Comparing biodiversity offset methodologies

Divergence in securing ‘no net loss’

There are at least 45 ‘biodiversity offset’ programmes and policies currently in place around the world. The common fundamental principle – that unites what is, in fact, a disparate array of related policies – is the need to ensure that there is ‘no net loss’ of biodiversity alongside economic development. That is to say, biodiversity gains achieved through offset interventions should, as a minimum, be greater than the residual biodiversity losses caused as a result of development impacts for a given project.

Despite the flat, open and rather bleak nature of the Ustyurt landscape, it hosts all kinds of species - such as this rather shy species of desert hedgehog, who reluctantly posed for photos after we fed it a hardboiled egg. (Photo by Joe Bull)

Despite the flat, open and rather bleak nature of the Ustyurt landscape, it hosts all kinds of species – such as this rather shy species of desert hedgehog, who reluctantly posed for photos after we fed it a hardboiled egg. (Photo by Joe Bull)

Whilst this concept of no net loss might seem relatively straightforward in theory, actually achieving it is something different entirely. One key challenge, assuming that the negative impacts of development are known with some degree of accuracy (which is certainly not always the case), is then deciding how to calculate the requisite biodiversity gains that would result in no net loss.

The various policies that do exist have established different methodologies that tell you exactly how large your gains need to be. Since the basic goal of all of these methodologies is the same – that is, no net loss – one might hope that they would give similar answers if they were applied to a common case study. Well, we tested this approach and it turns out they don’t (Bull et al., 2014). This analysis highlights how different the philosophy behind biodiversity offsetting in different countries can be.

Different approaches, same case study 

The comparison analysis came about because we had collected data on the habitat impacts of oil and gas development for a region in northwest Uzbekistan (discussed in Decision Point #17). A separate initiative is underway to develop a biodiversity offset policy for Uzbekistan, and colleagues from the United Nations Development Programme and state institutions were asking us what biodiversity offset methodology we would recommend (ie, how to calculate the requisite habitat restoration gains). In response to this, we decided to see what would happen if we applied a range of biodiversity offset calculation methods to the same Uzbek case study.

As offset methodologies are designed for different habitats in different countries, we focused on the general principles underlying each methodology. We chose to apply the following to Uzbekistan: US wetland banking, US conservation banking, Victorian native grassland compensation, a version of the Victorian native grassland compensation specially adapted to the Uzbek landscape, Canadian fish habitat compensation, and the relatively new UK biodiversity offset method. The details of how we applied each different approach, and estimated the impact on fauna and flora resulting from the oil and gas activities, are contained within the paper itself and supplementary materials.

The take away message is that applying different methods for calculating the required offset activities resulted in highly divergent outcomes for biodiversity (which was expressed as habitat condition x area, or ‘weighted area’ in Table 1). The differences are even starker if you calculate net biodiversity outcomes over time – we assumed a 40 year period, which is the same length of time as these oil and gas reserves have been exploited (Figure 1).

Table 1: Comparison of aggregated offset requirements across different methodologies, expressed as the weighted area within which conservation actions must be applied, under a static appraisal of 40 years of oil and gas development. Uncertainty represents the potential range in spatial extent of oil and gas infrastructure. Saiga habitat is, for context, the area of a proposed reserve available for restoration of habitat condition under species-based offsetting.

Table 1: Comparison of aggregated offset requirements across different methodologies, expressed as the weighted area within which conservation actions must be applied, under a static appraisal of 40 years of oil and gas development. Uncertainty represents the potential range in spatial extent of oil and gas infrastructure. Saiga habitat is, for context, the area of a proposed reserve available for restoration of habitat condition under species-based offsetting.

Explaining divergence 

There are various reasons for the divergence in outcomes (as seen in Figure 1). One is that some offset methodologies have multipliers incorporated into the basic metric, which require gains to be multiplied by some factor to account for restoration uncertainty and time lags (eg, the UK metric) whereas others do not (eg, US wetlands).

Figure 1: Plot of net weighted area of land at benchmark condition (in km2) against time (in years) resulting from hypothetical offsets in Uzbekistan, using different methodologies. The Canadian method applied to species and US Conservation methods are exactly aligned, and represent ‘out-of-kind’ offsetting here. Upper and lower bounds reflect uncertainty in both estimation of impacts and, for the lower bound, the possibility of up to 50% non-compliance.

Figure 1: Plot of net weighted area of land at benchmark condition (in km2) against time (in years) resulting
from hypothetical offsets in Uzbekistan, using different methodologies. The Canadian method applied to
species and US Conservation methods are exactly aligned, and represent ‘out-of-kind’ offsetting here. Upper
and lower bounds reflect uncertainty in both estimation of impacts and, for the lower bound, the possibility of
up to 50% non-compliance.

A second is that some metrics are highly prescriptive (eg, Victorian grasslands) whilst others are almost completely open to interpretation (eg, Canadian fish habitat – although note that this policy has recently been revised). This means that methods such as the one used in Victoria are strongly tied to Australian habitats, which in turn makes them relatively hard to transfer to a different set of habitats without adaptation – a finding that would seem trivial, were it not for the fact that there are instances of decision makers attempting to transfer the Victorian method wholesale to completely different regions.

A third reason for the divergence is that some of the methodologies permitted exploration of ‘out-of-kind’ offsets (eg, the application of the Canadian method to fauna): we allowed vegetation losses to be compensated with gains in habitat for priority fauna species, and calculated net outcomes using a rough equivalency scale. Although this analysis was illustrative only, it does provide at least some support to the argument for considering out-of-kind biodiversity offsetting.

These reasons for the divergence in outcomes link back ultimately to the philosophy behind offsetting, and how this varies between different jurisdictions.” 

The consequences of divergence 

All of these reasons for the divergence in outcomes link back ultimately to the philosophy behind offsetting, and how this varies between different jurisdictions. Whilst some societies might allow out-of-kind offsetting for practical reasons, others might simply deem it inappropriate or even potentially immoral; part of the reason for the recent fuss about offsetting in the UK media.

Alternatively, some social or institutional contexts might favour a more prescriptive approach (I get the impression from colleagues there that this would be the case in Uzbekistan), whereas others value the freedom to interpret policy individually.

In any case, the findings highlight that the method used to quantify losses and gains strongly influences the biodiversity outcomes of offsetting, which is interesting in itself. In turn, this may imply that offsets generated using different methodologies are not directly transferable between jurisdictions. The latter conclusion is not purely academic, as there have been mumblings for some time now about the potential for trans-jurisdictional (eg, international) biodiversity offset credit trades.

Oil and gas activity results in a variety of known impacts upon the Ustyurt habitat, including clearance of vegetation, disturbance to fauna, pollution of soils, increased natural resource use (particularly water, which is in short supply), and facilitation of poaching. However, it also provides sorely needed employment and income to those living in the region. (Photo by Joe Bull)

Oil and gas activity results in a variety of known impacts upon the Ustyurt habitat, including clearance of vegetation, disturbance to fauna, pollution of soils, increased natural resource use (particularly water, which is in short supply), and facilitation of poaching. However, it also provides sorely needed employment and income to those living in the region. (Photo by Joe Bull)

Again, the fundamental point of biodiversity offset policies is generally to ensure ‘no net loss’. But it is far from clear, or at least far from clearly stated, what exactly different jurisdictions are trying to achieve no net loss of. In fact, a lack of clarity and definition over what we actually want to conserve – species diversity, habitat diversity, ecosystem functions, ecosystem services, ecosystem stability or resilience, some or all of the above – is a wider challenge for conservation science.

By designing methodologies to achieve even a seemingly objective goal such as ‘no net loss’, we inevitably encode subjective values into the policy. This is perhaps not always recognized. Analyses such as ours highlight the importance of this topic.


More info: Joseph Bull j.bull10@imperial.ac.uk 

References 

Bull JW, EJ Milner-Gulland, KB Suttle & NJ Singh (2014). Applying different biodiversity offset calculation methodologies to a case study in Uzbekistan. Biological Conservation 178: 2–10.

 http://www.sciencedirect.com/science/article/pii/S0006320714002663

Leave a Reply