18.2 Models of Ecosystem Structure
was developed for the coarse-scale assessment, but
also was used for evaluating management alternatives as part of the mid-scale effort.
The basis for vegetation succession in CRBSUM
is a conceptual model of post-fire succession along
multiple pathways that converge to a 'climax'
community in the absence of disturbance (Kessell
and Fischer. 1981). This approach was expanded
for CRBSUM to include disturbances other than
fire (such as insect and disease attack. grazing, thinning, and cutting) and to incorporate nonforest vegetation types. The endpoint of a particular successional pathway is a potential vegetation type (PVT)
with a corresponding biophysical setting. Fifty-five
PVTs were defined for CRBSUM runs using expert workshops. For each PVT, a set of successional
classes, each described by a structural stage and
cover type. are linked along pathways converging
to the PVT. Rule-based functions determine
changes in successional class as a result of the occurrence of disturbances and successional development in the absence of disturbance. Disturbances
are implemented through management scenarios
comprised of sets of probabilities that a management action or disturbance will occur on a cell in
any given year. Probabilities are specified by geographic region, PVT, and successional class.
Maps of management regions, PVT, cover type,
initial structural stage, and initial succession age
are needed to run CRBSUM. The latter two maps
can be input to the model or can be generated stochastically by the model. A limitation of CRBSUM
is the simulation of disturbances as independent
events in each cell. which precludes propagation
of fire or insect attack across large areas or synchronization of disturbance occurrences. As a consequence, maps of disturbances generated by
CRBSUM are applicable only for projection of
general geographic trends. rather than for describing specific disturbance locations or year to year
variability in disturbance occurrence. Limitations
of the successional pathway approach include a deterministic specification of the endpoint of succession and lack of detail in defining some successional trajectories (Keane et al., 1996a). Many
parameters were required to implement succession
pathway models for each PVT and disturbance
probabilities. Parameters were quantified using
teams of experts; problems were noted with a lack
of consistency and level of expertise across vegetation types (Keane et aI., 1996a).
CRBSUM was used to predict coarse-scale vegetation changes for four management strategies
termed consumptive demand, historical, passive
management, and active management. Each man263
agement strategy was implemented by adjusting
disturbance probabilities to reflect long-term management goals. The objective of the simulations
was to contrast trends in landscape characteristics
(e.g., changes in cover type) among alternative land
management policies over a period of 100 years.
18.2.7 MAPSS
MAPSS is a biogeographic model constructed to
simulate potential vegetation type, vegetation leaf
area index (LAI), and site water balance and runoff
(Neilson, 1995). The model incorporates thermal
and site water balance constraints on vegetation.
Thermal constraints are simulated with simple
physiologically based rules. A process-based site
water balance model integrates vegetation leaf area
and stomatal conductance in canopy transpiration
and soil hydrology. Because the model simulates
potential climax vegetation, environmental constraints are implemented to determine vegetation
carrying capacity and potential type. Therefore, it
is assumed that in any given location vegetation
leaf area production will be maximized to just utilize available soil water.
MAPSS includes woody vegetation (trees and
shrubs) and grasses that compete for light and water. The model operates on a monthly time step.
Fire, the only disturbance included in the model, is
a rule-based function of available grass or shrub
fuel and high summer rainfall as an index of convective activity. Vegetation is classified by rules
incorporating thermal, humidity, and productivity
constraints. which translate modeled life form and
LAI into a vegetation type. The classification.
which includes biome physiognomy (trees, shrubs.
and grasses), leaf form. closed or open vegetation
based on LAI, and thermal zone (tundra, taigaboreal, boreal, temperate, subtropical, and tropical),
results in 35 possible vegetation types.
The model was run over the conterminous
United States and was found to simulate the distribution of forest, grassland, and desert areas with
reasonable accuracy, as well as reproducing
monthly runoff. The largest biogeographic error
was in the prediction of the occurrence of forest in
most of the prairie peninsula region of the central
United States. Simulation of episodic summer
droughts followed by fire may correct this problem.
MAPSS can be used across a broad range of spatial scales to predict changes in potential vegetation and water regimes under altered climates. An
advantage of the model is its incorporation of the
linkage between vegetation and water balance
was developed for the coarse-scale assessment, but
also was used for evaluating management alternatives as part of the mid-scale effort.
