Existence of non-constant positive stationary solutions of the shadow predator-prey systems with Allee effect

Abstract

In this paper, we consider the dynamics of the shadow system of a kind of homogeneous diffusive predator-prey system with a strong Allee effect in prey. We mainly use the time-mapping methods to prove the existence and non-existence of the non-constant positive stationary solutions of the system in the one dimensional spatial domain. The problem is assumed to be subject to homogeneous Neumann boundary conditions.

MSC: 35K57, 35B09.

Bao and Liu Advances in Dierence Equations 2014, 2014:226 http://www.advancesindifferenceequations.com/content/2014/1/226

Web End =http://www.advancesindifferenceequations.com/content/2014/1/226

R E S E A R C H Open Access

Existence of non-constant positive stationary solutions of the shadow predator-prey systems with Allee effect

Zhenhua Bao1* and He Liu2

*Correspondence: [email protected]

1School of Mathematics, Liaoning Normal University, Dalian, 116029, ChinaFull list of author information is available at the end of the article

Abstract

In this paper, we consider the dynamics of the shadow system of a kind of homogeneous diusive predator-prey system with a strong Allee eect in prey. We mainly use the time-mapping methods to prove the existence and non-existence of the non-constant positive stationary solutions of the system in the one dimensional spatial domain. The problem is assumed to be subject to homogeneous Neumann boundary conditions.

MSC: 35K57; 35B09

Keywords: homogeneous diusive predator-prey system; strong Allee eect in prey; shadow system; non-constant positive stationary solutions

1 Introduction

In this paper, we are mainly concerned with the following homogeneous diusive predator-prey system with a strong Allee eect in prey:

u

t duxx = u( u)(ub ) muva+u, t > , x (, h),

v

t dvxx = dv + muva+u, t > , x (, h), ux(x, t) = ux(x, t) = vx(x, t) = vx(x, t) = , t > , x = , h, u(x, ) = u(x) , v(x, ) = v(x) , x (, h).

(.)

Here u = u(x, t) and v = v(x, t) stand for the densities of the prey and predator at time t > and a spatial position x (, h) with h (, ), respectively; d, d > are the diusion coecients of the species; d is the death rate of the predator, a measures the saturation eect, m is the strength of the interaction. The Allee threshold b is assumed to be smaller than . The strong Allee eect introduces a population threshold, and the population must surpass this threshold to grow. The boundary condition here is assumed to be homogeneous Neumann type, which implies that there is no ux for the populations on the boundary. For more details on the problem (.), we refer interested readers to [] and references therein.

In [], the authors considered the traveling wave solutions of system (.). More precisely, they showed that there is a non-negative traveling wave solution of system (.) connecting the semi-trivial solution (b, ) and the positive equilibrium solution (u, v). They also

2014 Bao and Liu; licensee Springer. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0

Web End =http://creativecommons.org/licenses/by/2.0 ), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Bao and Liu Advances in Dierence Equations 2014, 2014:226 Page 2 of 7 http://www.advancesindifferenceequations.com/content/2014/1/226

Web End =http://www.advancesindifferenceequations.com/content/2014/1/226

proved that, under certain suitable conditions, there is a small traveling wave train solution of system (.).

In [], the authors considered the non-existence of non-constant positive steady state solutions, and bifurcations of spatially homogeneous and non-homogeneous periodic solutions as well as non-constant steady state solutions are studied. These results allow for the phenomenon that the rich impact of the Allee eect essentially increases the system spatiotemporal complexity.

Although the existence and non-existence of non-constant steady state solutions of the system (.) has been considered in [] for nite diusion coecients, no results have been reported to consider the existence and non-existence of the positive non-constant steady state solutions for the shadow system corresponding to the system (.). The shadow system we mentioned here stands for the system where one of the diusion coecients tends to innity. The readers are referred to [] for the earlier contributions on the shadow systems.

Thus, the purpose of this paper is to consider the existence and non-existence of positive non-constant solutions of the following elliptic equations:

where d .

The methods we used in the paper are standard time-mapping methods (see [] and references therein for precise details on time-mapping methods). We hope that the results in the paper will allow for a clearer understanding of the rich dynamics of this particular pattern formation system. In Section , we state the derivation of the shadow system of the original reaction-diusion system (.). In Section , we study the existence of the non-constant stationary solutions of the shadow system; in Section , we end up our discussions by drawing some conclusions.

2 Derivation of the shadow system

Firstly, we state the following useful a priori estimate for the non-negative solutions of system (.) obtained in []:

Lemma . Suppose that d, d, a, b, d, m, h > , and that (u(x), v(x)) is a non-negative steady state solution of (.). Then either (u, v) is one of constant solutions: (, ), or (b, ), or for x [, h], (u(x), v(x)) satises

< u(x) < and < v(x) < ( b)

bd +

muva + u dv = . (.)

duxx = u( u)(ub ) muva+u, x (, h), dvxx = dv + muva+u, x (, h), ux(x) = ux(x) = vx(x) = vx(x) = , x = , h,

(.)

dd . (.)

For later use in our discussion, we rewrite the second equation of system (.) in the following way:

vxx + d

Bao and Liu Advances in Dierence Equations 2014, 2014:226 Page 3 of 7 http://www.advancesindifferenceequations.com/content/2014/1/226

Web End =http://www.advancesindifferenceequations.com/content/2014/1/226

Lemma . implies that the non-negative solutions of the system (.) is bounded. Then we know that vxx as d . Since our problem is of Neumann boundary condition

type, v is a constant, say . As d , there exists a positive number C = C(d) > , such

that

( b) bd +

d d <

( b)

bd + C(d).

Thus,

< < ( b)

bd + C(d).

Integrating the second equation from to , we obtain

mua + u d

dx = . (.)

Thus, system (.) reduces to the following single -parameterized scalar reaction-diusion equation:

duxx = u( u)(ub ) mua+u, u () = u (h) = ,

(.)

subject to the additional condition (.) and the condition

< u, < C := max

, ( b)

bd + C(d) . (.)

3 Existence of non-constant positive stationary solutions of the shadow system

In this section, we mainly concentrate on the existence of the non-constant positive solutions of the reduced shadow system (.).

For the purpose of our investigations, we dene

f (u) = u( u)

ub

mua + u and F(u) =

u

f (s) ds. (.)

Then we introduce the following energy functional:

E(x) = d

ux(x)

. (.)

From (.), we can nd that, for any x (, h), E (x) , and F(u(x)) < F() = F(),

where := u(), and := u(h).

It follows that if u = u(x) is a solution of (.), then F(u) must attain its local minimal value at a point in (, ).

We rewrite f (u) as

f (u) = mu

a + u

(u)

+ F

u(x)

,

Bao and Liu Advances in Dierence Equations 2014, 2014:226 Page 4 of 7 http://www.advancesindifferenceequations.com/content/2014/1/226

Web End =http://www.advancesindifferenceequations.com/content/2014/1/226

where

(u) := (a + u)( u)(u b)

kb . (.)

We have the following lemma on the properties of the function (u) dened above.

Lemma . For any u (, b) (, ), we have (u) < , while (u) > for any u (b, ).

In particular, (b) = () = . Moreover,. Suppose that ab + a b < holds. Then for any u (, ), (u) is increasing, while

for any u (, ) (, ), (u) is decreasing, where

:= b + a + (b

b + a

. Suppose that ab + a b holds. Then, for any u (, ), (u) is increasing, while

for any u (, ), (u) is decreasing.

Proof It is obvious that for any u (, b) (, ), we have (u) < , while (u) > for any

u (b, ). We can directly check that

(u) = u (a b )u + (ab + a b)

kb

=

kb(u

) u

+ ) km , (.)

with b < u() < < u+() < .

Since f (u) < for < u < u(), f (u) > for u() < u < u+(), it follows that F(u) is convex in (, u+()), and concave in (u(), ), and F(u) taking its local minimum value

at u = u(). In other words, F(u) is decreasing in (, u()) (u+(), ), and increasing

in (u(), u+()).

+ a) + (ab + a b) ,

:=

(b + a) + (ab + a b)

.

(.)

. (.)

Because > > provided that ab + a b < , we conclude from (.) that (u) is increasing ( (u) > ) for u (, ), while for any u (, ) (, ), (u) is decreasing

( (u) < ). The second part of the lemma can be proved similarly.

From Lemma ., it follows that (u) attains its maximum value := () at u = . If

> holds, then f (u) < for all u (, ). Thus, F(u) does not has its minimal value

point in (, ), which implies that the shadow system (.) does not possess positive non-

constant stationary solutions. Similarly, if = holds, then system (.) does not also possess positive non-constant stationary solutions.

Thus, in order for the shadow system to have non-constant positive stationary solutions, we need to concentrate on the case when (, ).

In this case, there exist two zeros of f (u) = , and we can denote them by

u() =

k (k

Bao and Liu Advances in Dierence Equations 2014, 2014:226 Page 5 of 7 http://www.advancesindifferenceequations.com/content/2014/1/226

Web End =http://www.advancesindifferenceequations.com/content/2014/1/226

Thus, the problem admits solutions for some h > if and only if (, ) and we are

now deriving the precise information on the suitable h > such that the problem has positive non-constant stationary solutions.

If F() F(u+()) holds, then there exists a unique (, u()), such that F() =

F(u+()). Dene

:=

, if F() F(u+()), , if F() F(u+()).

(.)

Then for any (, u()), there exists a unique (u(), u+()), such that F() =

F().

By the denition of E(x) = d(ux(x))/ + F(u(x)), and the fact that E (x) , we have

ux(x) =

F() F(u) /d > , x (, h). (.)

Fix u(h) = , and integrate (.); we have

h() =

d

du

(F() F(u)). (.)

In the following, we want to show the limit of h() as u(). In fact, following the

same argument in [] (say, for example, pp.-), one can verify that

lim

u() h(

) =

d

f (u()). (.)

Also following [], for a given number u (, ), we dene u = g(s) by the relation

F

g(s) F

u()

= s

,

. (.)

Then s = g(u) is well dened and is strictly increasing in (, ), since in this interval F(u) is convex and takes a strict minimum at u = u().

Let p > be given by

p = F(

) F

sign s = sign

u u()

= sign

f (u)

u() > . (.)

Then we have

h() =

p

p

p s =

g (p

g (s) ds

cos t) dt, (.)

where we make another change of variable, s = p cos t, t .

Bao and Liu Advances in Dierence Equations 2014, 2014:226 Page 6 of 7 http://www.advancesindifferenceequations.com/content/2014/1/226

Web End =http://www.advancesindifferenceequations.com/content/2014/1/226

By the same argument as on p. of [], we have

g (s) =

F(u)

|f (u)|

, F(u) = F(u) F

u()

,

g (s) = f f F

f (u), g () =

f

f

u()

,

(.)

g (s) = g (s)

f (u)H(u),

g () = (f (u()))/

f

f f u ()

,

where

H(u) := f (u)f (u)F(u) + f (u)

f (u) f (u)F(u)

. (.)

By (.), we have

f (u) = u( u)

ub m

u

a + u.

Then after direct calculations, we have

f (u) := u

b +

b + u m

a

(a + u) ;

f (u) :=

b u +

b + +

ma (a + u) ;

(.)

Clearly, the function f (u) = has a unique root u > since f () > , f () < and limu+ f (u) = .

On the other hand, by the properties of f , we know that f has two critical points in (, u()), denoted by c and c, that is to say, f (c) = f (c) = . Since f () < , limx f (u) = , and f (u) has a unique positive critical point u > , it follows that

c < u < u.

Thus, the results in Lemma . in [] hold true in our problem. That is, for any u

(, u()), we have H(u) < , which together with (.) implies that g (s) > .

By (.), we have dp/d < . By (.), we have

h (p) =

f (u) := b

ma (a + u) .

cos tg (s) dt, h (p) =

cos tg (s) dt, s = p cos t. (.)

