1: %\documentclass[doublespacing, final, narrowdisplay, reviewcopy,seceqn]{elsart}
2: \documentclass[seceqn]{elsart5p}
3:
4: \bibliographystyle{elsart-harv}
5:
6: \usepackage{natbib}
7: \usepackage{graphicx}
8: \usepackage{amssymb}
9: \usepackage{lineno}
10:
11: \newcommand{\pderiv}[2]{\frac{\partial #1}{\partial #2}}
12:
13: %\linenumbers
14:
15: \begin{document}
16:
17: \begin{frontmatter}
18: \title{Allee Effects and Extinction in a Lattice Model}
19: \author{Alastair Windus and Henrik Jeldtoft Jensen\corauthref{cor1}}
20: \corauth[cor1]{Corresponding author.
21: Tel: +44 20 7594 8541, fax: +44 20 6594 8516, email address: h.jensen@imperial.ac.uk,
22: h.jensen@ic.ac.uk (HJ. Jensen).}
23: \address{Mathematics Department, Imperial College London, South Kensington Campus, London. SW7 2AZ.}
24:
25: \begin{abstract}
26: In the interest of conservation, the importance of having a large habitat available for a species is widely known. Here, we introduce a lattice-based model for
27: a population and look at the importance of fluctuations as well as that of the population density, particularly
28: with respect to Allee effects. We examine
29: the model analytically and by Monte Carlo simulations and find
30: that, while the size of the habitat is important, there exists a critical population density below which extinction is
31: assured. This has large consequences with respect to conservation, especially
32: in the design of habitats and for populations whose density has
33: become small. In particular, we find that the probability of survival for
34: small populations can be increased by a reduction in the size of the habitat and show that there exists an optimal size reduction.
35: \end{abstract}
36:
37: \begin{keyword}
38: Extinction; Allee effects; Critical population density; Habitat size; Fluctuations; Mean
39: field; Monte Carlo simulations
40:
41: % PACS codes here, in the form: \PACS code \sep code
42:
43: \end{keyword}
44:
45: \end{frontmatter}
46:
47: \section{Introduction} \label{Section: Introduction}
48: Extinction is becoming a greater and greater issue all over
49: the world and is a cause of extreme concern. It has been estimated
50: that anthropogenic extinctions are resulting in the loss of a few percent of the current world's biosphere, which is of magnitude 3 to 4 times the natural background rate \citep{May}.
51: The World Conservation Union
52: (IUCN), through its Species Survival Commission (SCC) develops criteria to assess the extinction rate for plants and animals
53: all over the world which enables them to keep a so-called \textit{Red List} (www.redlist.org) of species which are threatened with extinction in order to promote their
54: conservation. The list currently shows over 16,000 threatened species around the world - a 45\% increase on the figure from the year 2000.
55:
56: It has been shown by examining both discrete \citep{Escudero} and continuous
57: \citep{SKellam} populations that, at least analytically speaking, there exists a critical
58: habitat size $L_c$ above which survival of a population is assured.
59: Here we examine what role the population density plays since, intuitively,
60: one would expect that even for $L>L_c$, a sufficiently large
61: population would be needed for growth.
62:
63: Lattice based models are widely used in ecology \citep[see for example][]{Tainaka,
64: Durrett, Itoh} and so we introduce such a model that incorporates birth,
65: death and diffusion. Unlike other similar models such as the \textit{contact process} \citep[e.g.][]{Harris, Oborny}, here
66: two individuals must meet in order to reproduce whereas one individual can
67: die by itself. This results in negative growth rates for populations that
68: fall below a critical density due to reproduction opportunities becoming
69: rare. In real populations, the positive correlation between size
70: and per capita growth rate of a population is known as the Allee effect \citep{Allee},
71: which has recently received much interest \cite[e.g.][]{Dennis2, Hurford,
72: Johnson}.
73: If the Allee effect is strong enough, the population size may even decrease
74: for small population sizes as in our model. Due to this behaviour, the effect has been examined with respect to extinction
75: \citep[see for example][and references therein]{Amarasekare, Courchamp, Stephens} but {\em primarily} deterministically and so without fluctuations in the population density.
76: Since fluctuations are likely to be highly significant for small populations,
77: we include the effects of these by examining Monte Carlo (MC) simulations
78: in the hope to gain a more realistic picture of the importance of the population density on the chances of survival. Further to the stochastic methods used to study Allee effects, such as stochastic differential equations \citep[e.g.][]{Dennis},
79: discrete-time Markov-chains \cite[e.g.][]{Allen} or diffusion processes \citep[e.g.][]{Dennis2}, lattice based models have space, as well as time, as a variable and take into account individuals, rather than just the macroscopic view of the population.
