INVESTIGADORES
GHERMANDI luciana
artículos
Título:
A nonlinear mixed-effects modelling approach for ecological data: Using temporal dynamics of vegetation moisture as an example.
Autor/es:
ODDI F.; MIGUEZ F.; GHERMANDI L.; BIANCHI L.; GARIBALDI L.
Revista:
Ecology and Evolution
Editorial:
Wiley and Sons
Referencias:
Año: 2019
Resumen:
Abstract1. Increasingly, often ecologist collects data with nonlinear trends, heterogeneousvariances, temporal correlation, and hierarchical structure. Nonlinear mixed‐ef‐fects models offer a flexible approach to such data, but the estimation and inter‐pretation of these models present challenges, partly associated with the lack ofworked examples in the ecological literature.2. We illustrate the nonlinear mixed‐effects modeling approach using temporal dy‐namics of vegetation moisture with field data from northwestern Patagonia. Thisis a Mediterranean‐type climate region where modeling temporal changes in livefuel moisture content are conceptually relevant (ecological theory) and have prac‐tical implications (fire management). We used this approach to answer whethermoisture dynamics varies among functional groups and aridity conditions, andcompared it with other simpler statistical models. The modeling process is setout ?step‐by‐step?: We start translating the ideas about the system dynamics toa statistical model, which is made increasingly complex in order to include differ‐ent sources of variability and correlation structures. We provide guidelines and Rscripts (including a new self‐starting function) that make data analyses reproduc‐ible. We also explain how to extract the parameter estimates from the R output.3. Our modeling approach suggests moisture dynamic to vary between grasses andshrubs, and between grasses facing different aridity conditions. Compared tomore classical models, the nonlinear mixed‐effects model showed greater good‐ness of fit and met statistical assumptions. While the mixed‐effects approach ac‐counts for spatial nesting, temporal dependence, and variance heterogeneity; thenonlinear function allowed to model the seasonal pattern.4. Parameters of the nonlinear mixed‐effects model reflected relevant ecologicalprocesses. From an applied perspective, the model could forecast the time when