Since

h () = dh()

dp = g ()

cos t dt = ,

and h (p) > due to g (s) > , it follows that h (p) > or equivalently dh/dp > . This together with the fact that dp/d < , we can conclude that dh/d < .

Bao and Liu Advances in Dierence Equations 2014, 2014:226 Page 7 of 7 http://www.advancesindifferenceequations.com/content/2014/1/226

Web End =http://www.advancesindifferenceequations.com/content/2014/1/226

Summarizing the analysis above, we can conclude the following.

Theorem . Let (u), be dened in (.) and (.), and u() be dened in (.). Then the shadow system (.) with the condition (.) has no non-constant positive stationary solutions if h := (), and has non-constant positive stationary solutions if and only

if (, min{ , C}) and

h > h :=

4 Conclusions

In this paper, we studied the existence and non-existence of the positive non-constant stationary solutions of a shadow system corresponding to a kind of diusive homogeneous predator-prey system with Holling type-II functional response and strong Allee eect in prey. We hope that the results in the paper will allow for the clearer understanding of the rich dynamics of this particular pattern formation system. Future work might include considering the qualitative behavior of the parabolic shadow system.

Competing interests

The authors declare that they have no competing interests.

Authors contributions

All authors have equal contributions. All authors read and approved the nal manuscript.

Author details

1School of Mathematics, Liaoning Normal University, Dalian, 116029, China. 2School of Mathematics, Physics and Biological Engineering, Inner Mongolia University of Science and Technology, Baotou, 014010, China.

Acknowledgements

The authors are very grateful to the anonymous referee for his/her valuable comments and suggestions, which led to an improved presentation of the manuscript.

Received: 23 April 2014 Accepted: 30 July 2014 Published: 15 August 2014

References

1. Boukal, DS, Sabelis, MW, Berec, L: How predator functional responses and Allee eects in prey aect the paradox of enrichment and population collapse. Theor. Popul. Biol. 72, 136-147 (2007)

2. Holling, CS: The components of predation as revealed by a study of small mammal predation of the Euorpean pine swalfy. Can. Entomol. 91, 293-320 (1959)

3. Lewis, M, Karevia, P: Allee dynamics and the spread of invading organisms. Theor. Popul. Biol. 43, 141-158 (1993)4. Morozov, A, Petrovovskii, S, Li, B: Spatiotemporal complexity of patchy invasion in a predator-prey system with the Allee eect. J. Theor. Biol. 238, 18-35 (2006)

5. Nisbet, RM, Gurney, WSC: Modelling Fluctuating Populations. Wiley, New York (1982)6. Petrovovskii, S, Morozov, A, Li, B: Regimes of biological invasion in a predator-prey system with the Allee eect. Bull. Math. Biol. 67, 637-661 (2005)

7. Wang, J, Shi, J, Wei, J: Dynamics and pattern formation in a diusive predatory-prey system with strong Allee eect in prey. J. Dier. Equ. 251, 1276-1304 (2011)

8. Hsu, C, Yang, C, Yang, T, Yang, T: Existence of traveling wave solutions for diusive predator-prey type systems.J. Dier. Equ. 252, 3040-3075 (2012)9. Nishiura, Y: Global structure of bifurcating solutions of some reaction-diusion systems. SIAM J. Math. Anal. 13, 555-593 (1982)

10. Opial, Z: Sur les priodes des solutions de Iquation direntielle x + g(x) = 0. Ann. Pol. Math. 10, 49-72 (1961)

11. Keener, J: Activators and inhibitors in pattern formation. Stud. Appl. Math. 59, 1-23 (1978)12. Jang, J, Ni, W, Tang, M: Global bifurcation and structure of Turing patterns in the 1-D Lengyel-Epstein model. J. Dyn. Dier. Equ. 16, 297-320 (2004)

doi:10.1186/1687-1847-2014-226Cite this article as: Bao and Liu: Existence of non-constant positive stationary solutions of the shadow predator-prey systems with Allee effect. Advances in Dierence Equations 2014 2014:226.

f (u()) and d =

h

mua + u dx. (.)