Finite size scaling of survival statistics in metapopulation models

Alice Doimo, Giorgio Nicoletti, Davide Bernardi, Prajwal Padmanabha · 2024

Read Finite size scaling of survival statistics in metapopulation models by Alice Doimo, Giorgio Nicoletti, Davide Bernardi, Prajwal Padmanabha (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-18448.txt.

Introduction

Doimo and co-authors restore chance to a classic ecology model. Metapopulation theory tracks species spread across habitat patches, but its standard form assumes huge local populations, washing out the randomness of births, deaths, and dispersal. Real patches hold finite numbers, and luck can kill a population that averages say should live.

The team derives how survival odds scale with patch capacity near the knife-edge between persistence and extinction. The deterministic capacity, the old model's summary number, still governs the statistics: it sets both survival probability and the moments of extinction time. The formulas give ecologists a way to fold demographic luck into landscape models without abandoning their structure.

Catalog record

CREATORAlice Doimo, Giorgio Nicoletti, Davide Bernardi, Prajwal Padmanabha
YEAR2024
COLLECTIONPAPERS
FORMATTXT
FILE SIZE67 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2412-18448

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Finite size scaling of survival statistics in metapopulation models Alice Doimo, Giorgio Nicoletti, Davide Bernardi, Prajwal Padmanabha Abstract Spatial metapopulation models are fundamental to theoretical ecology, enabling to study how landscape structure influences global species dynamics. Traditional models, including recent generalizations, often rely on the deterministic limit of stochastic processes, assuming large population sizes. However, stochasticity - arising from dispersal events and population fluctuations - profoundly shapes ecological dynamics. In this work, we extend the classical metapopulation framework to account for finite populations, examining the impact of stochasticity on species persistence and dynamics. Specifically, we analyze how the limited capacity of local habitats influences survival, deriving analytical expressions for the finite-size scaling of the survival probability near the critical transition between survival and extinction. Crucially, we demonstrate that the deterministic metapopulation capacity plays a fundamental role in the statistics of survival probability and extinction time moments. These results provide a robust foundation for integrating demographic stochasticity into classical metapopulation models and their extensions. Preprint: APS/123-QED I introduction Understanding the dynamics of populations distributed across fragmented habitats is a longstanding challenge in ecology and conservation biology. Metapopulation theory provides a foundational framework for examining how local extinctions and landscape-mediated colonization processes shape the collective dynamics of spatially structured populations over time [ 1 , 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 ] . The concept of metapopulation, first introduced by Levins [ 11 ] , describes a “population of populations”, i.e., a set of distinguished subpopulations that are spatially separated, but interconnected by the ongoing exchange of individuals. This exchange occurs across a spatial network of habitat patches, varying in quality, connectivity, and area [ 2 ] . Both field and theoretical studies have highlighted the validity of this approach, showing that landscape topology significantly influences the flow of individuals between habitats, ultimately determining the survival or extinction of the metapopulation [ 12 , 13 , 14 , 15 , 16 ] . In particular, to capture the effects of spatial structure and habitat fragmentation on species persistence, Hanski and Ovaskainen introduced a fundamental measure known as metapopulation capacity [ 17 , 18 , 19 ] , which determines the survival of a focal species within a given landscape. This measure is defined as the maximum eigenvalue of an appropriate landscape matrix, and allows for a direct comparison between theoretical models and real-world networks of habitat fragments [ 20 ] . Although the application of network theory to spatial ecology [ 21 , 22 , 23 ] has advanced the study of complex dispersal network structures [ 5 , 24 , 25 , 26 , 27 , 28 ] , metapopulation capacity has traditionally been defined within deterministic models. Crucially, such models overlook fluctuations arising from demographic and environmental stochasticity [ 29 , 30 , 31 ] . Early contributions by Lande [ 32 ] and Hanski [ 33 ] pioneered the integration of stochasticity into metapopulation dynamics, emphasizing the increased risk of extinction posed by random fluctuations at low population levels. Migration through spatially connected patches, by contrast, provides insurance against simultaneous global extinction, underscoring the fundamental role of dispersal network topology. Classical approaches to studying these dynamics, such as Stochastic Patch Occupancy Model [ 34 , 35 ] , the Incidence Function Model [ 33 ] and their extensions [ 36 , 37 , 38 , 39 ] , have enabled the computation of extinction times and quasi-stationary occupancy distributions. However, these approaches predominantly rely on purely numerical methods due to the complexity of stochastic processes, spatial heterogeneity, and variability in the parameter space [ 40 ] . As a consequence, in metapopulation dynamics, qualitative results are often obtained from deterministic approximations of stochastic processes, which lead to theoretically tractable models and provide broad qualitative insights [ 4 ] . In this work, we rigorously extend the traditional deterministic metapopulation framework by explicitly incorporating the stochasticity of individual-level processes to investigate analytically their potential impact on metapopulation dynamics and persistence. To incorporate stochasticity from the ground up, we focus on the role of finite carrying capacity, i.e., the maximum size of the subpopulations that compose the metapopulation. This approach offers a precise way to control fluctuations, allowing us to disentangle the effects of local demographic stochasticity and dispersal network topology. Specifically, we examine whether critical thresholds in carrying capacity exist beyond which stochastic fluctuations become catastrophic, or conversely, whether there are local system sizes beyond which population persistence is significantly enhanced. To address this question, we build upon our previous work, which extends the deterministic spatial metapopulation model developed by Hanski and Ovaskainen to arbitrary landscape structures [ 5 ] . We adopt a bottom-up approach, starting from a microscopic model describing colonization, extinction, and dispersal of a focal species at the individual level. To analytically study the stochastic dynamics of the system, we perform a Van Kampen inverse system-size expansion of the corresponding master equation [ 41 ] , selecting the inverse of the carrying capacity of the patches - i.e., the maximum local population size supported by each patch - as the expansion parameter. We then introduce a series of simplifying hypotheses to derive analytical insights from the resulting Fokker-Planck equation. In particular, we assume translational invariance within a complete network and marginalize to capture the effective behavior of representative single-site variables. By imposing a separation of timescales [ 42 ] between dispersal and local processes, we derive an effective one-dimensional quasi-stationary (QSS) Langevin equation that accurately captures the overall metapopulation dynamics. In the deterministic limit, we recover the expected absorbing phase transition [ 43 , 44 ] between extinction and survival regimes, as determined by the system’s metapopulation capacity. Stochasticity, however, renders the survival regime metastable - a key result especially relevant for systems with small carrying capacity. Moreover, the effective QSS equation enables us to analytically study the general scaling behavior of the survival probability in the vicinity of the extinction transition. We demonstrate that near this transition point, the survival probability for different carrying capacities exhibits a universal behavior consistent with finite-size scaling [ 45 , 46 ] . Furthermore, we analytically derive the scaling of the first and second moments of extinction time with varying carrying capacity, showing that they similarly collapse to a universal form. Through this analysis, we bridge results from the deterministic metapopulation framework with its stochastic counterpart. We demonstrate that the metapopulation capacity - a deterministic measure - still plays a crucial role in the statistics of survival probability and extinction time moments. II model We introduce a model that describes the microscopic dynamics of a single species within a metapopulation. Individuals undergo processes such as birth, death, reproduction, and dispersal across a network of habitat patches. This model was proposed in [ 5 ] , where the connection between its deterministic limit and the Hanski and Ovaskainen model [ 2 , 18 ] was demonstrated. In this section, we first provide a detailed description of the microscopic processes governing the dynamics and then briefly recall the derivation of the macroscopic deterministic equations, as outlined in [ 5 ] . This is instrumental in the analysis of the full

Read the complete text