rcontroll: An R interface for the individual-based forest dynamics simulator TROLL
A central challenge in ecology is understanding the emergence of patterns as the result of interactions among individuals. Dynamic forest models can provide a fine-scale description of the ecological, physiological and environmental processes that explain the demography of coexisting tree species. This in turn helps predict changes under future scenarios. However, model accessibility is a major obstacle to a wide use and communication across scientific disciplines and for educational purposes. Here, we present the R package rcontroll, which provides access to the TROLL forest simulator in the R environment. TROLL is individual-based and spatially explicit and leverages knowledge of ecology, biogeochemistry and tree ecophysiology through a trait-based parameterisation. TROLL has been used to simulate carbon fluxes and tree diversity in tropical and subtropical forests and to explore forest resilience to disturbance and environmental changes more generally. rcontroll provides a user-friendly environment to set up and analyse TROLL simulations with varying community compositions, ecological parameters and climate conditions. We show how to test parameter sensitivity in TROLL using the rcontroll R package. We also demonstrate the flexibility and ease of use of rcontroll by replicating a previously published study based on the TROLL simulator. Both examples are included with reproducible code documents. Complex forest simulators are important scientific tools for science and education, and wide access to these tools is an important condition for their adoption. TROLL is designed to address a wide range of ecological and environmental questions, and the new R package rcontroll is designed to be an entry point for TROLL model users.
| Summary: | A central challenge in ecology is understanding the emergence of patterns as the result of interactions among individuals. Dynamic forest models can provide a fine-scale description of the ecological, physiological and environmental processes that explain the demography of coexisting tree species. This in turn helps predict changes under future scenarios. However, model accessibility is a major obstacle to a wide use and communication across scientific disciplines and for educational purposes. Here, we present the R package rcontroll, which provides access to the TROLL forest simulator in the R environment. TROLL is individual-based and spatially explicit and leverages knowledge of ecology, biogeochemistry and tree ecophysiology through a trait-based parameterisation. TROLL has been used to simulate carbon fluxes and tree diversity in tropical and subtropical forests and to explore forest resilience to disturbance and environmental changes more generally. rcontroll provides a user-friendly environment to set up and analyse TROLL simulations with varying community compositions, ecological parameters and climate conditions. We show how to test parameter sensitivity in TROLL using the rcontroll R package. We also demonstrate the flexibility and ease of use of rcontroll by replicating a previously published study based on the TROLL simulator. Both examples are included with reproducible code documents. Complex forest simulators are important scientific tools for science and education, and wide access to these tools is an important condition for their adoption. TROLL is designed to address a wide range of ecological and environmental questions, and the new R package rcontroll is designed to be an entry point for TROLL model users. |
|---|