1: \documentclass[twocolumn,showpacs,preprintnumbers,amsmath,amssymb]{revtex4}
2: %\documentclass[preprint,showpacs,preprintnumbers,amsmath,amssymb]{revtex4}
3: \usepackage{graphicx}% Include figure files
4: \usepackage{dcolumn}% Align table columns on decimal point
5: %\usepackage{cite}
6: \usepackage{bm}% bold math
7: \begin{document}
8:
9: \title{THE SIMPLEST MODEL OF SPATIALLY DISTRIBUTED POPULATION WITH REASONABLE MIGRATION OF ORGANISMS}
10:
11: \author{Michael G.Sadovsky}%
12: \email{msad@icm.krasn.ru}
13: \affiliation{%
14: Institute of biophysics of SD of RAS;\\ 660036 Russia, Krasnoyarsk, Akademgorodok.
15: }%
16:
17: \begin{abstract}
18: The simplest model of a smart spatial redistribution of individuals is proposed. A
19: single-species population is considered, to be composed of two discrete subpopulations
20: inhabiting two stations; migration is a transfer between them. The migration is not
21: random and yields the maximization of net reproduction, with respect to the transaction
22: costs. The organisms are supposed to be globally informed. Discrete time model is
23: studied, since it shows all the features of a smart migrations, while the continuous time
24: case brings no new knowledge but the technical problems. Some properties of the model are
25: studied and discussed.
26: \end{abstract}
27:
28: \pacs{87.23.Cc, 87.23.Kg, 87.23.-n}
29:
30: \maketitle
31:
32: \section{\label{introd}Introduction}
33: Modelling of the dynamics of biological takes start from the works by V.Volterra
34: \cite{volt,volt1} (see also \cite{lotka}). The approach follows, in brief, a chemical
35: kinetics, where reproduction, death and other types of interaction of organisms are
36: described with law of mass action. A spatially distributed population is modelled with
37: the ``reaction~--~diffusion'' system, where diffusion is supposed to describe the
38: redistribution of organisms over a space. Good coincidence of the solutions of such
39: ``reaction~--~diffusion'' systems to observed dynamics of populations and communities
40: just masks the serious problem.
41:
42: The point is that the diversity and abundance of possible (structurally stable) regimes
43: of such dynamic systems exceed drastically any really available family of trajectories of
44: real systems. Thus, one always is able to match a differential equation (or a system of
45: differential equation, either ordinary, or partial differential equations) to the dynamic
46: behavior of any really observed system. Doubtlessly, living organisms, even
47: microorganisms, differ drastically in their ``microscopic'' behavior from the chemical
48: substances and relevant chemical reactions.
49:
50: Vito Volterra, the founder of the mathematical biology of populations understood pretty
51: well this discrepancy. Later, the comprehension of limitations of this chemical
52: methodology fell off. The situation is going worse when one tends to model a dynamics of
53: spatially distributed populations and communities. Basically, ``reaction -- diffusion''
54: systems make the basis for modelling of such spatially distributed entities
55: \cite{lotka,1,ber,ber3,xehi,alstad,murray,thieme,edel2,bern,turch,gi-co04}. Famous soviet
56: mathematician Andrew Kolmogorov studied in detail such type of equations \cite{kolmog}.
57:
58: A methodology of modelling of spatially distributed populations based of the reaction --
59: diffusion system has the great discrepancy. It constrains significantly the individual
60: (behavioral) properties of organisms under consideration: one must suppose that the
61: organisms move over a space randomly and aimlessly \cite{ecol2003}. Obviously, such
62: assumption does not hold true (see, e.g. \cite{sci2000,insect,annals}), even for
63: microorganisms \cite{bolgar,zelkniga,levitt,ameba}. The assumption towards the idle
64: transfers of organisms in space is obviously less favorable for the higher organised
65: species.
66:
67: Modelling methodology based on evolution principles is the way to pass over the
68: discrepancy mentioned above. This approach takes the origin from the evolution studies of
69: J.B.S.Haldane \cite{haldane}. This is the most general principle prescribing the way
70: biological systems evolve. In brief, it force to evolve a system toward the maximization
71: of net reproduction. This latter is an average number of {\sl per capita} descendants
72: survived at the course of a series of reproductions over an arbitrary long generation
73: line \cite{otbor1,otbor2,otbor3,otbor4}. Later, they found this principle to be even more
74: general, than just a biological one. Indeed, the principle holds true for any system
75: where the inheritance takes place \cite{rozonoer,zah-usp,zah-spek,lvov,ezer,na2}.
76:
77: A consistent and rational implementation of this principle faces the problem of a lack of
78: knowledge of how the specific biological issues impact the survival of a species. In
79: turn, the question arises, what is an entity to be evolving? An ordinary answer on this
80: question is that the species is an evolving entity. Actually, the situation looks more
81: complicated; not discussing this problem in detail, further we shall follow this idea.
82: The principle formulated above yields the following rule for the model implemented below:
83: evolution optimality in spatial distribution of organisms is equivalent to the
84: maximization of (an average) net reproduction over space, with respect to the evolution
85: trade off for such redistribution. Some further details on this issue could be found in
86: \cite{otbor1,otbor4,na2,na4}.
87:
88: \section{Basic model of the smart migration}
89: Consider a population inhabiting two stations; hence, the population consists of two
90: subpopulations. Any movements of individuals within a station must be neglected; thus,
91: only the transfer from station to station is considered as a migration act. No spatial
92: effects in the population dynamics are presumed, for each subpopulation, as soon as no
93: migration occurs. We shall consider a discrete time model; continuous modelling is
94: possible, as well, but it brings no new issues but the serious technical difficulties.
95:
96: Further, the dynamics of each subpopulation is supposed to follow the Verchulst's
97: equation \cite{verh,sharkov1,sharkov,cyc,progress}. Namely, let $N_t$ and $M_t$ denote
98: the abundance of the first subpopulation (of the second one, respectively), so that
99: \begin{subequations}\label{eq:1}
100: \begin{equation}\label{eq:1-1}
101: N_{t+1} = a \cdot N_t - b \cdot N_t^2
102: \end{equation}
103: \textrm{and}
104: \begin{equation}\label{eq:1-2}
105: M_{t+1} = c \cdot M_t - d \cdot M_t^2\;,
106: \end{equation}
107: \end{subequations}
108: respectively, for the case of the absence of migration. We shall implement the migration
109: into consideration later. Here $a$ and $c$ represent fertility of the relevant
110: subpopulation, while $b$ and $d$ show the effect of density dependent competition within
111: a subpopulation, each. It must be said, that unlike for the Verchulst's equation of a
112: single population, a student may not change the equations (\ref{eq:1-1}, \ref{eq:1-2})
113: for the dimensionless form, eliminating the coefficients $b$ and $d$. The point is that
114: the migration effects would break down such transformation. The functions
115: \begin{subequations}\label{eq:2}
116: \begin{equation}\label{eq:2-1}
117: k_r(N_t) = a - b \cdot N_t
118: \end{equation}
119: \textrm{and}
120: \begin{equation}\label{eq:2-2}
121: k_l(M_t) = c - d \cdot M_t
122: \end{equation}
123: \end{subequations}
124: are the reproduction coefficients, in relevant stations, respectively.
125:
126: Migration is a transfer of a part of subpopulation from one station into the other.
127: Migration itself affects the reproduction, survival and other vital functions of an
128: organism. All these issues will be integrated into the parameter called the {\sl cost of
129: migration} $\mu$: no negative impact on the reproduction, survival and other vital
130: functions of an organism is observed, as $\mu = 0$. Otherwise, the growth of $\mu$ yields
131: a decay in survival of organisms. Rather simple way to introduce the migration cost is
132: proposed in \cite{zob89,monit89,moya}: $$p = \exp\{- \,\mu\}\;,$$ so that $p$, $0 < p
133: \leq 1$ becomes a probability of the transfer from station to station with no damages for
134: further reproduction.
135:
136: Migration starts, as soon as the conditions of life (measured in the units of
137: reproduction coefficient (\ref{eq:2})) ``there'' becomes better, than ``here''. This
138: narrative issue could be easily transformed into the formal way. Migration from
139: $N$-station (from $M$-station, respectively) starts, if $p \cdot k_l(M) > k_r(N)$ (if
140: $k_l(M) < p \cdot k_r(N)$, respectively) for the given time moment $t$, $t = 1, 2, 3,
141: \ldots\,$. We shall suppose, the migration runs as a single act, entirely, and the
142: reproduction always follows the migration. Thus, a life cycle of a population consists of
143: two stages: the former is a migration (if it takes place at the given time moment $t$),
144: and the latter is the reproduction ran according to (\ref{eq:1}), with the abundances
145: appeared due to migration, at each station, independently.
146:
147: Schematically, the model works in three steps. \hfil \hfil \hfil \linebreak \textbf{The
148: first step} consists in the determination of the fact of migration, i.e. in the
149: comparison of reproduction coefficients ``there'' and ``here'':
150: \begin{itemize}
151: \item if $p \cdot k_l(M) > k_r(N)$, then the migration runs from $N$-station to
152: $M$-station;
153: \item if $k_l(M) < p \cdot k_r(N)$, then the migration runs from $M$-station
154: to $N$-station;
155: \item otherwise no migration takes place.
156: \end{itemize}
157: \textbf{The second step} consists in the determination of the migration flux. Migration
158: yields the change of life conditions measured in the units of reproduction coefficient
159: (\ref{eq:2}). An emigration results in the growth of the coefficient, reciprocally,
160: immigration results in the decrease of that latter, since the coefficients (\ref{eq:2})
161: are supposed to be a linear descending function of the abundance $N$ (or $M$). The number
162: of migrating individuals tends to equalize the reproductions coefficients, due to an
163: abundance redistribution: $$k_l(M - \Delta) = p \cdot k_r(N+ p \cdot \Delta)$$ for the
164: case of migration from $M$-station to $N$-station, and vice versa: $$p \cdot k_l(M+
165: p\cdot \Delta) = k_r(N- \Delta)\;.$$ Here the term $p\cdot \Delta$ represents the fact of
166: mortality of individuals at the course of migration; this is the way to account the
167: transfer cost, for this model. The migration flux, then, is determined according to
168: \begin{subequations}\label{eq:3}
169: \begin{equation}\label{eq:3-1}
170: \Delta_{MN} = \frac{pc - a +bN - pdM}{b+p^2d}
171: \end{equation}
172: \textrm{for the migration from $N$-station to $M$-station, and}
173: \begin{equation}\label{eq:3-2}
174: \Delta_{NM} = \frac{pa - c + dM - pbN}{d + p^2b}
175: \end{equation}
176: \end{subequations}
177: for the inverse migration.
