Mountain Wild Horse Range (Coughenour, 1999) and to a variety of other large herbivore ecosystems in the western United States, East Africa, and elsewhere.

To be useful in informing and assessing AMLs, key capabilities of such an ecosystem model would include

  • Prediction of plant-biomass dynamics and production responses to climatic variations and soils. Dynamics must be represented at least seasonally and ideally on a weekly or even daily basis. Dry, wet, and average years should be realistically simulated. Seasonal dynamics are important because forage biomass varies greatly owing to intraseasonal and interseasonal precipitation patterns and herbivore offtake on different parts of the landscape at different times throughout the year.
  • Realistic simulation of plant-production responses to herbivory, including under-compensatory and overcompensatory responses.
  • Simulation of changes in vegetation cover over multiyear periods.
  • Differentiation of simulated plants into functional groups, including preferred and nonpreferred species for herbivores.
  • Representation of spatially variable patterns of precipitation and temperature and their effects on vegetation. Spatial patterns of precipitation can be thought of as dynamic precipitation maps in the model.
  • Simulation of dynamic snowpack distributions across the landscape because these affect forage availability and herbivore distributions.
  • Simulation of dynamic herbivore habitat selection and resulting spatial distributions in response to water, forage, topography, cover, and barriers.
  • Simulation of herbivore forage intake and resulting effects on herbivore body condition.
  • Representation of key nutrient cycles, particularly nitrogen and soil-carbon dynamics.
  • Representation of key hydrological responses, particularly runoff and infiltration responses to changes in vegetation cover, which may result from herbivory.
  • Simulation of interactions with other species via competition for forage, water, and habitat and effects on other species resulting from equid-induced habitat alterations. Ecosystem modeling can represent forage competition, and effects on habitats could be represented by linkages to habitat models for other species.

With those modeling capabilities, it would be possible to predict the effects of different horse or burro densities and distributions on ecosystem dynamics and to assess whether horse or burro densities are sustainable in the long term. It would also be possible to infer or directly represent interactions with other species, including wildlife and livestock. Competition between livestock, wildlife, and horses or burros is affected by the degree of overlap in species forage preferences and spatial distributions. Modeling could also be used to assess the effects of restrictions on horse or burro movements that arise from fencing and other land uses. Such habitat fragmentation results in reduced opportunities for herbivores to access key grazing areas in times of food shortages on primary ranges. Restrictions of movement can also result in higher herbivore densities and grazing pressures than would occur if the animals could disperse or migrate. Vegetation or ecosystem models must be verified through comparisons with monitoring data described above. It is recognized that no single model is completely accurate; however, iterative adjustment of a model on the basis of data will improve it and make it more useful.

The National Academies of Sciences, Engineering, and Medicine
500 Fifth St. N.W. | Washington, D.C. 20001

Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement