A model solution for good conservation

Making models indispensable in conservation decision-making

Conservation managers often have to make decisions in uncertain and complex situations. This uncertainty can be paralysing: “Do I choose option A or option B? Both have so much uncertainty around them that I just can’t decide!”

One way of dealing with this uncertainty is by modelling the different choices on offer to see what type of results they might yield. The correct use of the appropriate model not only helps in making robust, transparent and defensible conservation decisions, it often generates insights on the nature of the system being managed. Examples where models have helped in making conservation management decisions include: the re-introduction of populations of hihi in New Zealand (Armstrong et al., 2007); the re-introduction of gray wolves in Yellowstone National Park, USA (Varley & Boyce, 2006); and protecting Kemp’s Ridley sea turtle in the Gulf of Mexico from habitat degradation (Crowder & Heppell, 2011).

So there’s no question that, when used well, models can deliver good outcomes. However, despite their demonstrated benefits, models are often mis-used or not used at all to support conservation decisions. Instead, decisions are frequently based on intuition, personal experience or unaided expert opinion; and this can lead to biased decisions that rest on hidden assumptions and individual agendas. This in turn can lead to poor outcomes with little capacity to learn.

I recently led an investigation on why models are still not used in many conservation decisions (Addison et al., 2013). We searched the scientific and grey literature for evidence of views and attitudes towards the use of models by those who commonly participate in environmental decision-making. We then divide these objections up into three separate categories that related to the role of models in making a decision, modelling practice and model outputs. Common objections to the use of models in environmental decision-making are summarised in Table 1. (Also see ‘Five objections to using decision science in conservation’)

Table 1. Common objections to the use of models in decision-making

The role of models in decision-making
We don’t need models for decision-making, we have experts
Developing and using models in decision-making is too resource intensive
Modelling practice
Models do not represent my conceptual understanding of the decision context
Models focus on environmental considerations of the decision context, but fail to capture the social, economic and political factors which influence conservation management options
Models are either too complicated or too simple
There are insufficient data to do quantitative modelling
Inadequate data quantity/quality leads to inaccurate model predictions
Model outputs
I don’t understand the way scientists communicate
Model outputs are too uncertain for decision-making

Here are some quotes that represent these common objections:

  • “In most countries conservation is grossly under-funded, and for many organizations the cost of hardware, an expert operator, and the experimentation required may inhibit the use of reserve selection algorithms (even if the software itself is free)” (Prendergast et al., 1999)

“Despite their demonstrated benefits, models are often mis-used or not used at all to support conservation decisions. Instead, decisions are frequently based on intuition, personal experience or unaided expert opinion.”

  •  Models “remain so complex that they are seen as black boxes instead of transparent analytical tools” (De Smedt, 2010)
  • “Neither are they [water managers and researchers] sharing the same language for expressing results or the requirements of models” (Borowski & Hare, 2007)

How do you deal with common objections such as these? We gave this some thought and would like to suggest five practical solutions to help modellers improve the effectiveness and relevance of their work in conservation decision-making:

Solution 1: Dispel common misconceptions

Modellers need to appreciate that there are common objections to the use of models. Some of these can be misconceptions and should be addressed. Modellers can begin to anticipate common objections to models and prepare responses by using the list of objections in Table 1. A simple response to the objection that models diminish the autonomy of decision-makers could be: Models are tools for helping us think, they provide decision support and they do not replace decision-makers.

Solution 2: Guide good modelling practice: Structured Decision Making

Simply using a model alone to solve a conservation management decision is not enough. A framework is needed to guide good modelling practice, and we believe structured decision making (SDM) and adaptive management (AM) frameworks are well suited to do this. These frameworks also enable the participation of decision-makers, stakeholders and experts in the decision-making process (see related stories in this issue).

There are a variety of decision-analysis tools to aid rigorous, transparent and logical decision-making. In Figure 1 we list some techniques that are particularly useful in engaging participants and assisting with the decision-making process. These include:

  • Qualitative techniques such as values-focused thinking. These can help participants develop a shared understanding of the problem, clarify the decision context and develop management alternatives.
  • Cause-and-effect models: These help explore the consequences of management alternatives. As models can be confusing to participants without modelling experience, models that retain a visual conceptual form, such as Bayesian Networks, can be useful. These can help avoid participants’ perceptions that models are black boxes.
  • Simple visual management tools such as control charts. These can be used to interpret monitoring data. Such tools assist with the learning and review feedback loops of adaptive management.
Good decision making involves the appropriate use of the right model. Figuring out what’s appropriate is all about sharing, collaboration and communication. Pictured here are Parks Australia staff working with decision scientists with the aim of providing decision support for resource allocation on Christmas Island (see Decision Point #61).