178:
179: \noindent Finally, \textbf{the third step} consists in the reproduction of organisms in
180: both subpopulations independently, with respect to the abundance ($\widetilde{N}_t$, or
181: $\widetilde{M}_t$, respectively) resulted from the migration:
182: \begin{subequations}\label{eq:4}
183: \begin{equation}\label{eq:4-1}
184: N_{t+1} = a \cdot \widetilde{N}_t - c \cdot \widetilde{N}_t^2
185: \end{equation}
186: \textrm{and}
187: \begin{equation}\label{eq:4-2}
188: M_{t+1} = c \cdot \widetilde{M}_t - d \cdot \widetilde{M}_t^2\;.
189: \end{equation}
190: \end{subequations}
191:
192: It should be stressed, that a reproduction in a station (i.e., at the subpopulations)
193: runs independently, with respect to the abundances occurred due to the migration. If no
194: migration takes place for some time moment $t$, then $\widetilde{N}_t = N_t$
195: ($\widetilde{M}_t = M_t$, respectively).
196:
197: \section{Some properties of the model}
198: The model (\ref{eq:1}~--~\ref{eq:4}) of the smart migration exhibits various dynamic
199: properties, not observed at the similar population dynamics models with no migration. The
200: dynamics of two (independent) subpopulations runs inside the rectangular $[0, a/b] \times
201: [0, c/d]$, if no migration occurs. As soon as $p
202: > 0$, the dynamics runs at the dovetail shown in Fig.\ref{f1}.
203: \begin{figure*}
204: \includegraphics{xvost.eps}
205: \caption{\label{f1}Space available for the dynamic trajectories of the model, in case of
206: non-zero migration; $\mathsf{1}$ -- no migration area, $\mathsf{2, 5}$ -- areas of
207: migration from $N$-station to $M$-station, $\mathsf{3, 4}$ -- areas of migration from
208: $M$-station to $N$-station. Horizontal axis shows the abundance at $N$-station, vertical
209: one shows similar abundance in $M$-station.}
210: \end{figure*}
211: If a trajectory (i.e. a point representing a couple of abundances $(N_t, M_t)$) reaches
212: the area $\mathsf{1}$, then no migration occurs at that time moment $t$, and no migration
213: will take place, while the trajectory remains at this area. The area is cut off with two
214: solid bold lines. These are determined by the equations $$\Delta_{MN} = 0 \qquad
215: \textrm{and} \qquad \Delta_{NM} = 0\;.$$ Areas $\mathsf{2}$ and $\mathsf{3}$ are the
216: space dynamically available by each subpopulation, independently, when no migration takes
217: place.
218:
219: On the contrary, the migration expands the dynamically reachable area. Such expansion
220: results from the smart migration: individuals emigrate from the overpopulated station
221: improving the survival of the entire population. Areas $\mathsf{2}$ and $\mathsf{5}$ in
222: Fig.\ref{f1} show this expansion. For any point $(N_t, M_t) \in \Omega$, where $\Omega$
223: is a union of these two areas, the migration gives the projection of this point on the
224: bold solid line bordering the area $\mathsf{1}$, parallel to the solid dashed line shown
225: in this Figure. Thin arrow in this Figure illustrates such projection. Similarly, if a
226: point $(N_t, M_t) \in \Phi$ belongs to the union $\Phi$ of areas $\mathsf{3}$ and
227: $\mathsf{4}$, then the migration maps it on the (lower) solid bold line bordering area
228: $\mathsf{1}$. Similarly, it is executed in parallel to the thin arrow, shown in the
229: figure.
230:
231: The tangent of the dashed line bordering the area $\mathsf{5}$ is equal $-p^{-1}$;
232: similarly, for the area $\mathsf{4}$ the tangent is equal to $-p$. The area $\mathsf{1}$
233: expands, as $p \rightarrow 0$ occupying the entire rectangular $[0, a/b] \times [0,
234: c/d]$, for $p = 0$. The areas $\mathsf{4}$ and $\mathsf{5}$ become a (semi-infinite)
235: strip each, of the permanent width. This fact differs the situation of the complete
236: absence of a migration from the infinitely big migration cost $\mu$. This area collapses
237: into the line (into the intercept, to be exact) defined the the equation $$bN - dM + a -
238: c =0\;,$$ as $p \rightarrow 1$.
239:
240: Next, the migration expands the allowed parameter values. The parameters $a$ and $c$ must
241: meet the constraint $$1 \leq a \leq 4 \qquad \textrm{and} \qquad 1 \leq c \leq 4\;,$$ in
242: migration free models \cite{lande,sharkov,sharkov1,ecol2003,2,3,progress}. The equation
243: (\ref{eq:1-1}) (or (\ref{eq:1-2}), respectively) exhibits elimination of a population, as
244: $a$ ($c$, respectively) exceeds $4$: maximum of $aN_t - bN_t^2$ may be greater than
245: $a/b$. The non-zero migration expands the range of the parameter, provided that the
246: overpopulation in one station will be compensated by the emigration into another.
247:
248: Equation (\ref{eq:1-1}) (or (\ref{eq:1-2}), respectively) shows various dynamic patterns,
249: for various parameter $a$ (or parameter $c$, respectively) value. The diversity of limit
250: regimes of (\ref{eq:1}) varies from a stable fixed point to a strange attractor as a
251: limit manifold. Similar diversity of regimes could be found at the equation system
252: (\ref{eq:1-1}, \ref{eq:1-2}), when the migration occurs. In addition, the system
253: (\ref{eq:1}~--~\ref{eq:4}) exhibits some other regimes, that are not possible with no
254: migration.