80: \\ \\
81: After introducing the model in the next section, we examine the Allee effects
82: present in our model in Section \ref{Section: Allee Effects}, particularly
83: with respect to a sudden decrease in population in Section \ref{Section: Decrease in PopDens}. The effects of the fluctuations are examined in Section
84: \ref{Section: Fluctuations}.
85:
86: \section{The Model} \label{Section: The Model}
87: We have a $d$-dimensional square lattice of linear length $L$ where each site is either
88: occupied by a single particle (1) or is empty (0). A site is chosen at random. If the site is occupied, the particle is removed with probability $p_d$,
89: leaving the site empty. If the particle does not die, a nearest neighbour site is randomly chosen. If the neighbouring site
90: is empty, the particle moves to that site. If however the
91: neighbouring site is occupied, with probability
92: $p_b$\footnote{We
93: note that the birth rate is actually given by $p_b(1-p_d)$ and not $p_b$ only.}, the particle reproduces, producing a new particle on another randomly selected neighbouring
94: site, conditional on that chosen square being empty. We therefore have the
95: following reactions for a particle $A$:
96: \begin{equation}
97: A\phi \longleftrightarrow\phi A,\quad A+A \longrightarrow 3A \quad\mbox{and}\quad A\longrightarrow\phi,
98: \end{equation}
99: where $\phi$ represents an empty site. A time step is defined
100: as the number of lattice sites and so is equal to approximately one update
101: per site. We use nearest neighbours and, throughout most of the paper,
102: periodic boundary conditions
103: which, although more unrealistic than, say, reflective boundary conditions, allow for better comparison with analytical results, since periodic systems
104: remain homogenous. We later, however, examine some results with reflective boundary conditions.
105:
106: Due to the conflict between the growth and decay processes in the model, we expect that with certain values of $p_b$ and $p_d$, extinction of the
107: population would
108: occur. Indeed, many models displaying such a conflict \citep[see for example ][]{Vespignani, Dammer, Oborny,
109: Peters} show a critical parameter value separating
110: an \textit{active} state and an \textit{inactive} or \textit{absorbing}
111: state that, once reached, the system cannot leave. As the rate of decay
112: increases, the so-called \textit{order parameter} (often the density of active sites)
113: decreases, becoming zero at a critical point, marking a change in phase
114: or \textit{phase transition}. In our case, the absorbing
115: state would represent an empty lattice and so extinction of the population.
116:
117: To show that this is indeed the case for our model, we derive a so-called
118: mean field equation \citep[e.g.][]{Opper} for the density of occupied sites $\rho(t)$. Assuming the particles are spaced homogeneously in an infinite system we have
119: \begin{equation} \label{mean field}
120: \pderiv{\rho(t)}{t} = p_b(1-p_d)\rho(t)^2(1-\rho(t))-p_d\rho(t). \end{equation}
121: The first term is the proliferation term and so is proportional to $\rho^2$,
122: the probability that the particle does not die $(1-p_d)$, the probability that the next randomly chosen site to give birth on is empty $(1-\rho)$ and
123: finally the probability that it gives birth if this is the case, $p_b$. The second term represents
124: particle annihilation and so is proportional to both $\rho$ and $p_d$, the probability
125: that the chosen particle dies.
126:
127: Eq. (\ref{mean field}) has three steady states,
128: \begin{equation} \label{steady states}
129: \bar\rho_0 = 0, \quad \bar\rho_\pm=\frac{1}{2}\left(1\pm\sqrt{1-\frac{4p_d}{p_b(1-p_d)}}\right).
130: \end{equation}
131: For $4p_d > p_b(1-p_d)$, $\bar\rho_\pm$ are imaginary, resulting in $\bar\rho_0$
132: being the only real stationary state and so, here, extinction occurs in all
133: circumstances. Keeping $p_b$ a constant from now on, we then have that our critical death rate is given by $p_{d_c} = p_b/(4+p_b)$ which separates the active
134: phase representing
135: survival and the absorbing state of extinction.
136:
137: Clearly Eq. (\ref{mean field}) is limited by the exclusion of diffusion and
138: noise as well as the false assumption of a homogenous population density. We do
139: however find at least good qualitative support for our mean field analysis
140: through numerical simulations. Fig. \ref{Critical Parameters} a)
141: shows the critical values of $p_d$ and $p_b$ separating the regions with
142: one and three stationary states according to both the mean field equation
143: and numerical simulations for $1$, $2$ and $3$ dimensions.