The basis for vegetation succession in CRBSUM
is a conceptual model of post-fire succession along
multiple pathways that converge to a 'climax'
community in the absence of disturbance (Kessell
and Fischer. 1981). This approach was expanded
for CRBSUM to include disturbances other than
fire (such as insect and disease attack. grazing, thinning, and cutting) and to incorporate nonforest vegetation types. The endpoint of a particular successional pathway is a potential vegetation type (PVT)
with a corresponding biophysical setting. Fifty-five
PVTs were defined for CRBSUM runs using expert workshops. For each PVT, a set of successional
classes, each described by a structural stage and
cover type. are linked along pathways converging
to the PVT. Rule-based functions determine
changes in successional class as a result of the occurrence of disturbances and successional development in the absence of disturbance. Disturbances
are implemented through management scenarios
comprised of sets of probabilities that a management action or disturbance will occur on a cell in
any given year. Probabilities are specified by geographic region, PVT, and successional class.
Maps of management regions, PVT, cover type,
initial structural stage, and initial succession age
are needed to run CRBSUM. The latter two maps
can be input to the model or can be generated stochastically by the model. A limitation of CRBSUM
is the simulation of disturbances as independent
events in each cell. which precludes propagation
of fire or insect attack across large areas or synchronization of disturbance occurrences. As a consequence, maps of disturbances generated by
CRBSUM are applicable only for projection of
general geographic trends. rather than for describing specific disturbance locations or year to year
variability in disturbance occurrence. Limitations
of the successional pathway approach include a deterministic specification of the endpoint of succession and lack of detail in defining some successional trajectories (Keane et al., 1996a). Many
parameters were required to implement succession
pathway models for each PVT and disturbance
probabilities. Parameters were quantified using
teams of experts; problems were noted with a lack
of consistency and level of expertise across vegetation types (Keane et aI., 1996a).
CRBSUM was used to predict coarse-scale vegetation changes for four management strategies
termed consumptive demand, historical, passive
management, and active management. Each man263
agement strategy was implemented by adjusting
disturbance probabilities to reflect long-term management goals. The objective of the simulations
was to contrast trends in landscape characteristics
(e.g., changes in cover type) among alternative land
management policies over a period of 100 years.
18.2.7 MAPSS
MAPSS is a biogeographic model constructed to
simulate potential vegetation type, vegetation leaf
area index (LAI), and site water balance and runoff
(Neilson, 1995). The model incorporates thermal
and site water balance constraints on vegetation.
Thermal constraints are simulated with simple
physiologically based rules. A process-based site
water balance model integrates vegetation leaf area
and stomatal conductance in canopy transpiration
and soil hydrology. Because the model simulates
potential climax vegetation, environmental constraints are implemented to determine vegetation
carrying capacity and potential type. Therefore, it
is assumed that in any given location vegetation
leaf area production will be maximized to just utilize available soil water.
MAPSS includes woody vegetation (trees and
shrubs) and grasses that compete for light and water. The model operates on a monthly time step.
Fire, the only disturbance included in the model, is
a rule-based function of available grass or shrub
fuel and high summer rainfall as an index of convective activity. Vegetation is classified by rules
incorporating thermal, humidity, and productivity
constraints. which translate modeled life form and
LAI into a vegetation type. The classification.
which includes biome physiognomy (trees, shrubs.
and grasses), leaf form. closed or open vegetation
based on LAI, and thermal zone (tundra, taigaboreal, boreal, temperate, subtropical, and tropical),
results in 35 possible vegetation types.
The model was run over the conterminous
United States and was found to simulate the distribution of forest, grassland, and desert areas with
reasonable accuracy, as well as reproducing
monthly runoff. The largest biogeographic error
was in the prediction of the occurrence of forest in
most of the prairie peninsula region of the central
United States. Simulation of episodic summer
droughts followed by fire may correct this problem.
MAPSS can be used across a broad range of spatial scales to predict changes in potential vegetation and water regimes under altered climates. An
advantage of the model is its incorporation of the
linkage between vegetation and water balance