Good decisions for the environment need an eye on the longer term
Long-term monitoring provides essential evidence on which to base good environmental decisions.
Good design is essential for effective long-term monitoring.
Things change over time; to remain effective, long-term monitoring needs to adapt around these changes.
Partnerships are crucial for ensuring long-term monitoring is maintained and listened to.
Long-term monitoring is most effective where it is complemented by other value frames (such as economics).
Effective long-term environmental monitoring is difficult and challenging; it requires good design, careful review, long-term commitment, and often gets overlooked when resources are handed out by our political leaders. Given this, why bother? We bother because long-term monitoring is the cornerstone of effective environmental policy and management. In a ‘post-truth’ age witnessing a crisis in biodiversity decline, long-term monitoring is something we can’t afford not to do.
However, if you are going to do it, it needs to be done properly. Literally thousands of scientific articles and dozens of books have been written on almost all aspects of long-term monitoring. This chapter does not attempt to repeat or even briefly summarise this vast body of work (though for a good guide to it see Lindenmayer & Likens, 2018). Rather, here we focus on five key learnings from the body of work on longterm monitoring as they relate to making good environmental decisions.
1. Evidence-based policy needs long-term monitoring
The mantra of modern governments and other bodies responsible for managing natural resources (including biodiversity) is that both management and policy must be ‘evidence based’. In a world in which ‘truth’ is constantly under attack the need is only greater, but where does that evidence come from. Long-term monitoring is often the essential source.
In terms of biodiversity, long-term monitoring is often needed to measure change in a given entity (such as a population of a species or the condition of an ecosystem), but also to measure how those entities change in response to some kind of management intervention (like pest control or habitat enhancement). Long-term monitoring is essential to determine if actions taken to manage the environment are effective, and therefore whether decisions made to invest in particular actions are vindicated (or whether different interventions are needed).
Consider the case of controlling Bitou bush, one of Australia’s worst invasive plant species (ironically, it was deliberately introduced in the first instance to help with sand-dune stabilisation). Long-term monitoring enabled us to determine the effectiveness of a program to control the Bitou bush in Booderee National Park in southern New South Wales. A significant proportion of the management budget for the park is dedicated to controlling this weed so dealing with it comes with a high opportunity cost (in terms of other work not being funded).
Despite the considerable expense, prior to us monitoring efforts at control, it wasn’t known whether the program was either ecologically effective or cost effective. Our monitoring work, carried out in collaboration with Booderee National Park management (see the box ‘15 years at Booderee’), revealed that the currently applied method of spraying Bitou bush, then burning the dead canes, following by respraying was the most ecologically effective treatment protocol for removing the invasive bush. Not only did it control the weed, this treatment also stimulated the restoration of native plant cover (Lindenmayer et al, 2015c).
The spray-burn-spray protocol was also found to be more cost-effective than other kinds of management interventions. Moreover, very few native animal species were found to be disadvantaged by the spray-burn-spray treatment protocol (Lindenmayer et al, 2017). This is an important outcome as Booderee National Park is a stronghold for populations of endangered species such as the eastern bristlebird. Such taxa were found to benefit from the removal of Bitou bush and post-treatment recovery of native vegetation (Lindenmayer et al, 2017).
The problems of not conducting long-term monitoring are evident from many failed environmental programs, including those in which very large investments were made.
For example, despite billions of dollars of investment in river restoration programs in the USA, a paucity of robust long-term monitoring made it impossible to determine whether such restoration efforts had been effective and successful (Bernhardt et al, 2005).
This is a far from isolated case. The effectiveness of billion dollar agri-environment schemes to better manage biodiversity and other conservation values in farming landscapes in Europe and North America is poorly known because of a lack of long-term monitoring. Similarly, large scale vegetation restoration and salinity mitigation programs funded by the Australian and State Governments remain poorly monitored (if monitored at all). This fundamental oversight leads to ineffective programs, vast amounts of wasted taxpayer funding and a public misperception that environmental problems cannot be resolved (Hajkowicz 2009).
Policy makers and politicians can certainly make poor environmental decisions, even in the face of overwhelming evidence of the need for alternative decisions to the ones they have made. We trust, however, that such poor decisions will be rarer when evidence is available than when there is no available evidence. Put another way, good evidence, based on long-term monitoring should provide scientists with the persuasive power to influence decisions, and even if not an ‘optimal’ decision, then perhaps ‘less bad’ than it might otherwise have been.
2. Effective long-term monitoring is built on good design
Long-term monitoring programs need to be underpinned by good design if they are to generate data that can guide effective environmental decisions. And that design begins with asking what the monitoring will actually be used for. And, if there is no intention on the part of managers to change their management or if there is no capacity to learn from the monitoring results, then a monitoring program may not even be appropriate (see Decision Point #52 p4-7).
If there is capacity to learn and a willingness of managers to respond, then there are some fundamental ingredients which contribute to good monitoring design. These include:
a. Careful articulation of the objectives of monitoring, with all partners being clear about the aims and objectives;
b. Good and tractable questions of management relevance (often being informed by a well-developed conceptual model of the system being monitored);
c. Implementation of a robust statistical design (that answers key questions);
d. Regular assessment of the data gathered (to ensure errors in a dataset are corrected or key missing variables can be gathered); and
e. The inclusion of trigger points for action if major changes occur in the system being monitored (Lindenmayer et al, 2013b).
Conversely, long-term monitoring programs established without these considerations can result in an expensive waste of resources. An example is the Alberta Monitoring Program (Alberta Biodiversity Monitoring Institute 2009) in the Canadian Province of Alberta in which millions of dollars are expended each year on a passive monitoring program that lacks questions or a robust underlying statistical design (see Lindenmayer & Likens, 2010a).
3. Adaptive monitoring can be essential
Things change, it’s a given. It’s better to adapt to changes than stick with a monitoring program that is no longer relevant. Often there is a need to change the questions being posed over time and/or change the underlying experimental design in response to those changed questions. Or there might be other reasons for change like the development of new technology that requires altered field-based measurement protocols.
Poor earlier policy and/or management decisions also might create extra cascading environmental problems, demanding a reworking of the original scope of a monitoring program. An adaptive monitoring approach (as described in Lindenmayer and Likens 2009) may be required to redesign a pre-existing monitoring program so that it can answer new key questions of management relevance that are useful in guiding environmental decisions.
An example of adaptive monitoring comes from monitoring the Environmental Stewardship Program (see the box on ‘Adaptive monitoring and the ESP’).
4. Partnerships are critical
Effective long-term monitoring programs need good partners – partners that will help frame the purpose of the program, assist in translating the monitoring results into effective management decisions, and act as champions for the program to ensure it has a long-term future.
Partnerships between scientists, resource managers and policy makers can ensure that the key questions being addressed in a long-term monitoring program are management relevant but at the same time scientifically tractable. Partnerships also provide a vehicle for regular exchange of information and the opportunity to build a broad constituency to maintain longterm work. Such partnerships are essential to ensure that the evidence gathered from long-term monitoring can be widely communicated to those responsible for decision making; this may include engagement with the political process to inform ministers (and minster’s advisors) on what the results of longterm monitoring are showing.
Considerable effort is needed to maintain the array of partnerships which underpin long-term monitoring and its link with effective environmental decision making. For example, the rapid turnover (churn) of staff within government agencies poses a particular challenge as champions for particular projects are needed to maintain them in the long term.
Considerable time often needs to be expended by the scientific leader of a long-term project to explain what the work aims to do, why it is important, why it is relevant to informed policy and decision making. This may need to be done repeatedly as new staff are recruited. Field trips to long-term monitoring sites can sometimes be particularly effective as these provide a practical and tangible context for how particular management problems are being examined and tackled through science-manager partnerships (Lindenmayer et al, 2013a).
5. But remember, ‘it’s the economy, stupid!’
Many long-term monitoring programs focus on threatened species and ecosystems and we know from experience this is a good basis for deciding how to effectively manage these systems. However, when it comes to our political representatives, long-term biophysical evidence is often of secondary significance in the political calculus. They are more interested in what it means for their voters which is why when considering the outputs of long-term monitoring programs it’s always valuable to consider how they can be integrated with other metrics relating to your system of interest. The environment is important but the social and economic dimensions of the system are possibly of greater significance when it comes to policy and decision making.
As an example, much has been written about the results of long-term ecological and environmental monitoring in the montane ash forests of the Central Highlands of Victoria (home to the Critically Endangered Leadbeater’s possum and several other threatened species; (Lindenmayer et al, 2015a). Unfortunately, much of the conservation science generated over many years remains ignored (Lindenmayer et al, 2015b). However, monitoring may have more traction with decision makers when key natural assets are monetized in economic and environmental frameworks like those developed by the United Nations, for example the System of Economic and Environmental Accounting (or SEEA).
The SEEA framework enables the ‘value-added value’ of industries based on natural resources like tourism, carbon, water and timber to be compared in a formal and internationally accepted accounting framework. When applied to the forests of the Central Highlands it showed that the value-added value of the native forest timber industry ($12 million) was a fraction of the water industry ($310 million) and the tourism sector ($260m) (Keith et al, 2017). Decisions to maintain timber production (which undermines the value of the water and tourism industries) are therefore based on something other than rational economics.