144: \begin{figure}[tb]
145: \centering\noindent
146: \begin{tabular}{c}
147: \tiny{a)} \\
148: \includegraphics[width=8cm]{Figure_1a.eps}
149: \\ \tiny{b)} \\
150: \includegraphics[width=8cm]{Figure_1b.eps}
151: \end{tabular}
152: \caption{a) Parameter domain for the number of steady states with the points showing the parameters giving 2 steady states and b) the steady state population densities with $p_b = 0.5$ for the mean field (line),
153: 1 ($+$), 2 ($\times$) and 3 ($\bullet$) dimensional simulations.}
154: \label{Critical Parameters}
155: \end{figure}
156: We see convincing agreement between our analytical and numerical results,
157: particularly for higher dimensions. The MC simulations were carried out on an initially fully occupied lattice with linear sizes $L=1000$,
158: $32$ and $10$ for each dimension respectively and we observed whether extinction
159: occurred during $10^5$ time steps. For each birth rate, the simulation
160: was repeated 500 times. If a single run survived, $p_d$ was increased, whereas
161: if extinction occurred in all runs, $p_d$ was reduced. Using the same initial
162: seed for the random number generator, an iterative procedure produced a
163: critical value with accuracy $\pm 2^{-11}$. This iterative procedure
164: was then repeated 5 times with different seeds and the average taken. Only a small number of repeats
165: were needed since the largest variance of the values obtained was of the
166: order of $10^{-8}$. From the figure we find that to 3 d.p. for $p_b=0.5$,
167: $p_{d_c} = 0.073 $ , 0.098 and 0.105 in 1,2 and 3 dimensions respectively. Due
168: to the finite size of the lattices and the finite time used for the above simulations, the actual critical death rates are likely to differ slightly
169: from
170: those given and more accurate techniques would have to be used to obtain
171: them \citep[see][for examples of such techniques]{Hinrichsen_Non}.
172:
173: With $\rho(t = 0) = 1$, as $p_d$ is increased, the steady-state population density decreases, becoming zero at $p_{d_c}$ as shown in Fig. \ref{Critical Parameters}
174: b), marking the phase transition. We see that the steady
175: state population density {\em appears} to change continuously in 1 dimension, whilst
176: discontinuously in 2 and 3 dimensions in agreement with the mean field results.
177: If indeed this is the case, we call such phase transitions \textit{continuous} and \textit{first-order} respectively. In both cases, the phase transition is marked by a very rapid
178: decrease in population density.
179: \\ \\
180: Briefly relating our model to biology, we note that the death rate for a given species may fluctuate for any number of
181: reasons but it must certainly
182: be true that at least the average value of $p_d$ must be less than $p_{d_c}$ for the species to have ever been in existence. However, as we have shown, even a temporary increase in $p_d$ above $p_{d_c}$, due to deforestation
183: or disease for example, will cause a very rapid and perhaps unrecoverable decrease in population. Extinctions however may also occur for reasons other than having
184: a super-critical death rate. We investigate the roles
185: of Allee effects and that of fluctuations in the next three sections where we examine simulations in the sub-critical or active phase and use the constant value \(p_{b}=0.5\).
186:
187: \section{Allee Effects} \label{Section: Allee Effects}
188: One reason we observe a decline in population growth at low densities is
189: due to individuals finding it harder to find a mate. This is empirically known to occur in both plant \citep[e.g.][]{Aizen} and animal \citep[e.g.][]{Lande2} populations. In our model, this aspect is incorporated by the fact that
190: two individuals are required for reproduction whereas an individual can die
191: by itself. As density decreases, each
192: individual therefore finds it increasingly difficult
193: to find another for reproduction before they die.
194: To examine this, we return to our mean field equation (\ref{mean
195: field}).
196:
197: It is easy to show that whereas $\bar\rho_+$ and $\bar\rho_0$ are stable stationary points of Eq. (\ref{mean field}), $\bar\rho_-$ is unstable. Since in the active phase, $\bar\rho_0 < \bar\rho_- < \bar\rho_+$, any population whose density $\rho(t)< \bar\rho_-$ will be driven to
198: extinction by the dynamics of the system. In fact
199: we find that for $p_d < p_{d_c}$,
200: \begin{equation} \label{Long term behaviour}
201: \rho(t) \longrightarrow \quad \left\{
202: \begin{array}{cl}
203: 0 & \mbox{ for } \rho(t) < \bar\rho_- \\
204: \bar\rho_+ & \mbox{ for } \rho(t) > \bar\rho_-
205: \end{array} \right.