255:
256: Migration provides a redistribution of individuals over a space. The redistribution is
257: not random; it results in maximization of the average (over two stations) reproduction
258: coefficient $k_r(\widetilde{N}) + k_l(\widetilde{M})$. The reproduction coefficient
259: reaches the maximum in one step; this point results both from the global information
260: accessibility of the data concerning the environmental conditions (coefficients $a$, $b$,
261: $c$ and $d$), population density (these are $N_t$ and $M_t$ in both stations, at every
262: time moment $t$), and the transfer cost $\mu$ (or the probability of the successful
263: transfer $p$). If $0 < p < 1$, the migration results a decrease of a general abundance,
264: since a part of individuals (namely, $p\cdot \Delta$) is eliminated, at each time step
265: $t$. Migration yields no elimination of individuals, as $p=1$.
266:
267: Various limit regimes occurred due to smart migration effect may be observed. There
268: exists the stable permanent one-side directed flux of individuals. For example, if $c$ is
269: big enough, and simultaneously $b$ is sufficiently small (thus increasing the
270: environmental capacity of $N$-station), one may observe the permanent one-way migration
271: flux limit regime. Indeed, such regime is determined by the equation
272: \begin{equation}\label{onemig}
273: \left\{
274: \begin{array}{c}
275: N^{\ast} = a\cdot \left(N^{\ast} + p\Delta^{\ast}\right) - b\cdot \left(N^{\ast} +
276: p\Delta^{\ast}\right)^2\\
277: M^{\ast} = c\cdot \left(M^{\ast} - \Delta^{\ast}\right) - d\cdot \left(M^{\ast} -
278: \Delta^{\ast}\right)^2\;,
279: \end{array}
280: \right.
281: \end{equation}
282: with $\Delta^{\ast}$ determined by (\ref{eq:3-2}): $$\Delta^{\ast} = \frac{pa - c +
283: dM^{\ast} - pbN^{\ast}}{d + p^2b}\,.$$ To figure out the impact of the smart migration on
284: the population dynamics, let's consider a particular case of $b = d = 1$, and $p =1$.
285: Such choice of the parameters means that the subpopulations differ in their growth rate,
286: only, and no losses of an abundance are resulted from a migration. The equality $p = 1$
287: also means a supreme mobility of an individual. Here $\Delta$ becomes equal to $$\Delta =
288: a-c+M-N\;,$$ and the system (\ref{onemig}) changes for
289: \begin{equation}\label{onemig1}
290: \left\{
291: \begin{array}{rcl}
292: N & = & a\cdot \left(M + \lambda\right) - \left(M + \lambda\right)^2\\
293: M & = & c\cdot \left(N - \lambda\right) - \left(N - \lambda\right)^2\;,
294: \end{array}
295: \right.
296: \end{equation}
297: with $\lambda = a - c$. Here the asterisks are omitted, since it makes no confusion. This
298: is the system of two polynomials of two variables $\left(M, N\right)$ of (formally) power
299: $2$ each. To solve a system of two polynomials of two variables, one must develop the
300: resultant of the system. This former is the determinant of the following matrix:
301: \begin{widetext}
302: \begin{equation}\label{detr1}
303: \left(
304: \begin{array}{cccc}
305: 0 & 1 & - \left[a(M+\lambda) - (M+\lambda)^2 \right] & 0\\
306: 0 & 0 & 1 & - \left[a(M+\lambda) - (M+\lambda)^2 \right]\\
307: 1 & -(2\lambda + c) & \lambda^2 + c\lambda + M & 0\\
308: 0 & 1 & -(2\lambda + c) & \lambda^2 + c\lambda + M\\
309: \end{array}
310: \right)
311: \end{equation}
312: \end{widetext}
313: for the case (\ref{onemig1}). Some roots of (\ref{detr1}) are the solution
314: of (\ref{onemig}). It should be stressed, that the first term at the first line of the
315: matrix (\ref{detr1}) becomes non-zero, for the case $b \neq d$.
316:
317: Suppose, $(N^{\ast},\ M^{\ast})$ is the solution of (\ref{onemig1}), and $N^{\ast} > 0$,
318: $M^{\ast}> 0$. Here the question arises, whether this solution is stable. The answer on
319: this question could be obtained due to linear approximation analysis. Let $N_t = N +
320: \nu_t$ and $M_t = M + \mu_t$, where $\nu_t, \mu_t$ are the small corrections;
321: substituting such $(N_t,\ M_t)$ into (\ref{onemig}) and omitting the terms of the second
322: and higher orders, one gets the following matrix for linear approximation:
323: \begin{equation}\label{ust1}
324: \left(
325: \begin{array}{cc}
326: 0 & -2\cdot \left(M+\lambda\right)\\
327: -2\cdot \left(N-\lambda\right) & 0\\
328: \end{array}
329: \right)\;.
330: \end{equation}
331: The eigenvalues of matrix (\ref{ust1}) are less, than $1$, when $\left|
332: \left(M+\lambda\right) \cdot \left(N-\lambda\right) \right| < 1/4$.
333:
334: Similarly, numerous other limit regimes of different complexity and structure could be
335: found, as well. One hardly can figure them out explicitly; moreover, there is no much
336: sense in detailed determination of these complex regimes. Computer simulation makes them
337: rather obvious.
