1. General Model Information

Name: Savanna - Landscape and Regional Ecosystem Model

Acronym: SAVANNA

Main medium: terrestrial
Main subject: biogeochemistry, agriculture
Organization level: ecosystem
Type of model: difference equations (2D)
Main application: research
Keywords: vegetation, grassland, shrubland, savanna, soil water, arid system, pastoralism, forage, primary production, herbivore, spatially explicit, process-based


Coughenour, Michael B.
Swift, David M.
Galvin, Kathleen A.
Ellis, James E.

Natural Resource Ecology Laboratory
Natural and Environmental Sciences Building,
Colorado State University
Fort Collins, Colorado 80523-1499


Coughenour, Michael B.
Swift, David M.
Galvin, Kathleen A.
Ellis, James E.


Savanna is a spatially explicit, process-oriented model of grassland, shrubland, savanna and forested ecosystems. The Savanna model simulates processes at landscape through regional spatial scales over annual to decadal time scales. The model is composed of hydrologic, plant biomass production, plant population dynamics, ungulate herbivory, ungulate spatial distribution, ungulate energy balance, ungulate population dynamics and wolf predation submodels . Since Savanna is process-oriented rather than empirical or rule-based, it aims toward realistic, general and explanatory representations of ecological change as opposed to descriptions of ecological states or prescribed responses.

Every simulation model has a certain level of tradeoff between mechanistic detail and model simplicity. While highly aggregated or simplified models are easier to use, they are less realistic, less generalizable, and less explanatory. On the other hand, highly mechanistic models are more difficult to implement and the marginal costs of added complexity are high. Highly resolved models are more computationally demanding, which may prohibit their implementation at large spatial or long temporal scales. Therefore, Savanna treats ecological processes at an intermediate level of resolution. The time step of the Savanna model is a week, which allows simulations over longer time scales and larger spatial scales. This time step allows Savanna to simulate landscapes composed of 100-1000 grid cells over 5-50 year time spans in a reasonable amount of time on the current generation of microprocessors.
Source: Coughenour M.B. (1994). Savanna - Landscape and Regional Ecosystem Model (Model Description)

Overall structure of Savanna, a spatially explicit, dynamic ecosystem model that was originally developed for studies of African pastoralism. It has also been applied to western U.S. and Canadian national parks as an ecosystem management model. Ungulates and their interactions with plants are a principal concern. Savanna is unique in representing linkages between plant production and population processes as well as between ungulate energy balance and their population processes. Primary production is tied directly to the soil water budget. The model uses geographic information systems for both data input and output.

II. Technical Information

II.1 Executables:

Operating System(s): Operating Systems: DOS

II.2 Source-code:

Programming Language(s):

II.3 Manuals:

II.4 Data:

III. Mathematical Information

III.1 Mathematics

Model equations

III.2 Quantities

III.2.1 Input

III.2.2 Output

IV. References

Coughenour, M. B. 1991. Dwarf shrub and graminoid responses to clipping, nitrogen, and water: simplified simulations of biomass and nitrogen dynamics.
Ecological Modelling 54:81-110.

Coughenour, M. B. 1992. Spatial modeling and landscape characterization of an African pastoral ecosystem: a prototype model and its potential use for monitoring drought.
pp. 787-810 In: D.H. McKenzie , D.E. Hyatt and V.J. McDonald (eds.). Ecological Indicators, Vol. I. Elsevier Applied Science, London and New York.

Ellis, J.E., J.A. Weins, D.F. Rodell and J.C. Anway. 1976. A conceptual model of diet selection as an ecosystem process.
J.. Theor. Biol. 60:93-108.

Jenkins, K.T., P.J. Happe, and R.G. Wright. 1990.Evaluating above-snow browse availability using nonlinear regression.
Wildl. Soc. Bull. 18:49-55.

Hobbs, N.T. 1989. Linking energy balance to survival in mule deer: development and test of a simulation model.
Wildl. Monogr. No. 101.

Minson, J. 1981. Nutritional differences between tropical and temperate pastures.
pp. 143-157 In: F,H.W. Morley (ed.), Grazing Animals, World Animal Science. B.1 (A. Neimann-Sorensen and D.E. Tribe (eds.). Elsevier, Amsterdam.

Parton, W.J., J.M.O. Scurlock, D.S. Ojima, T.G. Gilmanov, R.J. Scholes, D.S.Schimel, T. Kirchner, J-C. Menaut, T. Seastedt, E. Garcia Moya, Apinan Kamnalrut and J.I. Kinyamario. 1993. Observations and modeling of biomass and soil organic matter dynamics for the grassland biome worldwide.
Global Biogeochem. Cycles 7:875-809.

Priestly, C.H.B. and R.J. Taylor. 1972. On the assessment of surface heat flux and evaporation using large-scale parameters.
Mon. Weather Rev. 100:81-92.

Ritchie, J.T. 1972. A model for predicting evaporation from a row crop with incomplete cover.
Water Resources Res. 8:1204-1213.

Shugart, H.H. 1984. A theory of forest dynamics.: the ecological implications of forest succession models.
Springer-Verlag, New York. 278pp.

Wight, J.R. and J.W. Skiles (eds.) 1987. SPUR: Simulation of production and utilization of rangelands. Documentation and user guide.
U.S. Dept. of Agric., Agric. Res. Serv., ARS 63. 372pp.

V. Further information in the World-Wide-Web

VI. Additional remarks

Last review of this document by: T. Gabele: Jan 27 1998
Status of the document:
last modified by Tobias Gabele Wed Aug 21 21:44:49 CEST 2002

Go back to Register of Ecological Models (R E M)