Model Use
This section guides readers through a generic set of steps involved in initializing a landscape, describing a suite of natural and anthropogenic disturbances and exploring the environmental and economic characteristics of the future landscape.
Step 1. Identify stakeholders and establish ALCES committees
The first step towards conducting a cumulative effects assessment in a regional setting is to identify the various stakeholders whose interests need to be represented. Identification of who should be a stakeholder is not an easy task, but selection of stakeholders should consider a broad range of industrial, environmental, governmental, aboriginal, and public groups affected by landuse decisions within the study area. As a general rule, if you need the consent of a definable group to implement the findings of your regional assessment, you should probably have them on board as part of the formal stakeholder team from the onset of the initiative.
In most cases, several committees will need to be established to ensure that all aspects of the project (data assembly and coordination, indicator selection, scenario selection, presentations of results, and exploration of alternative scenarios) are properly attended to. Generally, a Technical Committee (aka Core Team) is struck that contains the person(s) responsible for the customization of the ALCES model, the actual running of ALCES, archiving of incoming data assumptions, and the preparation of initial geo-spatial data. The Technical Committee answers to a broader Steering Committee (Domain Experts) populated by individuals who are experts and represent each of the major landuses in the study area (forestry, agriculture, energy, transportation, human settlements, mining, aboriginal, recreation/tourism). Members of the Steering Committee are responsible for providing (often a delegating exercise) information on the trajectories, metrics, and relevant indicators of each landuse. The Steering Committee generally reports to the broad stakeholder group that contains all participants in the exercise.
A key individual in the Technical Committee (Core Team) is the project coordinator. This person has the important job of keeping the project "on the rails", ensuring completion of assigned tasks, and that all data are properly assembled, organized and archived. This person must have a solid understanding of the strategic and operational aspects of ALCES and be able to work closely with the Forem Technologies team if model customization is required. Minimally, the Project Coordinator must have a good working relationship with the Steering Committee and have well established skill sets in the following areas: interpersonal communication, data organization, attentive to operational detail yet able to discuss strategic level issues, familiarity with managing a time and money budget, and general familiarity with each of the major landuses occurring on the study area.
It is common for Steering Committees to establish an External Validation Committee whose purpose is to provide an independent assessment to the rigor and plausibility of the ALCES inputs and outputs.
Step 2. Define study area and acquire spatial data
Once the composition of the stakeholder community has been established, this group can begin the task of defining the geographic perimeter of their Study Area, for it will be within this geographical region that various combinations of natural disturbances and human landuse scenarios will be explored and simulated in ALCES. There exist several criteria (ecological, social, administrative, and political) that can be used to establish the basis for the boundary of the study area. It is the responsibility of the ALCES user to select which criterion (or combination of criteria) will be used to define the study area boundary. The ALCES user is encouraged to consider the appropriate spatial scale of the study area relative to the types of questions that are being asked by the stakeholder community. In general, large scale disturbance regimes (such as wildfire) or large landuse practices (i.e., agriculture, land-based forest allocations) require an exploration of larger regions in ALCES. Once completed, the study area boundaries will be provided to analysts of spatial datasets (GIS technicians), who will begin the tasks of describing the initial composition of the study area.
Once the GIS technician has received the boundary of the study area, the task of describing the size and current composition of the landbase can begin. It is important to consider which set of landuse practices will be examined through scenarios analyses, for it will be the footprints associated with these landuses that the GIS technicians must locate recent and accurate spatial information for.
The ALCES datasheet (ALCESINPUT) is built in Microsoft Excel. Once you have opened the file (ALCESINPUT), you will follow a set of instructions on summarizing initial landscape and landuse information and entering these values in the datasheet. Once this datasheet is completed, these data will be manually entered in the ALCES model.
The landscape in ALCES is broken into a nested hierarchy. At the crudest level, landscape and disturbance types are categorized into vegetated and non-vegetated types. Beneath this stratum lies the next break-down of the landscape into native and human-made (anthropogenic) features. Beneath this hierarchy lies landscape types that share a common structure (i.e., forests, wetlands, shrublands, water). ALCES considers various agricultural practices to be landscape types. Disturbance types are divided into general categories of vegetated or non-vegetated.
