Quo vadis, agent-based modelling tools?

Daly, Aisling J.; Visscher, Lander De; Baetens, Jan; Baets, Bernard De

doi:10.1016/j.envsoft.2022.105514

Cited by 11 publications

(3 citation statements)

References 78 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Currently, these approaches face various challenges due to the complexity surrounding IBM development and analysis. However, current trends in next-generation IBMs increasingly promote standardization of both development and analysis, paving the way for making these parameter inference approaches more widely applicable (Daly et al 2022).…”

Section: Extending Deb-ibm: Modelling Behaviours and Data Needsmentioning

confidence: 99%

Environment‐dependent population dynamics emerging from dynamic energy budgets and individual‐scale movement behaviour

Barhdadi,

Daly,

Baetens

et al. 2023

Oikos

Self Cite

View full text Add to dashboard Cite

Individual biology influences environment‐dependent population dynamics through life history. Population models that consider individual physiology are therefore popular for modelling dynamics under various environments. In recent years, a quantitative framework integrating metabolic theory (dynamic energy budget theory) into individual‐based models (DEB‐IBMs) has emerged to investigate the link from individual physiology to demography. However, this link shows substantial variation, some of which may be explained by individual behaviours operating in local environments. Current DEB‐IBMs focusing on population modelling do not consider individual‐scale behaviours and instead resort to imposed population‐level relationships. We therefore propose to extend the DEB‐IBM approach to consider the role of individual‐scale behaviour by replacing the functional response – a population‐averaged phenomenological relationship – with individual‐scale foraging mechanisms in a spatially heterogeneous environment. Using this model, we simulate consumer dynamics in a consumer‐resource system for different individual behaviours across a range of temperature, resource carrying capacity and individual variability values. We further illustrate the model in a case study by comparing simulated population dynamics with both the classical DEB‐IBM and experimental data for a laboratory Daphnia magna population. Simulations reveal that temperature‐ and resource‐dependent consumer extinction probability patterns change with individual behaviour. Moreover, simulations agree with experimental data on D. magna populations: dynamics after the initial growth peak were better captured under random walk movement behaviour compared to the classical DEB‐IBM. Both the simulation and case study showed how fine‐scale behaviour mediates the metabolism's impact on population dynamics by allowing for the emergence of different functional responses. Our model thus provides a link between metabolism, life history and population dynamics by centring behavioural mechanisms and environmental heterogeneity at the individual scale. This expansion of the modelling toolbox for physiologically structured populations can boost theory development by bridging various fields in ecology, and contribute to our understanding of environment‐dependent ecological patterns.

show abstract

Section: Extending Deb-ibm: Modelling Behaviours and Data Needsmentioning

confidence: 99%

Environment‐dependent population dynamics emerging from dynamic energy budgets and individual‐scale movement behaviour

Barhdadi,

Daly,

Baetens

et al. 2023

Oikos

Self Cite

View full text Add to dashboard Cite

show abstract

“…Some can be measured directly, while others have to be inferred by comparing the emerging simulated dynamics to those observed in reality (Thiele et al, 2014). Of the many problems that may arise when doing parameter inference of complex, non-linear models (e.g., equifinality (Beven and Freer, 2001), dependencies between parameters (Li and Vu, 2013), sloppy parameter sensitivities (Gutenkunst et al, 2007)), one is particularly cumbersome for many ABM applications: the lack of a tractable likelihood (Daly et al, 2022). With the latter we refer to the fact that the likelihood function of most ABMs -which plays a key role in parameter inference in both frequentist and Bayesian statistics (Gelman et al, 2013) -can rarely be evaluated in practice.…”

Section: Introductionmentioning

confidence: 99%

A critical review of common pitfalls and guidelines to effectively infer parameters of agent-based models using Approximate Bayesian Computation

De Visscher,

De Baets,

Baetens

2024

Environmental Modelling & Software

Self Cite

View full text Add to dashboard Cite

“…Developing strategies to maximise the efficacy of these tools is critical, and such strategies should be well-informed by predictions made from detailed, quantitative genetic models. Here we introduce the resevol R package as a tool for building individual-based models and simulating pest management [ 21 ].…”

Section: Introductionmentioning

confidence: 99%

resevol: An R package for spatially explicit models of pesticide resistance given evolving pest genomes

Duthie,

Mangan,

McKeon

et al. 2023

PLoS Comput Biol

View full text Add to dashboard Cite

The evolution of pesticide resistance is a widespread problem with potentially severe consequences for global food security. We introduce the resevol R package, which simulates individual-based models of pests with evolving genomes that produce complex, polygenic, and covarying traits affecting pest life history and pesticide resistance. Simulations are modelled on a spatially-explicit and highly customisable landscape in which crop and pesticide application and rotation can vary, making the package a highly flexible tool for both general and tactical models of pest management and resistance evolution. We present the key features of the resevol package and demonstrate its use for a simple example simulating pests with two covarying traits. The resevol R package is open source under GNU Public License. All source code and documentation are available on GitHub.

show abstract

Quo vadis, agent-based modelling tools?

Cited by 11 publications

References 78 publications

Environment‐dependent population dynamics emerging from dynamic energy budgets and individual‐scale movement behaviour

Environment‐dependent population dynamics emerging from dynamic energy budgets and individual‐scale movement behaviour

A critical review of common pitfalls and guidelines to effectively infer parameters of agent-based models using Approximate Bayesian Computation

resevol: An R package for spatially explicit models of pesticide resistance given evolving pest genomes

Contact Info

Product

Resources

About