206: \quad\mbox{ as } t \longrightarrow \infty.
207: \end{equation}
208: We test this numerically in 1, 2 and 3 spatial dimensions by finding the
209: value of $p_d$ that separates the active and absorbing states for different initial conditions. The MC simulations were carried out and the
210: critical death rate found iteratively in the same fashion as in Section \ref{Section:
211: The Model}. The results
212: are shown in Fig. \ref{phase diagram}
213: \begin{figure}[tb]
214: \centering\noindent
215: \includegraphics[width=8cm]{Figure_2.eps}
216: \caption{Phase diagram showing the critical values of $p_d$
217: separating the 2 long-term outcomes of the system for different initial population density according to the mean field (line) and
218: the 1 (+), 2 ($\times$) and 3 ($\bullet$) dimensional MC simulations.}
219: \label{phase diagram}
220: \end{figure}
221: and clearly show the importance of the initial population density for survival.
222: The density dependence appears to increase with
223: dimensionality, which we expect, since two individuals meeting becomes progressively
224: harder as the dimensionality of the system increases.
225:
226: The existence of this critical population density is highly significant to the conservation of species. It is clear that a sufficiently small population
227: will not grow, regardless of how much space and resources are available.
228: It also has repercussions if a population density were to suddenly decrease
229: due to disease or particularly harsh meteorological conditions, for example. We examine
230: this further in Section \ref{Section: Decrease in PopDens} after examining
231: the role of fluctuations.
232:
233: \section{Fluctuations} \label{Section: Fluctuations}
234: We expect extinction due to fluctuations in the population density to occur
235: when the order of the fluctuations approaches the mean population density.
236: Empirically, demographic stochasticity (that is, chance events of mortality and reproduction) is known to be greater in smaller populations \citep{Lande}
237: than in larger ones. Population and habitat size are, on average, positively correlated and so, particularly with the existence of the critical population density,
238: we expect extinction due to fluctuations to occur for smaller lattice sizes as has been suggested
239: by others \citep[e.g.][]{Pimm, Escudero_Buceta}.
240:
241: We see in Fig. \ref{Fluctuations}
242: \begin{figure}[tb]
243: \centering\noindent
244: \includegraphics[width=8cm]{Figure_3.eps}
245: \caption{Log-log plot of the standard deviation of the population
246: density versus the number of sites in the 1 (+), 2 ($\times$)
247: and 3 ($\bullet$) dimensional systems. The hashed line has gradient
248: -0.5 for the eye and indicates the power law behaviour. Insert: The fluctuations v.s. $p_d$ dimensional case with the same symbol
249: notation.}
250: \label{Fluctuations}
251: \end{figure}
252: that, numerically, the fluctuations in the population density
253: $\Delta\rho$ decrease with the number of lattice sites $N$ through a power law with exponent -0.50 in all dimensions, which is what we would expect
254: from the \textit{central limit theorem}. Simulations were carried out for fixed $p_d=0.03$ and $p_b=0.5$ and the standard deviation obtained from $5\times 10^3$ surviving runs for each lattice size. The insert in Fig. \ref{Fluctuations} shows how the size
255: of the fluctuations also increase as the critical point is approached. These larger fluctuations will also increase
256: the probability of extinction as indicated in Fig. \ref{IncreasingL}
257: \begin{figure}[tb]
258: \centering\noindent
259: \includegraphics[width=8cm]{Figure_4.eps}
260: \caption{How $P_s$ varies with $L$ for the 1 dimensional model with
261: (from left to right), $p_d$ = 0.04, 0.05 and 0.06. Similar results
262: are seen in 2 and 3 dimensions.}
263: \label{IncreasingL}
264: \end{figure}
265: where we examine the probability of survival $P_s$, that is, the probability
266: that extinction has not occurred up to some time $t_m$. We examine the 1
267: dimensional case only using three different values of $p_d$ with $t_m = 10^3$
268: and repeat the simulation $5\times10^4$ times for each lattice size. The figures
269: clearly show how the probability of survival increases with $L$, yet decreases
270: as $p_d$ increases. Indeed, as $p_{d_c}$ is approached,
271: population density decreases and fluctuation size increases resulting in
272: species with higher death rates being more susceptible to extinction.
273: This is indeed observed in nature where long-lived species are known,
274: in general, to have a higher chance of survival than short-lived ones \citep{Pimm}.
