What is SORTIE?
SORTIE is an individual-tree, spatially-explicit model of forest dynamics. It originated from the small-scale disturbance model SORTIE developed and tested in the early 1990s for transitional oak-northern hardwood forests in the northeastern US (Pacala et al. 1996). In 1995 Dave Coates, Phil LePage, Elaine Wright and other scientists from the Research Section of the British Columbia Forest Service in Smithers, BC began collaborating with Charles Canham and Lora Murphy of the Cary Institute of Ecosystem Studies, Millbrook, New York to start a SORTIE research program in the transitional interior cedar hemlock forests of northwestern BC as part of the Date Creek Experiment.
As part of the development of SORTIE/BC, further sub-models were added, including a disturbance module that allowed different types of partial cutting and planting. Because of the many changes being made to the model, SORTIE/BC was restructured and reprogrammed in C++ in the early 2000s. The result is SORTIE-ND where ND signifies the model’s focus on local neighbourhood dynamics. The SORTIE-ND model, code, user manual, and other useful information reside here. Users and developers of the model can keep in contact through this site.
SORTIE-ND simulates changes in tree populations over time. The model uses a combination of empirical and mechanistic behaviours to predict forest dynamics based on field experiments that measure fine-scale and short-term interactions among individual trees. All field studies operate at the neighbourhood scale of forest dynamics. SORTIE-ND is designed to provide growth predictions for individual trees in multi-species complex structured stands. It has a much higher degree of flexibility in terms of which processes (termed behaviours) are set to act on a population of trees. SORTIE-ND encapsulates the theory of neighbourhood dynamics, where interactions among individual trees and their spatially heterogeneous environment are inherently local in nature acting at a scale over restricted distances. All model behaviours and related parameters are user-specified and the model can be fit to a wide range of specific conditions. SORTIE-ND has an intermediate position between purely empirical and process-based models.
Why use SORTIE?
With the ability to capture the interaction between individuals across heterogeneous environments, SORTIE is an effective tool to better understand and test our assumptions of how forests act as complex adaptive systems. As management (harvest and planting) are built into the model, SORTIE allows us to test how forest management can impact future forest recovery.
The interactions among individual trees and their spatially heterogeneous environment are inherently local in nature acting at a neighbourhood scale over restricted distances. Unfortunately, traditional growth and yield models that are deterministic and non-spatial are not very useful in this context. Models that use trees as individual modeling agents, and are spatially explicit, represent a significant advancement.
In SORTIE-ND, many of the interactions have non-linear relationships and/or have random events associated with them. The non-linearity of interactions, stochastic behaviour processes, and a large number of behaviours makes SORTIE-ND a good example of a modeling approach aimed at being able to represent complex behaviour in forests.
The SORTIE model is available in two formats: a GUI and an R package. Regardless of which format you prefer, you must download and interact with the GUI.
Download the SORTIE-ND GUI below:
Modelling Concepts
The basic SORTIE model state is defined by the plot, trees, and grids. The plot is the underlying location in which the simulation takes place. It has a particular size and shape, and attributes for climate and geographic location. The trees are the individuals making up the forest on the plot. Grids hold additional data that varies from place to place, such as soil chemistry or light level at the forest floor. All of these together define the model state at a particular time.
The processes that act to change the model state are called behaviors. Behaviors often correspond to biological processes. They are individually contained units, but often work together to create a complex, interacting system. For instance, a simulation might consist of three behaviors: a behavior to calculate light levels for trees, one to determine the amount of tree growth as a result of the amount of light, and one to select trees to die if they grow too slowly. Behaviors are placed in a certain order to correctly structure their interactions.
The unit of time in SORTIE is the timestep. It represents a set of one or more years. A single timestep consists of each behavior acting according to their parameters, in a defined order. The process is repeated for the total number of timesteps, which creates a single simulation, or run.
The basic structure of the SORTIE system is very simple. Its power lies in its incredible flexibility. Almost every aspect of the model is under direct user control.
When you start the SORTIE software, you are using a tool that helps you to define the state data and behaviors that will make up a simulation. Once you have done this, you have created a parameter file. The parameter file completely defines a run. You can load and run your parameter file any time.
What is rSORTIE?
The SORTIE model is also available as an R package.
rsortie is an R interface that allows users to model forest stand dynamics in an R environment.
Follow the link below to learn more about rsortie and the benefits of using the SORTIE model in R.