Volume 2011, Issue 1 354016
Research Article
Open Access

Positive Almost Periodic Solutions for a Time-Varying Fishing Model with Delay

Xia Li

Xia Li

Department of Mathematics, Yunnan University, Yunnan, Kunming 650091, China ynu.edu.cn

Search for more papers by this author
Yongkun Li

Corresponding Author

Yongkun Li

Department of Mathematics, Yunnan University, Yunnan, Kunming 650091, China ynu.edu.cn

Search for more papers by this author
Chunyan He

Chunyan He

Department of Mathematics, Yunnan University, Yunnan, Kunming 650091, China ynu.edu.cn

Search for more papers by this author
First published: 16 October 2011
Citations: 4
Academic Editor: Dexing Kong

Abstract

This paper is concerned with a time-varying fishing model with delay. By means of the continuation theorem of coincidence degree theory, we prove that it has at least one positive almost periodic solution.

1. Introduction

Consider the following differential equation which is widely used in fisheries [14]:
(1.1)
where N = N(t) is the population biomass, L(t, N) is the per capita fecundity rate, M(t, N) is the per capita mortality rate, and F(t) is the harvesting rate per capita.
In (1.1), let L(t, N) be a Hills′ type function ([1, 2])
(1.2)
and take into account the delay and the varying environments; Berezansky and Idels [5] proposed the following time-lag model based on (1.1) [16]
(1.3)
where b(t) = M(t, N) + F(t).

The model (1.3) has recently attracted the attention of many mathematicians and biologists; see the differential equations which are widely used in fisheries [1, 2]. However, one can easily see that all equations considered in the above-mentioned papers are subject to periodic assumptions, and the authors, in particular, studied the existence of their periodic solutions. On the other hand, ecosystem effects and environmental variability are very important factors and mathematical models cannot ignore, for example, reproduction rates, resource regeneration, habitat destruction and exploitation, the expanding food surplus, and other factors that affect the population growth. Therefore it is reasonable to consider the various parameters of models to be changing almost periodically rather than periodically with a common period. Thus, the investigation of almost periodic behavior is considered to be more accordant with reality. Although it has widespread applications in real life, the generalization to the notion of almost periodicity is not as developed as that of periodic solutions; we refer the reader to [718].

Recently, the authors of [19] proved the persistence and almost periodic solutions for a discrete fishing model with feedback control. In [20, 21], the contraction mapping principle and the continuation theorem of coincidence degree have been employed to prove the existence of positive almost periodic exponential stable solutions for logarithmic population model, respectively. A primary purpose of this paper, nevertheless, is to utilize the continuation theorem of coincidence degree for this purpose. To the best of the authors’ observation, there exists no paper dealing with the proof of the existence of positive almost periodic solutions for (1.3) using the continuation theorem of coincidence degree. Therefore, our result is completely different and presents a new approach.

2. Preliminaries

Our first observation is that under the invariant transformation N(t) = ey(t), (1.3) reduces to
(2.1)
for γ > 0, with the initial function and the initial value
(2.2)
For (2.1) and (2.2), we assume the following conditions:
  • (A1)

    a(t), b(t) ∈ C([0, +), [0, +)) and K(t) ∈ C([0, +), (0, +));

  • (A2)

    θ(t) is a continuous function on [0, +) that satisfies θ(t) ≤ t;

  • (A3)

    ϕ(t):(−, 0)→[0, ) is a continuous bounded function, ϕ(t) ≥ 0, y0 > 0.

By a solution of (2.1) and (2.2) we mean an absolutely continuous function y(t) defined on (−, +) satisfying (2.1) almost everywhere for t ≥ 0 and (2.2). As we are interested in solutions of biological significance, we restrict our attention to positive ones.

According to [22], the initial value problem (2.1) and (2.2) has a unique solution defined on (−, ).

Let X, Y be normed vector spaces, L : Dom  LXY be a linear mapping, and N : XY be a continuous mapping. The mapping L will be called a Fredholm mapping of index zero if dim   Ker L = codim  ImL < + and Im  L is closed in Y. If L is a Fredholm mapping of index zero and there exist continuous projectors P : XX and Q : YY such that Im  P = Ker   L, Ker Q = ImL = Im (IQ), it follows that the mapping L|Dom   L∩Ker P : (IP)XIm  L is invertible. We denote the inverse of that mapping by KP. If Ω is an open bounded subset of X, then the mapping N will be called L-compact on , if is bounded and is compact. Since ImQ is isomorphic to Ker L, there exists an isomorphism J : ImQ → Ker L.

Theorem 2.1 (see [19].)Let Ω ⊂ X be an open bounded set and let N : XY be a continuous operator which is L-compact on . Assume that

  • (1)

    LyλNy for every yΩ∩Dom  L and λ ∈ (0,1);

  • (2)

    QNy ≠ 0 for every yΩ∩Ker  L;

  • (3)

    the Brouwer degree deg {JQN, Ω∩Ker  L, 0} ≠ 0.

Then Ly = Ny has at least one solution in .

3. Existence of Almost Periodic Solutions

Let AP() denote the set of all real valued almost periodic functions on , for fAP() we denote by
(3.1)
the set of Fourier exponents and the module of f, respectively. Let K(f, ɛ, S) denote the set of ɛ-almost periods for f with respect to SC((−, 0], ), l(ɛ) denote the length of the inclusion interval, and denote the mean value of f.

Definition 3.1. y(t) ∈ C(, ) is said to be almost periodic on if for any ɛ > 0 the set K(y, ɛ) = {δ:|y(t + δ) − y(t)| < ɛ, ∀t} is relatively dense; that is, for any ɛ > 0 it is possible to find a real number l(ɛ) > 0 for any interval with length l(ɛ); there exists a number δ = δ(ɛ) in this interval such that |y(t + δ) − y(t)| < ɛ for any t.

Throughout the rest of the paper we assume the following condition for (2.1):
  • (H)

    a(t), b(t), K(t), tθ(t) ∈ AP(), m(b/Kγ) ≠ 0  and  m(a) ≠ m(b).

In our case, we set
(3.2)
where
(3.3)
where
(3.4)
and α is a given constant; define the norm
(3.5)

Remark 3.2. If f is ɛ-almost periodic function, then is ɛ-almost periodic if and only if m(f) = 0. Whereas fAP() does not necessarily have an almost periodic primitive, m(f) = 0. That is why we can not make V1 = {zAP() : m(z) = 0} and let V1 = {yAP() : mod (y) ⊂ mod (F), ∀μ ∈ Λ(y)  satisfy   | μ | > α}.

We start with the following lemmas.

Lemma 3.3. X and Y are Banach spaces endowed with the norm ∥·∥.

Proof. If yn ∈ V1 and yn converge to y0, then it is easy to show that y0AP() with mod (y0) ⊂ mod (F). Indeed, for all we have

(3.6)
Thus
(3.7)
which implies that y0V1. One can easily see that V1 is a Banach space endowed with the norm ∥·∥. The same can be concluded for the spaces X and Y. The proof is complete.

Lemma 3.4. Let L : XY and

(3.8)
where Ly = y = dy/dt. Then L is a Fredholm mapping of index zero.

Proof. It is obvious that L is a linear operator and Ker  L = V2. It remains to prove that Im  L = V1. Suppose that ϕ(t) ∈ Im  LY. Then, there exist ϕ1V1 and ϕ2V2 such that

(3.9)
From the definitions of ϕ(t) and ϕ1(t), one can deduce that and are almost periodic functions and thus ϕ2(t) ≡ 0, which implies that ϕ(t) ∈ V1. This tells us that
(3.10)
On the other hand, if φ(t) ∈ V1∖{0} then we have . Indeed, if then we obtain
(3.11)
It follows that
(3.12)
Thus
(3.13)
Note that is the primitive of φ in X; therefore we have φIm  L. Hence, we deduce that
(3.14)
which completes the proof of our claim. Therefore,
(3.15)
Furthermore, one can easily show that Im  L is closed in Y and
(3.16)
Therefore, L is a Fredholm mapping of index zero.

Lemma 3.5. Let N : XY, P : XX, and Q : YY such that

(3.17)
Then, N is L-compact on (Ω is an open and bounded subset of X).

Proof. The projections P and Q are continuous such that

(3.18)
It is clear that
(3.19)
Therefore
(3.20)
In view of
(3.21)
we can conclude that the generalized inverse (of L) KP : Im  L → Ker  P∩Dom  L exists and is given by
(3.22)
Thus
(3.23)
where f[y(t)] is defined by
(3.24)

The integral form of the terms of both QN and (IQ)N implies that they are continuous. We claim that KP is also continuous. By our hypothesis, for any ɛ < 1 and any compact set SC((−, 0], ), let l(ɛ, S) be the inclusion interval of K(F, ɛ, S). Suppose that {zn(t)} ⊂ ImL = V1 and zn(t) uniformly converges to z0(t). Because , there exists ρ(0 < ρ < ɛ) such that . Let l(ρ, S) be the inclusion interval of K(F, ρ, S) and

(3.25)
It is easy to see that l is the inclusion interval of both K(F, ɛ, S) and K(F, ρ, S). Hence, for all t ∉ [0, l], there exists such that t + μt ∈ [0, l]. Therefore, by the definition of almost periodic functions we observe that
(3.26)
By applying (3.26), we conclude that is continuous and consequently KP and KP(IQ)Ny are also continuous.

From (3.26), we also have that and KP(IQ)Ny are uniformly bounded in . In addition, it is not difficult to verify that is bounded and KP(IQ)Ny is equicontinuous in . Hence by the Arzelà-Ascoli theorem, we can immediately conclude that is compact. Thus N is L-compact on .

Theorem 3.6. Let condition (H) hold. Then (2.1) has at least one positive almost periodic solution.

Proof. It is easy to see that if (2.1) has one almost periodic solution , then is a positive almost periodic solution of (1.3). Therefore, to complete the proof it suffices to show that (2.1) has one almost periodic solution.

In order to use the continuation theorem of coincidence degree theory, we set the Banach spaces X and Y the same as those in Lemma 3.3 and the mappings L, N, P, Q the same as those defined in Lemmas 3.4 and 3.5, respectively. Thus, we can obtain that L is a Fredholm mapping of index zero and N is a continuous operator which is L-compact on . It remains to search for an appropriate open and bounded subset Ω. Corresponding to the operator equation

(3.27)
we may write
(3.28)
Assume that y = y(t) ∈ X is a solution of (3.28) for a certain λ ∈ (0,1). Denote
(3.29)
In view of (3.28), we obtain
(3.30)
and consequently,
(3.31)
which implies from (H) that
(3.32)
Similarly, we can get
(3.33)
By inequalities (3.32) and (3.33), we can find that there exists t1 such that
(3.34)
where
(3.35)
Then from (3.26), we have
(3.36)
or
(3.37)
Choose the point νt2 ∈ [l, 2l]  K(F, ρ, S), where ρ(0 < ρ < ɛ) satisfies K(F, ρ) ⊂ K(z, ɛ). Integrating (3.28) from t2 to ν, we get
(3.38)
However, from (3.28) and (3.38), we obtain
(3.39)
Substituting back in (3.37) and for νt2 + l, we have
(3.40)
where
(3.41)
Let . Obviously, it is independent of λ. Take
(3.42)
It is clear that Ω satisfies assumption (1) of Theorem 2.1. If yΩ    Ker  L, then y is a constant with . It follows that
(3.43)
which implies that assumption (2) of Theorem 2.1 is satisfied. The isomorphism J : Im  Q → Ker   L is defined by J(z) = z for z. Thus, JQNy ≠ 0. In order to compute the Brouwer degree, we consider the homotopy
(3.44)
For any yΩ∩Ker  L, s ∈ [0,1], we have H(y, s) ≠ 0. By the homotopic invariance of topological degree, we get
(3.45)
Therefore, assumption (3) of Theorem 2.1 holds. Hence, Ly = Ny has at least one solution in . In other words, (2.1) has at least one positive almost periodic solution. Therefore, (1.3) has at least one positive almost periodic solution. The proof is complete.

4. An Example

Let . Then (1.3) has the form
(4.1)
One can easily realize that m(b(t)/[K(t)] γ) > 0 and m(a(t)) > m(b(t)); thus condition (H) holds. Therefore, by the consequence of Theorem 3.6, (4.1) has at least one positive almost periodic solution (Figure 1).
Details are in the caption following the image
Transient response of state N(t) when γ = 2.

Acknowledgment

This work is supported by the National Natural Sciences Foundation of People’s Republic of China under Grant 10971183.

      The full text of this article hosted at iucr.org is unavailable due to technical difficulties.