When location matters: proximity of extractive projects to biodiversity sensitive sites and environmental risk exposure

The missing link between biodiversity and business conduct risk

Increasing interest in biodiversity has brought attention to environmentally sensitive sites, including Key Biodiversity Areas and protected areas. Key Biodiversity Areas are sites contributing significantly to the global persistence of biodiversity, in terrestrial, freshwater, and marine ecosystems. Protected areas are geographical spaces, recognized, dedicated and managed, through legal or other effective means, to achieve the long-term conservation of nature with associated ecosystem services and cultural values.

The 30x30 target adopted as part of the Global Biodiversity Framework by 188 nations aims to protect 30% of land and water by the year 2030 and will significantly increase the area covered by protected areas. However, recent research has shown that human pressure within these areas is on the rise and mere designation without proper management is often insufficient.¹ While many Key Biodiversity Areas are not currently covered by protected areas, as countries designate areas in line with the 30x30 target, Key Biodiversity Areas will likely guide the location of these new protected areas. As such, establishing proper governance in these sites will be critical, and in many cases, companies will need to consider new compliance requirements and their role in conservation.

In the extractive sector, pledges such as the “no-go” commitment of several oil and gas companies to refrain from operating within the boundaries of UNESCO World Heritage Sites indicate a trend towards the protection of nature. Ipieca, the global oil and gas association for advancing environmental and social performance across the energy transition, has stated its support for the Global Biodiversity Framework, encouraging industry action on biodiversity. In the mining sector, the International Council on Mining and Metals (ICMM) Principle 7 commits to the conservation of biodiversity and guides respect for legally designated protected areas. These commitments also benefit companies, shielding them from the harshest legal, financial, or reputational turmoil that could arise from damage to biodiversity attributed to their projects.

Intuition and experience tell us that operating within the vicinity of an environmentally sensitive site introduces a potential risk to both the site and the project’s owners and operators to a degree higher than if the site remained untouched, or if the project was situated elsewhere. However, the extent of such a change in environmental and reputational risk has not previously been measured for a large sample of companies and projects. While it may be expected that project risk carries up to the company level, the ESG risks associated with a project do not always impact the companies that own or operate them equally. Large-scale projects can have several owners and operators, some of whom may be largely unknown to the public and may not be mentioned in a criticism of the project.

Environmental and social impacts can translate into tangible financial risks, whether through reputational, legal, or systemic events like biodiversity collapse. Companies, and those that invest in them, need to understand the degree to which they expose themselves to legal, financial, and reputational risk by operating projects near environmentally sensitive sites.

Several studies have reviewed the direct overlap between extractive activities and certain environmentally sensitive sites. A 2015 study performed by WWF, Aviva Investors, and Investec Asset Management² analyzed the threat posed by the extractive sector to UNESCO World Heritage Sites, some of the most high profile and protected sites in the world, such as the Great Barrier Reef and Yellowstone National Park. It found that 31% of Natural World Heritage Sites were exposed to risk from extractive projects. A 2020 study by Sonter et al.³ indicated that the global area occupied by mines overlaps 8% with protected areas and 7% with Key Biodiversity Areas. Their research also suggested that the protected area designation seems to prevent mining activities within their limits to some extent, when compared to non-mining activities. While immediate overlap is a primary focus, other research has indicated how the impacts of extractive projects can extend beyond their direct footprint. For instance, in the oil and gas sector, research has helped to understand the downstream impacts of pipeline placement, such as evaluating the spread of pollution to nearby protected areas in the event of a pipeline rupture.⁴ A literature review of buffer distances by UN Environment Programme World Conservation Monitoring Centre (UNEP-WCMC) found that the Area of Influence considered by researchers is shaped by the type of operations and habitat at risk. They found that the extent of direct impacts varies between project contexts, and very often stretches past the immediate site boundaries.⁵

Extractive activities often take place near or within environmentally sensitive sites. While researchers voice concern over ensuing environmental risks, no research to date has explored the relationship between ESG risk and operations near environmentally sensitive sites.

Overview of ESG risk data

RepRisk’s corporate and project risk dataset provides information on material ESG and business conduct risks. RepRisk screens, on a daily basis, over 150,000 public sources and stakeholders in 23 languages. These include print media, online media, social media, government bodies, regulators, think tanks, newsletters, and other online sources. This list of sources is reviewed regularly and extended according to daily searches, RepRisk’s own research, and through client feedback. RepRisk’s methodology is issues- and events-driven rather than company-driven – i.e., RepRisk screens sources and stakeholders for ESG risk incidents in accordance with the RepRisk research scope, not a defined list of companies.

RepRisk’s environmental incident methodology captures instances where a company is publicly criticized for negative environmental impact.⁶ Environmental incidents include both direct impacts on Key Biodiversity Areas and protected areas and broader environmental impacts such as contributions to global pollution. These risks could materialize as direct impact to nature, for example, with reports of fish die-off as a result of pollution from a tailings dam collapse, or could indicate potential future risk to biodiversity, as with protests of a planned pipeline project that would traverse the habitat of endangered species. In both scenarios, RepRisk considers the project or company of interest to be exposed to environmental risk.

Each risk incident is labeled with a degree of severity, determined as a function of three dimensions: first, what are the consequences of the risk incident; second, what is the extent of the impact; and third, was the risk incident caused by an accident, by negligence, by intent, or in a systematic way. Consideration of these dimensions yields three levels of severity: low severity, medium severity, and high severity. Since 2007, RepRisk has captured over 31,000 unique ESG risk incidents linked to extractive infrastructure projects in the mining and oil and gas sectors. To provide a sense of how these incidents are distributed across sectors and regions, Table 1 shows the average count of ESG risk incidents for these project types across different locations.

Table 1. Number of projects and environmental risk incidents by region

Mining projects in Latin America and oil and gas projects in North America are, on average, associated with more ESG risk incidents, likely a result of both the prevalence of these activities in the respective regions and the attention they receive.

Overview of geospatial risk data

RepRisk Geospatial offers a first-of-its-kind solution to biodiversity risk assessment, linking projects and the companies that own or operate them to environmentally sensitive sites based on their proximity.

RepRisk Geospatial provides proximity data to shed light on potential biodiversity risks using a methodology developed with the Integrated Biodiversity Assessment Tool (IBAT) Alliance. First, RepRisk geolocates mining and oil and gas projects. Next, RepRisk detects environmentally sensitive sites near each project, using IBAT's authoritative dataset of over 300,000 protected areas and Key Biodiversity Areas. IBAT’s Key Biodiversity Area data comes from the World Database of Key Biodiversity Areas, which is managed by BirdLife International on behalf of the KBA Partnership, while the protected areas data is maintained by UNEP-WCMC through the World Database on Protected Areas. With these datasets, it can be determined whether a mine, pipeline, or other extractive project is in or near an environmentally sensitive site. RepRisk then attributes those insights to the projects’ owner and operator companies, identifying owners that own as little as 0.5% of an asset. Where multiple owners or operators are identified, project-level insights are attributed to all of them. When RepRisk detects an environmentally sensitive site near a project, RepRisk assigns the project a Proximity Level to capture the project’s potential impacts on the sensitive site. RepRisk and IBAT defined the Proximity Levels based on UNEP-WCMC research about the geospatial extent of impacts of mining and oil and gas projects.

There are three Proximity Levels:

1. Immediate (within 1 km) covers the direct footprint of most operations – the area covered by the project’s infrastructure and associated habitat clearance.
2. Close (within 1 – 10 km) captures most direct impacts in terrestrial environments, including effluents, pollutants, and noise disturbance.
3. General (within 10 – 30 km) captures direct impacts in environments where emissions, noise disturbance, and sedimentation can travel larger distances, such as in freshwater and marine environments.

Different project types have different physical footprints and other considerations, impacting their placement. In the oil and gas sector, for instance, the majority of pipelines weave in or near at least one environmentally sensitive site, while other types of extractive projects like mines generally have more contained operations. Figure 1 below shows the proportion of projects in each category that fall within the different Proximity Levels.

Figure 1. Proximity to environmentally sensitive site by project type

---

We used multiple regression to test which project and company attributes significantly predicted environmental risk. A set of 3,447 extractive infrastructure projects and 17,031 owner or operator companies were used for this analysis. The projects and companies included in this study are those having both spatial and ESG risk information in the RepRisk dataset. Projects with spatial information and no history of ESG risk were excluded, as were projects with ESG risk information and no spatial information. The ESG risk data used in the analysis is composed of incidents from 2007-2022.

In the regression model, the response variable is a severity-weighted count of environmental risk incidents, calculated as the logarithm of the incident count, weighted by the severity (harshness) of each risk incident. For projects, this represents all environmental risk incidents linked to the project within the period of consideration. For companies, this represents all environmental risk incidents linked to the company within the period of consideration, which may or may not also be linked to individual projects the company owns or operates.

The geospatial Proximity Levels are represented as binary variables, indicating whether a project is within the given range. For companies, this binary variable indicates whether the company owns or operates at least one project within the given range. The full list of factors considered in the regressions is detailed in the appendix.

Descriptive analysis

Descriptive analysis of the dataset yielded insights into the associated risk posed to a project or company by operating near different types of environmentally sensitive sites. The International Union for Conservation of Nature (IUCN) Management Categories classify protected areas, ranging from protected areas with sustainable use of natural resources (category VI) to strict nature reserves (category Ia). Shown in Figure 2, the data suggest that operating near some IUCN Management Categories, like national parks or monuments, is associated with a higher degree of risk to the project. Generally, categories Ia and Ib aim to have stricter protection with largely unmodified areas, and the results indicate that fewer incidents are recorded in relation to these sites. While 42% of projects operating in range of the lower IUCN Management Categories are active, only 26% of projects operating in range of the higher IUCN Management Categories (IUCN Ia and Ib) are active.

Figure 2. Risk of environmental incident by proximity and IUCN Management Category

Three important designations of protected areas are explored in Figure 3. Notably, activities in immediate proximity to a World Heritage Site (natural or mixed) are associated with the highest possibility of a project experiencing legal, financial, or reputational issues. Closer proximity is linked to a higher risk for projects operating near World Heritage Sites and Ramsar sites (wetlands of international importance). UNESCO-MAB (Man and the Biosphere Reserves), locations recognized to promote sustainable development through conservation and sustainable use, on the other hand, show no pattern with respect to project proximity.

Figure 3. Risk of environmental incident by proximity and designation

Alliance for Zero Extinction sites are Key Biodiversity Areas which are home to 95% or more of the remaining population of one or more species listed as Endangered or Critically Endangered on the IUCN Red List of Threatened Species. Operation closer to these sites is associated with higher environmental risk to the project.

Figure 4. Risk of environmental incident risk by proximity to Alliance for Zero Extinction site

Regression model results: project-level

The project-level regression results show that immediate (within 1 km) proximity to an environmentally sensitive site had a significant effect on the environmental risk of a project, with this degree of proximity associated with a 30% increase in environmental risk incidents above projects more than 30 km removed from these sites. Proximity to certain types of environmentally sensitive sites also indicates higher environmental risk, with World Heritage Sites and Alliance for Zero Extinction sites associated with a 36% and 35% increase in risk incidents, respectively.

Table 1. in the Appendix provides the full regression results for projects.

Regression model results: company-level

The company analysis serves as a robustness check to confirm the assumption that project risk spills over to the company. At the company level, ownership or operation of a project within immediate proximity to an environmentally sensitive site was found to have a significant effect on the environmental risk of a company, associated with a 77% increase in environmental risk above companies with no such spatial risk exposure for publicly traded companies, and 27% for private companies. Close and general proximity were not found to be significant at the company level.

As the model separately accounts for whether a company is publicly traded or not, the results indicate there is a gap in risk shouldered by public and private companies that are associated with immediate proximity to environmentally sensitive sites. Public companies are not only more exposed to ESG risk overall, but also tend to be held to a higher standard when it comes to biodiversity risk.

Whether a company is considered large, owning, or operating more projects than 90% of the companies, was also found to have a significant relationship with environmental risk. Larger public companies are more likely to manage more projects and be more exposed to public criticism, resulting in more incidents being reported.

Proximity of managed projects to certain types of environmentally sensitive sites also indicates higher environmental risk for owners and operators. World Heritage Sites are associated with a 69% increase in the risk response, and Alliance for Zero Extinction sites a with a 58% increase.

Table 2. in the Appendix provides the full regression results for owners and operators.

Implications for companies and investors

Biodiversity loss and related impacts are often viewed as indirect risks, however association with these issues has serious reputational implications for businesses today. Using RepRisk’s corporate risk and Geospatial datasets we examined the relationship between the placement of extractive projects and their environmental risk.

For projects, we observed that immediate proximity to environmentally sensitive sites is associated with a higher degree of risk. The level of risk increases for certain categories of protected areas and Key Biodiversity Areas such as UNESCO World Heritage Sites and Alliance for Zero Extinction sites. These results were echoed at the company level, with additional factors like company size and whether the company is publicly listed playing a role. The company level results indicate a relationship between company involvement in projects that pose a risk to biodiversity and higher overall environmental risk exposure. While the risk incidents at the company level are not necessarily linked to individual projects, this finding could be interpreted as companies associated with projects with known proximity risk tend to have higher environmental risk exposure overall. As such, owning or operating one of these geospatially high-risk projects could serve as a red flag that the company is failing to manage environmental risks and point to wider environmental negligence.

In short, these results indicate that proximity matters: the operation of an extractive project within the immediate range of an environmentally sensitive site is associated with significantly greater environmental risk. For investors and companies who own, finance, and invest in extractive projects, these results can support informed decision-making by guiding investors and companies away from the riskiest project placements, reducing damage to both nature and bottom-line. This general guidance, paired with detailed project- and company-specific business conduct risk data can support due diligence processes and shape investment decisions.

For projects, the following factors were considered:

• The project is within immediate proximity of at least one environmentally sensitive site
• The project is within close proximity of at least one environmentally sensitive site
• The project is within general proximity of at least one environmentally sensitive site
• The project is in the oil and gas sector
• The project is in the mining sector
• The owner is listed (public)
• The project is within immediate proximity of at least one UNESCO World Heritage site
• The project is within immediate proximity of at least one Alliance for Zero Extinction site

For companies, the following factors were also considered:

• The company owns or operates at least one project within immediate proximity of at least one environmentally sensitive site
• The company owns or operates at least one project within close proximity of at least one environmentally sensitive site
• The company owns or operates at least one project within general proximity of at least one environmentally sensitive site
• The company owns more than 7 projects, representing the top 10% of companies in terms of number of projects owned.
• The company operates more than 4 projects, representing the top 10% of companies in terms of projects operated.
• The company owns or operates at least one project with immediate proximity to an Alliance for Zero Extinction site
• The company owns or operates at least one project with immediate proximity to a UNESCO World Heritage site

Table 2. Project-level results: Environmental incident count (severity-weighted, log)

Coefficients have been transformed (e^(b)-1) for easier interpretability (e.g., Per regression model III, on average, a project in immediate proximity to an environmentally sensitive site sees 30% increase in environmental risk compared to projects more than 30 km removed from these sites). Standard errors are reported below the coefficients. The dataset contains 3447 unique projects; the N=4129 observations in the regression are unique project-company (operator or owner) pairs as the regression includes company-level characteristics such as listing status.

Table 3. Company-level results: Environmental incident count (severity-weighted, log)

Coefficients have been transformed (e^(b)-1) for easier interpretability (e.g., Per regression model III, on average, companies who own or operate a project in immediate proximity to an environmentally sensitive site are associated with 12% higher environmental risk over companies who don’t). Standard errors are reported below the coefficients.