New Features in RAMAS GIS 4.0

Note: If you are still using version 2 (1997), also see New Features in RAMAS GIS 3.0

• The maximum numbers are increased for stages (50), populations (500), and population management actions (1000).
• The manual is expanded, with a new section on using the model results in species conservation and management.

Spatial data and habitat dynamics:

• The program can model dynamic patch structures (temporally changing number and location of patches; e.g., a forest being fragmented by logging or roads). The program analyzes time series of habitat maps to identify and simulate splitting, merging, appearing and disappearing habitat patches.
• Dispersal can be modeled as a function of habitat characteristics, based on a "friction" map (requires Idrisi32).
• Probability and impact of catastrophes can be functions of habitat maps. For example, a "fire history" map can be used to calculate population-specific fire probability.
• The program can import maps in ERDAS and Idrisi32 formats. It can export habitat suitability and patch maps in Idrisi32 and ArcInfo raster (grid) formats.

Metapopulation model (RAMAS Metapop):

• Transparency: The program has several new features that are specifically designed to make it more transparent.
• New results: (1) The expected minimum abundance (minimum abundance during a trajectory, averaged over all trajectories). (2) The median, quartiles, and 90% confidence interval of the final metapopulation abundance. (3) Risk of low harvest (probability that the harvest will be below a range of thresholds).
• Modeling sex structure: The model can include separate stages for males and females. The mating system can be monogamous, polygynous, or polyandrous. For polygynous and polyandrous systems, the degree of polygamy is set with a "number of mates per individual" parameter.
• Density dependence: The user can select which stages to include as the basis of density dependence. For example, modeling a territorial bird species with Ceiling model may be better if "ceiling" is defined only in terms of adults.
• User-defined density dependence function allows modelers to put in any DD function they want.
• Genetics: Inbreeding depression can be modeled using the "user-defined density dependence function" option described above. An example function is provided with the program.
• Dispersal into a population (immigration) can be modeled as a function of the carrying capacity of that population. This can be used for species that can detect habitat quality while dispersing between patches.
• Spreading catastrophe: The model can include a catastrophe that spreads to other populations, with two options: (1) Carried by dispersers (specifying probability of infection per disperser), or (2) Spread by a vector (based on a probability-distance function).
• Catastrophes that affect dispersal rates can be included (e.g., large dispersal after a 10-yr flood in a river fish species or a riparian metapopulation).
• A catastrophe can have multiple effects, for example it can affect both vital rates and dispersal (emigration) rates.
• Correlated catastrophes: If both catastrophes are local (or both are regional), they can be independent (as in version 3), or positively or negatively correlated.
• Time-varying catastrophe probability: The program allows specifying catastrophe probability as a function of the time since the last catastrophe. This feature can be used to model, for instance, fire risk that increases with time (fuel accumulation) until a fire happens, when it drops to zero.
• Time-varying catastrophe effect: The program allows specifying catastrophe effects (local multiplier) as a function of the time since the last catastrophe. This feature can be used to model, for instance, to model fire impact that increases with time (hotter fire with more fuel accumulation) until a fire happens, when it drops to the normal level.
• A new parameter lets users specify the number of years since the last catastrophe at the start of the simulation.
• Conditional population management: The conditions for population management (harvest, translocation, introduction) include three new options that increase the flexibility of modeling strategies based on the current abundance (for example harvest rate as a function of total abundance).
• Observation error: The program allows the user to specify an observation error. When a population management (such as Harvest) is based on a specified proportion of existing individuals, the calculation of the number to be harvested is based on an "estimated" population size, which includes an observation error. Similarly, when population management actions are conditional (e.g., Harvest only if abundance>X) the "estimated" abundance includes an observation error. This allows realistic simulation of hunting, harvest and fishery management.
• Harvest from all stages: When population management is specified as a constant number (e.g., harvest X individuals every other year), a new option lets the specified number to be taken from each stage (as in the previous version) or from all stages, proportional to their abundances at that time step.
• The program allows negative correlation between survivals. The new option will be useful in many cases, e.g., to model seed bank in plants (proportion that remains dormant and proportion that germinates may be negatively correlated).
• The program provides more flexibility in modeling deterministic change in stage matrix elements. A new file structure for time series of the "relative survival" and "relative fecundity" parameters allows them to be stage-specific. Another option allows these time-series to be reset to beginning after a catastrophe that affects vital rates.
• A new "Constraints" matrix is used to specify proportion of each stage matrix element that is survival (as opposed to fecundity) and thus should add up to 1 for the column sums. This allows survivals to be constrained even if there is recruitment to more than one stage, e.g., in the case of plants or in models with sex structure (see above). (Version 3.0 assumed that fecundities are in the first row of the stage matrix).