275:
276: \section{A Decrease in Population Density} \label{Section: Decrease in PopDens}
277: Apart from the initial conditions, it is certainly conceivable that the population
278: density could fall below the critical value due to a reduction in population
279: size. From Eq. (\ref{Long
280: term behaviour}) we expect that the population will survive only as long as $\rho(t)>\rho_c$. We simulate this by increasing $p_d$ to
281: 1 at some $t=t_k$ and then returning $p_d$ to what it was before, once a density $\rho_s$
282: has been reached. We examine this here in 2 dimensions with now reflective
283: rather than the previously used periodic boundary conditions. Qualitatively,
284: all previous results have been very similar when using reflective boundary
285: conditions but here we want to increase this degree of realism in our model.
286:
287: For 2 dimensional simulations with an initial population density $\rho(0) = 0.128$, the
288: critical death rate is 0.093 (3 d.p.) as shown in Fig. \ref{phase diagram}. So, for $p_d = 0.093$, we would expect that if $\rho_s
289: > 0.128$, the population will survive, with the population density returning
290: to what it was before, whereas for $\rho_s
291: < 0.128$, extinction will occur. Due to the fluctuations in the population that occur in the simulations, we expect more of an increase in the likelihood
292: of extinction as $\rho \longrightarrow \rho_c^+$ rather than the definite
293: survival/extinction result that the mean field predicts.
294: \begin{figure}[tb]
295: \centering\noindent
296: \begin{tabular}{c}
297: \tiny{a)} \\
298: \includegraphics[width=8cm]{Figure_5a.eps}
299: \\ \tiny{b)} \\
300: \includegraphics[width=8cm]{Figure_5b.eps}
301: \end{tabular}
302: \caption{a) The average population density of the surviving runs only.
303: b) The average population density of all the runs (solid line) and the survival probability $P(t)$ (hashed line), i.e. the probability
304: that extinction has not occurred up to time $t$.}
305: \label{Disease}
306: \end{figure}
307: Fig. \ref{Disease} shows the results for $\rho_s = 0.13$, where we see that for those runs
308: that {\em did} survive, the population density does indeed return to what it once
309: was. We also see, as expected, that most of the runs did result in extinction.
310: In fact the survival rate was 0.004.
311:
312: From Fig. \ref{Disease} b) we observe that there is a time delay of approximately
313: 40 time steps between the
314: sudden decrease in population and when
315: the survival probability begins to fall. Assuming a particle that dies the $n$th time it is picked, survives
316: $n-1$ time steps, it is easy to show that the expected lifetime (in time steps) of an individual is given by $(1-p_d)/p_d$. We therefore have a time delay of approximately four lifetimes (recall $p_d = 0.093$) which, for a lot of species, is ample time to act.
317:
318: In order to prevent extinction in such a case, the population density must be increased beyond $\rho_{c}$. This has important ecological implications
319: since it shows that the probability of extinction can be decreased, not only
320: by increasing the population (which is of course not always possible), but
321: also by a {\em decrease} in habitat size.
322:
323: To see whether this hypothesis holds, we simulate this again using $p_d
324: = 0.093$ but this time $\rho_s =\rho_c = 0.128$ so that the chance of survival is negligible. This time however, once the population density has been reduced,
325: the area covered by the lattice is reduced by half. The organisms in the half that remains are left where they are, whereas those
326: in the half that is removed are randomly placed in the remaining half. This
327: then doubles the population density, bringing the population out
328: of the sub-critical population density. Once the population has recovered
329: and stabilised, the lattice size is returned to how it once was. The results
330: are shown in Fig. \ref{Disease Recovery}
331: \begin{figure}[tb]
332: \centering\noindent
333: \includegraphics[width=8cm]{Figure_6.eps}
334: \caption{Plot showing the recovery of the population $n(t)$ for
335: the surviving runs only after a disease
336: breakout at $t = 3000$ due to the re-sizing of the lattice. The lattice
337: is returned to how it was originally at $t=6000$ and the population recovers its original value.}
338: \label{Disease Recovery}
339: \end{figure}
340: and clearly show the recovery of the population once the lattice
341: size has been reduced. In fact, out of 1000 runs, the probability of survival
342: rose from 0.003 to 0.281.
