Nova OnLine Model Library Examples

Each of the Webpage links below is a version of a Nova model designed to run in a Web browser. Each model is expressed in a Novascript variant that is equivalent to the script produced by the Nova Desktop capture process. A Nova runtime engine written entirely in Javascript is used to exectute the model in the client browser.

Nova Online uses HTML 5 controls for input. Visualization components designed to interface with Nova Online models (graphs, tables, agentviewers, etc.) have been built using the D3 platform. A rudimentary set of such visualizers is used here; future work will incorporate more advanced use of the D3 platform.

The first two are models used in published research.

Ebola Epidemic Model

An agent-based model described in Getz, et. al, "Tactics and Strategies for Managing Ebola Outbreaks and the Salience of Immunization", Computational and Mathematical Methods in Medicine, Volume 2015 (2015), Article ID 736507, http://dx.doi.org/10.1155/2015/736507.

Agent-based Model using a Genetic Algorithm

This is a stochastic, agent-based model to study how genetic traits and experiential changes in the state of agents and available resources influence individuals' foraging and movement behaviors. It is described fully in Getz, et. al., "Panmictic and Clonal Evolution on a Single Patchy Resource Produces Polymorphic Foraging Guilds", PLOS ONE, in press.

The following are drawn from the Nova Model Library.

Double Exponential Growth

This model shows 2 instances of an exponential growth population model with different growth rates. It is implemented as two "chip" instances of a single Nova submodel with growth-rate inputs.

Lotka

Lotka-Volterra predator-prey model.

Diffusion

In this version of diffusion, temperature flows between neighboring cells at a rate proportional to the difference in temperature between the cells.

Life

This model implements Conway's Game of Life on a bounded 50x50 board. The initial configuration can be altered by left-clicking on a cell to change its color and dragging that cell to change the colors of neighboring cells.

Fire Spread

Each cell in this model is one of three states: burnable-green (0), burning-red (1) or burned-black (2). If a burnable cell has a burning neighbor, it will become burning in the next generation with the probability given by the slider's value. Each burning cell burns for exactly 1 generation, and then changes to burned. Each cell can also be part of a firewall (state 3). Firewalls (blue) remain in a fixed state and can prevent fires from spreading. The initial configuration can be altered by left-clicking on a cell to change its color and dragging that cell to change the colors of neighboring cells.

Agent Motion Model

This model contains agents moving over a toroidal landscape. Agents may die and disappear, and new agents are born and added to the landscape, with the same probability (0.2).

Agent/Cell Motion Model

This model demonstrates the interaction between agents and cells. Agents move across a toroidal landscape As them move they cause the color of the cells over which they run to gradually change from green to white.

Agent SIR

In this agent-based world individuals are either infected (red) or susceptible (green). Agents move one step at at time in a random walk except that they cannot traverse the blue barriers. Any susceptible agent within 1 unit of an infected agent becomes infected. Agents randomly recover based on the Recovery Probability set by the user.

Agent SIR II

A second, "cuter" version of Agent SIR.

Measles or Flu SIR model

Another agent-based version of SIR, with controls for different properties. Comparison graphing is also supported.

Network Diffusion (Grid Version)

This model demonstrates diffusion of a quantity through a directed network. The quantity moves among nodes in the network only along established, directed links between two nodes. The simple rules that drive this diffusion lead to intertesting patterns related to the topology, density and stability of the network (derived from Stonedahl & Wilensky, 2008). This version places the nodes in a grid and limits connections to nearest neighbors only.

Network Diffusion (General Version)

Same as the previous model, only without the grid limitation. Uses d3.force-directed graph layout (Bostok, http://github.com/mbostok/d3/wiki) to display the network.

Network SIR Model

The model demonstrates the spread of a virus through a network. Although the model is somewhat abstract, one interpretation is that each node represents a computer, and we are modeling the progress of a computer virus (or worm) through this network. Each node may be in one of three states: susceptible, infected, or resistant. In the academic literature such a model is sometimes referred to as an SIR model for epidemics. (derived from Stonedahl & Wilensky, 2008).

Localized Network SIR Model

In this version of the previous model a network of 400 nodes is divided into 4 "localities" of 100 nodes each. Within each locality the nodes form a grid with as many as 8 immediate neighbors. Local connections are only possible within immediate neighborhoods. In addition, a small number of "non-local" connections join nodes in different localities

Susceptible/Infected/Recovered Simulation Games

set of three "circle" games which demonstrate the spreading of infection and recovery from that infection.