338:
339: Here we present the simplest model of a smart migration, strongly opposing the ``reaction
340: -- diffusion'' methodology. The model is rather clear and apparent. The simplicity
341: results from the peculiar feature of the model; this is the case of globally informed
342: individuals. Indeed, an individual makes a decision whether it would migrate, or would
343: stay in the station, referring to the available information concerning the life
344: conditions. So, the key issue is what information towards that matter is available.
345:
346: Let's concentrate on some mathematical issues followed from the smart migration. The
347: first one is that the dynamics of a population with smart migration is irreversible in
348: time. Indeed, the migration at the model (\ref{eq:1}~--~\ref{eq:4}) is a projection, from
349: mathematical point of view; thus, a set of different states are transformed into unique
350: one, and there is no regular way to figure out which one was preceding the observed
351: abundance resulted from the migration act.
352:
353: Another important issue is that it expands both permissible phase space, and the
354: parameters values, in comparison to similar models with no migration. It should be said,
355: that such effects, probably, also could be observed (at least, for some peculiar
356: combinations of the parameters) for the systems with random, or aimless migration. We
357: doubt that the direct comparison of the areas of permissible phase variables, or the
358: parameters, for the case of smart vs. the aimless migration makes much sense. Obviously,
359: the model is rather simple and specific to pretend to describe properly any real
360: biological system. The specificity here manifests in the global information
361: accessibility; in occupation not more than two stations; in the absence of
362: ``microscopic'' consideration of dynamics within a station.
363:
364: The main purpose of this paper is to show the methodology of the modelling of spatially
365: distributed populations (and other biological communities) with no chemical analogies,
366: rather than to find out some peculiar dynamic regime pretending to match exactly a
367: dynamics of a real biological community.
368:
369: The model provides that the life conditions at the residence station are converted into a
370: couple of parameters $a$ and $b$ ($c$ and $d$, respectively). Besides, it is supposed
371: that the density of the subpopulation (or its total abundance $N$, or $M$, respectively)
372: is known, as well. Not discussing at the moment the details of the detection of
373: population density, or other conditional parameters {\sl per se}, suppose that the
374: environmental abundance (density, indeed) and other parameters are detectable for an
375: individual. Besides, the model (\ref{eq:1}~--~\ref{eq:4}) suggests that similar
376: parameters, and the abundance are known, at the distant station. Such suggestion makes
377: the individuals to be globally informed. A feasibility of such presumption is doubtful,
378: nevertheless, this assumption is a common place for mathematical population biology
379: \cite{levin97,ecol2003,lande,ber3,progress,edel2,alstad}.
380:
381: The methodology presented above is rather powerful, and provides a researcher with the
382: tool for studying spatially distributed populations with no artificial and absolutely
383: unrealistic hypotheses towards the microscopic behavior of individuals, i.e. the
384: randomness and aimlessness of their transfers over a space. One sees the following
385: furthering of the approach described here. First, a two-species (or several species)
386: communities could be described within the framework of the methodology. Again, one should
387: consider a two-station model, where each species (say, predators and preys) migrate from
388: station to station and back. A dynamics within a station might be modelled with the the
389: most common equation (say, with Lotka-Volterra equation), thus explicating the effect of
390: smart migration in the dynamics of a multi-species community.
391:
392: Both single-species, and multi-species models with smart migration may incorporate
393: various patterns of information accessibility, for individuals. The model presented above
394: is based on the hypothesis of the global accessibility of information to an individual.
395: It means, that an individual knows the conditions of life (expressed in the coefficients,
396: at the case of the model (\ref{eq:1} -- \ref{eq:4}), both at the station of residence,
397: and the station of immigration; it knows the density of each subpopulation. Finally, the
398: an individual knows the transfer cost, in this case.
399:
400: The hypothesis of total lack of information available to individuals opposes the idea
401: mentioned above. Here an individual operates with the inner, extremely local information,
402: when makes a decision on the change of the location for another one. In brief, such
403: situation could be described like a threshold migration, where the transfer act takes
404: place only when the local conditions become worse than some individually defined level.
405: Still, the situation of the total lack of information is not equivalent to random and
406: aimless migration. The difference becomes clear, if one considers the situation of the
407: transfer act occurrence, while the local density at the occupancy place is still very
408: low. The smart migration under the total lock of information would start up, while
409: chemically-like diffusive migration will not take place, in such case.
410:
411: All these assumptions seem to be too strong and specific. In general, the individuals
412: operate with a part of information. There are several problems here, both of mathematical
413: origin, and of biological essence. The first one consists in exact and comprehensive
414: definition of what exactly is known to individuals. Next one is the discretion between
415: the behavioral patterns supported by the reasonable choice of the way to behave, and
416: those determined instinctively. Consider the seasonal bird migration. Surely, the fact
417: that some species change a reproduction site for a winter spending site, falls beyond the
418: will of a bird, it does not make a matter of reasonable of self-made choice: that is the
419: instinct forcing them to fly away. On the other hand, the choice of a peculiar site to
420: spend a night (if any) is made by the birds in a flock reasonably, with respect to the
421: detail features and circumstances of the current situation. Finally, the problem arises
422: when one tends to determine how far (in space) the individuals are able to collect and
423: process the information (concerning the living conditions ``there'').
424:
425: The model (\ref{eq:1}~--~\ref{eq:4}) implies that the transfer cost $\mu$ is symmetrical,
426: and does not depend on the direction of migration. It might be so, while more realistic
427: idea is that the transfer cost should be unsymmetrical. Evidently, the simplest way to
428: figure out the transfer cost is to split it into three parts:
429: \begin{equation}\label{eq:9}
430: \mu = \mu_{\textrm{out}}(\mathsf{A}) + \mu_{\textrm{in}}(\mathsf{B}) +
431: \delta(\mathsf{A},\mathsf{B})\;.
432: \end{equation}
433: Here $\mu_{\textrm{out}}(\mathsf{A})$ is the transfer cost of successful emigration from
434: the station~$\mathsf{A}$; $\mu_{\textrm{in}}(\mathsf{B})$ is the transfer cost of
435: successful introduction into the station~$\mathsf{B}$, respectively; finally,
436: $\delta(\mathsf{A},\mathsf{B})$ is the pure transfer cost from station~$\mathsf{A}$ to
437: station~$\mathsf{B}$. Obviously, one should expect to face the asymmetry
438: $$\mu_{\textrm{out}}(\mathsf{A}) \neq \mu_{\textrm{out}}(\mathsf{B}) \qquad \textrm{and}
439: \qquad \mu_{\textrm{in}}(\mathsf{A}) \neq \mu_{\textrm{in}}(\mathsf{B})\;,$$ in general.
440: Also, a symmetry of pure transfer cost $\delta(\mathsf{A},\mathsf{B}) =
441: \delta(\mathsf{B},\mathsf{A})$ is doubtful. It should be said, that this point makes no
442: problem in its implementation at the model (\ref{eq:1}~--~\ref{eq:4}).
443:
444: Another significant constraint of the model (\ref{eq:1}~--~\ref{eq:4}) is the spatial
445: structure limited with two stations. Indeed, an expansion of the approach presented above
446: for the case of several stations, and, ultimately, for a continuous, or quasi-continuous
447: case of a habitat is strongly desirable. Suppose, a population inhabits three stations;
448: here we presume the global information accessibility, as well. Suppose, further, the
449: conditions (i.e., the abundances and the parameters) make the situation when the
450: individuals from the station $\mathsf{A}$ must migrate either to station $\mathsf{B}$, or
451: to station $\mathsf{C}$. No one knows exactly, in advance, what is a proportion of
452: individuals immigrating into the station $\mathsf{B}$ vs. those immigrating the station
453: $\mathsf{C}$. This is the main obstacle here. There exists the approach withdrawing this
454: discrepancy; it is based on the interval mathematics \cite{int,int1,int2,int3,int4}. The
455: detailed discussion of that issue falls beyond the scope of this paper.
456:
457: \section{Conclusion}
458: The model described above implements the methodology of evolution optimality into the
459: problem of the modelling of spatially distributed populations. Migration causes the
460: growth on net reproduction (which is a reproduction rate, in our case), in average, over
461: the space. The model comprises the simplest case of two stations, where the spatial
462: distribution is restricted to a transfer of individuals from station to station and back;
463: the transfer cost is supposed to be symmetrical one. The model shows the expansion of the
464: environment capacity, in comparison to the case of the migration absence.
465:
466: \begin{acknowledgments}
467: I am thankful to Prof. Alexander Gorban from Liechester University for the general
468: presentation of the problem, and for permanent cooperation. The work was partially
469: supported be Krasnoyarsk science foundation, grant 6F050C.
470: \end{acknowledgments}
471:
472: \begin{thebibliography}{99}
473:
474: \bibitem{alstad} Alstad~D. (2001) Basic Populus Models of Ecology, Prentice
475: Hall.
476:
477: \bibitem{bern} Bernstein~R. Population Ecology: An Introduction to
478: Computer Simulations. Wiley \& Sons.
479:
480: \bibitem{ber} Berryman~A.A. (2002) Population cycles: the case for
481: trophic interactions. New York: Oxford University Press, 321 p.
482:
483: \bibitem{ber3} Berryman~A.A. (2003) On principles, laws and theory in
484: population ecology // \textit{Oikos}, \textbf{103}, pp.695 -- 701.
485:
486: \bibitem{cyc} Castro-e-Silva~A., Bernardes~A.T. (2001) Analysis of chaotic
487: behaviour in the population dynamics // \textit{Physica}~\textbf{A 301}, pp.63 -- 70.
488:
489: \bibitem{sci2000} Condit~R., Ashton~P.S., Baker~P.,
490: Bunyavejchewin~S., Gunatilleke~S., Gunatilleke~N., Hubbell~S.P., Foster~R.B., Itoh~A.,
491: LaFrankie~J.V., Hua Seng Lee, Losos~E., Manokaran~N., Sukumar~R., Yamakura~T. (2000)
492: Spatial patterns in the distribution of tropical tree species // \textit{Science}
493: \textbf{288}, pp.1414 -- 1418.
494:
495: \bibitem{edel2} Edelstein~L., Edelstein~K. (1988)
496: Mathematical Models in Biology. Birkheauser Mathematics Series.
497:
498: \bibitem{ezer} Ezersky~A.B., Rabinovich~M.I. (1990) Nonlinear-wave
499: competition and anisotropic spectra of spatiotemporal chaos of Faraday ripples //
500: \textit{Europhysics Letters}, \textbf{13}(3), pp.243 -- 249.
501:
502: \bibitem{2} Gilpin~M.E., Ayala~F.J. (1973) Global models of growth
503: and competition // \textit{Proc.Natl. Acad.Sci.}, \textbf{70}, pp.3590 -- 3593.