Notably, in a communique from a 2016 COAG meeting, the Commonwealth Minister for the Environment and Energy and his State and Territory colleagues recommended that environmental accounting be widely applied and adopted in Australia. We suggest that the approach has the potential to add considerable value to datasets that are being gathered in environmental monitoring programs and provides a new way that such programs can help influence decision making.
Decisions and long-term monitoring
Long-term monitoring programs are often linked with many kinds of decisions; some associated with better informing on-the-ground management, others linked with changes in policies. There are also scientific decisions associated with the ‘inner workings’ of long-term monitoring programs such as the way they are designed or re-designed and how protocols for field measurements might be altered on the basis of the development of new techniques or the discovery of new problems (such as the colonization of new species of invasive organisms).
The five major themes discussed here can potentially influence each of these kinds of decisions and vice-versa. That is, management, policy and scientific decisions can influence (and also be influenced by) the need to make good environmental decisions, the importance of good study design in monitoring, the need for adaptive management, the fundamental importance of partnerships, and recognition of the potential value of extending biodiversity monitoring into other domains.
Regardless of what influences what, the case for good longterm monitoring as an (evidence) base for better decision making is indisputable.
What is long-term monitoring?
There are many formal definitions of what constitutes longterm monitoring but a good rule of thumb I apply is that it is any investigation involving repeat measurement that has been running continuously for ten or more years. Ten years is not a magical number separating ‘short-term’ from ‘longterm’, however monitoring programs that have run for longer than ten years usually have a ‘long-term’ framing aimed at capturing trends and variability that are often not evident in shorter programs.
Adaptive monitoring and the ESP
The Environmental Stewardship Scheme (ESP) was established to assess the effectiveness of management interventions associated with a major agri-environment scheme in the temperate woodlands of eastern Australia. Farmers are paid to carry out specific management actions to improve the condition of patches of endangered Box Gum Grassy Woodlands (BGGWs) on their land. The ESP comprises a total of 158 farms in a region stretching over 2000 km (south to north) (Burns et al, 2016). A patch of BGGW targeted for stewardship management on each of the 158 farms is also targeted for monitoring with a matched control patch (where no stewardship management occurs) also monitored on each farm.
The budget for the monitoring was initially substantial as the government agency responsible for the ESP demanded that every patch on every farm be monitored every 1-2 years. However, funding cuts occurred four years into the 15-year program.
Adaptive monitoring had to be adopted to prevent the entire monitoring effort collapsing. A rotating sampling monitoring approach was introduced in which 65% of farms were surveyed in any given year with a complete ‘census’ of all farms undertaken twice in four years (Lindenmayer et al, 2012).
The emphasis of monitoring switched from an assessment of compliance for implementing particular kinds of conservation management to an estimation of changes in condition across a large ‘population’ of sites.
Unfortunately, even this adaptive change was insufficient to enable a long term assessment of the effectiveness of the scheme because, subsequent to this, the Australian Government changed the monitoring approach using a different group and methodology. This meant that what had been done before does not match with the limited effort that is being done now, thereby breaching the integrity of the time-series data gathered over the past 8 years and precluding robust assessment of whether the programme is effective (or otherwise) (Lindenmayer & Likens, 2018; Lindenmayer et al, 2018). All of which highlights the many challenges of running an effective long-term monitoring program.
15 years at Booderee
Strong and enduring partnerships have been at the heart of the success of the 15-year monitoring program at Booderee National Park (Lindenmayer et al, 2013a) on the south coast of NSW. The data from the monitoring program have underpinned approaches to fire management by the resource managers of the park.
The partnership has focussed on three key issues within Booderee: the impacts of fire on native biota, the response of vertebrates to feral animal control and the control of Bitou bush.
In regards to fire, a new understanding of the relationships between bird persistence and recovery following fire (derived from empirical research) has resulted in a change from uniform prescribed burning of entire compartments of native vegetation to patchy fires across a maximum proportion of a given compartment.
Monitoring has also demonstrated the value of feral animal control showing it substantially increases populations of some animals such as the common brushtail possum, the long-nosed bandicoot and the endangered eastern bristlebird. On this basis, an intensified approach to feral animal control in Booderee National Park is now well established as a key and ongoing conservation activity recognised formally within the official management plan for the reserve.
Despite the extensive work completed on long-term monitoring and how it might be better designed and implemented, some important challenges remain. First, monitoring programs need to better account for cumulative effects of multiple, interacting disturbances in ecosystems and, for example, help us better predict major problems such as the risk of ecosystem collapse (Sato and Lindenmayer 2018).
In addition, more information is needed on the effectiveness of monitoring in terms of return on financial and logistical investment to help better answer questions like: What is the value of new information gained from monitoring? and How does that information help contribute to better decision making? New insights on return on investment are also critical for determining when it is time to stop monitoring as well as when not to start new monitoring programs (see also (McDonald-Madden et al, 2010).
Finally, there is a need to revisit the already large literature on the connections between science and policy and, in a post-truth era (Lubchenko 2017). We need to explore:
- How to reinstate the value of empirical evidence in making better decisions and
- How can we reduce the amount of time between gathering and analysing data in monitoring programs and making better decisions
More info: David Lindenmayer David.Lindenmayer@anu.edu.au
References and further reading
Alberta Biodiversity Monitoring Institute (2009). Program overview. Alberta Biodiversity Monitoring Institute, Edmonton.
Bernhardt ES, MA Palmer & JD Allan (2005). Synthesizing US river restoration projects. Science 308: 636-637.
Hajkowicz S (2009). The evolution of Australia’s natural resource management programs: Towards improved targeting and evaluation of investments. Land Use Policy 26:471-478.
Kay G, PS Barton, D Driscoll, S Cunningham, W Blanchard, S McIntyre & DB Lindenmayer (2016). Incorporating regional-scale ecological knowledge to improve the effectiveness of large-scale conservation programmes. Animal Conservation 10: 515-525.
Keith H, M Vardon, J Stein, J Stein & DB Lindenmayer (2017). Ecosystem accounts define explicit and spatial trade-offs for managing natural resources. Nature Ecology and Evolution 1: 1683–1692.
Lindenmayer DB & GE Likens (2009). Adaptive monitoring: a new paradigm for long-term research and monitoring. Trends in Ecology and Evolution 24:482-486.
Lindenmayer DB & GE Likens (2010). The science and application of ecological monitoring. Biological Conservation 143:1317-1328.
Lindenmayer DB, C Zammit, SA Attwood, E Burns, CL Shepherd, GE Kay & J Wood (2012). A novel and cost-effective monitoring approach for outcomes in an Australian biodiversity conservation incentive program. PLOS One 7:e50872.
Lindenmayer DB, C MacGregor, N Dexter, M Fortescue & P Cochrane (2013a). Booderee National Park management: Connecting science and management. Ecological Management & Restoration 14:2-10.
Lindenmayer DB, M Piggott & B Wintle (2013b). Counting the books while the library burns: Why conservation monitoring programs need a plan for action. Frontiers in Ecology and the Environment 11:549-555.
Lindenmayer DB, D Blair, L McBurney & S Banks (2015a). Mountain Ash. Fire, Logging and the Future of Victoria’s Giant Forests. CSIRO Publishing, Melbourne.
Lindenmayer DB, D Blair, L McBurney & SC Banks (2015b). Ignoring the science in failing to conserve a faunal icon – major political, policy and management problems in preventing the extinction of Leadbeater’s possum. Pacific Conservation Biology 21:257- 265.
Lindenmayer DB, J Wood, C MacGregor, Y Buckley, N Dexter, M Fortescue, RJ Hobbs & J Catford (2015c). A long-term experimental case study of the ecological effectiveness and cost effectiveness of invasive plant management in achieving conservation goals; Bitou Bush control in Booderee National Park in eastern Australia. PLOS One 10:e0128482.
Lindenmayer DB, J Wood, C MacGregor, RJ Hobbs & J Catford (2017). Non-target impacts of weed control on birds, mammals and reptiles. Ecosphere 8: https://doi.org/10.1002/ecs2.1804
Lindenmayer DB & GE Likens (2018). Effective Ecological Monitoring. (Second Edition) CSIRO Publishing, Melbourne.
Lindenmayer DB, D Michael, M Crane & D Florance (2018). Ten lessons in 20 years: Insights from monitoring fauna and temperate woodland revegetation. Ecol Manag & Restor 19: 36-43.
Lubchenko J (2017). Environmental science in a post-truth world. Frontiers in Ecology and Environment 15:3.
McDonald-Madden E, PWJ Baxter, RA Fuller, TG Martin, ET Game, J Montambault & HP Possingham (2010). Monitoring does not always count. Trends in Ecology and Evolution 25:547-550.
Sato C & DB Lindenmayer (2018). Meeting the global ecosystem collapse challenge. . Conservation Letters 11 https://doi.org/10.1111/conl.12348
Sato C, J Wood, JA Stein, M Crane, S Okada, D Michael, G Kay, D Florance, J Seddon, P Gibbons & DB Lindenmayer (2016). Natural tree regeneration in agricultural landscapes: The implications of intensification. Agriculture, Ecosystems & Environment 230:98-104. United Nations (2012). System of Environmental-Economic Accounting Central Framework. United Nations, New York.
Welsh AH, RB Cunningham & RL Chambers (2000). Methodology for estimating the abundance of rare animals: seabird nesting on North East Herald Cay. Biometrics 56:22-30.