343:
344: We expect there to be an optimal habitat reduction size - too large a reduction
345: and the population will be in danger from large fluctuations associated with
346: smaller habitat sizes whereas too small
347: a reduction and the density will not be increased sufficiently. We therefore
348: plot in Fig. \ref{HabitatReduction}
349: \begin{figure}[tb]
350: \centering\noindent
351: \includegraphics[width=8cm]{Figure_7.eps}
352: \caption{How the probability of survival changes with different reductions
353: in $L$, starting from $L=32$.}
354: \label{HabitatReduction}
355: \end{figure}
356: the probability that the system does not go extinct up to some $t=t_{max}$,
357: $P_s$, against the reduction in $L$, $\Delta L$. We again use $p_d = 0.093$
358: and $\rho_s = 0.128$. For small $\Delta L$, $P_s$ changes little
359: due to the density not being reduced enough, yet for larger $\Delta L$, the
360: larger fluctuations resulting from the smaller value of $L$ also cause $P_s$
361: to be small.
362:
363: Whilst reflective boundary conditions were used
364: here, very similar results were obtained using periodic boundary conditions.
365: In fact, with periodic boundary conditions, the probability of survival increased more significantly by
366: the decrease in $L$ due to the population being able to grow in two directions
367: rather than in just one after the habitat size has been returned to what
368: it once was. This of course could be achieved in reality
369: by reducing the habitat from more than one direction.
370: \\ \\
371: This model was proposed to represent how the area in
372: which a population is found could be reduced in real-life. The species could
373: be driven towards one end of the habitat with a boundary placed to prevent
374: them leaving the desired area. This boundary could then be removed once the
375: population has recovered. Clearly this is easier for larger, land-based animals
376: but in principle, at least, could be achieved for all species.
377:
378: \section{Conclusions} \label{Section: Conclusions}
379: Allee effects are certainly observed in nature \cite[][et al]{Stephens2,Pederson,Gyllenberg}
380: and have been studied with respect to extinction. Using a lattice model,
381: we have observed Allee effects together with the role of fluctuations, with the advantage of being able to examine the effects of habitat size. Being
382: able to model the population as a group of \textit{individuals} which move, breed and die, rather than as a variable in an equation, has enabled us to gain a more realistic insight into how real populations behave.
383:
384: Rather than the clear-cut conclusions that deterministic models produce,
385: conservationists often examine the \textit{probability}
386: that a population will maintain itself without significant demographic or
387: genetic manipulation for the foreseeable ecological future \citep{Soule}.
388: In this spirit, for a sufficiently large population density we have shown that the probability of survival does increase with habitat
389: size due to the smaller fluctuations. However, far more important are the death rate and population density since if these fall on the wrong side of their critical values, extinction is almost a certainty.
390:
391: Our findings are certainly significant for the design of habitats. The notion of a critical habitat size, mentioned in the Introduction, is misleading, since, it is certainly not true that for a fixed population size, the larger the habitat size the better. Regardless of the amount of space and resources available,
392: a population will only grow if the density is above its critical
393: value. We also proposed, in the last section, a method for greatly reducing the probability of extinction by reducing the habitat size once a species has become rare.
394:
395: Our notion of density has been that of the number of individuals per unit
396: area. While we assumed this to be constant in space when deriving our mean
397: field equation (\ref{mean field}), clearly this will vary amongst real populations. In fact, for populations that are found in patches, the value
398: of the density will depend very much on the scales used. The same is true
399: of the MC results as shown in Figure \ref{ClusteringPicture},
400: \begin{figure}[tb]
401: \centering\noindent
402: \includegraphics[width=8cm]{Figure_8.eps}
403: \caption{Snapshot of the output from a 2 dimensional lattice with
404: $L=100$. A value of $p_d = 0.1$ was used and the picture was taken
405: at $t = 600$ when $\rho = 0.2223.$}
406: \label{ClusteringPicture}
407: \end{figure}
408: where we see clear examples of clustering. In nature, species will cluster
409: to varying degrees and hence the value of the critical population density
410: will also vary and would need to be estimated in each case.
411: \\ \\
412: Compared to other stochastic models, we claim the use of lattice models gives
413: a more realistic insight into the way in which real populations behave. We
414: do still however, recognise the inaccuracies in our model and the difficulties in
415: implementing the observations. We believe the model to be valid, to a greater
416: or lesser degree to all species which rely on others for growth, perhaps
417: particularly those who live alone yet sexually reproduce. In fact, due to the great
418: variety of species, we have presented the above as ideas which may be of qualitative, rather than quantitative, relevance to conservation management.
419: \section*{Acknowledgments}
420: We would like to thank Be\'ata Oborny for very helpful discussions and references. Alastair Windus would also like to thank the Engineering
421: and Physical Sciences Research Council (EPSRC) for his Ph.D. studentship.
