Model Development

Pinyon-Juniper Woodland Model

Fig. 1 - Mapped risk of displacement
by pinyon-juniper

Variables selected for the pinyon-juniper risk model included landcover, (i.e., sagebrush taxon), elevation, potential for seed dispersal (based on proximity of pinyon-juniper woodlands to sagebrush), precipitation, and landform. Variables were parameterized differently for groups of ecological provinces within the Great Basin. Evaluation of the landcover map indicated that, within the study area, the spatial representation of pinyon-juniper woodlands was most accurate in three provinces in the eastern Great Basin. Consequently, our application of the risk model was restricted to those provinces (Figure 1).

Risk classes were defined as follows:

• Low—the probability that pinyon-juniper woodlands will displace existing sagebrush cover types within 30 years is minimal; currently, little or no pinyon-juniper is likely to be present in the overstory of these sagebrush stands.
• Moderate—the probability of displacement within 30 years is likely, but less so than for sagebrush at high risk; currently, pinyon-juniper is likely to be a minor to common component of the overstory of these stands. This class represents a transition phase in the conversion of sagebrush cover types to pinyon-juniper woodlands (Miller and others 1999a).
• High—the probability of displacement within 30 years is high; currently, pinyon-juniper is likely to be a common to dominant component of the overstory of these stands.

Cheatgrass Model

Fig. 2 - Mapped risk of displacement
by cheatgrass

Variables selected for the cheatgrass model included aspect, slope, elevation, and landform by ecological province. Recently published research based on remotely sensed data from the Great Basin supports the use of several of these variables (Bradley and Mustard 2006, Peterson 2006). The model was applied to all susceptible landcover types within the Great Basin ecoregion (Figure 2). Nonsusceptible landcover types, (e.g., marsh/wetland) were assigned to the none-risk level. The remaining risk classes were defined as follows:

• Low—the probability that cheatgrass will displace existing sagebrush or other susceptible cover types within 30 years is minimal; currently, native plants are likely to dominate the understory of these stands.
• Moderate—the probability of displacement within 30 years is moderate, but lower than for types at high risk; currently, either cheatgrass or native plants can dominate the understory.
• High—the probability of displacement within 30 years is very likely; currently, cheatgrass is likely to dominate the understory.