- Research
- Open access
- Published:
Spreading-vanishing dichotomy in a degenerate logistic model with general logistic nonlinear term
Boundary Value Problems volume 2014, Article number: 180 (2014)
Abstract
In this paper, we study the degenerate logistic equation with a free boundary and general logistic term in higher space dimensions and heterogeneous environment, which is used to describe the spreading of a new or invasive species. We first prove the existence and uniqueness of the local solution for the free boundary problem by the contraction mapping theorem, then we show that the solution can be expanded to all time using suitable estimates. Finally, we prove the spreading-vanishing dichotomy.
1 Introduction
It is an important problem to study the spreading of the invasive species in invasion ecology, which is an interesting branch of ecology. Using differential equations to study ecology becomes a main approach in ecological research. Most of the ecological phenomena such as species extinction can be explained by the nature of the differential equations. In the research of the spreading of the muskrat in Europe, Skellam observed the well-known phenomenon that many animal species spread to a new environment in a linear speed, which means the spreading radius eventually shows a linear growth speed against times [1]. Firstly, he calculated the square root of the area of the muskrat range from a map, which gives the spreading radius. Then he plotted it against times and observed the data points lay on a straight line. Several mathematical models have been proposed to discuss this phenomenon (see [2]).
The most successful mathematical model to describe the problem is the following logistic equation over the entire space :
where stands for the population density of a spreading species, is the diffusion rate, is the intrinsic growth rate, means the habitat carrying capacity. Fisher [3] and Kolomogorov et al.[4] made a pioneering contribution on this problem. They proved the problem admits traveling wave solutions of the problem (1) for space dimension : for any , there is a solution satisfying
or there is no such solution if . The constant is regarded as the minimal speed of the traveling waves. Fisher claimed that the constant is the spreading speed for the advantageous gene and proved it by a probabilistic argument. Then Aronson and Weinberger gave a clearer description and a rigorous proof for this phenomenon (see [5]).
Although the approach predicts the successful spreading and the establishment of a new species with a nontrivial initial population , it has the obvious shortcoming that it is regardless of its initial population and the initial area, which is in sharp contrast with the real-world observations. It is a well-known conclusion that the large time behavior of a species determined by its initial size. In the real-world environment, an animal species either profits from the same species, or it will be hurt by the competition with the same species. This phenomenon is called ‘Allee effect’, namely there exists a critical population density such that the species can establish itself when the density is greater than the critical value, or it will die out on the other hand [6]. To include the ‘Allee effect’, we usually replace the logistic term in (1) by a bistable function as follows:
In 2010, Du used a free boundary problem to describe the spreading of species in [7], [8]. First, Du studied the problem in one space dimension and with a homogeneous environment in [7] and then extended the conclusions to higher space dimensions and a heterogeneous environment in [8]. Considering the following free boundary problem with logistic term in the same way as the problem (1), Du proved both spreading and vanishing can happen depending on the initial size:
where , , (), , and is the moving boundary to be determined, and the initial function satisfies
Problem (2) describes spreading of a new species over an -dimensional habitat with an initial population density , which occupies an initial region . (Here stands for the ball with the center at 0 and radius .) The free boundary stands for the spreading front, which is the boundary of the ball . The radius of the free boundary increases with a speed that is proportional to the population gradient at the front: . In the same way as (1), the coefficient function means an intrinsic growth, represents an intra-specific competition, and is the diffusion rate.
The free boundary is governed by the equation , which is a special case of the well-known Stefan condition. The condition has been applied in a number of problems. For example, it was used to describe the melting of ice in contact with water [9], in the modeling of oxygen in the muscle [10], and in wound healing [11].
Du has proved that the problem (2) admits a unique solution for all the with and . Moreover, compared with the traditional logistic equation, the solution of the free boundary problem (2) is typical of the spreading-vanishing dichotomy. All this means that as the species either successfully spreads to the entire new environment and stabilizes at a positive equilibrium (called spreading), in the case that and , or it fails to establish itself and dies out in the long run (called vanishing), in the sense that and . The criteria for spreading or vanishing are as follows: If the radius of the initial region is greater than a critical size , namely , then the spreading always occurs for all the initial function satisfying (3). On the other hand, if , whether spreading or vanishing happens is determined by the initial population and the coefficient in the Stefan condition.
Compared with the free boundary problem (2) and the problem (1), (2) is more similar to the spreading process in real world. At first, compared with the persistent spreading in the model (1), both spreading and vanishing can occur in the model (2) depending on the initial size. Next, for any finite , the solution of the problem (2) is supported on a finite domain of , which expands with the increase of . However, in the problem (1), the solution is always positive for all the as .
The logistic term of the form has been thoroughly discussed by Du in [7], [8]. In ecology, this logistic term is too simple to describe the phenomenon in the real world. Thus, we will study a more complex logistic term as follows:
Here, the main condition is the same as the problem (2), where , , (), , is a moving boundary to be determined, , and are given positive constants, , , and there exist positive constants such that
The initial function satisfies
Moreover, the logistic nonlinear term satisfy the conditions (A1) and (A2) listed below:
(A1) and is increasing on ;
(A2), where .
Keller [12] and Osserman [13] proposed these conditions in 1957. These conditions have been used widely to study those functions which behave like (). We can easily obtain , from condition (A2). Clearly, is a special case.
In Section 2, we first prove the existence and uniqueness of the local solution for the free boundary problem (4) (Theorem 2.1) by the contraction mapping theorem, then we show that the solution can be expanded to all using suitable estimates (Theorem 2.3). Finally, we prove the spreading-vanishing dichotomy in Section 3.
2 Existence and uniqueness for the free boundary problem
In this section, we will prove the existence and uniqueness for the problem (4). The approaches were introduced in [7] and some changes on it are needed.
Theorem 2.1
For any givensatisfying (5) and any constant, there is asuch that problem (4) has a unique solution
moreover,
where, andonly depend on, and.
Proof
At first, we follow [8] and [11] to straighten the free boundary. Then the problem (4) becomes
where , , , , , .
We denote and for ,
and
It is easily seen that is a complete metric space with the following metric:
For , due to , we have
Next, we use the contraction mapping theorem to prove the existence and uniqueness of the local solution. Firstly, for any given , we have
Thus, for , we have and for such ,
therefore
So, although is singular at ,
still represents an elliptic operator acting on () over the ball , whose coefficients are continuous in when .
Applying theory and the Sobolev imbedding theorem [14], we find that the following initial boundary value problem:
has a unique solution for any and
where is a constant dependent on , and .
Let
we have
and with
Now, we define ,
Clearly is a solution of (8) if and only if it is the fixed point of ℱ.
Thus, if we let , ℱ maps into itself.
Next, we will prove that if is sufficiently small, ℱ is a contraction mapping on . In fact, if we let and denote , we obtain
Setting , then satisfies
Using the estimates and the Sobolev imbedding theorem, we have
where depends on , and , , , . Taking the difference of the equations for , results in
Combining (9), (14), and (15), and assuming , we get
where depends on and . Therefore for
we have
This means that ℱ is a contraction mapping on . By the contraction mapping theorem, we find that ℱ has a unique fixed point in . Moreover, it follows that we have the Schauder estimates and . Moreover, we have (11) and (13). This shows that is a unique local classical solution of the problem (8). □
Next, we will use some suitable estimates to show that the solution can be extended to all .
Lemma 2.2
Letbe the solution to the problem (2) defined onfor some. Then there exist constantsandindependent ofsuch that
Proof
By the strong maximum principle, we have
Thus for .
Due to (5), using the comparison principle, we have for , , where is the solution of following problem:
By the condition (A2), we easily obtain , . Thus, there exists an such that . Clearly, the is the supremum of the problem (16). Therefore, we have
Next, using the approach in [8], it is easy to prove that for , where is independent of . Then the proof is complete. □
Theorem 2.3
The solution of the problem (4) exists and is unique for all. Moreover, the unique solutiondepends continuously onand the parameters appearing in (4).
The proof is the same as Theorem 4.3 in [8]. So we omit the details.
3 Spreading-vanishing dichotomy
By Lemma 2.2, we see that is monotonic and therefore there exists such that . Let be the principal eigenvalue of the problem
It is well known that is a strictly decreasing continuous function and
Thus, for fixed and , there is a unique such that
and
By the following two lemmas, we can obtain the spreading-vanishing dichotomy.
Lemma 3.1
If, then, and.
Proof
We first prove . Arguing by contradiction, we suppose and there is a such that for all the . Therefore, for all , we have
Moreover, for any sufficiently small , there exists a such that
Consider the following problem:
which is a logistic problem . Clearly, the problem (19) has a unique positive solution (see Proposition 3.3 in [10]). We have
where is the unique positive solution of the following problem:
Using the comparison principle
which implies that
On the other hand, consider the problem
where
Clearly, the problem (24) also has a unique positive solution
where is a unique positive solution of the following problem:
In the same way, the comparison principle implies that
and
Using a compactness and uniqueness argument, we can easily obtain
Therefore, it follows from (23), (28) and the arbitrariness of that
By the argument of Lemma 2.2 in [8], we obtain
Thus
which implies that
Hence , which contradicts our assumption . So we have .
Next, we will prove that as . Let be the unique positive solution of the following problem:
where
The comparison principle implies for and . Due to , we have and it follows from a well-known conclusion about logistic equation that uniformly for as (see [10]). Therefore, we get . □
Lemma 3.2
If, then
whereis the unique positive solution of the following problem:
Proof
By [15], we find that the problem (32) has a unique positive solution. Moreover, the solution must be radially symmetric since (32) is invariant under rotations around the origin of .
To prove (31), we use a squeezing argument in [16]. Consider the Dirichlet problem
and the boundary blow-up problem
Clearly, these problems have positive radial solutions and for large . It follows from the comparison principle in [15] that increases to the unique positive solution of (32) as and decreases to .
Choose an increasing sequence such that as , and for all . Then both and converge to as . For each , we can find a such that for . Note that the following problem:
has a unique positive solution , and
Using the comparison principle, we have
Thus
Let , we have
Similarly, by the proof of Theorem 4.1 in [16], we obtain
Let , we have
Therefore, (31) follows from (35) and (36). □
Combining Lemmas 3.1 and 3.2, we can easily obtain the spreading-vanishing dichotomy as follows.
Theorem 3.3
Letbe the solution of the free boundary problem (2). Then the following alternative holds:
-
1.
Spreading: and
-
2.
Vanishing: and .
Next, we will discuss when the two alternatives occur exactly. We divide the argument into two cases:
In case (a), we can easily obtain since for all . Therefore, the following conclusion follows from Lemma 3.1.
Theorem 3.4
If, then.
In the same way as in the discussion in [8], we need a comparison principle which can be used to estimate both and the free boundary to study case (b).
Lemma 3.5
(Comparison principle)
Suppose, , with, and
If we have
then the solutionof the free boundary problem (4) satisfies
Proof
For small , in the problem (4), let replace , replace , and replace , where satisfying
and
We denote by the unique solution of the above problem.
We claim that for all . Clearly, it is true for small . If our claim is wrong, then we will find a first such that for and , which implies
Now, we compare and over the region . Using the strong maximum principle in , we have . Thus with . It follows that . Then we obtain . Due to , it follows that , which contradicts (37). This shows our claim is correct. Then applying the usual comparison principle for , we obtain .
Due to the unique solution depending continuously on the parameters in (4), as , converges to the unique solution of the problem (4). Setting , we can get the conclusion. □
Now, we consider case (b). As in [8], we first examine the case that is large, then we investigate the case is small. Finally, we use Lemma 3.5 to show that there exists a critical such that spreading occurs when and vanishing happens when .
Lemma 3.6
Suppose, then there existsdepending onsuch that spreading occurs when.
Proof
We prove it by contradiction. Suppose that there is an increasing sequence satisfying as such that the unique solution of the problem (4) with satisfies for all the . Thus, by Lemma 3.1, we have and hence
where is the unique solution of following problem:
where
Due to the fact that , we have
Combining (38) and (39), we obtain uniformly for . Therefore, it follows from conditions (A1) and (A2) that there exists a independent of such that
To simplify the discussion, we omit from , , , and in the following argument.
One calculates directly
Integrating from to implies
For and , due to , we have
Then
Let , since (38) and (39), we obtain
and hence
Using Lemma 3.5, and are increasing in . Thus
Thus, from (40) we deduce
This contradicts as . □
Lemma 3.7
Suppose, then there existsdepending onsuch that vanishing occurs when.
Proof
At first, we will construct a suitable upper solution of the problem (2), then we use Lemma 3.5 to obtain the conclusion. For and , define
where , , are positive constants to be chosen later and is the first eigenfunction of the following problem:
with and . Due to , we have
Since we have the fact that and
we have
Let , then . Direct calculation gives
Due to condition (A1), we easily obtain . Since , we can choose sufficiently small such that
Let , we have
Now, we choose sufficiently large such that
Direct calculation implies
Thus if we let
then for any , we have
hence satisfies
Hence, from Lemma 3.5, we have and for , . This implies . □
In the same way as the proof of Theorem 2.1 in [8], we can prove the following theorem.
Theorem 3.8
If, then there exists adepending onsuch that spreading occurs when, and vanishing happens when.
References
Skellam JG: Random dispersal in theoretical populations. Biometrika 1951, 38: 196-218. 10.1093/biomet/38.1-2.196
Shigesada N, Kawasaki K: Biological Invasions: Theory and Practice. Oxford University Press, Oxford; 1997.
Fisher RA: The wave of advance of advantageous genes. Annu. Eugen. 1937, 7: 335-369.
Kolmogorov AN, Petrovsky IG, Piskunov NS: Étude de l’équation de la diffusion avec croissance de la quantité de matière et son application a un problème biologique. Bull. Univ. État Mosc. Sér. Int. A 1937, 1: 1-26.
Aronson DG, Weinberger HF: Nonlinear diffusion in population genetics, combustion, and nerve pulse propagation. Partial Differential Equations and Related Topics 1975, 5-49. 10.1007/BFb0070595
Taylor CM, Hastings A: Allee effects in biological invasions. Ecol. Lett. 2005, 8: 895-908. 10.1111/j.1461-0248.2005.00787.x
Du Y, Lin ZG: Spreading-vanishing dichotomy in the diffusive logistic model with a free boundary. SIAM J. Math. Anal. 2010, 42: 377-405. 10.1137/090771089
Du Y, Guo ZM: Spreading-vanishing dichotomy in the diffusive logistic model with a free boundary, II. J. Differ. Equ. 2011, 250: 4336-4366. 10.1016/j.jde.2011.02.011
Rubinstein LI: The Stefan Problem. Am. Math. Soc., Providence; 1971.
Cantrell RS, Cosner C: Spatial Ecology via Reaction-Diffusion Equations. Wiley, New York; 2003.
Chen XF, Friedman A: A free boundary problem arising in a model of wound healing. SIAM J. Math. Anal. 2000, 32: 778-800. 10.1137/S0036141099351693
Keller JB:On solutions of . Commun. Pure Appl. Math. 1957, 10: 503-510. 10.1002/cpa.3160100402
Osserman R:On the inequality . Pac. J. Math. 1957, 7: 1641-1647. 10.2140/pjm.1957.7.1641
Ladyzenskaja OA, Solonnikov VA, Ural’ceva NN: Linear and Quasilinear Equations of Parabolic Type. Am. Math. Soc., Providence; 1968.
Dong W, Liu L:Uniqueness and existence of positive solutions for degenerate logistic type elliptic equations on . Nonlinear Anal. 2007, 67: 1226-1235. 10.1016/j.na.2006.07.009
Du Y, Ma L:Logistic type equations on by a squeezing method involving boundary blow-up solutions. J. Lond. Math. Soc. 2001, 64: 107-124. 10.1017/S0024610701002289
Acknowledgements
The authors thank the referees much for their helpful suggestions. This work was supported by the China Postdoctoral Science Foundation and the Heilongjiang Province Postdoctoral Science Foundation.
Author information
Authors and Affiliations
Corresponding author
Additional information
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
XD provided the main ideas and lead to write the paper. YL and XJ proved the local existence part. LM and KS proved the rest. HQ checked all the arguments. All authors read and approved the final manuscript.
Authors’ original submitted files for images
Below are the links to the authors’ original submitted files for images.
Rights and permissions
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0), which permits use, duplication, adaptation, distribution, and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.
About this article
Cite this article
Dong, X., Li, Y., Jiang, X. et al. Spreading-vanishing dichotomy in a degenerate logistic model with general logistic nonlinear term. Bound Value Probl 2014, 180 (2014). https://doi.org/10.1186/s13661-014-0180-9
Received:
Accepted:
Published:
DOI: https://doi.org/10.1186/s13661-014-0180-9