422: %\bibliography{bibliography}
423: \begin{thebibliography}{31}
424: \expandafter\ifx\csname natexlab\endcsname\relax\def\natexlab#1{#1}\fi
425: \expandafter\ifx\csname url\endcsname\relax
426: \def\url#1{\texttt{#1}}\fi
427: \expandafter\ifx\csname urlprefix\endcsname\relax\def\urlprefix{URL }\fi
428:
429: \bibitem[{Aizen and Feinsinger(1994)}]{Aizen}
430: Aizen, M., Feinsinger, P., 1994. Forrest fragmentation, pollination and plant
431: reproduction in {C}haco dry forrest, {A}rgentina. Ecology 75~(2), 330--351.
432:
433: \bibitem[{Allee(1931)}]{Allee}
434: Allee, W., 1931. Animal Aggregation: A Study in General Sociology. University
435: of Chicago Press.
436:
437: \bibitem[{Allen et~al.(2005)Allen, Fagan, H\"ogn\"as, and Fagerholm}]{Allen}
438: Allen, L., Fagan, J., H\"ogn\"as, G., Fagerholm, H., 2005. Population
439: extinction in discrete-time stochastic population models with an allee
440: effect. J. Differ. Equ. Appl. 11~(4-5), 273--293.
441:
442: \bibitem[{Amarasekare(1998)}]{Amarasekare}
443: Amarasekare, P., 1998. Allee effects in metapopulation dynamics. Am. Nat.
444: 152~(2), 298--302.
445:
446: \bibitem[{Courchamp et~al.(1999)Courchamp, Clutton-Brock, and
447: Grenfell}]{Courchamp}
448: Courchamp, F., Clutton-Brock, T., Grenfell, B., 1999. Inverse density
449: dependence and the allee effect. Trends Ecol. Evol. 14~(10), 405--410.
450:
451: \bibitem[{Dammer and Hinrichsen(2003)}]{Dammer}
452: Dammer, S., Hinrichsen, H., 2003. Epidemic spreading with immunization and
453: mutations. Phys. Rev. E 68~(1), 016114.
454:
455: \bibitem[{Dennis(1989)}]{Dennis}
456: Dennis, B., 1989. Allee effects: population growth, critical density, and the
457: chance of extinction. Nat. Res. Model. 3, 481--538.
458:
459: \bibitem[{Dennis(2002)}]{Dennis2}
460: Dennis, B., 2002. Allee effects in stochastic populations. OIKOS 96~(3),
461: 389--401.
462:
463: \bibitem[{Durrett and Levin(1998)}]{Durrett}
464: Durrett, R., Levin, S., 1998. Spatial aspect of interspecific competition.
465: Theor. Popul. Biol. 53~(1), 30--43.
466:
467: \bibitem[{Escudero(2005)}]{Escudero}
468: Escudero, C., 2005. Particle statistics and population dynamics. Physica A 354,
469: 371--380.
470:
471: \bibitem[{Escudero et~al.(2004)Escudero, Buceta, de~la Rubia, and
472: Lindenberg}]{Escudero_Buceta}
473: Escudero, C., Buceta, J., de~la Rubia, F., Lindenberg, K., 2004. Extinction in
474: population dynamics. Phys. Rev. E 69, 021908.
475:
476: \bibitem[{Gyllenberg et~al.(1997)Gyllenberg, Hanski, and Hastings}]{Gyllenberg}
477: Gyllenberg, M., Hanski, I., Hastings, A., 1997. Structured {M}etapopulation
478: {M}odels. In: Hanski, I., Gilpin, M. (Eds.), Metapopulation {B}iology:
479: {E}cology, {G}enetics and {E}volution. Academic Press, London, pp. 93--122.
480:
481: \bibitem[{Harris(1974)}]{Harris}
482: Harris, T.~E., 1974. Contact interactions on a lattice. Ann. Probab. 2~(6),
483: 969--988.
484:
485: \bibitem[{Hinrichsen(2000)}]{Hinrichsen_Non}
486: Hinrichsen, H., 2000. Non-equilibrium critical phenomena and phase transitions
487: into absorbing states. Adv. Phys. 49~(7), 815--958.
488:
489: \bibitem[{Hurford et~al.(2006)Hurford, Hebblewhite, and Lewis}]{Hurford}
490: Hurford, A., Hebblewhite, M., Lewis, M., 2006. A spatially explicit model for
491: an allee effect: Why wolves recolonize so slowly in {G}reater {Y}ellowstone.