504:
505: \bibitem{gi-co04} Ginzburg~L.R., Colyvan~M. (2004) Ecological
506: Orbits: how planets move and populations grow. New York: Oxford University Press, 354 p.
507:
508: \bibitem{3} Gromov~M. (2000) A dynamical model for synchronisation and for
509: inheritance in microevolution: a survey of papers of A.Gorban, The talk given in the IHES
510: seminar, ``Initiation to functional genomics: biological, mathematical and algorithmical
511: aspects'', Institut Henri Poincar\'{e}, November 16, 2000.
512:
513: \bibitem{otbor3} Gorban~A.N. (1984) Equilibrium encircling. Equations of
514: chemical kinetics and their thermodynamic analysis, Novosibirsk: Nauka plc., 1984.
515:
516: \bibitem{otbor1} Gorban~A.N. (1992) Dynamical systems with inheritance, In:
517: Some problems of community dynamics, R.G. Khlebopros (ed.); Novosibirsk: Nauka plc.,
518: pp.40 -- 72.
519:
520: \bibitem{otbor4} Gorban~A.N. (2005) Systems with inheritance: dynamics of
521: distributions with conservation of support, natural selection and finite-dimensional
522: asymptotics // arXiv:cond-mat/0405451
523:
524: \bibitem{na2} Gorban~A.N., Karlin~I.V. (2003) Family of additive
525: entropy functions out of thermodynamic limit // Physical Review~\textbf{E 67}, 016104.
526:
527: \bibitem{na4} Gorban~A.N., Khlebopros~R.G. (1988) Demon of
528: Darwin: Idea of optimality and natural selection, Moscow: Nauka (FizMatGiz).
529:
530: \bibitem{bolgar} Gorban~A.N., Sadovsky~M.G. (1987) Population
531: mechanisms of cell aggregation in continuous cultivation systems // \textit{Biotechnology
532: and Biotechnique}, \textbf{2}(5), pp.34 -- 36.
533:
534: \bibitem{zob89} Gorban~A.N., Sadovsky~M.G. (1989) Optimization
535: models of spatially distributed populations: Alle's effect // \textit{Rus.J.General
536: Biol.}, \textbf{50}(1), pp.66 -- 72.
537:
538: \bibitem{monit89} Gorban~A.N., Sadovsky~M.G. (1989)
539: Optimization models: the case of globally informed individuals. In: Problems of
540: environmetnal monitoring and modelling of ecosystems, vol.\textbf{11}, Leningrad:
541: Gidrometeoizdat, pp.198 -- 203.
542:
543: \bibitem{haldane} Haldane~J.B.S. (1990) The Causes of Evolution, Princeton
544: Science Library, Princeton University Press.
545:
546: \bibitem{int} Jaulin~L., Kieffer~M., Didrit~O., Walter~E. (2001)
547: Applied Interval Analysis, with Examples in Parameter and State Estimation, Robust
548: Control and Robotics. Heidelberg, Hamburg: Springer-Verlag.
549:
550: \bibitem{kolmog} Kolmogorov~A.N., Petrowsky~I.G., Piscounov~N.A.
551: (1937) \'{E}tude de l'\'{e}quation de la diffusion avec croissance de la quantit\'{e} de
552: mati\`{e}re et son application \`{a} un probl\`{e}me biologique //
553: \textit{Mosc.Univ.Bull.Math.} \textbf{1}, pp.1 -- 25.
554:
555: \bibitem{int2} Kr\"{a}mer~W., von~Gudenberg~J.W. (2001)
556: Scientific Computing, Validated Numerics, Interval Methods, Kluwer,
557: Boston/Dordrecht/London.
558:
559: \bibitem{int3} Kulisch~U.W. (2002) Advanced Arithmetic for the Digital
560: Computer, Springer-Verlag, Wien.
561:
562: \bibitem{lande} Lande~R., Engen~S., Saether~B.-E., Sther~B.-E.
563: (2003) Stochastic Population Dynamics in Ecology and Conservation (Oxford Series in
564: Ecology and Evolution), Oxford Univ.Press.
565:
566: \bibitem{ecol2003} Law~R., Murrell~D.J., Dieckmann~U. (2003) Population
567: growth in space and time: spatial logistic equations // \textit{Ecology}, \textbf{84}(1),
568: pp.252 -- 262.
569:
570: \bibitem{levin97} Levin~S.A., Grenfell~B., Hastings~A.,
571: Perelson~A.S. (1997) Mathematical and Computational Challenges in Population Biology and
572: Ecosystems Science // \textit{Science}, \textbf{275}, pp.334 -- 43.
573:
574: \bibitem{levitt} Levitt~P.R. (1975) General kin
575: selection models for genetic evolution of sib altruism in diploid and haplodiploid
576: species // \textit{Proc.Nat.Acad.Sci. USA}, \textbf{72}(11), pp.4531 -- 4535.
577:
578: \bibitem{lotka} Lotka~A.J. (1925) Elements of Physical Biology.
579: Baltimore: Williams \& Wilkens.
580:
581: \bibitem{lvov} L'vov~V.S. (1994) Wave turbulence under parametric excitation
582: applications to magnets, Springer, Berlin, Heidelberg.
583:
584: \bibitem{insect} Maron~J.L., Harrison~S. (1997) Spatial Pattern
585: Formation in an Insect Host-Parasitoid System // \textit{Science}, \textbf{278}, pp.1619
586: -- 1621.