Good decision making involves the appropriate use of the right model. Figuring out what’s appropriate is all about sharing, collaboration and communication. Pictured here are Parks Australia staff working with decision scientists with the aim of providing decision support for resource allocation on Christmas Island (see Decision Point #61).

Solution 3: Improve the social process

A SDM/AM framework on its own does not facilitate the social process of decision-making. There are a number of ways to improve the interactions and dynamics between participants. Many of these relate to genuinely including participants in the decision-making process and ensuring there is a balanced representation of participants. When genuinely engaged in the decision-making process, participants often feel a greater sense of ownership of decisions. However, participatory model building can become challenging when participants hold divergent views. There may be little that modellers can do to remedy such challenges through the model building process, although improving communication and building trust will help.

Solution 4: Improve communication

We have all been told that good communication is an essential skill to have, but very few modellers have had comprehensive training in this area. Two-way face-to-face communication, active listening, and demonstrating respect towards participants can create a more productive modelling process. However, we recognise that some modellers may lack the communication skills to competently engage participants. If communication is not a strength, we suggest modellers should seek specialist training in communication or employ a skilled workshop facilitator. Independent facilitators are particularly useful in high-stakes cases involving conflict, as it is easier for them to remain objective and not become emotionally involved in workshop discussions.

Solution 5: Build trust

Figure 1: A Structured Decision Making/Adaptive Management framework with modelling techniques that can assist the decision-making process.

Figure 1: A Structured Decision Making/Adaptive Management framework with modelling techniques that can assist the decision-making process.

Trust is an essential element of effective decision-making. To build trust, a modeller should demonstrate both their professional credibility and that of their modelling technique. Investing time in frequent personal contact, such as face-to-face meetings, telephone calls and emails, will help modellers foster interpersonal ties. An important aspect of building trust is that a modeller should not allow their values to influence the essential objectivity of their work. This can be very challenging as modellers who work closely with decision-makers are bound to form value judgments about the decisions in which they are involved.

Making models indispensable

If all five of the solutions outlined here were a regular part of a modellers’ practice, we believe most of the obstacles to models being used would be overcome. And the result of this would be more effective decision making and better conservation outcomes.

Of course, implementing some of the solutions will challenge many modellers as they require skills outside of their core training and experience. However, if the aim is to achieve better conservation outcomes, then it’s definitely worth considering.

We hope that our recommendations help broaden the use of models, forging deeper and more appropriate linkages between science and management. If we can make models indispensable in conservation decision-making, everyone is a winner.

More info: Prue Addison p.addison@student.unimelb.edu.au

References

Addison PFE, L Rumpff, SS Bau, JM Carey, YE Chee, FC Jarrad, MF McBride & MA Burgman (2013). Practical solutions for making models indispensable in conservation decision-making. Diversity and Distributions 19: 490–502.

Armstrong DP, I Castro & R Griffiths (2007). Using adaptive management to determine requirements of re-introduced populations: the case of the New Zealand hihi. Journal of Applied Ecology 44: 953–962.

Borowski I & M Hare (2007). Exploring the gap between water managers and researchers: difficulties of model-based tools to support practical water management. Water Resources Management 21: 1049–1074.

Crowder L & S Heppell (2011). The Decline and Rise of a Sea Turtle: How Kemp’s Ridleys Are Recovering in the Gulf of Mexico. Solutions 2: 67-73.

De Smedt P (2010). The use of impact assessment tools to support sustainable policy objectives in Europe. Ecology and Society 15: 30.

Prendergast JR, RM Quinn & JH Lawton (1999). The gaps between theory and practice in selecting nature reserves. Conservation Biology 13: 484–492.

Varley N & MS Boyce (2006). Adaptive management for reintroductions: updating a wolf recovery model for Yellowstone National Park. Ecological Modelling 193: 315–339.

Leave a Reply