The social life of science in policy and planning

This article first appeared in the Autumn 2016 edition of Fire Australia magazine. By Dr Timothy Neale, Researcher, Bushfire and Natural Hazards CRC and Western Sydney University. 

Quite rightly, people in the natural hazards sector hold science in high regard. Scientific research is crucial to predicting and preparing for events whose behaviours and occurrence are both high consequence and highly uncertain. As such, it is unsurprising that government agencies often emphasise their commitment to having ‘science-led’ policies. But the routes between science, policy and planning are complex and variable. What is clear is that there is no simple relationship between having ‘more’ science and ‘less’ uncertainty, or that more or less of either leads to action (the relationship between scientific research and climate change policy are prime examples of this). I could list dozens of reasons why the interface between science, policy and practice does not run smoothly, but let us just focus briefly on two.

First, science is a diverse world of knowledge and as such, it is ripe for debate, whether by scientists, politicians, policymakers or others. The fact that the research process is open-ended, in which uncertainties can often be reduced by not resolved, means there are often abundant reasons to delay decisions about how to proceed.

Second, there are many obstacles to integrating science within government agencies, not the least of which are resource constraints. Other factors of institutional culture also influence how, and if, new research is utilised.

None of this is anyone’s fault - they are simply the conditions in which we operate. In the Scientific Diversity, Scientific Uncertainty and Risk Mitigation Policy and Planning project at the Bushfire and Natural Hazards CRC, the research team is looking at three case studies where scientific knowledge is changing how natural hazard risk is mitigated. The driving questions of the project are about the science, policy and practice interface. Given that uncertainty is an inherent part of scientific practice and method, for example, how do those engaged in risk mitigation manage these uncertainties in their decision-making? What do practitioners think are the keys to bringing new scientific knowledge in? What else is in play beyond the given technical innovation? We have been very fortunate to find some great partners in the sector interested in understanding more about this space. Our aim is that this project will support the capacity of practitioners to explain and justify what they do to others, whether those others are other risk mitigation professionals, the public, the media, or courts and inquiry processes.

In each case study we begin by interviewing practitioners in the area to gain an understanding from them of how science and other forms of knowledge inform their work and what they feel are the key issues and uncertainties that they face. We then hold a workshop to discuss these factors using scenario exercises where practitioners are given different scenarios, or predictions, of what the area they work in might be like in 20 years time. Understandably, people who work in the natural hazards sector are often focused upon the immediate context: what is going to happen this season? What is happening in the community this year? A scenario exercise is a good way to move outside these parameters to think, in this case, about longer trends, how we are going to prepare for these futures, and how science can and should inform these preparations. So far, we have held workshops for two case studies, one in the Barwon-Otway area of south west Victoria and one in the Greater Darwin area of the Northern Territory.

To tell you a little about these case studies, let me start in Victoria. For the last several years, the Barwon-Otway area has been the site of a pilot – led by the Department of Environment, Land, Water and Planning – to test an alternative strategy to how to mitigate bushfire risk. To simplify significantly, the risk-based strategy involves, first, the generation of loss estimates from suites of bushfires simulated within PHOENIX RapidFire (a 2-dimensional bushfire simulator) and, second, the comparison of asset losses between those suites. This might involve, for example, simulating fires under worst case (i.e. Black Saturday conditions) weather conditions, in which a) no planned or unplanned fires have occurred for several decades, b) all public land has been prescribed burned, as well as c) some accidental fires and some prescribed burning have occurred. Given the model’s ability to predict house losses from the intensity of each fire, the three suites can therefore be compared to reveal the benefit of fire in the landscape and the residual risk that remains. A more complex arrangement, also trialled, might compare multiple asset losses across multiple suites, each comprising thousands of simulations using random ignition and weather scenarios. In short, it is a system for calculating bushfire risk and measuring the benefits (or not) of intervening in the landscape.

Of course, some things fit into modelling better than others. One of the primary drivers of risk management anywhere is reducing losses of human life, but human behaviour – particularly human behaviour in rare and extreme events – does not map well onto algorithms. The practitioners we have met have often stressed the importance of pairing advanced tools like PHOENIX with professional experience, local knowledge, and interpersonal trust.   

The Victorian government has recently announced that it would be moving to this risk-based strategy (also known as Bushfire Risk Landscapes) across the state in mid-2016. This is a “brave and positive step”, to quote Dr Trent Penman, moving away from the existing focus on burning a percentage (five per cent) of public lands each year to reduce the risk to life, property and the environment. The hope of the research team is that our work with fire, land and emergency managers in the Barwon-Otway area will provide some insight into one of the key steps in this policy transformation and the new questions it has raised for practitioners. Once agencies have a new level of information about where risk lies in the landscape, and the benefit and potential of mitigation, what is the best strategy for passing that information on to the wider public? While, as Eburn and Handmer argue (2012: 19), there “is no legal impediment to releasing reasonably accurate hazard information,” there are clear disincentives to freely releasing information that is highly complex and has the potential to be re-used in negative ways.

The Northern Territory is quite a different context to ‘down south,’ as Territorians often point out, though it also presents interesting parallels in natural hazard management. As part of the tropical savanna, the Greater Darwin area has an annual bushfire season in which approximately 40 per cent of the total area is typically burnt. Every year, as the wet season subsides around April, bushfire practitioners work to burn off the new grass with the aim of reducing fuel loads during the late dry season.

Unlike ‘down south,’ this abundance of fire in the landscape is widely accepted as part of Territory life and its environment. However, there are several trends at work in the Greater Darwin area that are now changing the bushfire risk and its mitigation.

One primary driver in this situation is Gamba grass, a pasture species introduced in the 1970s and 1980s that grows tall, thick and flammable if not grazed intensively. Gamba is very invasive and has, over the past decade, turned parts of the Greater Darwin area to monoculture, creating high fuel loads that, in the right conditions, produce up to eight times more heat that native grassfires. Another driver of change is the increased level of subdivision and housing development surrounding Darwin, as areas such as Palmerston and Litchfield take up some of the city’s population pressure. Bushfires, previously understood as a minor risk in the Northern Territory, are beginning to claim houses and other assets in the areas infested with gamba.

Our work with practitioners has given us a new understanding of how crucial science and practitioners have been to understanding this risk and responding to it through policy and planning. For example, demonstrating the characteristics of gamba, tracking its progress, and illustrating its economic and environmental costs have all been necessary to attracting the attention of policymakers and the public in the Northern Territory, a place in which budgets are routinely strained and fire management is not – for various reasons – a top government priority.

In 2016, the team is moving onto our third and final case study, hoping to draw links across the different contexts we have engaged with. What is clear is that scientific research, whether in a laboratory or a landscape, is never simply technical. Nor is there a single stable entity we might call ‘science’; it is instead, as van Kerkhoff and Lebel state (2006: 454), a “permeable, changeable, and contestable” thing. As such, the ways in which decision-makers and practitioners integrate and utilise science is a thoroughly social question, shaped by the capacities and affordances of the contexts in which they operate. While it is important to continue to place a high value on scientific research in the natural hazards sector, it is also important to remember that this research is embedded in social dynamics and social networks – a ‘social life’ which we are, at present, only beginning to understand.