Workshop Proceedings Evaluating the Effectiveness of Forest Management Guidelines for Conserving Biodiversity and Ecological Processes

Rob Rempel
Research Scientist, Spatial Ecology
Ontario Ministry of Natural Resources
PDF (580KB)

Science - a way of knowing

Science is a body of knowledge based on a structured approach to learning that uses exact observation, experimentation (manipulative or mensurative) and logic to test hypotheses of cause and effect. The Scientific Method utilizes both induction and deductive logic, and requires experimentation to isolate confounding effects, but is incapable of proving hypotheses to be absolutely true.

The scientific method never proclaims that a hypothesis is true, rather it attempts to determine if there is necessary and sufficient evidence to reject a hypothesis as true. There are logical weaknesses with the scientific method, so rejection of a hypothesis is never unequivocal. There is no distinction between applied science or pure science, no distinction between management level science or laboratory level science. The only distinction that counts is well-designed scientific studies versus poorly designed studies.

Finally, science-based management emphasizes the scientific method to support development and evaluation of management options, and is better known as adaptive management. This presentation examines the application of the scientific method to evaluating the effectiveness of Ontario's forest management guidelines for the goal of conserving biodiversity and ecological processes.

Decision Analysis and Adaptive Management

Following principles developed in part by Nudds et al. (2003), twelve key steps have been identified that were followed, or will be followed, for development of the new forest management guides for landscapes. Science plays a key role at many points in this process, but some people stop at about Step 4, and consider that they have achieved a science-based approach.

1. Involve as many parties as possible.

2. Specify management objectives and options.

3. Identify the critical uncertainties, as hypotheses.

4. Critical, rigorous examination of evidence for alternative hypotheses.

5. Develop models to forecast outcomes, given different hypotheses.

6. Evaluate and rank competing hypotheses by likelihood in light of uncertainty.

7. Evaluate alternative management options.

8. Select options for inclusion in the guide.

9. Identify the highest uncertainties to further evaluate through effectiveness monitoring.

10. Design and implement a hypothesis-based monitoring program to evaluate effectiveness of policy options, according to sound principles of experimental design.

11. Monitor key responses.

12. Update ranking of competing hypotheses by likelihood given monitoring results. Section 7 of the landscape guide outlines the planned guide revision process.[1]

Linking Policy to Hypothesis

The Forest Management Planning Manual shall provide for determinations of the sustainability of Crown forests in a manner consistent with the following principles:

1. Large, healthy, diverse and productive Crown forests and their associated ecological processes and biological diversity should be conserved.

2. The long term health and vigour of Crown forests should be provided for by using forest practices that, within the limits of silvicultural requirements, emulate natural disturbances and landscape patterns while minimizing adverse effects on plant life, animal life, water, soil, air and social and economic values, including recreational values and heritage values (Crown Forest Sustainability Act 1994).

While not using the term "integrity" specifically, the Crown Forest Sustainability Act (CFSA) requires that "associated ecological processes and biological diversity should be conserved", and thereby implicitly establishes the goal of maintaining ecological integrity. Therefore, maintaining ecological integrity becomes the key direction to meet the intent of the CFSA. There is an emphasis on pattern and process, meaning that we need to evaluate both pattern and process.

Policy as Hypothesis

In this case, the policy is that emulation of natural disturbances and landscape patterns will form the basis of the "coarse filter" approach to maintaining ecological integrity. The policy hypothesis is that emulating natural disturbances and landscape patterns will result in the conservation of biodiversity and ecological processes.

Effectiveness Monitoring

Effectiveness monitoring is the evaluation phase of an adaptive management approach to resource management. The focus is on Outputs, rather than Inputs. Effectiveness monitoring is science-based and hypothesis driven. It attempts to answer if given a specific direction, was that direction effective in meeting its stated objective?

Predictions Arising from the Emulation of Natural Disturbance Hypothesis

A key set of predictions (Figure 3) arising from the forest management guides is that across spatial scales and over time, the following will be similar between habitats that have arisen from natural disturbance versus habitats that have arisen through application of the forest management guides:

Community Structure: The community composition and diversity of focal species (taxonomic and functional composition).

Population Trends: Trends in abundance for selected focal species (at the stand and large landscape areas).

Ecological Processes: The biotic and abiotic indicators of ecological processes (primary production, nutrient cycling, energy flow, and hydrological cycles).

Figure 3. Predicted relationships between pattern, composition, and structure and forest management disturbances versus natural disturbances.

When Does "Minimize Adverse Effects" Take Precedence?

There are two exceptions to the natural disturbance approach: 1) Where no natural analog exists (e.g. soil rutting or specific mercury levels); and 2) Where natural disturbance approach might result in an effect that society does not condone (e.g., violates Migratory Bird Act or Fisheries Act) or does not meet its goals (e.g., conservation of Species at Risk or species of economic or recreational value). This results in a different set of hypotheses than those arising from the "emulation of natural disturbance" paradigm.

An example of hypothesis-based monitoring: Songbird Community Structure

Policy Hypothesis: Songbird community structure will be similar in areas managed under the new forest management guides and in areas subject to natural disturbance processes (natural reference areas).

Identifying a Focal Songbird Community for Evaluating Policy Hypotheses

The correlation of the songbird community data with the explanatory forest variables reveals relatively clear associations between groups of species and characteristics of the forest. As seen in Figure 4, older stands and conifer are associated with BRCR and EVGR, higher levels of young forest are associated with COYE, and higher levels of age-class interspersion are associated with CSWA and VEER.

Figure 4. The correlation of the songbird community data with the explanatory forest variables.

Apply the models to make spatial predictions through time (scenario analysis)

Spatial maps produced by the spatial harvest program at different periods in time were run under different option sets, brought into a spatial analysis program, and then applied to the habitat models. This resulted in predictions of habitat occupancy. Depending on habitat element, it is either assessed at a 50 or 5000 ha scale.

For spatial habitat models, we use input data assessed at different spatial scales, apply these models called resource selection functions, and from this make predictions of expected patterns of habitat use.

Two-level approach to monitoring

The two-level approach to monitoring includes: 1) Broad-scale monitoring that allows us to assess cumulative effects of forest management and natural processes (what changes are happening); and 2) Mensurative experimental monitoring that allows us to assess whether forest management is causing changes (why changes are happening).

Formalized Reporting

Adaptive management often fails in that feedback loops are not identified and formalized.

The Declaration Order Conditions made under the Environmental Assessment Act (32 [b] parts xiii and xiv; 33 [b] part iii) (1990) require summaries of the monitoring projects described in this document to be included in the Provincial Annual Report on Forest Management and the State of the Forest Report, respectively. Results of these effectiveness monitoring studies will also be used during guide reviews to help assess the need for revisions (Condition 38[d]).

References

Crown Forest Sustainability Act (1994). Ontario Ministry of Natural Resources. Accessed on August 23, 2009. URL: http://www.e-laws.gov.on.ca/html/statutes/english/elaws_statutes_94c25_e.htm

Environmental Assessment Act (1990). Ontario Ministry of the Environment. Accessed on August 23, 2009. URL: http://www.e-laws.gov.on.ca/html/statutes/english/elaws_statutes_90e18_e.htm