You can navigate to the landscape initialization panel by clicking on the “Initial & Future Landscape Composition & Metrics” button from the Home panel and then click on the Initial Landscape Composition Area button. If you are unfamiliar with the appearance of different landscape and disturbance types, you are encouraged to view the graphic library embedded in the ALCES help file found within the Stella help button found in the menu above.
In general, the completed ALCES datasheet will include information on the size and composition of the study area, the age class structure of the forest landbase, plant community dynamics relationships, the area, length, and spatial distribution of various human landuse footprints.
Step 3. Quantify meteorology and natural disturbances on landscape
The importance of the hydrological cycle and disturbance regimes (i.e., fire, insect outbreaks) to ecological function and landscape structure is immense. Collectively, these perturbation regimes help define the composition, function, and age class structure of the landscape.
Since precipitation, temperature, fire occurrence, and insect outbreaks can be highly variable in time, it is important to define its temporal variance to assist managers in exploring the consequences to landscape composition. For this reason, ALCES allows the user to explore climate and perturbation regimes as either constant or stochastic regimes. ALCES contains a meteorological / hydrological model requiring data input (both means and variances) for precipitation, sublimation, evapotranspiration, and surface runoff rates for each landscape type. These data can be obtained from summary analyses of meteorological stations found within the landscape.
The fire and insect regime module in ALCES allows the user to explore different rates of fire and insect outbreak on the vegetated landscape and to run perturbations as either a constant or stochastic regime. ALCES allows the user to explore relationships between forest landscape age and fire rate and to impose different levels of fire suppression on forest landbases. Rates of historic fire and suppression levels can be obtained from the literature.
ALCES allows users to explore relationships involving climate change on regional landscapes. If the user has access to information from regional climate models defining expected future change in temperature and precipitation, then this information can be entered in Panel #6.1. Values entered in the graphic input devices represent departures from longterm averages. If a climate change scenario is entered into ALCES, the following variables can be affected: -average annual precipitation -average annual temperature -plant productivity (native and agricultural) -probability (and severity) of fire events -carbon pool dynamics
Relationships between climate change and fire rate are described in Panel 6.
Step 4. Identify those human landuses occurring on your landscape
Landscapes vary considerably in terms of what landuses exist and at what level of intensity they have expressed themselves. For example, a landscape may have a fledgling transportation network (eg., portions of the northern boreal landscape) or may possess a mature network of roads (eg., the agricultural portion of the white zone).
To conduct realistic cumulative effects analyses, it is important that the ALCES user do a thorough job of identifying those landuse practices that exist on the landscape. Those landuses currently accommodated in ALCES are: - Forestry - Agriculture - Energy (Petrochemical) - Transportation - Humans and Their Settlements (including activities specific to First Nations Peoples) - Mining - Trapping - Recreational Activities (hunting, golfing, hiking, skiing)
One helpful strategy to identifying the occurrence of given landuses is to examine aerial photography of the study area for the tell-tale footprints of each landuse type.
The GIS analyst will be of great help in identifying current landuses, as the spatial dataset for the study area will contain features relevant to landuses (examples would be high grade roads, cutblocks, seismic lines, wellsite, human settlements, ski resorts, golf courses, etc.)
Step 5. Define future scenarios of landuse on your landscape
Although Step 3 (acquiring spatial data for your landscape study area) will inform you about the type, area, and spatial distribution of anthropogenic footprints on your current landscape, it will not tell you what the future development pathways of various landuses will be. The domain experts in the Steering Committee (or representatives of the industrial sectors) are the logical source to better understand future expectations of resource flow and landscape composition. For example, if a single large forestry company dominates your landscape, approach that company for information on expected harvest levels and the constraints that will determine the detailed forest management plan. Similarly, assembling information from an energy sector company can provide a strategic-level insight into the future footprints of the hydrocarbon industry. In situations where multiple companies occur on your study area, the ALCES user may be better served by approaching relevant government agencies (Alberta Energy, Sustainable Resource Management, Alberta Agriculture), regulatory boards (e.g., AEUB), or industry associations (e.g., CAPP, AFPA, ACC), as these groups have a better synoptic overview of the industry and likely future trajectories.
The type of questions to be asked vary, but in general, often include: *How much area (ha) and length (km) for a given disturbance type (eg., seismic line, road, cutblock) can be expected each year into the future? What level of inter-annual variation is expected around these temporal trends? *What is the likely spatial distribution of these footprints? i.e., In what landscape types will the disturbance types occur? Is the historic distribution of landuse footprints indicative of the future or will different spatial patterns likely occur? *What is the anticipated lifespan (in years) of each footprint? This will define how many years elapse between the creation of a particular industrial footprint and its subsequent disappearance if the feature is not a permanent one. It is important to determine if reclaimed features are to return to the landscapes from which they came or are they likely to be reclaimed to a different landscape type. *Does the industrial sector anticipate changing metrics to footprints through time? For example, a seismic line may be 5 meters in width today, but could become more narrow in Decade 2 as new technologies arise.
The user is encouraged to closely inspect the ALCES dataset and each landuse panel for specific questions intended for each industrial sector. Future development trajectories of industrial footprints for each landuse type can be entered as annual averages for each decade interval or can be entered as variable annual values.
Step 6. Define wildlife habitat and community richness
ALCES currently simulates habitat availability, quality, and effectiveness, and population dynamics for 2 selected species of wildlife. To calculate habitat availability, ALCES requires the user to define what proportion of each landscape type is available to each wildlife species and to define the width and use response of buffers surrounding such features as roads, seismic lines, wellsites, pipelines, and cutblocks. The proportional use of wildlife species within each buffer can also be defined.
Wildlife habitat quality is calculated by ALCES based on the following:
- Identification & weighting of various wildlife habitat elements tracked by ALCES. Attributes currently tracked include density of natural edge, density of anthropogenic edge, forest age class structure, herb phytomass, shrub phytomass, snag phytomass, litter phytomass, snowpack depth, and stream continuity
- The availability of weighted habitat elements on each landscape type occupied by each wildlife species
- The response curve defining the relationship between the relative availability of a wildlife habitat element and habitat quality (scale of 0 to 1)
Step 7. Identify landscape, economic and environmental targets
Most landscapes contain innumerable ecological and landuse attributes. For this reason, it is important that stakeholders narrow the range of ecological and industrial values into a manageable number. Careful consideration to the selection of ecological variables is required, for some elements offer special insight into ecological function and integrity.
ALCES currently contains four panels that focus on Targets and Thresholds. One is these deals with landscape composition of the boreal forest. A second one deals with flows of landuse and natural resource flows. A third has been customized for the prairie and agricultural landscape. The fourth focuses on issues relevant to mountain landscapes where protection and recreational landuses dominate.
Within each Threshold panel is a brown input table where stakeholders can define their chosen target and threshold values for each threshold attribute. A target value is that which stakeholders want a resource or landscape value to achieve. A threshold value is that level that a resource or landscape resource is not to exceed.
Once target and threshold values have been identified, and a suite of landuse and natural disturbance regimes (eg., fire) have been turned off or on in the "Landuse Switches" panel, the user can then simulate the future landscape and determine if threshold or target variables have been achieved or violated. These panels provide information on the status of each variable, the amount it has changed since the beginning of the simulation, and whether it satisfies or violates the defined threshold levels.
Users can also specify maximum density levels for such features as seismic lines, pipelines, wellsite and roads in Panel 8.14. These values can be stipulated for each landscape type. If these values are exceeded, the simulation will stop and the value that has been exceeded will be indicated.
