A sustainability condition for stochastic forest model (Q507308)

From MaRDI portal
scientific article
Language Label Description Also known as
English
A sustainability condition for stochastic forest model
scientific article

    Statements

    A sustainability condition for stochastic forest model (English)
    0 references
    0 references
    0 references
    0 references
    0 references
    3 February 2017
    0 references
    The authors study a stochastic forest model given by the two-dimensional stochastic differential system \[ \begin{split} du_t &= \{ \rho v_t - [a(v_t- b)^2 + c + f] u_t \} dt \\ dv_t &= (f u_t - h v_t) dt+ \sigma v_t dw_t. \end{split} \] Here $u$ and $v$ denote the tree densities of young and old age classes. The parameters $a,b,c,\rho,f > 0$ are positive constants. The parameter $\rho$ is the reproduction rate of old trees, the parameter $h$ is the mortality of old trees, and the parameter $f$ is the aging rate of young trees. Furthermore, the function $v \mapsto a(v-b)^2 + c$ is the mortality of young trees, which may depend on the old-tree density. The process $\{ w_t, t \geq 0 \}$ is a one-dimensional Brownian motion, and the constant $\sigma > 0$ is the intensity of the white noise. \par The authors prove existence and uniqueness of non-negative solutions for every starting point, and that these solutions are bounded. More precisely, let \[ \mathbb{R}_+^2 = \{ (u,v) \in \mathbb{R}^2 : u > 0 \text{ and } v > 0 \}, \] and denote by $\overline{\mathbb{R}_+^2}$ its closure in $\mathbb{R}^2$. Then for each starting point $(u_0,v_0) \in \overline{\mathbb{R}_+^2}$ there is a unique global $\overline{\mathbb{R}_+^2}$-valued solution starting in $(u_0,v_0)$, and if $u_0 + v_0 > 0$, then the solution is $\mathbb{R}_+^2$-valued; see Theorem 2.2. In either case, the solution is bounded; see Theorem 2.3 for this result and further statements. \par The authors also provide sufficient conditions in terms of the parameters for sustainability and for decline of the forest. More precisely, if \[ h < \frac{\rho f}{ab^2 + c + f} \quad \text{and} \quad \sigma^2 < 2 \bigg( \frac{\rho f}{a b^2 + c + f} - h \bigg), \] then the stochastic differential equation is sustainable, which means that \[ \limsup_{t \to \infty} \mathbb{E} u_t > 0 \quad \text{and} \quad \limsup_{t \to \infty} \mathbb{E} v_t > 0; \] see Theorem 3.2. Furthermore, in this case there exists a Borel invariant measure; see Theorem 3.5 for details. Concerning decline of the forest, the authors provide in Theorem 4.1 sufficient conditions such that \[ \lim_{t \to \infty} \mathbb{E} u_t = 0 \quad \text{and} \quad \lim_{t \to \infty} \mathbb{E} v_t = 0, \] and in Theorem 4.2 sufficient conditions such that almost surely \[ \lim_{t \to \infty} u_t = 0 \quad \text{and} \quad \lim_{t \to \infty} v_t = 0. \] Some numerical examples conclude the paper.
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    forest model
    0 references
    sustainability
    0 references
    stochastic differential equation
    0 references
    Markov process
    0 references
    0 references
    0 references