Boundedness and stabilization in a two-species chemotaxis-competition system of parabolic-parabolic-elliptic type
Masaaki Mizukami

TL;DR
This paper establishes conditions for global bounded solutions and convergence rates in a two-species chemotaxis-competition system, extending previous results to cases where parameters are greater than or equal to one.
Contribution
It introduces new conditions ensuring global existence and stability for all positive parameters, including the previously unaddressed cases where parameters are at least one.
Findings
Global existence of bounded solutions for all positive parameters.
Convergence rates for solutions when parameters are in specific ranges.
Extended stability analysis beyond previous parameter restrictions.
Abstract
This paper deals with the two-species chemotaxis-competition system , , , where is a bounded domain in with smooth boundary, ; and are constants satisfying some conditions. The above system was studied in the cases that and , and it was proved that global existence and asymptotic stability hold when are small. However, the conditions in the above two cases strongly depend on , and have not been obtained in the case that . Moreover, convergence rates in the cases that and have not been studied. The purpose of this work is to…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
0002010Mathematics Subject Classification. Primary: 35K51; Secondary: 92C17, 35B40. 000*Key words and phrases: chemotaxis; Lotka–Volterra; global existence; stabilization. *
**Boundedness and stabilization in a two-species chemotaxis-competition system of parabolic-parabolic-elliptic type **
Masaaki Mizukami 000 E-mail: [email protected]
Department of Mathematics, Tokyo University of Science
1-3, Kagurazaka, Shinjuku-ku, Tokyo 162-8601, Japan
- Abstract. This paper deals with the two-species chemotaxis-competition system
[TABLE]
where is a bounded domain in with smooth boundary , ; and are constants satisfying some conditions. The above system was studied in the cases that and , and it was proved that global existence and asymptotic stability hold when are small ([5, 32, 34]). However, the conditions in the above two cases strongly depend on , and have not been obtained in the case that . Moreover, convergence rates in the cases that and have not been studied. The purpose of this work is to construct conditions which derive global existence of classical bounded solutions for all which covers the case that , and lead to convergence rates for solutions of the above system in the cases that and .
1 Introduction
Many phenomena, which appear in natural science, especially, biology, chemistry and physics, support animals’ lives. In this paper we focus on chemotaxis which is one of the important properties and is related to e.g., movement of sperm, migration of neurons or lymphocytes and tumor invasion. Chemotaxis is the property such that species move towards higher concentration of the chemical substance when they plunge into hunger.
A mathematical problem which describes a part of the life cycle of cellular slime molds with chemotaxis is called the Keller–Segel system:
[TABLE]
where and . Moreover, the chemotaxis system with growth terms
[TABLE]
was proposed by [25, 31], where and . After the pioneering work of Keller–Segel [19], the Keller–Segel system and the chemotaxis system are intensively studied (see e.g., [2, 13, 15]). A generalized problem of Keller–Segel systems, which means a two-species chemotaxis system, was proposed in [36] and also has studied (see e.g., [3, 4, 7, 8, 10, 21, 37]; global existence was proved in [7, 8, 37]; and thier asymptotic stability was shown in [37]; related works which deal with blow-up of solutions can be seen in [3, 4, 7, 8, 10, 21]). Recently, a two-species chemotaxis system with competitive kinetics
[TABLE]
with some and , which describes the evolution of two competing species which react on a single chemoattractant, was proposed by Tello–Winkler [34] and was studied (see [1, 22, 26, 27, 28, 29, 30, 38]). About this problem with , global existence and boundedness was obtained in the 2-dimensional case ([1]) and the -dimensional setting ([22]); moreover, asymptotic behavior of solutions was established in [1, 27]. Related works which dealt with global existence and boundedness in this two-species problem with sensitivity functions can be found in [27, 38]; and related works which treated the non-competition case are in [26, 28, 29, 30]. These results in the case are motivated by the results ([5, 32, 34]) in the case . Therefore the parabolic-parabolic-elliptic problem reduced by letting seems to be helpful to analyze the fully parabolic case.
In this paper we consider the two-species chemotaxis system with competitive kinetics of parabolic-parabolic-elliptic type
[TABLE]
where is a bounded domain in () with smooth boundary and is the outward normal vector to . The constants and are positive. The initial data are assumed to be nonnegative functions. The unknown functions and represent the population densities of two species and shows the concentration of the chemical substance at place and time .
The problem (1.1) is a problem on account of the influence of chemotaxis, diffusion, and the Lotka–Volterra competitive kinetics, i.e., with coupling coefficients in
[TABLE]
The mathematical difficulties of the problem (1.1) are to deal with the chemotaxis term and the competition term . To overcome these difficulties, in the case that and in (1.1), Tello–Winkler [34] applied comparison methods to this problem and obtained global existence of classical bounded solutions and their asymptotic behavior under the conditions that
[TABLE]
However, if or , then these conditions break down. Recently, it was shown that the conditions
[TABLE]
lead to global existence and asymptotic stability in (1.1) in the case that ([5]). The conditions (1.4)–(1.5) partially relax (1.3) in view of the point mentioned above. On the other hand, in the case that and in (1.1) Stinner–Tello–Winkler [32] established global existence and stabilization of global classical solutions when
[TABLE]
are satisfied. In summary the two-species chemotaxis-competition model (1.1) were studied in the cases that and , and it was proved that global existence and same asymptotic behavior as solutions to the Lotka–Volterra competition model (1.2) hold when are small. However, the conditions in the above two cases strongly depend on , and have not been obtained in the case that . Moreover, convergence rates in the cases that and have not been studied.
The purpose of this work is to construct conditions which derive global existence of classical bounded solutions for all which covers the case that , and lead to convergence rates for solutions of (1.1) in the cases that and .
For establishing global existence and boundedness we shall suppose that and satisfy the following conditions:
[TABLE]
We assume that the initial data satisfy
[TABLE]
Now the main results read as follows. The first one is concerned with global existence and boundedness in (1.1).
Theorem 1.1**.**
Let , , , , and let be a bounded domain with smooth boundary. Assume that (1.6) are satisfied. Then for any satisfying (1.7) with some , there exists an exactly one pair of nonnegative functions
[TABLE]
which satisfy (1.1). Moreover, the solutions are uniformly bounded, i.e., there exists a constant such that
[TABLE]
and the solutions are the Hölder continuous functions, i.e., there exist and such that
[TABLE]
Remark 1.1**.**
This result give the existence of global classical bounded solutions in the case that . Moreover, the condition (1.6) relaxes (1.4) which assumed for global existence of solutions in [5]. Indeed, if and satisfy the condition (1.4), then and satisfy the condition (1.6). However, the condition (1.6) does not always relax those assumed in [32] and [34]; in the case that the condition (1.6) relaxes (1.3) under the condition
[TABLE]
and in the case that the condition (1.6) relaxes the condition
[TABLE]
which was used to obtain global existence in [32], when
[TABLE]
hold.
The main theorem tells us the following result in the 2-dimensional case.
Corollary 1.2**.**
Let , , , , and let be a bounded domain with smooth boundary. Then for any satisfying (1.7) with some , (1.1) possesses a unique global bounded classical solution.
In the case asymptotic behavior of solutions to (1.1) will be discussed under the following additional conditions: there exists such that
[TABLE]
and
[TABLE]
The second theorem gives asymptotic behavior in (1.1) in the case .
Theorem 1.3**.**
Let , , , , and let be a bounded domain with smooth boundary. Assume that there exists a unique global classical solution of (1.1) satisfying
[TABLE]
with some . Then under the conditions (1.8)–*(1.10), satisfies that there exist and such that *
[TABLE]
where
[TABLE]
Remark 1.2**.**
If the assumption of Theorem 1.1 and (1.8)–(1.10) are satisfied, then Theorem 1.3 gives the convergence rates for solutions of (1.1) in the case that . Moreover, the conditions (1.8)–(1.10) are the same conditions as that assumed in [27] in the case that and .
In the case asymptotic behavior of solutions to (1.1) will be discussed under the following additional conditions: there exist and such that
[TABLE]
The third one gives asymptotic behavior in (1.1) in the case .
Theorem 1.4**.**
Let , , , , , and let be a bounded domain with smooth boundary. Assume that there exists a unique global classical solution of (1.1) such that
[TABLE]
with some . Then under the conditions (1.11)–(1.12), has the following properties:**
- (i)
If and take in (1.11)–(1.12), then there exist and satisfying
[TABLE] 2. (ii)
If , then there exist and satisfying
[TABLE]
Remark 1.3**.**
If the assumption of Theorem 1.1 and (1.11)–(1.12) are satisfied, then Theorem 1.4 gives the convergence rates for solutions in the cases that and . Moreover, the conditions (1.11)–(1.12) are the same conditions as that assumed in [27] in the case that and .
Remark 1.4**.**
Stabilization in the case that is a still open question. In the case that a Lotka–Volterra competition model with diffusion term was studied; however, its analysis is difficult and it is known that solutions have complicated structures (see cf. [9, 16, 17, 18, 20, 23, 24]).
The strategy of the proof of Theorem 1.1 is to extend a method in [33] to a two-species case. We first aim to establish the -estimate for with some from the following derivative of :
[TABLE]
Since the third equation in (1.1) derives that
[TABLE]
we shall show that a combination of (1.13) and (1.14), along with the condition (1.6) implies
[TABLE]
which leads to -estimate for . Then aided by standard semigroup estimates, we can obtain the -estimate for . On the other hand, one of the keys for the proof of Theorems 1.3 and 1.4 is to derive the following energy estimate:
[TABLE]
for all with some positive function and some constant , where is a solution of (1.1). Thanks to (1.15), we can obtain that there exists such that
[TABLE]
which together with the regularity of the solution leads to Theorems 1.3 and 1.4.
This paper is organized as follows. In Section 2 we prove global existence and boundedness (Theorem 1.1) through a series of lemmas. Section 3 is devoted to the proof of asymptotic stability (Theorems 1.3 and 1.4); we first provide some lemmas which will be used later, and we next devide the section into Sections 3.1 and 3.2 according to the proof of Theorem 1.3 and that of Theorem 1.4, respectivly.
2 Global existence and boundedness
In this section we shall show global existence and boundedness in (1.1). First we will recall the known result about local existence of solutions to (1.1) ([5, Lemma 2.1], [32, Lemma 2.1]).
Lemma 2.1**.**
Let , , , , and let be a bounded domain with smooth boundary. Then for any satisfying (1.7) for some , there exist and an exactly one pair of nonnegative functions
[TABLE]
*which satisfy (1.1). Moreover, *
[TABLE]
We next give the -estimate for with some which plays an important role in deriving -estimate for . The proof is based on the proof of [33, Lemma 2.2].
Lemma 2.2**.**
Assume that (1.6)–(1.7) are satisfied. Then for all , there exists such that
[TABLE]
for all , where
[TABLE]
Proof.
We fix . Here we note from the condition (1.6) that . Multiplying the first equation in (1.1) by and integrating it over , we obtain that
[TABLE]
Then integration by parts and the third equation in (1.1) imply that
[TABLE]
Therefore a combination of (2.1) with (2.2) yields that
[TABLE]
Recalling that
[TABLE]
we establish from the Hölder inequality
[TABLE]
that there exists satisfying
[TABLE]
which implies that
[TABLE]
Thus we can attain the conclusion of this lemma. ∎
Similarly, we can confirm the following -estimate for with some .
Lemma 2.3**.**
Assume that (1.6)–(1.7) are satisfied. Then for all , there exists such that
[TABLE]
where .
Proof.
A similar argument as in the proof of Lemma 2.3 derives this lemma. ∎
Now we could construct all estimates which will enable us to obtain the estimate for the solution; Lemmas 2.2 and 2.3 lead to the following lemma. The proof is based on a known argument involving semigroup estimates which derive the -estimate for from -estimate with (see e.g., [2]).
Lemma 2.4**.**
Assume that (1.6)–(1.7) are satisfied. Then there exists such that
[TABLE]
Moreover, there exist and such that
[TABLE]
Proof.
We fix , where and are the intervals defined in Lemmas 2.2 and 2.3. Then thanks to Lemmas 2.2 and 2.3, we can find such that
[TABLE]
We first verify the -estimate for . Here for all , the standard elliptic regularity argument (see e.g., [11, Theorem 19.1]) leads to the existence of a constant satisfying
[TABLE]
Therefore a combination of (2.5) with (2.4) yields from the Sobolev embedding theorem that there exists such that
[TABLE]
since . We next establish the -estimate for . Since , we can take such that
[TABLE]
We take satisfying
[TABLE]
Then satisfies
[TABLE]
Now for all we note that
[TABLE]
is finite. To obtain the estimate for we put and represent according to
[TABLE]
for . In the case that , i.e., , from the order preserving property of the Neumann heat semigroup we see that
[TABLE]
In the case that using the - estimate for (see [35, Lemma 1.3]) yields that there is such that
[TABLE]
Next due to a known smoothing property of (see [12, Lemma 3.3]), we can find such that
[TABLE]
Noting from and (2.4) that
[TABLE]
with some , we establish that there exists such that
[TABLE]
Finally, the maximum principle together with the elementary inequality
[TABLE]
implies that there exists such that
[TABLE]
Therefore a combination of (2.6), the nonnegativity of with (2.7), (2.8), (2.9), (2.10) tells us that there exist such that
[TABLE]
which implies from that
[TABLE]
with some . Thus we obtain the -estimate for . Similarly, we can verify the -estimate for . Then invoking (2.5), we see that there exists such that
[TABLE]
which implies (2.3). Moreover, known regularity arguments (see [6, Proposition 2.3]) enable us to find and satisfying
[TABLE]
which implies the end of the proof. ∎
Proof of Theorem 1.1.
Lemma 2.4 directly shows Theorem 1.1. ∎
3 Stabilization
In this section we will establish stabilization of solutions to (1.1). Here we assume that there exists a unique global classical solution of (1.1) satisfying
[TABLE]
with some . we first recall a important lemma for the proof of Theorems 1.3 and 1.4 (see [14, Lemma 4.6]).
Lemma 3.1**.**
Let satisfy that there exist constants and such that
[TABLE]
Assume that
[TABLE]
with some constant . Then
[TABLE]
We next provide the following lemma which will be used to confirm that the assumption of Lemma 3.1 is satisfied.
Lemma 3.2**.**
Let . Suppose that
[TABLE]
Then
[TABLE]
holds for all .
Proof.
Straightforward calculations lead to the conclusion of this lemma (for more details, see [27, Lemma 3.2]). ∎
Finally, we give the following lemma which enables us to upgrade the -convergence rate to -convergence rate.
Lemma 3.3**.**
Let be a solution to (1.1). Assume that there exists a decreasing function satisfying
[TABLE]
Then there exists such that
[TABLE]
for all .
Proof.
For all we first obtain from the Hölder inequality that
[TABLE]
holds for all , which means from the boundedness of that
[TABLE]
with some . Here (2.5) enables us to see that
[TABLE]
Thus we have that there is such that
[TABLE]
Then by using a similar argument as in the proof of [1, Lemma 3.6] we infer that there exists such that
[TABLE]
Finally, since satisfies
[TABLE]
we can apply the maximum principle to
[TABLE]
and hence obtain the existence of a constant such that
[TABLE]
which concludes the proof of this lemma. ∎
3.1 Convergence. Case 1:
In this subsection we establish stabilization in the case that . We first confirm that the assumption of Lemma 3.1 are satisfied.
Lemma 3.4**.**
Assume that (1.8)–(1.10) are satisfied. Then there exist a nonnegative function and a constant such that
[TABLE]
holds for all . Moreover, there exists satisfying
[TABLE]
Proof.
Let be a constant defined in (1.8)–(1.10). First we shall show that the function defined as
[TABLE]
satisfies that (3.1) holds for all with some . From straightforward calculations we infer that
[TABLE]
Here in light of (1.8)–(1.10) we can take satisfying
[TABLE]
and
[TABLE]
Invoking the Young inequality, we obtain that
[TABLE]
and
[TABLE]
Therefore since the definition of yields
[TABLE]
a combination of (3.3) with (3.4) and (3.5) implies
[TABLE]
Noting from the third equation in (1.1) that
[TABLE]
we establish that
[TABLE]
where
[TABLE]
for all . In order to see (3.1) we will show that
[TABLE]
with some by using Lemma 3.2. To confirm that the assumption of Lemma 3.2 is satisfied we put
[TABLE]
for , and shall see that there exists such that for . Here obviously holds, and the condition (1.8) implies that
[TABLE]
Moreover, aided by the definition of , we can obtain that
[TABLE]
Therefore a combination of the above inequalities and the continuity argument implies that there exists such that hold for . Thus Lemma 3.2 derives that
[TABLE]
which yields that (3.1) holds for all . Then since from the Taylor formula is a nonnegative function for (more details, see [1, Lemma 3.2]), integrating (3.1) over concludes the proof of this lemma. ∎
Lemma 3.5**.**
Assume that (1.8)–(1.10) are satisfied. Then
[TABLE]
Proof.
A combination of Lemmas 3.1 and 3.4 implies this lemma. ∎
Next we desire to establish convergence rates for the solution of (1.1). We note that in view of Lemma 3.3 it is sufficient to confirm the -convergence rates for the solution.
Lemma 3.6**.**
Assume that (1.8)–(1.10) are satisfied. Then there exist and such that
[TABLE]
Proof.
Aided by Lemma 3.5 and the L’Hôpital theorem, a similar argument as in the proof of [1, Lemma 3.7] (or [27, Proof of Theorem 1.2]) derives that there exist and such that for all ,
[TABLE]
where is the function defined as (3.2). Therefore we obtain from (3.1) that
[TABLE]
with some , which implies that there exists such that
[TABLE]
Thus a combination of (3.6) and (3.7) yields that
[TABLE]
which concludes the proof of this lemma. ∎
3.2 Convergence. Case 2:
In this subsection we will obtain stabilization in the case that . In this case we also have to confirm that the assumption of Lemma 3.1 is satisfied.
Lemma 3.7**.**
Assume that (1.11)–(1.12) are satisfied. Then there exist a nonnegative function and constants and such that
[TABLE]
holds for all . Moreover, there exists satisfying
[TABLE]
Proof.
Let and be a constant defined in (1.11)–(1.12). We first show that the function defined as
[TABLE]
fulfils (3.8) for all with and some . Noting from the relation that
[TABLE]
from straightforward calculations we derive that
[TABLE]
for all . Here thanks to (1.11)–(1.12), we can take such that
[TABLE]
Invoking the Young inequality, we obtain that
[TABLE]
Therefore since the definition of yields
[TABLE]
the equation (3.10) implies that
[TABLE]
for all , where . Noting from the third equation in (1.1) that
[TABLE]
we establish that for all ,
[TABLE]
where
[TABLE]
Then by using the same argument as in the proof of Lemma 3.4 we can see that
[TABLE]
with some , which means that (3.8) holds with and . ∎
Then we will establish the convergence result for the solution to (1.1) in the case that .
Lemma 3.8**.**
Assume that (1.11)–(1.12) are satisfied. Then we have
[TABLE]
Proof.
A combination of Lemmas 3.1 and 3.7 implies this lemma. ∎
Finally, we shall show two lemmas which give asymptotic behavior in the case that .
Lemma 3.9**.**
Let and . Assume that (1.11)–(1.12) are satisfied with and . Then there exist and satisfying
[TABLE]
Proof.
In the case that and a similar argument as in the proof of [27, Lemmas 4.3] enables us to see that there exist and such that
[TABLE]
where is the function defined as (3.9) and
[TABLE]
Thus a combination of the above inequality and (3.8) means that
[TABLE]
holds for all , which together with the same argument as in the proof of Lemma 3.6 leads to the conclusion of this lemma in the case that and . ∎
Lemma 3.10**.**
Let and . Assume that (1.11)–(1.12) are satisfied. Then there exist and satisfying
[TABLE]
Proof.
First we can verify from the same argument as in the proof of [27, Lemma 3.7] that there exist and such that
[TABLE]
for all . Hence it follows from the Cauchy–Schwarz inequality and the boundedness of that
[TABLE]
which implies from (3.8) that
[TABLE]
Therefore we can find such that
[TABLE]
Therefore thanks to the boundedness of and (3.11), we obtain that
[TABLE]
which proves this lemma. ∎
Proof of Theorems 1.3 and 1.4.
A combination of Lemmas 3.6, 3.9, 3.10 and 3.3 immediately leads to the conclusions of these theorems. ∎
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1[1] X. Bai and M. Winkler. Equilibration in a fully parabolic two-species chemotaxis system with competitive kinetics. Indiana Univ. Math. J. , 65:553–583, 2016.
- 2[2] N. Bellomo, A. Bellouquid, Y. Tao, and M. Winkler. Toward a mathematical theory of Keller–Segel models of pattern formation in biological tissues. Math. Models Methods Appl. Sci. , 25:1663–1763, 2015.
- 3[3] P. Biler, E. E. Espejo, and I. Guerra. Blowup in higher dimensional two species chemotactic systems. Commun. Pure Appl. Anal. , 12:89–98, 2013.
- 4[4] P. Biler and I. Guerra. Blowup and self-similar solutions for two-component drift-diffusion systems. Nonlinear Anal. , 75:5186–5193, 2012.
- 5[5] T. Black, J. Lankeit, and M. Mizukami. On the weakly competitive case in a two-species chemotaxis model. IMA J. Appl. Math. , 81:860–876, 2016.
- 6[6] J. Cao, W. Wang, and H. Yu. Asymptotic behavior of solutions to two-dimensional chemotaxis system with logistic source and singular sensitivity. J. Math. Anal. Appl. , 436:382–392, 2016.
- 7[7] C. Conca and E. E. Espejo Arenas. Threshold condition for global existence and blow-up to a radially symmetric drift-diffusion system. Appl. Math. Lett. , 25:352–356, 2012.
- 8[8] C. Conca, E. E. Espejo Arenas, and K. Vilches. Remarks on the blowup and global existence for a two species chemotactic Keller–Segel system in ℝ 2 superscript ℝ 2 \mathbb{R}^{2} . European J. Appl. Math , 22:553–580, 2011.