Step 8. Conduct sensitivity analyses on landscape objectives
An important challenge to resource managers conducting cumulative effects assessments is to determine the relative effects of different disturbance regimes and landuse practices on desired landscape objectives such as timber flow, wildlife habitat, hydrocarbon extraction, landscape fragmentation, and agricultural productivity. It is also important for the ALCES team to identify those model assumptions that cause the greatest effect on desired output variables, for it is these "driver" variables that need to be reasonably accurate to ensure simulation results that instill confidence. In a modelling world of limited time and funds, it is most wise to invest additional effort at refining estimates of "driving" variables, and not those that have minimal effect on chosen indicator variables. To facilitate fast identification of the major "drivers" affecting landscape composition and flows of natural resources, ALCES provides the user with a series of analytical levers to pull.
#1. Use the On/Off buttons in the Landuse Switches panel to selectively turn on and off each of the landuse practices and disturbance regimes. First, identify the graph or table or threshold panel in ALCES that best illustrates the variable(s) of interest, then proceed by progressively subjecting the landscape to a greater number of disturbance regimes and landuses. It is often highly illustrative to begin with no landuse or disturbance regimes. As you progressively turn on more landuses, you will gain an appreciation for those which have a major influence, those which have a minor influence, and those which have no influence. You are advised to try different combinations of landuses operating during a simulation run, as the results can lead to surprising non-linear results.
#2. Once you have determined which landuses and disturbance regimes are causing a dominant effect on a chosen landscape objective (ie., timber flow or wildlife habitat) by turning them off and on, you can now begin to explore the effects of more subtle changes to the behavior of a given landuse practice. For example, if a logging operation is affecting wildlife habitat in an adverse or beneficial manner, is it because of the type of forest trajectory harvested, the minimum volume requirement, the lower age of eligible seral stages, the harvest and regeneration strategy, or the cutblock size frequency? By systematically modifying these input variables prior to simulation runs, the user quickly begins to appreciate those elements of landuse practice whose modification can best affect a desired future condition.
Step 8, continued
#3. ALCES also provides an automated sensitivity analyses mode. It can be accessed through the "Sensi Spec" button under the "Run" button in the top menu. In this panel, the user can select a single or group of variables and request of ALCES that a number of consecutive simulations be run, each with a different level of a "driving" variable. An example would be exploring the effects of 5 different levels of fire (0, 0.001, 0.005, 0.01, 0.05) on the age class structure of mixedwood forests. This batch run feature allows for rapid exploration of the mitigation value of different landuse or management strategies. It is worth noting that variables that are expressed as graphic input devices can not be used directly with sensi specs. To overcome this limitation, ALCES has incorporated a series of "FT Development Trajectory" variables that can be modified using sensitivity analyses. The default values for the FT Development Trajectories is 1 (=100% of the values provided by the graphic input device), but these variables can be modified between 0 (=0% of graphic input value) and 2 (=200% of graphic input value).
#4. Remember that there are both positive and negative drivers that effect indicator variables. While it is instinctive to seek out negative drivers that cause problems, it is important that ALCES Users also identify positive drivers as increasing the level of these levers can assist in finding solutions.
#5. When exploring relationships between management levers and indicators, the response surfaces may not express themselves until an appropriate amount of time has expired. A good example may be the sustainable harvest of wood fiber, whereby a 50 year simulation may indicate a non-declining harvest of wood fiber, a simulation conducted over a 200 year period may indicate that a given harvest level is non-sustainable.
Step 9. Select disturbance regimes and landuses for simulation in ALCES
An effective strategy for exploring the relative importance of various industrial practices and disturbance regimes to various stakeholder targets of resource flows and landscape composition, is to systematically turn them off and on while running simulation scenarios. This approach reveals to the ALCES user which processes are “driving” the landscape, which are “followers”, and helps the stakeholder group properly identify which landuses assist in the attainment of resource flow targets and which jeopardize their attainment.
ALCES provides a simple system for allowing the user to switch on or off each of the landuse practices or disturbance regimes. At the upper right corner of each panel is a button called “Landuse Switches”. Clicking on this button navigates the user to a panel containing a suite of switches. The switches can be either up (on) or down (off). To provide users with a quick method of turning off all landuses at once, a master switch is provided. If clicked off, all landuses below are considered off, if clicked on, only those landuses below switched up are considered activated.
