Here a ‘future’ is what may happen to the Kimberly region to the year 2050. This depends on i) how the Kimberly system works, ii) what events may occur and iii) what we do and how we react to these events. We define these as follow:

We do not know precisely how the Kimberly system works, but we have a few working hypothesis. Each working hypothesis results in a model specification. It defines how the EwE and Alces models run and represents a snapshot of how we believe the system is and works now. Uncertainty is represented via alternative model specifications, which are implemented by varying the model structure, input parameters and specific model components.
A development scenario defineswhat event(s) may occur or how different trends may play out into the future. Unlike model specifications, scenarios describe events or processes considered external to the Kimberley system, like extreme natural events, demographic changes, global and national economic crises, industrial and infrastructure development, etc.
A management strategy is a course of action, policy or intervention we can use to respond to various scenarios. It constrains or promotes specific human activities to achieve agreed environmental, social or economic objectives.
A future is an estimation of what the system may look like several years down the track, according to the model. It results from the interplay between model specifications, scenarios and strategies.

It is important to evaluate how desirable a future is. Would we like to live in that future? Or could we live with it? To answer these questions and to compare alternative futures, we select and report on significant environmental, social and economic indicators. Notice that this evaluation is intrinsically subjective, since it is based on trade-offs in satisfying different indicators and expectations. Because of this, the evaluation of a future’s desirability is not carried out numerically, but via stakeholders’ engagement.

Asking questions about the ‘future’

What kinds of questions can we ask about the future? We cannot ask absolute questions like “what will happen?” Among other reasons, this is because the future is not pre-determined and our actions will affect it. We can ask hypothetical or conditional questions, like “if this event occurs, then what may the future look like?’ This is a more desirable approach to describing the “future” because it defines the events and conditions which may affect it.

We use models to answer these conditional questions and the elements we described above fit nicely into this question-answer structure. The scenario, management strategy and model description represent the conditioning of the question (If the scenario occur and we implement this management strategy and the system works as the model says…) and the model provides the conditional answer (… then the future may look like the model output).

Figure 1. Modelling is like asking the following question: ‘If a specific event occurs (scenario) and we take this action (strategy) and the system works this way (model description), then what may happen?’. The model output is a possible answer to the question.

Who asks the questions

Asking the questions involves two interrelated, but different, steps. First, we need to formulate the questions themselves - what we want to know about the future. Second, we need to turn the questions into something a computer model can process. The second step is technical, and involves modellers and experts. The first step is strategic and involves local stakeholders, people with local knowledge and people with a vision. Importantly, it should involve both people who will make decisions and people who will be affected by these decisions.

Not all questions can be answered by computer models of course. The interaction between modellers, stakeholders and decision makers will help refine the questions which can be answered and how they should be asked.