NetLogo User Community Models

(back to the NetLogo User Community Models)

Download If clicking does not initiate a download, try right clicking or control clicking and choosing "Save" or "Download".(The run link is disabled for this model because it was made in a version prior to NetLogo 6.0, which NetLogo Web requires.)

## WHAT IS IT?

The model's goal is to introduce an agent-based method for landscape diversity measurements. The landscape's structural complexity depends on the composition and the configuration as well. The composition refers to the number and occurrence of different landscape types while the configuration means the physical distribution or spatial character within a landscape. (McGarigal et al. 1994)
By the landscape discovery agents (scouts) we collect the different kind of land covers into a common sum as diversity potential. It gives a result in real complixity (on low abstraction level), which includes composition and configuration. The model describes the landscape diversity as a living creature, instead of decomposition the problem to different indices and metrics - like edge density, number of classes, patch number, Shannon index, Moran’s I.

## HOW IT WORKS

The scouts have a memory, since they individually stores those land cover what they have seen. In addition they search in a conic shape of view with a given distance. Their task is the collection of different land cover types as fast as possible. If they discover something new land cover, then one potential unit is added to the world's landscape diversity potential (as Monte Carlo integral).

## HOW TO USE IT

First you have to click ‘Setup world’, before this part you can predefine the number of classes, and the graininess (seed percent) of the landscape. Then the agents (here scouts) must be placed with the ‘Setup scouts’ button. At this step you can vary the parameters of the scout discovery, for example you can put a stochastic uncertainty on them. Finally you can run the simulation with the Scouting button.

## THINGS TO TRY

First you should generate a world with the initial setting, and test the model with only one scout. The scout's path helps you to understand the underlying behaviour. You can check that the last potential is equal with the number of the discovered land cover types (original land cover not included).

Then you can create a more complex world with a bit higher seed percent (e.g. 0.01) and land cover class (e.g. 10). Run the simulation with 50 scouts for 100 ticks, and note down the potential what you got. Without changing the agent parameters rerun the discovery on a less diverse landscape (have only 3 kind of land cover) and compare the results.

## EXTENDING THE MODEL

The model is expandable, we could order higher scores for particular land covers like forests.
An elevation model could be added to model. It could have an effect on the agents' speed or their visibility distance.
The simulation time could be changed to agent energy. For example, they would have an initial energy which increases by discovering land cover types, and decreases by moving or rotating.

The user can import and analyse real landscape models through NetLogo's gis extension.

## CREDITS AND REFERENCES

Wirth, E., Szabó, Gy., and Czinkóczky, A.: MEASURE OF LANDSCAPE HETEROGENEITY BY AGENT-BASED METHODOLOGY, ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., III-8, 145-151, http://dx.doi.org/10.5194/isprs-annals-III-8-145-2016, 2016.

McGarigal, K., Marks, B.: Fragstats: Spatial Pattern Analysis Program for Quantifying Landscape Structure. Forest Science Department, Oregon State University, Corvallis, 1994

Wilensky, U.: NetLogo, http://ccl.northwestern.edu/netlogo/. Center for Connected Learning and Computer-Based Modeling, Northwestern University, Evanston, IL., 1999