ALCES users are encouraged to begin their landscape analyses by exploring the effects of natural disturbance regimes (fire, insect outbreaks, flooding) in isolation to human landuses. This will provide the user with an appreciation of the natural dynamics of the landscape system. Once a comfort level has been achieved about how natural dynamics affect landscape composition, structure, and age class composition, the user is encouraged to begin introducing one landuse practice at a time. Explore how each landuse affects landscape objectives, targets and thresholds and a suite of landscape metrics such as age class structure, patch size, and gross composition.
Step 10. Identify landscape and resource issues. Seek mitigations.
As the ALCES user conducts a series of simulation scenarios, systematically turning off and on various disturbance regimes and landuse practices, the character of the future landscape will emerge. With each simulation run, the user should evaluate the performance of the scenario against the landscape and resource flow targets and thresholds identified by the stakeholder process. Where violations occur, work hard to identify the causal agents. For example, if density of linear corridors exceeds a threshold value, is the problem largely with seismic lines, pipelines, wellsite access roads, cutblock roads, or general transportation infrastructure. Examine different management scenarios (for example, narrower seismic lines that revegetate more quickly) and explore the ecological and economic implications of alternative management models.
ALCES contains several thousand outputs (either table or figure variables) and they have been arranged into convenient tables to allows users to extract information and evaluate relationships quickly. The major output panels in ALCES include:
Summary of Landscape Types (a hierarchical view of the transforming landbase)
Summary of Anthropogenic Features (selected areas and edges associated with human disturbance types)
Selected Landscape Features (profiling changes to landscape age, size, edge and recovery)
Once the major issues, and their causes, have been identified, the stakeholder group has some tangible basis for exploring alternative landuse practices. Don’t be surprised if there exists numerous ready solutions, ones that are attractive from both an ecological and economic standpoint.
On the contrary side, there exists basic limits to all resource flows and to landscape composition. In some cases, the stakeholder community may be able to suggest combinations of landuse practices that significantly reduce, but not remove, risk to resource flow targets. In these cases, the stakeholder community must consider trade-offs in the goals of different sectors of society.
Step 11. External model validation
Once the User has completed Steps 1 through 10 in ALCES, the overseeing committee may wish to have the regional assessment results critiqued by an external validation team. As long as the evaluator selection criteria, terms of reference, timelines, and costs are reasonable, the benefits of an external validation initiative are many, and include:
- an independent examination on the "reasonableness" of the initial landscape composition and future landuse trajectories
- commentary on the appropriateness of the environmental, social, and economic indicators selected
- a "sniff test" consideration of whether the indicators are responding to a transforming landscape in an expected fashion.
It is essential that the committee who selects the reviewers appreciate the importance of selecting individuals who have a strong strategic-level appreciation of the dynamics of the study area relative to the questions at hand. It is likely that several evaluators would be approached to ensure that the major social, economic, and environmental engines are addressed. The committee is really looking for individuals who have the ability to run a simple ALCES-like simulation in their head, so that they can compare their intuitive results to those simulated empirically in ALCES.
In the event that some elements of the regional assessment raise an alarm, the reviewer should be asked to comment on the following:
1. Is the initial landscape composition reasonable?
2. Are there any major operational landuses not being considered in the assessment?
3. Are the growth or loss trajectories of each landuse reasonable?
4. Are the defined response surfaces between indicators and landuse elements appropriate?
5. Are the treatment of natural disturbance regimes (fire, insect outbreaks, meteorology) logical?
6. Is the simulation length (number of years) appropriate for the various landuses and indicators being considered?
The commentary provided to questions 1 to 5 should prove helpful to the ALCES team as they consider whether a poor assumption has entered the modelling initiative, or whether they need to provide additional information to the assessor. If a pattern is confusing to an independent professional assessor, it is also likely to be confusing to the general public.