492: Theor. Popul. Biol. 70~(3), 244--254.
493:
494: \bibitem[{Itoh et~al.(2004)Itoh, Tainaka, Sakata, Tao, and Nakagiri}]{Itoh}
495: Itoh, Y., Tainaka, K., Sakata, T., Tao, T., Nakagiri, N., 2004. Spatial
496: enhancement of population uncertainty near the extinction threshold. Ecol.
497: Model. 174~(1-2), 191--201.
498:
499: \bibitem[{Johnson et~al.(2006)Johnson, Liebhold, Tobin, and
500: Bjornstad}]{Johnson}
501: Johnson, D., Liebhold, A., Tobin, P., Bjornstad, O., 2006. Allee effects and
502: pulsed invasion by the gypsy moth. Nature 444~(7117), 361--363.
503:
504: \bibitem[{Lande(1987)}]{Lande2}
505: Lande, R., 1987. Extinction thresholds in demographic models of territorial
506: populations. Am. Nat. 130~(4), 624--635.
507:
508: \bibitem[{Lande et~al.(2003)Lande, Engen, and Saether}]{Lande}
509: Lande, R., Engen, S., Saether, B., 2003. Stochastic Population Dynamics In
510: Ecology and Conservation. Oxford University Press.
511:
512: \bibitem[{May et~al.(1995)May, Lawton, and Stork}]{May}
513: May, R., Lawton, J., Stork, N., 1995. Assessing extinction rates. In: Lawton,
514: J., May, R. (Eds.), Extinction Rates. Oxford University Press, pp. 1--24.
515:
516: \bibitem[{Oborny et~al.(2005)Oborny, Mesz\'ena, and Szab\'o}]{Oborny}
517: Oborny, B., Mesz\'ena, G., Szab\'o, G., 2005. Dynamics of populations on the
518: verge of extinction. OIKOS 109~(2), 291--296.
519:
520: \bibitem[{Opper and Saad(2001)}]{Opper}
521: Opper, M., Saad, D. (Eds.), 2001. Advanced {M}ean {F}ield {M}ethods: {T}heory
522: and {P}ractice. MIT press.
523:
524: \bibitem[{Pedersen et~al.(2001)Pedersen, Hanslin, and Bakken}]{Pederson}
525: Pedersen, B., Hanslin, H., Bakken, S., 2001. Testing for positive
526: density-dependent performance in four bryophyte species. Ecology 82~(1),
527: 70--88.
528:
529: \bibitem[{Peters and Neelin(2006)}]{Peters}
530: Peters, O., Neelin, J., 2006. Critical phenomena in atmospheric precipitation.
531: Nat. Phys 2~(6), 393--396.
532:
533: \bibitem[{Pimm(1991)}]{Pimm}
534: Pimm, S., 1991. The Balance of Nature? Ecological Issues in the Conservation of
535: Species and Communities. The University of Chicago Press.
536:
537: \bibitem[{Skellam(1951)}]{SKellam}
538: Skellam, J.~G., 1951. Random dispersal in theoretical populations. Biometrika
539: 38~(1-2), 196--218.
540:
541: \bibitem[{Soul\'e(1987)}]{Soule}
542: Soul\'e, M. (Ed.), 1987. Viable Populations for Conservation. Cambridge
543: University Press.
544:
545: \bibitem[{Stephens and Sutherland(1999)}]{Stephens2}
546: Stephens, P., Sutherland, W., 1999. Consequences of the {A}llee effect for
547: behaviour, ecology and conservation. Trends Ecol. Evol. 14~(10), 401--405.
548:
549: \bibitem[{Stephens et~al.(1999)Stephens, Sutherland, and Freckleton}]{Stephens}
550: Stephens, P., Sutherland, W., Freckleton, R., 1999. What is the {A}llee effect?
551: OIKOS 87~(1), 185--190.
552:
553: \bibitem[{Tainaka(1988)}]{Tainaka}
554: Tainaka, K., 1988. Lattice model for the lotka-volterra system. J. Phys. Soc.
555: Jpn. 57~(8), 2588--2590.
556:
557: \bibitem[{Vespignani et~al.(2000)Vespignani, Dickman, Munoz, and
558: Zapperi}]{Vespignani}
559: Vespignani, A., Dickman, R., Munoz, M., Zapperi, S., 2000. Absorbing-state
560: phase transitions in fixed-energy sandpiles. Phys. Rev. E 62~(4), 4564--4582.
561:
562: \end{thebibliography}
563:
564:
565: \end{document}
566:
567: