Risk levels of LMEs based on multivariate indicators weighted by the Human Development Index.

Identifying patterns of risk among Large Marine Ecosystems using multiple indicators

Click here to choose an LME

K.M. Kleisner, L. Talaue-McManus , B.S. Halpern, P.J. Kershaw, V.W.Y. Lam, K. Sherman


Introduction

One of the two goals of the Transboundary Waters Assessment Programme (TWAP) is to conduct a baseline global assessment of five transboundary water system categories: (1) Large Marine Ecosystems (LMEs), (2) Open Oceans, (3) Aquifers, (4) River Basins, and (5) Lakes & Reservoirs, with the aim of providing guidance to the GEF and other stakeholders for prioritization of interventions within these water systems. With the exception of the Open Oceans, these assessments are comparative and group the transboundary water systems into five risk categories (from very high to very low) based on a suite of indicators. Risk is broadly defined as the probability of adverse consequences for humans and the environment in relation to the changing states of transboundary waters. The LMEs assessment is based on indicators under each of the five LME modules (Productivity, Fish & Fisheries, Pollution & Ecosystem Health, Socioeconomics, and Governance), for which global data sets are available. Results of the individual indicators are presented by module on this website. Triggers of risk are usually multiple factors, which may be biophysical, socioeconomic, or governance-related in some combination. To identify patterns of risk among LMEs using multiple indicators, a number of indicators with strong directionality in indicating ‘good’ or ‘bad' ecosystem states are used to identify groups of LMEs based on their similarities across the modules.

Multivariate and risk scoring techniques provided complementary approaches in delineating LMEs at risk through the simultaneous use of multiple indicators that measure biophysical, socioeconomic and governance pressures and states.

Approach

  1. The simultaneous use of multiple indicators allows for the classification of large marine ecosystems into thematic clusters, and along axes defined by a combination of dominant indicators, and for the ranking of LMEs using risk scores. Statistical classification and ordination techniques were used to identify clusters of LMEs subject to similar pressures and impacts;
  2. Indicators that were strongly directional in indicating ‘good’ or ‘bad' ecosystems states, and which were assessed for at least 60 of the 66 LMEs, were chosen for this ranking analysis and used to cluster the LMEs: ‘Demersal Non-destructive Low-Bycatch Fishing’, ‘Pelagic Low Bycatch’, ‘Proportion of Collapsed and Overexploited Stocks’, ‘Capacity enhancing subsidies as a fraction of the value of fisheries’, ‘Proportion of Catch from Bottom Impacting Gear’, ‘Index of Coastal Eutrophication Potential’, ‘Plastic Debris Density’, ‘Percentage Change in Area of MPAs’, ‘Shipping Pressure’, ‘Percentage Rural Population within 100km of the Coast’, and the ‘Night Light Development Index’;
  3. The Human Development Index was used as a measure of the socioeconomic status of an LME as well as a weighting factor in determining an overall risk score for each LME. The scores took into account the average of all fisheries indicators, and all metrics for pollution and ecosystem health used in the study. This approach is one of a number of ways to rank LMEs;
Within the 11-indicator domain used in this study, LMEs with developing economies showed highest risks in terms of coastal eutrophication and plastic litter density, and moderate to high risks from collapsed and overexploited fish stocks.

Results

Figure 1 below illustrates how the LMEs cluster according to these indicators and lists the indicator(s) that were dominant in driving the clusters:


Figure 1. The main LME clusters and associated indicators

LMEs encompassing developed nations suffered from high risks because of high shipping frequencies, high capacity-enhancing fisheries subsidies, and high catches from bottom-impacting gear

The nature of these clusters informs the identification of LMEs of potential interest for policy and management interventions. The risk categorization provides an interpretation of the priority status of the LMEs within a human developmental framework.

The dominant risks associated with combinations of indicators can be illustrated spatially, as shown in the following maps (Figures 2 and 3). Shipping pressure (brown colors) and coastal rural population density (blue colors) are two of the strongest indicators defining LME groupings (Figure 2). Heavily developed regions such as the North Sea, East China Sea and Northeast U.S. Continental Shelf LMEs have higher risks associated with shipping pressure. LMEs that have higher risk due to vulnerable rural populations in coastal areas include the High Arctic LMEs.

All LMEs, except for the Australian shelf LMEs, the Red Sea and the Gulf of California are at risk because of low percentage of established recovery zones such as marine protected areas.

Figure 2. Shipping pressure (highly positive/brown colors) and coastal rural population density (highly negative/blue colors)

High risk associated with pressures from catch from bottom impacting gear types (brown colors) and pressures due to demersal non-destructive low bycatch fishing (blue colors) were also strong indicators in defining the groupings (Figure 3). LMEs with high pressure from destructive bottom gears include the Southeast U.S. and the East Central Australian Shelf, Southwest Australian Shelf and Southeast Australian Shelf LMEs. Those with more pressure due to demersal non-destructive low bycatch fishing include LMEs in Asia like the Sulu-Celebes Sea, the Indonesian Sea and the South China Sea LMEs, as well as European LMEs like the Norwegian Sea, the Baltic Sea, and the Icelandic Shelf and Sea LMEs.

Caveats and limitations: This approach, like most indicator-based assessments is highly subjective and open to interpretation. The indicators selected here are a subset of the potentially relevant indicators, but were selected because they had the greatest information content. The approach outlined here provides one example of risk prioritization that places emphasis on the level of human development in a region and the suite of selected indicators. Patterns may change as more spatial data specific to the LMEs become available.

Figure 3. Catch from bottom impacting gear types (highly positive/brown colors) and pressures due to demersal non-destructive low bycatch fishing (highly negative/blue colors)

In order to provide an illustrative linear ranking of assessed units, which is often required by managers as a basis for setting priorities, the Human Development Index (HDI) was used to weight the LME risk scores. Other approaches could be taken, but the use of the HDI follows the premise that coastal populations of LMEs with lower socioeconomic development status will be at a higher risk for the same levels of environmental status, and may indicate situations where an LME population will have a limited ability to cope with degraded transboundary waters due to increased pressures and impacts on the coastal ecosystem and/or the increased dependence on ecosystem services. Low levels of human development based on educational attainment, life expectancy and per capita gross national income indicate few livelihood options and limited resources for basic existence, and much less for dealing with the impacts of natural disasters, climate change and degraded ecosystems. The results are shown in Figure 4, where LMEs with very high scores are colored red; those with medium risk scores in yellow, and LMEs with very low risk scores in blue. The LMEs that were at highest risk were those with the lowest HDI and included the Somali Coastal, Guinea and Canary Current LMEs (GEF-eligible LMEs). These LMEs had low fisheries related risks, but incurred high risks from pollution and low proportions of MPA Area coverage. LMEs with the lowest risk had high HDI status, but were also subject to pressures from pelagic low bycatch fishing levels and capacity-enhancing fisheries subsidies.

It should be noted that the LME rankings could change based on the indicators used as well as on the weighting or ranking scheme adopted. Results relate to the scale of the entire LME and do not reflect on any individual country's management of its coastal waters.

Figure 4. Risk levels of LMEs based on multivariate indicators weighted by the Human Development Index.

Click on the map for an interactive version

Click/right-click here for a fullscreen/downloadable version