Library Open Repository

Developing Integrated Performance Measures for Spatial Management of Marine Systems


Downloads per month over past year

Smith, D and Fulton, EA and Johnson, P and Jenkins, G and Buxton, CD and Barrett, NS (2011) Developing Integrated Performance Measures for Spatial Management of Marine Systems. Technical Report. CSIRO, Hobart.

[img] PDF
Integrated_Perf...pdf | Request a copy
Full text restricted
Available under University of Tasmania Standard License.


Outcomes Achieved
This study is, as far as the authors‟ are aware the first „whole of system‟ approach to performance evaluation of spatial management. The results of this study should improve effectiveness and efficiency of spatial management, through development of better performance assessment methods. More importantly, is should lead to greater uptake of spatial management approaches to achieve ESD for marine resources and ecosystems.
An important outcome of the project has been that our model (Atlantis-SM) and what we have learned from its development and application, provides a "transportable" framework for developing performance measures. The basic approach, both the telescoping treatment of habitats in and around spatial closures and the management strategy evaluation framework for representing the estimation of indicators, can be applied to other systems via new implementations of the Atlantis framework.
However, even without going that far, it is possible to take the lessons learnt in this case to other systems. In particular it contributes to the policy debate around the implementation of EBM. It highlights the potential for monitoring for EBM performance to be far from simple. Monitoring schemes with small spatial coverage or infrequent temporal repetition (on the order of 3-5 years or more) had no power to rapidly detect changes at the broader system level. In addition, the finding that there is likely to be system specific reference points is particularly important, as the tendency within the literature has been to try and find generic rules and approaches for universal application and broad scale comparisons. The indication from this study that universal reference points (analogous to B0.4 in fisheries) or directions, which do not take into account local specificity, may not be feasible has important consequences as is at odds with recent literature and suggests that adoption of such approaches could lead to a very misleading interpretation of management performance.
There is increasing interest in the spatial management of marine systems worldwide and it is seen as a crucial step towards implementation of ecosystem-based management. This has seen a growing focus on managing marine systems at various spatial scales and assessing the relative roles of different spatial components to large marine systems as a whole. However, the scientific basis for the spatial management of marine systems is limited, although spatial area management, as illustrated by Marine Protected Areas (MPAs), has received considerable attention as the „new‟ tool to control over-exploitation of fish stocks. To properly evaluate the potential merit of spatial management, there is a distinct need for such approaches to have clearly stated objectives, meaningful indicators and effective monitoring of performance with respect to management objectives. Where performance assessment has been undertaken, it is usually focuses on examining the consequences in the immediate area of the spatial management zone (e.g. what accumulates in a closed area) rather than examining the system-wide effects and benefits. In addition, while an enormous number of candidate ecological indicators have been proposed in the literature, these are generally at the whole of system level rather than being spatially explicit.
In this study we reviewed the available information on monitoring for spatial management and associated performance measures, for programs both in Australia and overseas. Overseas monitoring programs reviewed were from the Philippines, the Caribbean, Indonesia, California, New Zealand, South Africa, Kenya, France and Ecuador. Australian programs reviewed were from Queensland, Tasmania, New South Wales, Victoria, and the Great Australian Bight. The majority of these programs were associated with spatial management of marine protected areas (MPAs). In addition the review considered monitoring for social and economic objectives of spatial management, and observational approaches for the spatial management of marine systems. A key outcome from the study into performance measurements of spatial management are the implications for monitoring designs.
The Atlantis modelling framework provided the basis for a model developed explicitly for this study; Atlantis-SM. It was calibrated using time series data from Victoria and Tasmania and was able to spatially simulate MPAs in the south east of Australia. It was developed to evaluate indicators at various spatial scales and how well they perform under a range of specifications and scenarios. We do not address whether or not there should be MPAs, rather the model is designed to develop an effective means to assess the performance of indicators of the system and the spatial management within it. The rationale for this focus is that no-take MPAs are likely to show the strongest contrast in the influence of human activity and so would contain the greatest potential differential and signal strength. If indicators are not effective in evaluating performance here they are unlikely to be useful in other forms of spatial management. To the authors‟ knowledge it represents the first such „whole of system‟ study undertaken on appropriate indicators for assessing the performance of spatial management.
We applied a 3x4x4x4* (productivity x MPA size x sampling schemes x impact type, with the * indicating that one of the impacts (fisheries) was also considered at 3 levels) matrix of specifications and scenarios to assess indicator robustness (ie how well they perform under different conditions), giving 432 individual outputs. The other impacts considered included climate change, nutrients, and illegal, unreported and unregulated (IUU) fishing.
As we were not addressing a specific management objective we drew upon indicators commonly used to address a range of spatial management objectives. The indicators evaluated were drawn from previous studies on ecological indicators and from the results of a literature review undertaken as part of this study. We also chose indicators that could be feasibly calculated and tested in the Atlantis-SM model, which cover the majority of indicators that can be feasibly and repeatedly measured in reality. These indicators cover the primary indicators used to date to monitor MPAs, and the recommended set from past studies of ecological indicators of the effects of anthropogenic impacts (especially fishing).
Compared with previous studies, the indicators checked show that, in broad terms, overall indicator performance still holds. However it also highlighted that monitoring for EBM performance may be far from simple. While sampling schemes of low frequency or spatial coverage are acceptable for detecting change inside and outside closures (also needing a reasonable time series to enable causes of the signal to be evaluated) they have little power to detect signals at broader spatial scales. Monitoring schemes with small spatial coverage or infrequent temporal repetition (on the order of 3-5 years or more) had no power to rapidly detect changes in the system; while intensive sampling was confounded by natural system variation and shifts through time, unless carefully planned around stratified sampling schemes. Moreover, indicators, such as pelagic:demersal fish biomass, that have been found to be useful across different ecosystem types proved sensitive to scale. These indicators were informative in the immediate area of closures (as the data at this scale is within habitat patches and individual species ranges and so avoid species-scale mis-matches) and globally (because at such large scales the ratio integrates across many species effectively smoothing out any potential mis-matches). However, they do not work at intermediate spatial scales because these exceed the typical spatial range of activity of individual species, but are not yet at a point where they smoothly integrate across sufficient groups.
The ecology of the groups in the system also impacts the performance of individual indicators based on those groups. For example, signals for mobile species can be over-stated outside reserves, while signals for more sedentary species decay rapidly with distance from the closures.
Atlantis-SM also suggests that variation in community dynamics between regions can lead to locally specific indicator-attribute relationships; meaning that while indicator signals are representative of the attribute at a specific locale, they may not always be consistent site-to-site. For instance the relationship between the indicator “relative lobster biomass” and the attribute “diversity” was linear (with R2> 0.92), but in opposite directions (in one case there was a positive correlation and in one a negative) at sites less than 300km apart. This difference in direction of response is due to locally specific environmental drivers and community dynamics and has significant implications for monitoring and management, as it
Developing Integrated Performance Measures for Spatial Management of Marine Systems
shows that an understanding of system dynamics at regional scales will be necessary to understand the signal obtained from indicators. This suggests that universal reference points (analogous to B0.4 in fisheries) or directions, which do not take into account local specificity, may not be feasible. This finding is at odds with recent literature on indicators which not only recommends a definitive set of indicators across many systems and scales but also recommends the use of reference points that are intended to be consistent across systems. Instead, suites of indicators drawn from the main general classes of indicators noted above (e.g. relative biomass, biomass ratios, relative habitat cover) will need their associated reference points or directions adjusted to suit status and processes at the locations of interest (and potentially through time as the system changes). Crucially, this also means that a lack of a temporal dimension in monitoring cannot be completely compensated for by periodically applying very intensive surveys across broad spatial scales.The results of the study indicated that fisheries dependent indicators should not be used alone unless there is absolutely no alternative (industry independent data is much preferred). Fishery independent surveys using commercial or research vessels (often using trawl or other extractive methods), are commonly used around the world. However in some spatial zoning arrangements, such as no-take MPAs, extractive sampling methods may be prohibited. It has been argued that by using a combination of (non-extractive) observational techniques (eg underwater visual census, video, BRUVS, acoustics, „smart tags‟ etc) to target specific species or habitats, spatial monitoring surveys can provide information on the whole ecosystem. While this might be the case, the efficacy of many of these methods for the sustained observing required to monitor marine systems has still to be demonstrated.
Finally, these are strong ecological reasons why a suite of indicators will be needed to capture performance of spatial management. When moving to triple bottom line objectives, the inclusion of social and economic objectives is still more reason for use of a broad suite of indicators to cover all aspects of the system (and triple bottom line objectives).

Item Type: Report (Technical Report)
Keywords: Spatial management modelling modelling MPA Marine Protected Area
Publisher: CSIRO
Identification Number - DOI: ISBN 9781921826375
Date Deposited: 04 Nov 2011 03:58
Last Modified: 08 Nov 2011 00:19
Item Statistics: View statistics for this item

Actions (login required)

Item Control Page Item Control Page