Community dynamics and sensitivity to model structure: towards a probabilistic view of process-based model predictions

Abstract

ResearchCite this article: Aldebert C, Stouffer DB. 2018 Community dynamics and sensitivity to model structure: towards a probabilistic view of process-based model predictions. J. R. Soc. Interface 15: 20180741. http://dx.Statistical inference and mechanistic, process-based modelling represent two philosophically different streams of research whose primary goal is to make predictions. Here, we merge elements from both approaches to keep the theoretical power of process-based models while also considering their predictive uncertainty using Bayesian statistics. In environmental and biological sciences, the predictive uncertainty of process-based models is usually reduced to parametric uncertainty. Here, we propose a practical approach to tackle the added issue of structural sensitivity, the sensitivity of predictions to the choice between quantitatively close and biologically plausible models. In contrast to earlier studies that presented alternative predictions based on alternative models, we propose a probabilistic view of these predictions that include the uncertainty in model construction and the parametric uncertainty of each model. As a proof of concept, we apply this approach to a predator -prey system described by the classical Rosenzweig-MacArthur model, and we observe that parametric sensitivity is regularly overcome by structural sensitivity. In addition to tackling theoretical questions about model sensitivity, the proposed approach can also be extended to make probabilistic predictions based on more complex models in an operational context. Both perspectives represent important steps towards providing better model predictions in biology, and beyond.rsif.royalsocietypublishing.org J. R. Soc. Interface 15: 20180741

Hidden layers of density dependence in consumer feeding rates

“…What dynamical properties these varied consumer types may impart to their populations and food webs remains unknown. Recent analyses of existing functional-response models suggest their effects could be quite strong (Adamson & Morozov, 2013;2014;Aldebert & Stouffer, 2018;Coblentz & DeLong, 2020). In both the empirical and theoretical literatures, it is common for researchers to decide a priori which model is most appropriate given their focal consumer and to analyze their data accordingly.…”

Systematic bias in studies of consumer functional responses

“…Past assessments of interference strengths have relied on point estimates alone, just as is implicitly done when model performance is assessed using information criteria such as (Box 1). However, assessing the uncertainty of parameter estimates is equally important when drawing inferences (Osenberg et al ., 1999; Aldebert and Stouffer, 2018). We approximated standard errors in one of three ways: For non‐bootstrapped datasets, we first attempted the confint profiling method of bbmle (Bolker, 2020) assisted by the SEfromHessian function of the HelpersMG package (Girondot, 2020).…”

“…Applications of the functional‐response concept are found in pest control, invasive species management and conservation biology, and are fundamental to theory in population, community and evolutionary ecology. Much effort has therefore been devoted to experiments designed to measure and statistically compare the many mathematical models that have been developed to characterise consumer functional responses, with theory evidencing often dramatic differences in the biological inferences and predictions that alternative model forms and parameter values provide (Fussmann and Blasius, 2005; Arditi and Ginzburg, 2012; Aldebert and Stouffer, 2018; Coblentz and DeLong, 2020).…”

Emerging Frontiers in the Study of Molecular Evolution

“…Indeed, considering the influence of competition, predation or co-evolution in addition to the spatial arrangement of populations can provide new insights into the evolutionary processes that shape species' distributions (Hand et al 2015). Explicit models for species interactions in communities that interface with metagenomic and ultimately full genomic data are a futuristic area (Aldebert and Stouffer, 2018;Shoemaker et al 2019).…”