DDNSDiscrete Dynamics in Nature and Society1607-887X1026-0226Hindawi Publishing Corporation69347210.1155/2009/693472693472Research ArticleGlobal Dynamics Behaviors of Viral Infection Model for Pest ManagementWeiChunjin1ChenLansun2VecchioAntonia1Science CollegeJimei UniversityXiamen 361021Chinajmu.edu.cn2Department of Applied MathematicsDalian University of TechnologyDalian 116024Chinadlut.edu.cn20092422010200920092009181120092009Copyright © 2009This 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.

According to biological strategy for pest control, a mathematical model with periodic releasing virus particles for insect viruses attacking pests is considered. By using Floquet's theorem, small-amplitude perturbation skills and comparison theorem, we prove that all solutions of the system are uniformly ultimately bounded and there exists a globally asymptotically stable pest-eradication periodic solution when the amount of virus particles released is larger than some critical value. When the amount of virus particles released is less than some critical value, the system is shown to be permanent, which implies that the trivial pest-eradication solution loses its stability. Further, the mathematical results are also confirmed by means of numerical simulation.

1. Introduction and Model Formulation

Currently, applications of chemical pesticides to combat pests are still one of the main measures to improve crop yields. Though chemical crop protection plays an important role in modern agricultural practices, it is still viewed as a profit-induced poisoning of the environment. The nondegradable chemical residues, which accumulate to harmful levels, are the root cause of health and environmental hazards and deserve most of the present hostility toward them. Moreover, synthetic pesticides often disrupt the balanced insect communities. This leads to the interest in Biological control methods for insects and plant pests [1, 2].

Biological control is, generally, man's use of a suitably chosen living organism, referred as the biocontrol agent, to control another. Biocontrol agents can be predators, pathogens, or parasites of the organism to be controlled that either kill the harmful organism or interfere with its biological processes . In a large number of biopesticides, the insect virus pesticide because of its high pathogenicity, specificity, and ease production plays an important role in pest biological control. The control of rabbit pests in Australia by the virus disease called “myxomatosis” provides a spectacular example of a virus controlling pest . The insect viruses for the biological control of pests are mainly baculoviruses. Baculoviruses comprise a family of double-stranded DNA viruses which are pathogenic for arthropods, mainly insects. The polyhedral occlusion body (OB) is the characteristic phenotypic appearance of baculoviruses and in case of a nucleopolyhedrovirus (NPV) typically comprised of a proteinaceous matrix with a large number of embedded virus particles. Baculoviruses have a long history as effective and environmentally benign insect control agents in field crops, vegetables, forests, and pastures .

Transmission is also key to the persistence of baculoviruses in the environment [6, 7]. Transmission occurs primarily when an NPV-infected larva dies and lyses, releasing a massive number of OBs onto foliage and soil. Susceptible hosts become infected when they ingest OBs while feeding. Defecation and regurgitation by infected larvae have been reported as additional routes of contamination of host plants with viruses . Moreover, some studies suggest that cannibalism and predation may also be routes of virus transmission . Environmental factors such as rainfall, wind transport, and contaminated ovipositors of parasitic hymenopterans could contribute to NPV transmission as well [11, 12].

Insight in the epidemiological dynamics, it is necessary to predict optimal timing, frequency, and dosage of virus application and to assess the short and longer term persistence of NPV in insect populations and the environment. Modeling studies can help to obtain preliminary assessments of expected ecological dynamics at the short and longer term. There is a vast amount of literature on the applications of microbial disease to suppress pests , but there are only a few papers on mathematical models of the dynamics of viral infection in pest control [16, 17]. System with impulsive effects describing evolution processes is characterized by the fact that at certain moments of time they abruptly experience a change of state. Processes of such type are studied in almost every domain of applied science. Impulsive differential equations have been recently used in population dynamics in relation to impulsive vaccination, population ecology, the chemotherapeutic treatment of disease, and the theory of the chemostat .

In this paper, according to the above description, we should construct a more realistic model by introducing additional virus particles (i.e., using viral pesticide) to investigate the dynamical behavior of viruses attacking pests, which is described as follows: S'(t)=rS(t)(1-S(t)+I(t)K)-βS(t)I(t)-θS(t)V(t),I'(t)=βS(t)I(t)+θS(t)V(t)-λI(t),V'(t)=-θS(t)V(t)+bλI(t)-μV(t),tnT,ΔS(t)=0,ΔI(t)=0,ΔV(t)=p,t=nT,n=1,2,, where S(t), I(t), and V(t) denote the density of susceptible pests, infected pests, and virus particles at time t, respectively. T is the impulsive period, n={1,2,}, p is the release amount of virus particles, ΔS(t)=S(t+)-S(t),ΔI(t)=I(t+)-I(t), and ΔV(t)=V(t+)-V(t). The assumptions in the model are as follows.

We assume that only susceptible pests S are capable of reproducing with logistic law; that is, the infected pests I are removed by lysis before having the possibility of reproducing. However, they still contribute with S to population growth toward the carrying capacity. r is intrinsic birth rate and K(>0) is carrying capacity.

The term θVS denotes that the susceptible pests S become infected I as they ingest foods contaminated with virus particles, in which θ is positive constant and represents the “effective per pest contact rate with viruses.” And βSI denotes that the susceptible pests S become infected I by the transmission of infective pests I according to other ways; perhaps β is very small and close to zero.

An infective pest I has a latent period, which is the period between the instant of infection and that of lysis, during which the virus reproduces inside the pest. The lysis death rate constant λ. λ gives a measure of such a latency period T being λ=1/T. The lysis of infected pests, on the average, produces b virus particles (b>1). b is the virus replication factor.

The virus particles V have a natural death rate μ due to all kinds of possible mortality of viruses such as enzymatic attack, pH dependence, temperature changes, UV radiation, and photooxidation.

The paper is organized as follows: in Section 2, some auxiliary results which establish the a priori boundedness of the solutions, together with the asymptotic properties of certain reduced systems which are used throughout the paper as a basis of several comparison arguments, are stated. In Section 3, by using Floquet's theory for impulsive differential equations, small-amplitude perturbation methods, and comparison techniques, we provide the sufficient conditions for the local and global stability of the pest-eradication periodic solution and the conditions for the permanence of the system. Finally, a brief discussion and numerical examples are given. We also point out some future research directions.

2. Preliminary

We give some definitions, notations, and lemmas which will be useful for stating and proving our main results. Let R+=[0,),R+3={(x1,x2,x3)xi>0,i=1,2,3}. Denote by f=(f1,f2,f3)T the map defined by the right hand of the first three equations in system (1.1). Let V:R+×R+3R+, then VV0 if

V is continuous in (nT,(n+1)T]×R+3 and for each zR+3,nNlim(t,z)(nT+,z)V(t,z)=V(nT+,z) exists;

V is locally Lipschitzian in z.

Definition 2.1.

VV0, then for (t,z)(nT,(n+1)T]×R3, the upper right derivative of V(t,z) with respect to system (1.1) is defined as D+V(t,z)=limh0+sup1h[V(t+h,z+hf(t,z))-V(t,z)]. The solution of (1.1), denoted by z(t)=(S(t),I(t),V(t)), is a piecewise continuous function z(t): R+R+3, z(t) is continuous on (nT,(n+1)T], nN, and z(nT+)=limtnT+z(t) exists. Obviously, the existence and uniqueness of the solution of (1.1) is guaranteed by the smoothness properties of f (for more details see ).

Lemma 2.2.

Suppose that z(t) is a solution of (1.1) with z(0+)0, then z(t)0 for all t0. Moreover, if z(0+)>0, then z(t)>0 for all t0.

Lemma 2.3.