587:
588: \bibitem{progress} Matsuda~H.N., Ogita~A., Sasaki~A., Sat\~{o}~K. 1992.
589: Statistical mechanics of population: the lattice Lotka-Volterra model // \textit{Progress
590: in Theoretical Physics}, \textbf{88}, pp.1035 -- 1049.
591:
592: \bibitem{murray} Murray~J.D. (2002) Mathematical Biology, vols.
593: \textbf{I--II}, Third edition. Berlin: Springer-Verlag.
594:
595: \bibitem{int4} Neumaier~A. (2001) Introduction to Numerical Analysis.
596: Cambridge: Cambridge Univ. Press.
597:
598: \bibitem{int1} Petkovic~M.S., Petkovic~L.D. (1998) Complex
599: Interval Arithmetic and Its Applications, John Wiley.
600:
601: \bibitem{rozonoer} Rozonoer~L.I., Sedyh~E.I. On the mechanisms of
602: evolution of self-reproduction systems, $\mathsf{1}$, \textit{Automation and Remote
603: Control}, \textbf{40}(2), pp.243 -– 251; $\mathsf{2}$, ibid., \textbf{40}(3), pp.419 --
604: 429; $\mathsf{3}$, ibid, \textbf{40}(5), pp.741 -- 749.
605:
606: \bibitem{zelkniga} Sadovsky~M.G., Gurevich~Yu.L., Manukovsky~N.S.
607: (1989) Kinetics of cell aggregation in continuous cultivation. In: Dynamics of chemical
608: and biological systems, Novosibirsk: Nauka plc, pp.134 -- 158.
609:
610: \bibitem{moya} Sadovsky~M.G. (1992) Optimization modelling
611: of globally informed individuals. In: Mathematical modelling in biology and chemistry.
612: Evolution approach, Novosibirsk: Nauka plc, pp.36 -- 67.
613:
614: \bibitem{otbor2} Semevsky~F.N., Semenov~S.M. (1984.)
615: Mathematical modelling of ecological processes. Leningrad: Gidrometeoizdat, 426 p.
616:
617: \bibitem{sharkov1} Sharkovsky~A.N. (1965) On cycles and the structure of
618: continuous mapping // \textit{Ukranian mathematical journal}, \textbf{17}(3), pp.104 --
619: 111.
620:
621: \bibitem{sharkov} Sharkovsky~A.N. (1983) Difference equations and
622: population dynamics / In: Mathematical methds in biology. Proc.$2^{\textrm{nd}}$ Ukranian
623: Conf., Kiev: Naukova Dumka plc., pp.143 -- 156.
624:
625: \bibitem{xehi} Smitalova~K., Sujan~S. (1991) A mathematical
626: treatment of dynamical models in biological science. Ellis Horwood.
627:
628: \bibitem{annals} Soares~P., Tom\'{e}~M. (1999) Distance-dependent
629: competition measures for eucalyptus plantations in Portugal // \textit{Annals of Forestry
630: Science}, \textbf{56}, pp.307 -- 319.
631:
632: \bibitem{ameba} Strassmann~J.E., Yong~Zhu, Queller~D.C. (2000)
633: Altruism and social cheating in the social amoeba {\sl Dictyostelium discoideum} //
634: \textit{Nature}, \textbf{408}, pp.965 -- 967.
635:
636: \bibitem{1} The geometry of ecological interactions:
637: simplifying spatial complexity. Dieckmann~U., Law~R., Metz~J.A.J., editors. Cambridge
638: University Press, Cambridge, UK.
639:
640: \bibitem{thieme} Thieme~H.R. (2003) Mathematics in Population Biology
641: (Princeton Series in Theoretical and Computational Biology). Princeton Univ.Press.
642:
643: \bibitem{turch} Turchin~P. (2003) Complex Population Dynamics: A
644: Theoretical/Empirical Synthesis. Princeton University Press, 357 p.
645:
646: \bibitem{verh} Verhulst~P.(1845) Recherches math\'{e}matiques sur la loi
647: d'accroissement de la population // \textit{Nouv.M\'{e}m.de l'Academie Royale des Sci.et
648: Belles-Lettres de Bruxelles} \textbf{18}, pp.1 -- 41.
649:
650: \bibitem{volt} Volterra~V. (1926) Variazioni e fluttuazioni del numero
651: d'individui in specie animali conviventi. \textit{Mem.R.Accad.Naz.dei Lincei}, Ser.VI, 2.
652:
653: \bibitem{volt1} Volterra~V. (1931) Le\c{c}ons sur la th\'{e}orie
654: math\`{e}matique de la lutte pour la vie, Gauthier-Villars, Paris.
655:
656: \bibitem{zah-usp} Zakharov~V.E., L'vov~V.S., Starobinets~S.S. (1974)
657: Turbulence of spin-waves beyond threshold of their parametric-excitation //
658: \textit{Uspekhi Fizicheskikh Nauk} \textbf{114}(4), pp.609 -- 654; English translation
659: \textit{Sov.Phys.-Usp.} \textbf{17}(6) (1975), pp.896 -- 919.
660:
661: \bibitem{zah-spek} Zakharov~V.E., L'vov~V.S., Falkovich~G.E.
662: (1992) Kolmogorov spectra of turbulence, vol.\textbf{1} Wave Turbulence. Springer,
663: Berlin.
664:
665: \end{thebibliography}
666:
667: \end{document}
668:
669: