (r/k(r + 1)) � 1 and (r + 1) � a, then
The rest of the paper is organized as follows. In Section 2, from (2) we obtain the following proposed discrete-time
a discrete-time prey-predator model under nonoverlapping predator-prey system as follows:
generation with refuge is formulated. Analysis of the local
stability of the proposed model is dealt with in Section 3. The c(1 − b)xn yn
xn+1 � axn 1 − xn 􏼁 − ,
bifurcation analysis of the proposed model is studied 􏼈1 + α(1 − b)xn 􏼉 1 + βyn 􏼁
through period-doubling bifurcation and Neimark–Sacker (3)
bifurcation in Section 4. Sections 5 and 6 give Marottos d(1 − b)xn yn
yn+1 � ,
chaos and control chaos procedures, respectively. The 􏼈1 + α(1 − b)xn 􏼉 1 + βyn 􏼁
proposed model is supported with the help of numerical
simulations in Section 7. Section 8 makes the conclusion to where a, b, c, d, α, and β are all positive constants. By bi-
this paper. ological meaning of the model variables, we only consider
the system in the region Ω � 􏼈(x, y) : x ≥ 0, y ≥ 0􏼉 in the
(x, y)− plane.
2. Description of the Proposed Predator-
Prey Model
3. Fixed Points and Stability Analysis of the
The proposed model belongs to a generalized predator-prey Prey-Predator System
model which incorporates logistic growth having no stage
structure for neither the prey nor the predator, which is Fixed points of system (3) are determined by solving the
represented by the following differential equations: following nonlinear system of equations:
dx x c(1 − b)xy
� rx􏼒1 − 􏼓 − cφ(x, y)y, x � ax(1 − x) − ,
dt k {1 + α(1 − b)x}(1 + βy)
(1) (4)
dy d(1 − b)xy
� dφ(x, y)y − fy, y� .
dt {1 + α(1 − b)x}(1 + βy)
where x and y denote the density of prey and predator
By simple calculation, we get three nonnegative fixed
populations at any time t, respectively, and r, k, c, d, and f
points as follows: (i) P0 � (0, 0), (ii) P1 � ((a − 1/a), 0),
are all positive constants and have their biological meanings
a > 1, and (iii) P2 � (x2 , y2 ), where x2 is positive root of the
accordingly. The intrinsic per capita growth rate of prey
equation Ax3 + Bx2 + Cx + D � 0, where A � αβad(1 − b),
population is denoted by r, k is the environmental carrying
B � αβd(1 − a)(1 − b), C � βd(1 − a) + c(1 − b)(d − α),
capacity of prey population, c is the maximal per capita
and D � − c, and y2 � (1/β)[(d(1 − b)x2 /(1 + α(1 − b)x2 ))
consumption rate of predators, d is the efficiency with which
− 1].
predators convert consumed prey into new predators, and f
is the per capita death rate of predators. The term φ(x, y)
represents the functional response of the predator Lemma 1. Equation Ax3 + Bx2 + Cx + D � 0 has (i)
population. a unique positive root if a < 1, (ii) three positive roots if
According to Maynard Smith [4], there is a quantity xr of 1 < a < 1 + (c(1 − b)(d − α)/βd), d > α, and (iii) a unique
prey population which incorporates refuges. Thereafter, the positive root if a > 1 + (c(1 − b)(d − α)/βd), d > α.
Mathematical Problems in Engineering 3

Now, the local behavior of system (3) is studied for each Proposition 4. Interior equilibrium point P2 (x2 , y2 ) is
equilibrium point of the prey-predator system. The stability Source if one of the following conditions holds:
of system (3) is carried out by computing the Jacobian
(i) x2 < (1/2) and adR > max(((dR − cQ − 1)/(1 −
matrix corresponding to each equilibrium point. The Ja-
2x2 )) + a, ((cQ − dR − 1)/(1 − 2x2 )) − a, (1/(1 − 2
cobian matrix J for system (3) is
x2 )))
a11 a12
J �􏼢 􏼣, (5) (ii) x2 > (1/2) and adR < min(((dR − cQ − 1)/(1 −
a21 a22 2x2 )) + a, ((cQ − dR − 1)/(1 − 2x2 )) − a, (1/(1 − 2
x2 )))
where
c(1 − b)y
a11 � a(1 − 2x) − � a(1 − 2x) − cQ, Proposition 5. Interior equilibrium point P2 (x2 , y2 ) is
{1 + α(1 − b)x}2 (1 + βy)
Saddle if one of the following conditions holds:
c(1 − b)x
a12 � − � − cR, (i) x2 < (1/2) and ((dR − cQ − 1)/(1 − 2x2 )) + a <
{1 + α(1 − b)x}(1 + βy)2 adR < ((cQ − dR − 1)/(1 − 2x2 )) − a
d(1 − b)y (ii) x2 > (1/2) and ((cQ − dR − 1)/(1 − 2x2 )) − a <
a21 � � dQ , adR < ((dR − cQ − 1)/(1 − 2x2 )) + a
{1 + α(1 − b)x}2 (1 + βy)

d(1 − b)x
a22 � � dR , Proposition 6. Interior equilibrium point P2 (x2 , y2 ) is
{1 + α(1 − b)x}(1 + βy)2 Nonhyperbolic if one of the following conditions holds:
(6)
(i) x2 < (1/2) and ((cQ − dR − 1)/(1 − 2x2 )) − a �
where adR ≠ (2/(1 − 2x2 )) > ((dR − cQ − 1)/(1 − 2x2 )) + a
(1 − b)y (ii) x2 > (1/2) and ((cQ − dR − 1)/(1 − 2x2 )) − a �
Q� , adR ≠ (2/(1 − 2x2 )) < ((dR − cQ − 1)/(1 − 2x2 )) + a
{1 + α(1 − b)x}2 (1 + βy)
(7) (iii) x2 < (1/2) and adR � (1/(1 − 2x2 )) > max(((dR−
R�
(1 − b)x
. cQ − 1)/(1 − 2x2 )) + a, ([a(1 − 2x) − cQ + dR]2 /4
{1 + α(1 − b)x}(1 + βy)2 (1 − 2x2 )))
(iv) x2 > (1/2) and adR � (1/(1 − 2x2 )) < min(((dR−
The characteristic equation of matrix J is λ2 − Tr(J)λ +
cQ − 1) /(1 − 2x2 )) + a, ([a(1 − 2x) − cQ + dR]2 /4
Det(J) � 0, where
(1− 2x2 )))
Tr(J) � Trace of matrix � a(1 − 2x) − cQ + dR PD1 � 􏼈a, b, c, d, α, β: x2 <(1/2) and ((cQ − dR − 1)/(1−
Det(J) � Determinant of matrix � ad(1 − 2x)R 2x2 )) − a � adR ≠ (2/(1 − 2x2 )) > ((dR − cQ − 1)/(1−
2x2 )) + a}
Hence, model (3) is a dissipative dynamical system, if
|ad(1 − 2x)R| < 1, is a conservative dynamical one, if and PD2 � 􏼈a, b, c, d, α, β: x2 > (1/2) and ((cQ − dR − 1) /
only if |ad(1 − 2x)R| � 1, and is an undissipated dynamical (1 − 2x2 )) − a � adR ≠(2/(1 − 2x2 ))<((dR − cQ − 1)/
system otherwise. (1 − 2x2 )) + a}
NS1 � 􏼈a, b, c, d, α, β: x2 < (1 / 2) and adR � (1 /(1 −
Proposition 1. The equilibrium point P0 � (0, 0) is Sink if 2x2 )) > max(((dR − cQ − 1) / (1 − 2x2 )) + a, ([a(1 −
a < 1, Saddle if a > 1, and Nonhyperbolic if a � 1. 2x) − cQ + dR]2 /4(1 − 2x2 )))}
NS2 � 􏼈a, b, c, d, α, β: x2 > (1/2) and adR � (1/(1− 2x2 ))
< min(((dR − cQ − 1)/(1− 2x2 )) + a,([a(1 − 2x) − cQ +
Proposition 2. The equilibrium point P1 is Sink if |2 − a| < 1
dR]2 /4(1 − 2x2 )))}
and |d(1 − a)(1 − b)/(a − α(1 − a)(1 − b))| < 1, Source if
|2 − a| > 1 and |d(1 − a)(1 − b)/(a − α(1 − a)(1 − b))| > 1, The prey-predator model (3) undergoes period-doubling
Saddle if |2 − a| > 1 and |d(1 − a)(1 − b)/(a − α(1 − a)(1 − bifurcation at P2 (x2 , y2 ), when parameters vary in a small
b))| < 1 or |2 − a| < 1 and |d(1 − a)(1 − b)/(a − α(1 − a)(1 − neighborhood of PD1 or PD2 . Model (3) undergoes Nei-
b))| > 1, and Nonhyperbolic if |2 − a| � 1 or |d(1 − a)(1 − mark–Sacker bifurcation at P2 (x2 , y2 ), when parameters
b)/(a − α(1 − a)(1 − b))| � 1 vary in a small neighborhood of NS1 or NS2 .

Proposition 3. Interior equilibrium point P2 (x2 , y2 ) is Sink 4. Bifurcation Analysis

if one of the following conditions holds:
This section carries out an investigation of the conditions for
(i) x2 < (1/2) and max(((dR − cQ − 1)/(1 − 2x2 )) + a, the existence of Neimark–Sacker bifurcation (NSB) and
((cQ − dR − 1) /(1 − 2x2 )) − a) < adR < (1/(1 − 2x2 )) period-doubling bifurcation (PDB) at the positive fixed
(ii) x2 > (1/2) and (1/(1 − 2x2 )) < adR < min(((dR − point P2 (x2 , y2 ) of proposed system (3). There is an
cQ − 1)/(1 − 2x2 ))+ a, ((cQ − dR − 1)/(1 − 2x2 )) − a) emergence of different kinds of bifurcations from the fixed
4 Mathematical Problems in Engineering

point in dynamical systems, when a particular parameter c(1 − b)y2

α1 � fx x2 , y2 , 0􏼁 � a 1 − 2x2 􏼁 − ,
passes through its critical value. Many dynamical properties 􏼈1 + α(1 − b)x2 􏼉 1 + βy2 􏼁
2

of a system can be discussed owing to emergence of NSB and

PDB. We discuss NSB and PDB for the positive fixed point c(1 − b)x2
α2 � fy x2 , y2 , 0􏼁 � − 2,
P2 (x2 , y2 ) of the prey-predator system (3) taking b as 􏼈1 + α(1 − b)x2 􏼉 1 + βy2 􏼁
a bifurcation parameter.
2αc(1 − b)2 y2
α11 � fxx x2 , y2 , 0􏼁 � − 2a + 3 ,
􏼈1 + α(1 − b)x2 􏼉 1 + βy2 􏼁

4.1. Neimark–Sacker Bifurcation. This section presents NSB c(1 − b)

α12 � fxy x2 , y2 , 0􏼁 � − 2 2,
of the proposed prey-predator model (3) at P2 (x2 , y2 ) for 􏼈1 + α(1 − b)x2 􏼉 1 + βy2 􏼁
the parameters are located in the following set:
2βc(1 − b)x2
α22 � fyy x2 , y2 , 0􏼁 � 3,
1 1 􏼈1 + α(1 − b)x2 􏼉 1 + βy2 􏼁
NS1 � 􏼨a, b, c, d, α, β: x2 < and a dR �
2 1 − 2x2
6α2 c(1 − b)3 y2
α111 � fxxx x2 , y2 , 0􏼁 � − 4 ,
􏼈1 + α(1 − b)x2 􏼉 1 + βy2 􏼁
dR − cQ − 1 [a(1 − 2x) − cQ + dR]2
> max􏼠 + a, 􏼡􏼩.
1 − 2x2 4 1 − 2x2 􏼁 2αc(1 − b)2
α112 � fxxy x2 , y2 , 0􏼁 � 3 2,
(8) 􏼈1 + α(1 − b)x2 􏼉 1 + βy2 􏼁

Similar arguments can be applied to the other case: 2βc(1 − b)

α122 � fxyy x2 , y2 , 0􏼁 � 2 3,
􏼈1 + α(1 − b)x2 􏼉 1 + βy2 􏼁
1 1
NS2 � 􏼨a, b, c, d, α, β: x2 > and adR � 6β2 c(1 − b)x2
2 1 − 2x2 α222 � fyyy x2 , y2 , 0􏼁 � − 4,
􏼈1 + α(1 − b)x2 􏼉 1 + βy2 􏼁
dR − cQ − 1 [a(1 − 2x) − cQ + dR]2 d(1 − b)y2
< min􏼠 + a, 􏼡􏼩. β 1 � g x x 2 , y 2 , 0􏼁 � 2 ,
1 − 2x2 4 1 − 2x2 􏼁 􏼈1 + α(1 − b)x2 􏼉 1 + βy2 􏼁
(9) d(1 − b)x2
β2 � gy x2 , y2 , 0􏼁 � 2,
In analyzing the NSB, b is used as the bifurcation pa- 􏼈1 + α(1 − b)x2 􏼉 1 + βy2 􏼁
rameter. Furthermore, b∗ (|b∗ | ⋘ 1) is the perturbation of b,
2αd(1 − b)2 y2
and we consider a perturbation of the model as follows: β11 � gxx x2 , y2 , 0􏼁 � − 3 ,
􏼈1 + α(1 − b)x2 􏼉 1 + βy2 􏼁
c(1 − (b + b∗ ))xn yn
xn+1 � axn 1 − xn 􏼁 − d(1 − b)
􏼈1 + α(1 − (b + b∗ ))xn 􏼉 1 + βyn 􏼁 β12 � gxy x2 , y2 , 0􏼁 � 2 2,
􏼈1 + α(1 − b)x2 􏼉 1 + βy2 􏼁
≡ f xn , yn , b∗ 􏼁, 2βd(1 − b)x2
β22 � gyy x2 , y2 , 0􏼁 � − 3,
􏼈1 + α(1 − b)x2 􏼉 1 + βy2 􏼁
∗
d(1 − (b + b ))xn yn
yn+1 � ≡ g xn , yn , b∗ 􏼁. 6α2 d(1 − b)3 y2
􏼈1 + α(1 − (b + b∗ ))xn 􏼉 1 + βyn 􏼁 β111 � gxxx x2 , y2 , 0􏼁 � ,
4
􏼈1 + α(1 − b)x2 􏼉 1 + βy2 􏼁
(10)
Let un � xn − x2 and vn � yn − y2 , then the equilibrium 2αd(1 − b)2
β112 � gxxy x2 , y2 , 0􏼁 � − 3 2,
point P2 (x2 , y2 ) is transformed into the origin, and further 􏼈1 + α(1 − b)x2 􏼉 1 + βy2 􏼁
expanding f and g as a Taylor series at (un , vn ) � (0, 0) to the 2βd(1 − b)
third order, the proposed model (10) becomes β122 � gxyy x2 , y2 , 0􏼁 � − 2 3,
􏼈1 + α(1 − b)x2 􏼉 1 + βy2 􏼁
un+1 � α1 un + α2 vn + α11 u2n + α12 un vn + α22 v2n + α111 u3n + α112 u2n vn
6β2 d(1 − b)x2
􏼌􏼌 􏼌􏼌 􏼌􏼌 􏼌􏼌 4 β222 � gyyy x2 , y2 , 0􏼁 � 4.
+ α122 un v2n + α222 v3n + O􏼒􏼐􏼌􏼌un 􏼌􏼌 + 􏼌􏼌vn 􏼌􏼌􏼑 􏼓, 􏼈1 + α(1 − b)x2 􏼉 1 + βy2 􏼁
(12)
vn+1 � β1 un + β2 vn + β11 u2n + β12 un vn + β22 v2n + β111 u3n + β112 u2n vn
The characteristic equation associated with the lineari-
􏼌􏼌 􏼌􏼌 􏼌􏼌 􏼌􏼌 4 zation of model (11) at (un , vn ) � (0, 0) is given by
+ β122 un v2n + β222 v3n + O􏼒􏼐􏼌􏼌un 􏼌􏼌 + 􏼌􏼌vn 􏼌􏼌􏼑 􏼓,
λ2 − Tr(J1 (b∗ ))λ + Det(J1 (b∗ )) � 0. Roots of the charac-
(11) teristic equation are
where
Mathematical Problems in Engineering 5

􏽱�������������������������
Tr J1 (b∗ )􏼁 ± i 4Det J1 (b∗ )􏼁 − Tr J1 (b∗ )􏼁􏼁
2 xn+1 � δxn − cyn + f1 xn , yn 􏼁,
λ1,2 b∗ 􏼁 � . (14)
2 yn+1 � cxn + δyn + g1 xn , yn 􏼁,
(13)
where the functions f1 and g1 denote the terms in model
∗ ∗
From |λ1,2 (b )| � 1, when b � 0, we have |λ1,2 (b )| � ∗ (14) in variables (xn , yn ) with the order at least two.
[Det(J1 (b∗ ))]1/2 and l � [d|λ1,2 (b∗ )|/db∗ ]h∗ �0 ≠ 0. In order to undergo Neimark–Sacker Bifurcation, we
In addition, when b∗ � 0, λi1,2 ≠ 1, i � 1, 2, 3, 4, this is require the following discriminatory quantity Ω to be
equivalent to Tr(J1 (0)) ≠ − 2, − 1, 1, 2. nonzero:
Now to study the normal form, let c � Im(λ1,2 ) and 2
(1 − 2λ)λ 1􏼌􏼌 􏼌􏼌2 􏼌􏼌 􏼌􏼌2
0 1 Ω � − Re⎡⎢⎣ ξ 11 ξ 20 ⎤⎥⎦ − 􏼌􏼌ξ 11 􏼌􏼌 − 􏼌􏼌ξ 02 􏼌􏼌 + Re􏼐λξ 21 􏼑,
δ � Re(λ1,2 ). We define T � 􏼢 􏼣, and using the trans- 1− λ 2
c δ
u x (15)
formation 􏼢 n 􏼣 � T􏼢 n 􏼣, model (11) becomes
vn yn where

1 1 1
ξ 20 � δ 2β22 − δα22 − α12 + 4cα22 􏼁 + cα12 + δi 4cα22 − 2α22 − 2δα22 􏼁
8 4 8

1 1 δα − 2β11 + δ3 α22 − δ2 β22 − δ2 α12 + δβ12

+ i􏼐4cβ22 + 2c2 α22 − 2α11 􏼑 + β12 + 11 ,
8 8 4c

1 1 β − δα11 + δβ12 − δ2 α12 δ2 β22 − δ3 α22

ξ 11 � c β22 − δα22 􏼁 + i􏼐c2 α22 + α11 + δα12 + δ2 α22 􏼑 + 11 − ,
2 2 2c c

1 1 β − δα11 + δβ12 − δ2 α12

ξ 02 � c 2δα22 + α12 + β22 􏼁 + i β12 + 2δβ22 − 2δα12 − α11 􏼁 − 11 (16)
4 4 4c

1 δ2 β22 − δ3 α22
+ α22 i􏼐c2 − 3δ2 􏼑 + ,
4 4c

3 1 1 1 1 3 1 3 3 3
ξ 21 � β222 􏼐c2 + δ2 􏼑 + β112 + δα112 + δβ122 + α122 􏼒 c2 + δ2 − δ􏼓 + α111 + α222 i􏼐c2 + 2δ2 􏼑 + α122 cδi
8 8 4 4 8 8 4 8 8 8

1 3 3β111 − 3δα111 + 3δβ112 − 3δ2 α112 + 3δ2 β122 − 3δ3 α122 + 3δ3 β222 − 3δ4 α222
− β122 ci − β222 cδi − i.
8 8 8c

From the above analysis we have the following result. 4.2. Period-Doubling Bifurcation. It is clear that one of the
eigenvalues of the positive fixed point P2 (x2 , y2 ) is λ1 � − 1
Theorem 1. If Ω ≠ 0, then model (3) undergoes NSB at and the other λ2 is neither 1 nor − 1 if parameters of the
P2 (x2 , y2 ) when the parameter b∗ varies in a small neigh- model are located in the following set:
borhood of the origin. Moreover, if Ω < 0 (Ω > 0), then an
attracting (repelling) inariant closed curve bifurcates from
P2 (x2 , y2 ) for b∗ > 0 (b∗ < 0).

1 cQ − dR − 1 2 dR − cQ − 1
PD1 � 􏼨a, b, c, d, α, β: x2 < and − a � adR ≠ > + a􏼩
2 1 − 2x2 1 − 2x2 1 − 2x2
(17)
1 cQ − dR − 1 2 dR − cQ − 1
or PD2 � 􏼨a, b, c, d, α, β: x2 > and − a � adR ≠ < + a􏼩.
2 1 − 2x2 1 − 2x2 1 − 2x2

Here, we discuss PDB of the proposed model (3) at In analyzing the PDB, b is used as the bifurcation parameter.
P2 (x2 , y2 ), when parameters vary in a small neighborhood Furthermore, b∗ (|b∗ | ⋘ 1) is the perturbation of b, and we
of PD1 . Similar arguments can be applied to the other cases. consider a perturbation of the model as follows:
6 Mathematical Problems in Engineering

c(1 − (b + b∗ ))xn yn xn+1 � − xn + f1 un , vn , b∗ 􏼁,

xn+1 � axn 1 − xn 􏼁 − (20)
􏼈1 + α(1 − (b + b∗ ))xn 􏼉 1 + βyn 􏼁 yn+1 � λ2 yn + g1 un , vn , b∗ 􏼁,
≡ f xn , yn , b∗ 􏼁, where the functions f1 and g1 denote the terms in model
(20) in variables (un , vn , b∗ ) with the order at least two.
d(1 − (b + b∗ ))xn yn From the center manifold theorem, we know that there
yn+1 � ≡ g xn , yn , b∗ 􏼁.
􏼈1 + α(1 − (b + b∗ ))xn 􏼉 1 + βyn 􏼁 exists a center manifold Wc (0, 0, 0) of model (20) at (0, 0) in
(18) a small neighborhood of b∗ � 0, which can be approximately
described as follows:
Let un � xn − x2 and vn � yn − y2 , then the equilibrium
point P2 (x2 , y2 ) is transformed into the origin, and further Wc (0, 0, 0) � 􏽮 xn , yn , b∗ 􏼁εR3 : yn+1 � α1 x2n + α2 xn b∗
expanding f and g as a Taylor series at (un , vn , b∗ ) � (0, 0, 0) 􏼌􏼌 􏼌􏼌 􏼌􏼌 􏼌􏼌 3
+ O􏼒􏼐􏼌􏼌xn 􏼌􏼌 + 􏼌􏼌b∗ 􏼌􏼌􏼑 􏼓􏼛,
to the third order, model (18) becomes
un+1 � α1 un + α2 vn + α11 u2n + α12 un vn + α13 un b∗ + α23 vn b∗ (21)

+ α111 u3n + α112 u2n vn + α113 u2n b∗ + α123 un vn b∗ where

2
􏼌􏼌 􏼌􏼌 􏼌􏼌 􏼌􏼌 􏼌􏼌 􏼌􏼌 4 α2 􏼂 1 + α1 􏼁α11 + α2 β11 􏼃 β22 1 + α1 􏼁
+ O􏼒􏼐􏼌􏼌un 􏼌􏼌 + 􏼌􏼌vn 􏼌􏼌 + 􏼌􏼌b∗ 􏼌􏼌􏼑 􏼓, α1 � +
1 − λ22 1 − λ22
vn+1 � β1 un + β2 vn + β11 u2n + β12 un vn + β22 v2n + β13 un b∗
1 + α1 􏼁􏼂α12 1 + α1 􏼁 + α2 β12 􏼃
+ β23 vn b∗ + β111 u3n − ,
1 − λ22
+ β112 u2n vn + β113 u2n b∗ + β123 un vn b∗ + β223 v2n b∗
􏼌􏼌 􏼌􏼌 􏼌􏼌 􏼌􏼌 􏼌􏼌 􏼌􏼌 4 1 + α1 􏼁􏼂α23 1 + α1 􏼁 + α2 β23 􏼃 1 + α1 􏼁􏼂α13 + α2 β13 􏼃
+ O􏼒􏼐􏼌􏼌un 􏼌􏼌 + 􏼌􏼌vn 􏼌􏼌 + 􏼌􏼌b∗ 􏼌􏼌􏼑 􏼓, α2 � 2 − 2 .
α2 1 + λ2 􏼁 1 + λ2 􏼁
(19) (22)
where α1 � fx (x2 , y2 , 0), α2 � fy (x2 , y2 , 0), α11 � fxx (x2 , Model (20) which is restricted to center manifold
y2 , 0), α12 � fxy (x2 , y2 , 0), α13 � fxb∗ (x2 , y2 , 0), α23 � fyb∗ Wc (0, 0, 0) has the following form:
(x2 , y2 , 0), α111 � fxxx (x2 , y2 , 0), α112 � fxxy (x2 , y2 , 0),
α113 � fxxb∗ (x2 , y2 , 0), α123 � fxyb∗ (x2 , y2 , 0), β1 � gx (x2 , xn+1 � − xn + h1 x2n + h2 xn b∗ + h3 x2n b∗ + h4 xn b∗2 + h5 x3n
y2 , 0), β2 � gy (x2 , y2 , 0), β11 � gxx (x2 , y2 , 0), β12 � gxy (x2 , 􏼌􏼌 􏼌􏼌 􏼌􏼌 􏼌􏼌 3
y2 , 0), β22 � gyy (x2 , y2 , 0), β13 � gxb∗ (x2 , y2 , 0), β23 � gyb∗ + O􏼒􏼐􏼌􏼌xn 􏼌􏼌 + 􏼌􏼌b∗ 􏼌􏼌􏼑 􏼓 ≡ F xn , b∗ 􏼁,
(x2 , y2 , 0), β111 � gxxx (x2 , y2 , 0), β113 � gxxb∗ (x2 , y2 , 0), (23)
β123 � gxyb∗ (x2 , y2 , 0), and β223 � gyyb∗ (x2 , y2 , 0).
α2 α2 where
We define T � 􏼢 􏼣, and it is obvious that
− 1 − α1 λ2 − α1
u x
T is invertible. Using the transformation 􏼢 n 􏼣 � T􏼢 n 􏼣,
vn yn
then model (19) becomes

2
α2 􏼂 λ2 − α1 􏼁α11 − α2 β11 􏼃 β22 1 + α1 􏼁 1 + α1 􏼁􏼂 λ2 − α1 􏼁α12 − α2 β12 􏼃
h1 � − − ,
1 + λ2 1 + λ2 1 + λ2

λ2 − α1 􏼁α13 − α2 β13 1 + α1 􏼁􏼂 λ2 − α1 􏼁α23 − α2 β23 􏼃

h2 � − ,
1 + λ2 α2 1 + λ2 􏼁
2
􏼂 λ2 − α1 􏼁α23 − α2 β23 􏼃 λ2 − α1 􏼁α1 λ − α1 􏼁α1 α13 − α2 β13 + α2 􏼂 λ2 − α1 􏼁α113 − α2 β113 􏼃 − β223 1 + α1 􏼁
h3 � + 2
α2 1 + λ2 􏼁 1 + λ2

2α2 α2 􏼂 λ2 − α1 􏼁α11 − α2 β11 􏼃 1 + α1 􏼁􏼂 λ2 − α1 􏼁α123 − α2 β123 􏼃 2β22 α2 1 + α1 􏼁 λ2 − α1 􏼁

+ − −
1 + λ2 1 + λ2 1 + λ2

α2 􏼂 λ2 − α1 􏼁α12 − α2 β12 􏼃 λ2 − 1 − 2α1 􏼁

+ ,
1 + λ2
Mathematical Problems in Engineering 7

α2 􏼂 λ2 − α1 􏼁α13 − α2 β13 􏼃 􏼂 λ2 − α1 􏼁α23 − α2 β23 􏼃 λ2 − α1 􏼁α2 2α2 α2 􏼂 λ2 − α1 􏼁α11 − α2 β11 􏼃

h4 � + +
1 + λ2 α2 1 + λ2 􏼁 1 + λ2

2β22 α2 1 + α1 􏼁 λ2 − α1 􏼁 α2 􏼂 λ2 − α1 􏼁α12 − α2 β12 􏼃 λ2 − 1 − 2α1 􏼁

+ + ,
1 + λ2 1 + λ2

2α2 α1 􏼂 λ2 − α1 􏼁α11 − α2 β11 􏼃 + α22 􏼂 λ2 − α1 􏼁α111 − α2 β111 􏼃 2β22 α1 λ2 − α1 􏼁 1 + α1 􏼁

h5 � +
1 + λ2 1 + λ2

􏼂 λ2 − α1 􏼁α11 − α2 β11 􏼃 λ2 − 1 − 2α1 􏼁α1 α2 1 + α1 􏼁􏼂 λ2 − α1 􏼁α112 − α2 β112 􏼃

+ − . (24)
1 + λ2 1 + λ2

c(1 − b)xn yn
⎜ axn 1 − xn 􏼁 − ⎟
For flip bifurcation, we require the two discriminatory ⎛
⎜
⎜
⎜
⎜ 􏼈 1 + α(1 − b)xn 􏼉 1 + βyn 􏼁 ⎞
⎟
⎟
⎟
⎟
quantities ξ 1 and ξ 2 to be nonzero: ⎜
⎜ ⎟
⎟
⎟
⎜
F Xn 􏼁 � ⎜
⎜ ⎟
⎟ ,
⎜
⎜ ⎟
⎟
⎟
􏼌 ⎜
⎜ ⎟
z2 F 1 zF z2 F 􏼌􏼌􏼌􏼌 ⎝ d(1 − b)xn yn ⎟
⎠
ξ 1 �􏼠 + 􏼡 􏼌 ,
zxzb∗ 2 zb∗ zx2 􏼌􏼌(0,0) 􏼈1 + α(1 − b)xn 􏼉 1 + βyn 􏼁
􏼌 (25) T
3 2 2 􏼌􏼌 X n � xn y n 􏼁 .
1 z F 1 z F 􏼌
⎝
ξ 2 �⎛ ⎞􏼌􏼌􏼌 .
⎠
3 +􏼠 2􏼡 􏼌􏼌 (26)
6 zx 2 zx 􏼌(0,0)
The eigenvalues corresponding with the fixed point P2 �
Finally, from the above analysis, we have the following (x2 , y2 ) are given by
result. 􏽱�������������������
− p x2 , y2 􏼁 ± p2 x2 , y2 􏼁 − 4q x2 , y2 􏼁
λ1,2 � , (27)
Theorem 2. If ξ 1 ≠ 0 and ξ 2 ≠ 0, then model (3) undergoes 2
PDB at P2 (x2 , y2 ) when the parameter b varies. Moreover, if where
ξ 2 > 0 (ξ 2 < 0), then the period-2 points that bifurcation from
P2 (x2 , y2 ) are stable (unstable). p x2 , y2 􏼁 � − 􏼂a 1 − 2x2 􏼁 − cQ2 + dR2 􏼃,

q x2 , y2 􏼁 � ad 1 − 2x2 􏼁R2 ,
5. Existence of Marottos Chaos
(1 − b)x2 (28)
R2 � 2,
This section presents chaotic nature of system (3) in the 􏼈1 + α(1 − b)x2 􏼉 1 + βy2 􏼁
sense of Marrotto.
(1 − b)y2
Q2 � 2 .
n n
Definition 1. Let the function F: R ⟶ R be differentiable 􏼈1 + α(1 − b)x2 􏼉 1 + βy2 􏼁
in Br (Z). The point Z ∈ Rn is an expanding fixed point of F
in Br (Z), if F(Z) � Z and all eigenvalues of DF(X) exceed 1 Therefore, the fixed point P2 (x2 , y2 ) has a pair of
in norm for all X ∈ Br (Z). complex eigenvalues, and the norm of them exceeds unity if
p2 (x2 , y2 ) − 4q(x2 , y2 ) < 0 and q(x2 , y2 ) − 1 > 0
Definition 2. Assume that Z is an expanding fixed point of F i.e., 4ad(1 − 2x2 )R2 > max(4, [a(1 − 2x2 ) − cQ2 + d
in Br (Z) for some r > 0. Then, Z is said to be a snap-back R2 ]2 ).
repeller of F if there exists a point X0 ∈ Br (Z) with X0 ≠ Z, Thus, we can state the following theorem. □
FM (X0 ) � Z, and DFM (X0) ≠ 0 for some positive integer
M.
Theorem 4. P2 (x2 , y2 ) is a snap-back repeller in U2 .

Theorem 3. If P2 ∈ U2 � 􏼈(x, y): 4ad(1 − 2x2 )R2 > max

(4, [a(1 − 2x2 ) − cQ2 + dR2 ]2 )}, then P2 is an expanding Proof. According to the definition of a snap-back repeller,
fixed point of F. one needs to find one point P∗ (x0 , y0 ) ∈ U2 such that
P∗ ≠ P2 , FM (P∗ ) � P2 , and |DFM (P∗ )| ≠ 0, for some posi-
tive integer M, where Map F is defined by (3).
Proof. For map, To proceed, notice that
8 Mathematical Problems in Engineering

c(1 − b)x0 y0 y2 1 + βy1 􏼁

x1 � ax0 1 − x0 􏼁 − , x1 � ,
􏼈1 + α(1 − b)x0 􏼉 1 + βy0 􏼁 (1 − b)􏼂dy1 − αy2 1 + βy1 􏼁􏼃
(29) (31)
d(1 − b)x0 y0 􏼂1 + α(1 − b)x1 􏼃 1 + βy1 􏼁􏼂ax1 1 − x1 􏼁 − x2 􏼃
y1 � , y1 � .
􏼈1 + α(1 − b)x0 􏼉 1 + βy0 􏼁 c(1 − b)x1

c(1 − b)x1 y1 Substituting x1 and y1 into equation (29) and solving

x2 � ax1 1 − x1 􏼁 − , x0 , y0 , we have
􏼈1 + α(1 − b)x1 􏼉 1 + βy1 􏼁
(30) y1 1 + βy0 􏼁
d(1 − b)x1 y1 x0 � ,
y2 � . (1 − b)􏼂dy0 − αy1 1 + βy0 􏼁􏼃
􏼈1 + α(1 − b)x1 􏼉 1 + βy1 􏼁
(32)
Now, a map F2 has been constructed to map the point 􏼂1 + α(1 − b)x0 􏼃 1 + βy0 􏼁􏼂ax0 1 − x0 􏼁 − x1 􏼃
y0 � .
P∗ � (x0 , y0 ) to the fixed point P2 (x2 , y2 ) after two itera- c(1 − b)x0
tions if there are solutions different from P2 for equations
(29) and (30). The solutions different from P2 for equation By simple calculations, we obtain
(30) satisfy the following equation:

􏼌􏼌 2 􏼌
􏼌􏼌DF P∗ 􏼁􏼌􏼌􏼌 � 􏼢a(B − 2AB) − [1 + α(1 − b)A][1 + βD][BD + AE]c(1 − b) − c(1 − b)AD[[α(1 − b)B][1 + βD] +[1 + α(1 − b)A]βE]􏼣
[1 + α(1 − b)A]2 [1 + βD]2

[1 + α(1 − b)A][1 + βD]d(1 − b)[CD + AF] − d(1 − b)AD[[α(1 − b)C][1 + β D] +[1 + α(1 − b)A]βF]
×􏼢 􏼣
[1 + α(1 − b)A]2 [1 + βD]2

[1 + α(1 − b)A][1 + βD]c(1 − b)[CD + AF] − c(1 − b)AD[[α(1 − b)C][1 + βD] +[1 + α(1 − b)A]βF]
− 􏼢a(C − 2AC) − 􏼣
[1 + α(1 − b)A]2 [1 + βD]2

[1 + α(1 − b)A][1 + βD][BD + AE]d(1 − b) − d(1 − b)AD[[α(1 − b)B][1 + βD] +[1 + α(1 − b)A]βE]
×􏼢 􏼣,
[1 + α(1 − b)A]2 [1 + βD]2
(33)

where Theorem 5. If 4ad(1 − 2x2 )R2 > max(4, [a(1 − 2x2 ) −

c(1 − b)xy cQ2 + dR2 ]2 ), the solutions (x1 , y1 ) and (x0 , y0 ) of equations
A � ax(1 − x) − , (31) and (32) satisfy in addition (x0 , y0 ), (x1 , y1 ) ≠ (x2 , y2 ),
{1 + α(1 − b)x}(1 + βy)
(x0 , y0 ) ∈ U2 , (x0 , y0 ) � (0, 0), and |DF2 (P∗ )| ≠ 0, then
cy P2 (x2 , y2 ) is a snap-back repeller of Map (3), and hence Map
B � a(1 − 2x) − , (3) is chaotic in the sense of Marotto.
{1 + α(1 − b)x}2 (1 + βy)

c(1 − b)x 6. Chaos Control of the Proposed System

C�− ,
{1 + α(1 − b)x}(1 + βy)2
(34) Several methods can be used for obtaining chaos control in
d(1 − b)xy discrete-time models. A few of these methods are the state
D� , feedback method, pole-placement technique, and hybrid
{1 + α(1 − b)x}(1 + βy)
control method. We give a feedback control method to
dy stabilize chaotic orbits at an unstable positive fixed point of
E� , the prey-predator system (3), with the following controlled
{1 + α(1 − b)x}2 (1 + βy)
form:
d(1 − b)x c(1 − b)xn yn
F� . xn+1 � axn 1 − xn 􏼁 − + S,
{1 + α(1 − b)x}(1 + βy)2 􏼈1 + α(1 − b)xn 􏼉 1 + βyn 􏼁
The solutions of equations (31) and (32) will farther d(1 − b)xn yn
subject to (x0 , y0 ), (x1 , y1 ) ≠ (x2 , y2 ), (x0 , y0 ) ∈ U2 and yn+1 � ,
􏼈1 + α(1 − b)xn 􏼉 1 + βyn 􏼁
|DF2 (P∗ )| ≠ 0, then P2 (x2 , y2 ) is a snap-back repeller in U2 .
Thus, the following theorem is established. □ (35)
Mathematical Problems in Engineering 9

1 1
b = 0.00 b = 0.01

0.8 0.8
Predator

Predator
0.6 0.6

0.4 0.4

0.2 0.2
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Prey Prey
(a) (b)
1 1
b = 0.02 b = 0.03

0.8 0.8
Predator

Predator
0.6 0.6

0.4 0.4

0.2 0.2
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Prey Prey
(c) (d)

Figure 1: Phase portraits for b � 0.0, 0.01, 0.02, 0.03.

with the following feedback control law as the control force: λ1 + λ2 � a11 + a22 − q1 ,
(40)
S � − q1 xn − x2 􏼁 − q2 yn − y2 􏼁, (36) λ1 λ2 � a22 a11 − q1 􏼁 − a21 a12 − q2 􏼁.

where q1 and q2 are the feedback gain and (x2 , y2 ) is The lines of marginal stability are determined by solving
a positive fixed point of the model. the equations λ1 � ± 1 and λ1 λ2 � 1. These conditions
The Jacobian matrix J for system (35) at (x2 , y2 ) is guarantee that the eigenvalues λ1 and λ2 have modulus less
than 1.
a11 − q1 a12 − q2 Suppose λ1 λ2 � 1, from (40), we have line l1 as follows:
J �􏼢 􏼣, (37)
a21 a22 a22 q1 − a21 q2 � a22 a11 − a21 a12 − 1. (41)
where Suppose λ1 � ± 1, from (40), we have line l2 and l3 as
c(1 − b)y follows:
a11 � a(1 − 2x) − ,
{1 + α(1 − b)x}2 (1 + βy) 1 − a22 􏼁q1 + a21 q2 � a11 + a22 − 1 − a22 a11 + a21 a12 ,

c(1 − b)x 1 + a22 􏼁q1 − a21 q2 � a11 + a22 + 1 + a22 a11 − a21 a12 .
a12 � − , (42)
{1 + α(1 − b)x}(1 + βy)2
(38) The stable eigenvalues lie within a triangular region by
d(1 − b)y
a21 � , line l1 , l2 , and l3 .
{1 + α(1 − b)x}2 (1 + βy)

d(1 − b)x 7. Numerical Simulations

a22 � .
{1 + α(1 − b)x}(1 + βy)2 This section presents the qualitative dynamical behaviors of
The corresponding characteristic equation of matrix J is the system by presenting Lyapunov exponent, bifurcation
diagram, and phase plane for specific parameter values. We
λ2 − a11 + a22 − q1 􏼁λ + a22 a11 − q1 􏼁 − a21 a12 − q2 􏼁. present the numerical simulation to observe rich dynamics
(39) of our model, so we consider the hypothetical parameters
values of the proposed discrete prey-predator system. The
Let λ1 and λ2 are the eigenvalues of the above charac- value of the parameters are consider as a � 4.2, b � 0.12,
teristic equation: c � 3.0, d � 3.5, α � 0.1, and β � 0.1 with initial population
10 Mathematical Problems in Engineering

1 1
b = 0.05 b = 0.06

0.8 0.8
Predator

Predator
0.6 0.6

0.4
0.4
0.1 0.2 0.3 0.4 0.5 0.6 0.2 0.3 0.4 0.5 0.6
Prey Prey
(a) (b)
1 b = 0.10
b = 0.065
0.8

0.8 0.7

Predator
Predator

0.6
0.6
0.5

0.4
0.4 0.2 0.25 0.3 0.35 0.4 0.45 0.5
0.2 0.3 0.4 0.5 0.6
Prey
Prey
(c) (d)

Figure 2: Phase portraits for b � 0.05, 0.06, 0.065, 0.10.

0.75 b = 0.150 0.74

b = 0.152

0.72
Predator

Predator

0.7

0.7

0.65 0.68
0.34 0.36 0.38 0.4 0.36 0.37 0.38 0.39
Prey Prey

(a) (b)
0.72 0.71098215
b = 0.153 b = 0.160

0.715 0.7109821
Predator

Predator

0.71 0.71098205

0.705 0.710982

0.7 0.71098195
0.365 0.37 0.375 0.38 0.3758202 0.37582025 0.3758203 0.37582035
Prey Prey

(c) (d)

Figure 3: Phase portraits for b � 0.150, 0.152, 0.153, 0.160.

Mathematical Problems in Engineering 11

1
0.5

Predator density
Prey density

0.8
0.4
0.6

0.3 0.4

0.2 0.2
3 3.2 3.4 3.6 3.8 4 4.2 4.4 4.6 3 3.2 3.4 3.6 3.8 4 4.2 4.4 4.6
a (bifurcation parameter) a (bifurcation parameter)
(a) (b)
1

0.5 0.9

Predator density
Prey density

0.8
0.4
0.7
0.3
0.6
0.2
4.38 4.4 4.42 4.44 4.46 4.48 4.5 0.5
4.38 4.4 4.42 4.44 4.46 4.48 4.5
a (bifurcation parameter)
a (bifurcation parameter)
(c) (d)

Figure 4: Orbit diagram of species for parameter a and local amplification.

0.8 1.2

1
0.6
Predator density
Prey density

0.8
0.4
0.6
0.2
0.4

0 0.2
0 0.05 0.1 0.15 0.2 0 0.05 0.1 0.15 0.2
b (bifurcation parameter) b (bifurcation parameter)
(a) (b)
0.8 1

0.6 0.8
Predator density
Prey density

0.4 0.6

0.2 0.4

0 0.2
0.01 0.02 0.03 0.04 0.05 0.06 0.01 0.02 0.03 0.04 0.05 0.06
b (bifurcation parameter) b (bifurcation parameter)
(c) (d)

Figure 5: Orbit diagram of species for parameter b and local amplification.

(0.6, 0.3) for simulation and graphical presentation of dif- and predator for varying parameters a, b, c, d, α, and β,
ferent dynamical observations. Figures 1–3 are iterative plots respectively.
for different refuge parameters, and we observed that refuge The phase portraits associated with Figure 5 are dis-
has a stabilizing effect. Figures 4–9 are bifurcation diagrams played in Figures 1–3. A chaotic attractor for b � 0.00 and
and corresponding local amplification diagram of the prey 0.01. A five-pieces chaotic attractor for b � 0.02 and 0.03.
12 Mathematical Problems in Engineering

1.5
0.5

Predator density
Prey density

0.4
1

0.3

0.2 0.5
1.5 2 2.5 3 3.5
1.5 2 2.5 3 3.5
c (bifurcation parameter)
c (bifurcation parameter)
(a) (b)
0.5 1

0.45 0.9

Predator density
Prey density

0.4 0.8

0.35 0.7

0.3 0.6

0.25 0.5
2.8 2.85 2.9 2.95 3 3.05 2.8 2.85 2.9 2.95 3 3.05
c (bifurcation parameter) c (bifurcation parameter)
(c) (d)

Figure 6: Orbit diagram of species for parameter c and local amplification.

0.8 1.2

1
0.6
Predator density
Prey density

0.8
0.4
0.6
0.2
0.4

0 0.2
3 3.2 3.4 3.6 3.8 4 3 3.2 3.4 3.6 3.8 4
d (bifurcation parameter) d (bifurcation parameter)
(a) (b)
0.8 1.2

1
0.6
Predator density
Prey density

0.8
0.4
0.6
0.2
0.4

0 0.2
3.7 3.75 3.8 3.85 3.9 3.95 3.7 3.75 3.8 3.85 3.9 3.95
d (bifurcation parameter) d (bifurcation parameter)
(c) (d)

Figure 7: Orbit diagram of species for parameter d and local amplification.

Period-10 orbit for b � 0.05. Period-68 orbit for b � 0.06. The orbit diagram of species for a, which is a scaling
Attracting invariant cycle for b � 0.065, 0.10, 0.15, and 0.152. parameter of the intrinsic per capita growth rate of prey
Fixed point for b � 0.153 and 0.160. population. Figure 4 shows that the system trajectory evolves
Mathematical Problems in Engineering 13

0.9

0.5 0.8

Predator density
Prey density

0.7
0.4
0.6
0.3
0.5
0.2
0 0.05 0.1 0.15 0.2 0.25 0.3 0.4
0 0.05 0.1 0.15 0.2 0.25 0.3
Bifurcation parameter Bifurcation parameter
(a) (b)
0.6 0.9

0.8
0.5

Predator density
Prey density

0.7
0.4
0.6
0.3
0.5

0.2 0.4
0.014 0.016 0.018 0.02 0.022 0.014 0.016 0.018 0.02 0.022
Bifurcation parameter Bifurcation parameter
(c) (d)

Figure 8: Orbit diagram of species for parameter α and local amplification.

1
0.6
0.8
Predator density
Prey density

0.5
0.4
0.6
0.3
0.2 0.4

0.1
0 0.05 0.1 0.15 0.2 0.25 0.3 0.2
0 0.05 0.1 0.15 0.2 0.25 0.3
Bifurcation parameter
Bifurcation parameter
(a) (b)
0.6 0.9

0.5 0.8
Predator density
Prey density

0.4 0.7

0.3 0.6

0.2 0.5
0.095 0.1 0.105 0.11 0.095 0.1 0.105 0.11
Bifurcation parameter Bifurcation parameter
(c) (d)

Figure 9: Orbit diagram of species for parameter β and local amplification.

from a fixed point to NSB and finally into a chaotic attractor. Figure 10 is the corresponding Lyapunov exponents of
There is one visible periodic windows in the bifurcation Figure 4.
diagram for both prey and predator. We also give local Figure 5 shows the orbit diagram of prey and predator
magnification of the corresponding bifurcation figure. population for the refuge parameter (b) with other fixed
14 Mathematical Problems in Engineering

0.05
0
–0.05
–0.1
Lyapunov exponents

−0.15
–0.2
–0.25
–0.3
–0.35
–0.4
–0.45
–0.5
3 3.2 3.4 3.6 3.8 4 4.2 4.4 4.6 4.8
Parameter a
Figure 10: Lyapunov exponents of the system for a.

0.1

0
Lyapunov exponents

–0.1

–0.2

–0.3

–0.4
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2
Parameter b
Figure 11: Lyapunov exponents of the system for b.

0.05

–0.05

–0.1
Lyapunov exponents

–0.15

–0.2

–0.25

–0.3

–0.35

–0.4

–0.45
1.5 2 2.5 3 3.5
Parameter c
Figure 12: Lyapunov exponents of the system for c.
Mathematical Problems in Engineering 15

0.1

Lyapunov exponents
0

–0.1

–0.2

–0.3

–0.4
3 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 4
Parameter d
Figure 13: Lyapunov exponents of the system for d.

0.05

–0.05
Lyapunov exponents

–0.1

–0.15

–0.2

–0.25

–0.3

–0.35

–0.4
0 0.05 0.1 0.15 0.2 0.25 0.3
Parameter e
Figure 14: Lyapunov exponents of the system for α.

0.05

–0.05
Lyapunov exponents

–0.1

–0.15

–0.2

–0.25

–0.3

–0.35

–0.4
0 0.05 0.1 0.15 0.2 0.25 0.3
Parameter f
Figure 15: Lyapunov exponents of the system for β.
16 Mathematical Problems in Engineering

1 0.8

0.9
0.7

0.8
0.6
0.7
Population density

Population density
0.6 0.5

0.5 0.4

0.4
0.3
0.3

0.2
0.2

0.1 0.1
0 20 40 60 80 100 0 20 40 60 80 100
Time Time
Prey Prey
Predator Predator
(a) (b)

Figure 16: Chaos control.

parameter values. The system trajectory evolves into a periodic process for both prey and predator. We also give local
orbit from the chaotic attractor and finally into a fixed point. magnification of the corresponding bifurcation figure. Fig-
There is only one obvious periodic window in all the bi- ure 15 is the corresponding Lyapunov exponent of Figure 9.
furcation process for both prey and predator. We also give local
magnification of the corresponding bifurcation figure. Fig-
ure 11 is the corresponding Lyapunov exponent of Figure 5. 7.1. Chaos Control. We observe chaotic behavior of prey-
Figure 6 shows the orbit diagram of the population for the predator, as shown in Figure 16(a). In this case fixed point
parameter (c) of per capita consumption rate of predators. (0.3183, 0.6919) is unstable. In the feedback control method
The system trajectory evolves from a fixed point to NSB and for feedback gain q1 � 0.5 and q2 � − 0.5, we observe the
finally into a chaotic attractor. We also give local magnifi- fixed point (0.3183, 0.6919) is stable, as shown in
cation of the corresponding bifurcation figure. Figure 12 is the Figure 16(b).
corresponding Lyapunov exponent of Figure 6.
Figure 7 shows the orbit diagram of species for the pa-
7.2. Fractal Dimension. The fractal dimension is defined by
rameter (d) of efficiency with which predators convert the
using Lyapunov exponents as follows.
consumed prey into a new predator. The system trajectory
Let f be a map on Rm . Consider an orbit with Lyapunov
evolves from a fixed point to NSB and finally into a chaotic
exponents h1 ≥ h2 ≥ . . . . ≥ hm , and let p denote the largest
attractor. There is one big visible periodic window in the i�p
integer such that 􏽐i�1 hi ≥ 0.
bifurcation diagram for both prey and predator. We also give
local magnification of the corresponding bifurcation figure. ⎪
⎧
⎪ 0, if no such p exist,
⎪
⎪
Figure 13 is the corresponding Lyapunov exponent of Figure 7. ⎪
⎪
⎪
⎪
Figure 8 shows the orbit diagram of prey and predator ⎪
⎪ i�p
⎨ 􏽐 hi
population for the parameter α with other fixed parameter Fractal dimension dL � ⎪ p + 􏼌􏼌􏼌 i�1 􏼌􏼌􏼌 , if p < m,
⎪
⎪ 􏼌􏼌hp+1 􏼌􏼌
values. The system trajectory evolves into a fixed point from ⎪
⎪
⎪
⎪
the periodic orbit. We also give local magnification of the ⎪
⎪
⎪
⎩ m,
corresponding bifurcation figure. Figure 14 is the corre- if p � m.
sponding Lyapunov exponent of Figure 8. (43)
Figure 9 shows the orbit diagram of prey and predator
population for the parameter β with other fixed parameter Fractal dimension and Lyapunov exponents for different
values. The system trajectory evolves into a periodic orbit values of b for the value of the parameters as a � 4.2, c � 3.0,
from the chaotic attractor and finally into a fixed point. There d � 3.5, α � 0.1, and β � 0.1 with initial population (0.6, 0.3)
is only one obvious periodic window in all the bifurcation are presented in Table 1.
Mathematical Problems in Engineering 17

Table 1: Fractal dimension and Lyapunov exponents for different [7] T. W. Anderson, “Predator responses, prey refuges, and
values of b. density-dependent mortality of a marine fish,” Ecology,
vol. 82, no. 1, pp. 245–257, 2001.
Value of b Lyapunov exponents Fractal dimension
[8] E. González-Olivares and R. Ramos-Jiliberto, “Dynamic
b � 0.0 h1 � 0.1295 2 consequences of prey refuges in a simple model system: more
h2 � 0.0251 prey, fewer predators and enhanced stability,” Ecological
Modelling, vol. 166, no. 1-2, pp. 135–146, 2003.
b � 0.01 h1 � 0.0956 2
[9] Y. Huang, F. Chen, and L. Zhong, “Stability analysis of a prey-
h2 � 0.0381
predator model with holling type III response function in-
b � 0.02 h1 � 0.0538 2.2036 corporating a prey refuge,” Applied Mathematics and Com-
h2 � − 0.0447 putation, vol. 182, no. 1, pp. 672–683, 2006.
[10] S. Magalhães, P. C. J. van Rijn, M. Montserrat, A. Pallini, and
b � 0.03 h1 � 0.0383 1.3865 M. W. Sabelis, “Population dynamics of thrips prey and their
h2 � − 0.0991 mite predators in a refuge,” Oecologia, vol. 150, no. 4,
pp. 557–568, 2007.
[11] V. H. W. Rudolf and J. Armstrong, “Emergent impacts of
8. Conclusion cannibalism and size refuges in prey on intraguild predation
systems,” Oecologia, vol. 157, no. 4, pp. 675–686, 2008.
In this paper, we have studied the effects of refuges on [12] Z. Ma, W. Li, Y. Zhao, W. Wang, H. Zhang, and Z. Li, “Effects
a predator-prey interaction, by using the analytical and of prey refuges on a predator-prey model with a class of
graphical approach. The refuges are considered as prey functional responses: the role of refuges,” Mathematical
refuge proportional to prey density. We evaluate the effects Biosciences, vol. 218, no. 2, pp. 73–79, 2009.
with regard to the local stability of the interior equilibrium [13] F. Chen, L. Chen, and X. Xie, “On a Leslie-Gower predator-
point and the long-term dynamics of the interacting pop- prey model incorporating a prey refuge,” Nonlinear Analysis:
ulations. The results show that the effects of refuges can Real World Applications, vol. 10, no. 5, pp. 2905–2908, 2009.
stabilize the interior equilibrium point of the proposed prey- [14] R. Cressman and J. Garay, “A predator-prey refuge system:
predator discrete-time domain model. We have studied evolutionary stability in ecological systems,” Theoretical
bifurcations in a discrete predator-prey model with refuge. Population Biology, vol. 76, no. 4, pp. 248–257, 2009.
[15] Y. Tao, X. Wang, and X. Song, “Effect of prey refuge on
Furthermore, it is also shown that the model exhibits various
a harvested predator-prey model with generalized functional
bifurcations of codimension 1, including period-doubling
response,” Communications in Nonlinear Science and Nu-
bifurcation, and Neimark–Sacker bifurcation as the values of merical Simulation, vol. 16, no. 2, pp. 1052–1059, 2011.
parameters vary. Numerical simulations and graphical [16] Y. Wang and J. Wang, “Influence of prey refuge on predator-
presentation show the rich dynamics in bifurcation and prey dynamics,” Nonlinear Dynamics, vol. 67, no. 1,
chaotic nature of the system model. pp. 191–201, 2012.
[17] S. Sarwardi, P. K. Mandal, and S. Ray, “Analysis of a com-
Data Availability petitive prey-predator system with a prey refuge,” Biosystems,
vol. 110, no. 3, pp. 133–148, 2012.
No data were used to support this study. [18] J. Zhao, M. Zhao, and H. Yu, “Complex dynamical behavior of
a predator-prey system with group defense,” Mathematical
Problems in Engineering, vol. 2013, Article ID 910349, 8 pages,
Conflicts of Interest 2013.
[19] Z. Ma, S. Wang, W. Li, and Z. Li, “The effect of prey refuge in
The authors declare that there are no conflicts of interest a patchy predator-prey system,” Mathematical Biosciences,
regarding the publication of this paper. vol. 243, no. 1, pp. 126–130, 2013.
[20] D. Pal and G. S. Mahapatra, “A bioeconomic modeling of two-
References prey and one-predator fishery model with optimal harvesting
policy through hybridization approach,” Applied Mathematics
[1] A. A. Berryman, “The orgins and evolution of predator-prey and Computation, vol. 242, pp. 748–763, 2014.
theory,” Ecology, vol. 73, no. 5, pp. 1530–1535, 1992. [21] M. Haque, M. S. Rahman, E. Venturino, and B.-L. Li, “Effect
[2] A. R. Hausrath, “Analysis of a model predator-prey system of a functional response-dependent prey refuge in a predator-
with refuges,” Journal of Mathematical Analysis and Appli- prey model,” Ecological Complexity, vol. 20, pp. 248–256,
cations, vol. 181, no. 2, pp. 531–545, 1994. 2014.
[3] M. E. Hochberg and R. D. Holt, “Refuge evolution and the [22] R. N. Fan, “A predator-prey model incorporating prey refuge
population dynamics of coupled host-parasitoid associa- and Allee effect,” Applied Mechanics and Materials,
tions,” Evolutionary Ecology, vol. 9, no. 6, pp. 633–661, 1995. vol. 713–715, pp. 1534–1539, 2015.
[4] J. Maynard Smith, Models in Ecology, Cambridge University [23] D. Jana, R. Agrawal, and R. K. Upadhyay, “Dynamics of
Press, Cambridge, UK, 1974. generalist predator in a stochastic environment: effect of
[5] J. Guckenheimer and P. Holmes, Nonlinear Oscillations, delayed growth and prey refuge,” Applied Mathematics and
Dynamical Systems and Bifurcations of Vector Fields, Computation, vol. 268, pp. 1072–1094, 2015.
Springer-Verlag, Berlin, Germany, 1983. [24] G. S. Mahapatra and P. Santra, “Prey-predator model for
[6] G. D. Ruxton, “Short term refuge use and stability of predator- optimal harvesting with functional response incorporating
prey models,” Theoretical Population Biology, vol. 47, no. 1, prey refuge,” International Journal of Biomathematics, vol. 9,
pp. 1–17, 1995. no. 1, Article ID 1650014, 2016.
18 Mathematical Problems in Engineering

[25] D. Pal and G. S. Mahapatra, “Effect of toxic substance on [44] H. Baek, “Complex dynamics of a discrete-time predator-prey
delayed competitive allelopathic phytoplankton system with system with ivlev functional response,” Mathematical Prob-
varying parameters through stability and bifurcation analy- lems in Engineering, vol. 2018, Article ID 8635937, 15 pages,
sis,” Chaos, Solitons & Fractals, vol. 87, pp. 109–124, 2016. 2018.
[26] P. Santra, G. S. Mahapatra, and D. Pal, “Prey-predator [45] J. Huang, S. Liu, S. Ruan, and D. Xiao, “Bifurcations in
nonlinear harvesting model with functional response in- a discrete predator-prey model with nonmonotonic func-
corporating prey refuge,” International Journal of Dynamics tional response,” Journal of Mathematical Analysis and Ap-
and Control, vol. 4, no. 3, pp. 293–302, 2016. plications, vol. 464, no. 1, pp. 201–230, 2018.
[27] J. Ghosh, B. Sahoo, and S. Poria, “Prey-predator dynamics [46] R. K. Upadhyay and R. K. Naji, “Dynamics of a three species
with prey refuge providing additional food to predator,” food chain model with Crowley-Martin type functional re-
Chaos, Solitons & Fractals, vol. 96, pp. 110–119, 2017. sponse,” Chaos, Solitons & Fractals, vol. 42, no. 3,
[28] Z. Ma, S. Wang, T. Wang, and H. Tang, “Stability analysis of pp. 1337–1346, 2009.
prey-predator system with holling type functional response [47] J. Zhou, “Qualitative analysis of a modified Leslie-Gower
and prey refuge,” Advances in Difference Equations, vol. 2017, predator-prey model with Crowley-Martin functional re-
no. 1, p. 243, 2017. sponses,” Communications on Pure and Applied Analysis,
[29] D. Pal, P. Santra, and G. S. Mahapatra, “Dynamical behavior vol. 14, no. 3, pp. 1127–1145, 2015.
of three species predator-prey system with mutual support [48] J. P. Tripathi, S. Tyagi, and S. Abbas, “Global analysis of
between non refuge prey,” Ecological Genetics and Genomics, a delayed density dependent predator-prey model with
vol. 3–5, pp. 1–6, 2017. Crowley-Martin functional response,” Communications in
[30] Q. Wang, Z. Liu, X. Zhang, R. A. Cheke, and A. Cheke, Nonlinear Science and Numerical Simulation, vol. 30, p. 4,
“Incorporating prey refuge into a predator-prey system with 2016.
imprecise parameter estimates,” Computational and Applied [49] X. Li, X. Lin, and J. Liu, “Existence and global attractivity of
Mathematics, vol. 36, no. 2, pp. 1067–1084, 2017. positive periodic solutions for a predator-prey model with
[31] S. H. Levine, “Discrete time modeling of ecosystems with Crowley-Martin functional response,” Electronic Journal of
applications in environmental enrichment,” Mathematical Differential Equations, vol. 191, pp. 1–17, 2018.
Biosciences, vol. 24, no. 3-4, pp. 307–317, 1975.
[32] M. G. Neubert and M. Kot, “The subcritical collapse of
predator populations in discrete-time predator-prey models,”
Mathematical Biosciences, vol. 110, no. 1, pp. 45–66, 1992.
[33] J.-C. Huang, “Bifurcations and chaos in a discrete predator-
prey system with holling type-IV functional response,” Acta
Mathematicae Applicatae Sinica, English Series, vol. 21, no. 1,
pp. 157–176, 2005.
[34] X. Liu and D. Xiao, “Bifurcations in a discrete time Lotka-
Volterra predator-prey system,” Discrete and Continuous
Dynamical Systems-Series B, vol. 6, no. 3, pp. 559–572, 2006.
[35] X. Liu and D. Xiao, “Complex dynamic behaviors of a dis-
crete-time predator-prey system,” Chaos, Solitons & Fractals,
vol. 32, no. 1, pp. 80–94, 2007.
[36] C. Çelik and O. Duman, “Allee effect in a discrete-time
predator-prey system,” Chaos, Solitons & Fractals, vol. 40,
no. 4, pp. 1956–1962, 2009.
[37] X. Liu, “A note on the existence of periodic solutions in
discrete predator-prey models,” Applied Mathematical
Modelling, vol. 34, no. 9, pp. 2477–2483, 2010.
[38] Z. Hu, Z. Teng, and L. Zhang, “Stability and bifurcation
analysis of a discrete predator-prey model with non-
monotonic functional response,” Nonlinear Analysis: Real
World Applications, vol. 12, no. 4, pp. 2356–2377, 2011.
[39] Z. He and X. Lai, “Bifurcation and chaotic behavior of
a discrete-time predator-prey system,” Nonlinear Analysis:
Real World Applications, vol. 12, no. 1, pp. 403–417, 2011.
[40] A.-E. A. Elsadany, “Dynamical complexities in a discrete-time
food chain,” Computational Ecology and Software, vol. 2,
no. 2, pp. 124–139, 2012.
[41] Q. Din, “Dynamics of a discrete Lotka-Volterra model,” Ad-
vances in Difference Equations, vol. 2013, no. 1, pp. 1–13, 2013.
[42] M. Zhao and Y. Du, “Stability of a discrete-time predator-prey
system with allee effect,” Nonlinear Analysis and Differential
Equations, vol. 4, no. 5, pp. 225–233, 2016.
[43] M. Gámez, I. López, C. Rodrı́guez, Z. Varga, and J. Garay,
“Ecological monitoring in a discrete-time prey-predator
model,” Journal of Theoretical Biology, vol. 429, no. 21,
pp. 52–60, 2017.
Advances in Advances in Journal of The Scientific Journal of
Operations Research
Hindawi
Decision Sciences
Hindawi
Applied Mathematics
Hindawi
World Journal
Hindawi Publishing Corporation
Probability and Statistics
Hindawi
www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 http://www.hindawi.com
www.hindawi.com Volume 2018
2013 www.hindawi.com Volume 2018

International
Journal of
Mathematics and
Mathematical
Sciences

Journal of

Hindawi
Optimization
Hindawi
www.hindawi.com Volume 2018 www.hindawi.com Volume 2018

Submit your manuscripts at

www.hindawi.com

International Journal of
Engineering International Journal of
Mathematics
Hindawi
Analysis
Hindawi
www.hindawi.com Volume 2018 www.hindawi.com Volume 2018

Journal of Advances in Mathematical Problems International Journal of Discrete Dynamics in

Complex Analysis
Hindawi
Numerical Analysis
Hindawi
in Engineering
Hindawi
Differential Equations
Hindawi
Nature and Society
Hindawi
www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018

International Journal of Journal of Journal of Abstract and Advances in

Stochastic Analysis
Hindawi
Mathematics
Hindawi
Function Spaces
Hindawi
Applied Analysis
Hindawi
Mathematical Physics
Hindawi
www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018