Let V:R+×R+nR+ and VV0. Assume that D+V(t,z(t))()g(t,V(t,z)),tτk,V(t,z(t+))()Ψn(V(t,z(t))),t=τk,kN,z(0+)=z0, where g:R+×R+nRn is continuous in (τk,τk+1]×R+n and for each νR+n,nNlim(t,ι)(τk+,ν)g(t,ι)=g(τk+,ν) exists, and Ψn:R+nR+n is nondecreasing. Let (t)=(t,0,U0) be the maximal(minimal) solution of the scalar impulsive differential equation U'(t)=g(t,U),tτk,U(t+)=Ψn(U(t)),t=τk,kN,U(0+)=U0 existing on [0,). Then V(0+,z0)()U0 implies that V(t,z(t))()(t),t0, where z(t) is any solution of (1.1) existing on [0,].

Note that if one has some smoothness conditions of g to guarantee the existence and uniqueness of solutions for (2.5), then (t) is exactly the unique solution of (2.5).

Lemma 2.4.

There exists a constant M1>0,M2>0 such that S(t)M1,I(t)M1,V(t)M2 for each positive solution (S(t),I(t),V(t)) of (1.1) with t being large enough.

Proof.

Define a function L such that L(t)=S(t)+I(t). Then we have D+L(t)|(1.1)+λL(t)=-rKS2(t)+(λ+r)S(t)-rS(t)I(t)K-rKS2(t)+(λ+r)S(t),t(nT,(n+1)T]. Obviously, the right hand of the above equality is bounded; thus, there exists M0=K(λ+r)2/4r>0 such that D+L(t)-λL(t)+M0. It follows that lim inftL(t)lim suptL(t)M0/λ. Therefore, by the definition of L(t) we obtain that there exists a constant M1=K(λ+r)2/4rλ>0 such that S(t)M1,I(t)M1. From the third and sixth equations of system (1.1), we have V'(t)=-θS(t)V(t)+bλI(t)-μV(t)bλM1-μV(t),tnT,V(nT+)=V(nT)+p,t=nT. According to Lemma 2.3 in  we derive V(t)V(0)e-μt+0tbλM1e-μ(t-s)ds+0<kT<tpe-μ(t-kT)bλM1μ+peμTeμT-1ast. Therefore, there exists a constant M2>0 such that V(t)M2. The proof is complete.

Next, we give some basic property of the following subsystem: y(t)=-dy(t),tnT,Δy(t)=p,t=nT,v(t)=a-bv(t),tnT,Δv(t)=θ,t=nT.

Lemma 2.5.

System (2.10) has a positive periodic solution y*(t) and for every positive solution y(t) of system (2.10), |y(t)-y*(t)|0 as t, where y*(t)=pe-d(t-nT)/(1-e-dT) and y*(0+)=p/(1-e-dT).

Lemma 2.6.

System (2.11) has a positive periodic solution v*(t) and for every positive solution v(t) of system (2.11), |v(t)-v*(t)|0 as t, where v*(t)=a/b+θe-b(t-nT)/(1-e-bT) and v*(0+)=a/b+θ/(1-e-bT).

3. Main Results

When S(t)=0, from the second and fifth equations of system (1.1), we have limtI(t)=0. Further, from the third and sixth equations of system (1.1), we have

V(t)=-μV(t),tnT,ΔV(t)=p,t=nT, By Lemma (2.10), we can obtain the unique positive periodic solution of system (3.1): V*(t)=pe-μ(t-nT)/(1-e-μT),nT<t(n+1)T, with initial value V*(0+)=p/(1-e-μT). Thus the pest-eradication solution is explicitly shown. That is, system (1.1) has a so-called pest-eradication periodic solution (0,0,V*(t)). Next, we shall give the condition to assure its global asymptotic stability.

Theorem 3.1.

Let (S(t),I(t),V(t)) be any solution of system (1.1) with positive initial values. Then the pest-eradication periodic solution (0,0,V*(t)) is locally asymptotically stable provided that rT<θ0TV*(t)dt.

Proof.

The local stability of periodic solution (0,0,V*(t)) may be determined by considering the behavior of small amplitude perturbation of the solution. Let S(t)=u(t),I(t)=v(t),V(t)=w(t)+V*(t). The corresponding linearized system of (1.1) at (0,0,V*) is u'(t)=(r-θV*(t))u(t),v'(t)=θV*(t)u(t)-λv(t),w'(t)=-θV*(t)u(t)+bλv(t)-μw(t),tnT,u(t+)=u(t),v(t+)=v(t),w(t+)=w(t),t=nT,n=1,2,, Let Φ(t) be the fundamental matrix of (3.3), then Φ(t) satisfies dΦ(t)dt=(r-θV*(t)00θV*(t)-λ0-θV*(t)bλ-μ)Φ(t), and Φ(0)=E3 (unit 3×3 matrix). Hence, the fundamental solution matrix is Φ(t)=(e0t(r-θV*(t))dt00Δe-λt0ΔΔe-μt), where the exact expressions of Δ are omitted, since they are not used subsequently. The resetting impulsive condition of (3.3) becomes (u(nT+)v(nT+)w(nT+))=(100010001)(u(nT)v(nT)w(nT)). Hence, if all the eigenvalues of M=(100010001)Φ(T) have absolute values less than one, then the periodic solution (0,0,V*(t)) is locally stable. Since the eigenvalues of M are λ1=e-λT<1,λ2=e-μT<1,λ3=e0T(r-θV*(t))dt and |λ3|<1 if and only if (3.2) holds, according to Floquet's theory of impulsive differential equation, the pest-eradication periodic solution (0,0,V*(t)) is locally stable.

In fact, for condition (3.2), rT represents the normalized gain of the pest in a period, while θ0TV*(t)dt represents the normalized loss of the pest in a period due to viral disease. That is, this condition is a balance condition for the pest near the pest-eradication periodic solution, which asserts the fact that in a vicinity of this solution (0,0,V*(t)) the pest is depleted faster than they can recover and consequently the pest is condemned to extinction.

Theorem 3.2.

Let (S(t),I(t),V(t)) be any solution of system (1.1) with positive initial values. Then the pest-eradication periodic solution (0,0,V*(t)) is globally asymptotically stable provided that rT<θpμ+θK.

Proof.

From (3.9), we know that (3.2) also holds. By Theorem 3.1, we know that (0,0,V*(t)) is locally stable. Therefore, we only need to prove its global attractivity. Since rT<θp/(μ+θK), we can choose an ε1>0 small enough such that (r+θε1)T-θpμ+θ(K+ε1)η<0. From the first equation of system (1.1), we obtain S'(t)rS(t)(1-S(t)/K). Consider the comparison equation ω'(t)=rω(t)(1-ω(t)/K),ω(0)=S(0), then we have  S(t)ω(t) and ω(t)K as t. Thus, there exists an ε1>0 such that S(t)K+ε1 for t being large enough. Without loss of generality, we assume S(t)K+ε1 for all t>0.

Note that V'(t)-[μ+θ(K+ε1)]V(t); by Lemmas 2.3 and 2.5, there exists a n1 such that for all tn1TV(t)z*(t)-ε1, where z(t) is the solution of z'(t)=-[μ+θ(K+ε1)]z(t),tnT,Δz(t)=p,t=nT,z(0+)=V(0+)>0.z*(t)=pe-[μ+θ(K+ε1)](t-nT)1-e-[μ+θ(K+ε1)]T,t(nT,(n+1)T]. Thus we have S'(t)=rS(t)(1-S(t)+I(t)K)-βS(t)I(t)-θS(t)V(t)S(t)[r-θV(t)]S(t)[r-θ(z*(t)-ε1)]. Integrating the above inequality on ((n1+k)T,(n1+k+1)T],kN, yields S(t)S(n1T)en1T(n1+1)T[r-θ(z*(t)-ε1)]dtS(n1T)ekη. Since η<0, we can easily get S(t)0 as t. For ε2>0 small enough being (ε2<λ/β), there must exist an n2(n2>n1) such that 0<S(t)<ε2, for tn2T; then from the second equation of system (1.1), we have I'(t)(βε2-λ)I(t)+θM2ε2, so lim inftI(t)lim suptθM2ε2/(λ-βε2).

In the following, we prove V(t)V*(t), as t+. From system (1.1), we have (-μ-θε2)V(t)V'(t)bλθM2ε2λ-βε2-μV(t), and by Lemmas 2.3, 2.5, and 2.6, there exists an n3(n3>n2) such that V2*(t)-εV(t)V1*(t)+εtnT,n>n3, where V1*(t)=bλθM2ε2/(λ-βε2)μ+pe-μ(t-nT)/(1-e-μT),V2*(t)=pe-(μ+θε2)(t-nT)/(1-e-(μ+θε2)T). Let ε20, we have I(t)0,  V1*(t)V*(t),  V2*(t)V*(t). Therefore, (0,0,V*(t)) is globally attractive. This completes the proof.

Corollary 3.3.

If p>p1*=rT(μ+θK)/θ or T<T1*=θp/r(μ+θK), then the pest-eradication periodic solution (0,0,V*(t)) is globally asymptotically stable.

We have proved that, if p>p1*=rT(μ+θK)/θ or T<T1*=θp/r(μ+θK), the pest-eradication periodic solution (0,0,V*(t)) is globally asymptotically stable; that is, the pest population is eradicated totally. But in practice, from the view point of keeping ecosystem balance and preserving biological resources, it is not necessary to eradicate the pest population. Next we focus our attention on the permanence of system (1.1). Before starting our result, we give the definition of permanence.

Definition 3.4.

System (1.1) is said to be permanent if there are constants m,M>0 (independent of initial value) and a finite time T0 such that all solutions z(t)=(S(t),I(t),V(t)) with initial values z(0+)>0, mz(t)M hold for all tT0. Here T0 may depend on the initial values z(0+)>0.

Theorem 3.5.

Let (S(t),I(t),V(t)) be any positive solution of (1.1) with positive initial values z(0+)>0. Then system (1.1) is permanent provided that rT>θ0TV*(t)dt.

Proof.

Suppose that z(t)=(S(t),I(t),V(t)) is a solution of system (1.1) with initial values z(0+)>0. By Lemma 2.4, there exists positive constants M1,M2 such that S(t)M1,I(t)M1, and V(t)M2 for t being large enough. We may assume S(t)M1,I(t)M1,V(t)M2 for all t0. From (3.11), we know that V(t)z*(t)-ε1pe-[μ+θ(K+ε1)]T1-e-[μ+θ(K+ε1)]T-ε1=̇m>0 for t being large enough. Thus we only need to find m1>0,m2>0 such that S(t)m1, I(t)m2 for t being large enough. We shall do it in two steps.

Step 1.

Since rT>θ0TV*(t)dt, that is rT>θp/μ, we can select m3>0(m3<λ/β),ε>0 small enough such that δ=̇rT-[rm3TK+rTK(θm3ηλ-βm3+ε)+βT(θm3ηλ-βm3+ε)+θεT+θbλm3Tμ+θpμ]>0. We shall prove that S(t)+I(t)<m3 cannot hold for all t>0. Otherwise, we have that S(t)<m3 for all t>0. Then from the third equation of system (1.1), we get V(t)=-θS(t)V(t)+bλI(t)-μV(t)bλm3-μV(t). Then V(t)u(t) and u(t)u*(t) as t, where u*(t) is the solution of u'(t)=bλm3-μu(t),tnT,Δu(t)=p,t=nT,u(0+)=V(0+)>0,u*(t)=bλm3μ+pe-μ(t-nT)1-e-μT,t(nT,(n+1)T]. Therefore, there exists a T̃>0 such that V(t)u(t)u*(t)+εbλm3μ+p1-e-μT+ε=̇η for t>T̃. From the second equation of system (1.1), we have I'(t)=θS(t)V(t)+βS(t)I(t)-λI(t)θm3η+(βm3-λ)I(t). Thus, I(t)θm3η/(λ-βm3)+ε for t being large enough. Therefore, there exists T1>T̃ such that S'(t)=rS(t)(1-S(t)+I(t)K)-βS(t)I(t)-θS(t)V(t)S(t)[r-rm3K-rK(θm3ηλ-βm3+ε)-β(θm3ηλ-βm3+ε)-θV(t)]S(t)[r-rm3K-rK(θm3ηλ-βm3+ε)-β(θm3ηλ-βm3+ε)-θ(u*(t)+ε)] for all t>T1. Let N0N such that (N0-1)TT1. Integrating the above inequality on ((n-1)T,nT],nN0, we have S(nT)S((n-1)T)e(n-1)TnT[r-rm3/K-(r/K)(θm3η/(λ-βm3)+ε)-β(θm3η/(λ-βm3)+ε)-θ(u*(t)+ε)]dt=S((n-1)T)erT-[rm3T/K+(rT/K)(θm3η/(λ-βm3)+ε)+βT(θm3η/(λ-βm3)+ε)+θεT+θbλm3T/μ+θp/μ]=S((n-1)T)eδ. Then S((n+k)T)S(nT)ekδ as k, which is a contradiction to the boundedness of S(t). Thus, there exists a t1>0 such that S(t1)m3.

Step 2.

If S(t)m3 for all tt1, then our aim is obtained. Hence we need only to consider the situation that S(t)m3 is not always true for tt1, and we denote t*=inftt1{S(t)<m3}. Then S(t)m3 for t[t1,t*) and S(t*)=m3, since S(t) is continuous. Suppose t*(n1T,(n1+1)T],n1N. Select n2,n3N such that n2T>T2=ln((M+u0*)/ε)μ,eδ1(n2+1)Teδn3>1, where u0*=p/(1-e-μT)+bλm3/μ,δ1=r-r(m3+M1)/K-βM1-θM2<0. Let T̂=(n2+n3)T. We claim that there must be a t2[(n1+1)T,(n1+1)T+T̂] such that S(t2)m3. Otherwise S(t)<m3,t2[(n1+1)T,(n1+1)T+T̂]. Consider (3.21) with u((n1+1)T+)=V((n1+1)T+). We have u(t)=(u(n1+1)T+-u0*)e-μ(t-(n1+1)T)+u*(t),t(nT,(n+1)T],n1+1nn1+1+n2+n3. Thus |u(t)-u*(t)|(M+u0*)e-μn2T<ε,V(t)u(t)u*(t)+ε,(n1+1+n2)Tt(n1+1)T+T̂. Thus, we have S'(t)=rS(t)(1-S(t)+I(t)K)-βS(t)I(t)-θS(t)V(t)S(t)[r-rm3K-rK(θm3ηλ-βm3+ε)-β(θm3ηλ-βm3+ε)-θV(t)]S(t)[r-rm3K-rK(θm3ηλ-βm3+ε)-β(θm3ηλ-βm3+ε)-θ(u*(t)+ε)] for (n1+1+n2)Tt(n1+1)T+T̂. As in Step 1, we have S((n1+1+n2+n3)T)S((n1+1+n2)T)eδn3. On the interval t[t*,(n1+1+n2)T], we have S'(t)=rS(t)(1-S(t)+I(t)K)-βS(t)I(t)-θS(t)V(t)S(t)[r-r(m3+M1)K-βI(t)-θV(t)]S(t)(r-r(m3+M1)K-βM1-θM2),S((n1+1+n2)T)S(t*)et*(n1+1+n2)T(r-r(m3+M1)/K-βM1-θM2)dtm3e(r-r(m3+M1)/K-βM1-θM2)(n2+1)T=m3eδ1(n2+1)T. Thus S((n1+1+n2+n3)T)m3eδ1(n2+1)Teδn3>m3, which is a contradiction. Let t¯=inftt*{S(t)m3}, then S(t¯)m3, for t[t*,t¯), and we have S(t)S(t*)e(t-t*)δ1m3e(1+n2+n3)Tδ1m1. For t>t¯, the same arguments can be continued, since S(t¯)m3, and m1,m3 are t1-independent. Hence S(t)m1 for all tt1. In the following, we shall prove that there exists m2>0 such that I(t)m2 for t being large enough. From the third equation of system (1.1), we have V(t)=-θS(t)V(t)+bλI(t)-μV(t)-θM1V(t)-μV(t). Then V(t)u1(t) and u1(t)u1*(t) as t, where u1*(t) is the solution of u1(t)=-θM1u1(t)-μu1(t),tnT,Δu1(t)=p,t=nT,u(0+)=V(0+)>0,u1*(t)=pe-(θM1+μ)(t-nT)1-e-(θM1+μ)T,      t(nT,(n+1)T]. Therefore, there exists a T1̃>t1>0 such that V(t)u1(t)u1*(t)-εp1-e-(θM1+μ)T-εζ for t>T1̃. Then from the second equation of system (1.1), we have I'(t)=βS(t)I(t)+θS(t)V(t)-λI(t)-λI(t)+θζm1, and then we have lim suptI(t)lim inftI(t)θζm1/λm2. Therefore I(t)m2 for t being large enough. The proof is complete.

Corollary 3.6.

If p<p2*=rμT/θ or T>T2*=θp/rμ, then system (1.1) is permanent.

Example 3.7.

Let us consider the following system: S'(t)=1.8S(t)(1-S(t)+I(t)2)-0.2S(t)I(t)-0.6S(t)V(t),I(t)=0.2S(t)I(t)+0.6S(t)V(t)-0.8I(t),V'(t)=-0.6S(t)V(t)+2.4I(t)-0.8V(t),tnT,ΔS(t)=0,ΔI(t)=0,ΔV(t)=p,t=nT,n=1,2,, According to Corollaries 3.3, and 3.6, we know that if T<0.67, then (0,0,V*(t)) is globally asymptotically stable (see Figure 1), and if T>1.67, then the system is permanent (see Figure 2).

Dynamical behavior of the system (1.1) with r=1.8,K=2,θ=0.6,β=0.2,p=4,λ=0.8,μ=0.8,b=3, and T=0.5.

Dynamical behavior of the system (1.1) with r=1.8,K=2,θ=0.6,β=0.2,p=4,λ=0.8,μ=0.8,b=3, and T=3.

4. Numerical Simulations and Discussion

In this paper, we have investigated the dynamical behavior of a pest management model with periodic releasing virus particles at a fixed time. By using Floquet's theorem, small-amplitude perturbation skills and comparison theorem, we establish the sufficient conditions for the global asymptotical stability of the pest-eradication periodic solution as well as the permanence of the system (1.1). It is clear that the conditions for the global stability and permanence of the system depend on the parameters p,T, which implies that the parameters p,T play a very important role on the model.

From Corollary 3.3, we know that the pest-eradication periodic solution (0,0,V*(t)) is globally asymptotically stable when p>p1* or T<T1*. In order to drive the pests to extinction, we can determine the impulsive release amount p such that p>p1* or the impulsive period T such that T<T1*. If we choose parameters as r=1.8,K=2,θ=0.6,β=0.2,p=4,λ=0.8,μ=0.8,b=3, and p=4, then we have T1*=0.67; so we can make the impulsive period T smaller than 0.67 in order to eradicate the pests (see Figure 1). In the same time T2*=1.67, so we can make the impulsive period T larger than 1.67 in order to maintain the system permanent (see Figure 2). Similarly, we can fix T and change p in order to achieve the same purpose. However, from a pest control point of view, our aim is to keep pests at acceptably low levels, not to eradicate them, only to control their population size. With regard to this, the optimal control strategy in the management of a pest population is to drive the pest population below a given level and to do so in a manner which minimizes the cost of using the control and the time it takes to drive the system to the target. We hope that our results provide an insight to practical pest management. However, in the real world, for the seasonal damages of pests, should we consider impulsive releasing virus particles on a finite interval? Such work will be beneficial to pest management, and it is reasonable. We leave it as a future work.

Acknowledgment

This work was supported by National Natural Science Foundation of China (10771179).

DebachP.Biological Control of Insect Pests and Weeds1964London, UKChapman & HallKurstakE.Microbial and Viral Pesticide1982New York, NY, USAMarcel DekkerGrasmanJ.Van HerwarrdenO. A.HemerikL.A two-component model of host-parasitoid interactions: determination of the size of inundative releases of parasitoids in biological pest controlMathematical Biosciences20011962207216DavisP. E.MyersK.HoyJ. B.Biological control among vertebratesTheory and Practices of Biological Control1976New York, NY, USAPlenum PressMoscardiF.moscardi@cnpso.embrapa.brAssessment of the application of baculoviruses for control of LepidopteraAnnual Review of Entomology1999442572892-s2.0-000946413210.1146/annurev.ento.44.1.257CoryJ. S.jsc@ceh.ac.ukMyersJ. H.myers@zoology.ubc.caThe ecology and evolution of insect baculovirusesAnnual Review of Ecology, Evolution, and Systematics2003342392722-s2.0-0020048380ZhouM.SunX.SunX.VlakJ. M.just.vlak@wur.nlHuZ.van der WerfW.Horizontal and vertical transmission of wild-type and recombinant Helicoverpa armigera single-nucleocapsid nucleopolyhedrovirusJournal of Invertebrate Pathology20058921651752-s2.0-2344444946610.1016/j.jip.2005.03.005VasconcelosS. D.CoryJ. S.WilsonK. R.Modified behavior in baculovirus-infected lepidopteran larvae and its impact on the spatial distribution of inoculumBiological Control19967299306YoungS. Y.Transmission of nuclear polyhedrosis virus prior to death of infected loblolly pine sawfly, Neodiprion taedae linearis Ross, on loblolly pineJournal of Entomological Science1998331152-s2.0-0024924150DhandapaniN.JayarajS.RabindraR. J.Cannibalism on nuclear polyhedrosis virus infected larvae by Heliothis armigera (Hubn.) and its effect on viral-infectionInsect Science and Its Application199314427430FuxaR. J.RichterA. R.Selection for an increased rate of vertical transmission of Spodoptera frugiperda (Lepidoptera: Noctuidae) nuclear polyhedrosis virusEnvironmental Entomology199120603609HammJ. J.StyerE. L.LewisW. J.A baculovirus pathogenic to the parasitoid Microplitis croceipes (Hymenoptera: Braconidae)Journal of Invertebrate Pathology19885211891912-s2.0-38249027669FalconL. A.Problems associated with the use of arthropod viruses in pest controlAnnual Review of Entomology197621305324BurgesH. D.HusseyN. W.Microbial Control of Insects and Mites1971New York, NY, USAAcademic PressTanadaY.Epizootiology of insect diseasesBiological Control of Insect Pests and Weeds1964London, UKChapman & HallGhoshS.BhattacharyyaS.BhattacharyaD. K.The role of viral infection in pest control: a mathematical studyBulletin of Mathematical Biology200769826492691MR235384910.1007/s11538-007-9235-8BhattacharyyaS.BhattacharyaD. K.Pest control through viral disease: mathematical modeling and analysisJournal of Theoretical Biology20062381177197MR220658810.1016/j.jtbi.2005.05.019LakshmikanthamV.BainovD. D.SimeonovP. S.Theory of Impulsive Differential Equations1989SingaporeWorld ScienceWeiC.ChenL.A delayed epidemic model with pulse vaccinationDiscrete Dynamics in Nature and Society200820081274695110.1155/2008/746951MR2403197ZBL1149.92329MengX.ChenL.The dynamics of a new SIR epidemic model concerning pulse vaccination strategyApplied Mathematics and Computation20081972582597MR240068010.1016/j.amc.2007.07.083ZBL1131.92056ZhangH.GeorgescuP.ChenL.An impulsive predator-prey system with Beddington-DeAngelis functional response and time delayInternational Journal of Biomathematics200811117MR241964310.1142/S1793524508000072ZBL1155.92045WangF.PangG.ChenL.Study of a Monod-Haldene type food chain chemostat with pulsed substrateJournal of Mathematical Chemistry2008431210226MR244944010.1007/s10910-006-9189-3ZBL1135.92035