Biodiversity offsets: possible methods for measuring biodiversity losses and gains for
use in the UK
Jo Treweek CEnv, MIEEM and Bill Butcher, Treweek Environmental Consultants
Helen Temple, The Biodiversity Consultancy
There is increasing interest in biodiversity offsetting to achieve no net loss (or preferably a net
gain) of biodiversity, when development might otherwise result in residual adverse effects
despite appropriate avoidance and mitigation measures. There are now laws requiring
enhancement policies. Development of similar systems in the UK and the EU is being explored
(Defra, 2009; EU, 20101), making it important to consider the risks and opportunities presented
by biodiversity offsetting. There is a growing body of theoretical and practical experience to
draw on: an extensive overview can be found on the Business and Biodiversity Offset
In determining what constitutes a reasonable and fair offset, we need to consider the extent to
which losses (due to impacts) and gains (due to offsets or compensation) balance out to
as a minimum. An essential ingredient of a robust biodiversity offset
system is therefore a credible method for measuring biodiversity losses and gains. This is the
main focus of this paper.
The challenge is to develop methods which will be straightforward to apply in practice without
sacrificing the precision required to ensure that all important aspects of biodiversity are
captured. In the absence of an agreed, universal measure of biodiversity and for entirely
pragmatic reasons, most offset systems are habitat-based. Compared with species populations,
habitats are relatively stable over time, can be adequately described with fewer types and are
normally used as the primary focus of biodiversity conservation.
Recognising that some important ecological attributes will not be adequately reflected using
habitat as a surrogate (population decline caused by traffic collisions, for example), we suggest
a habitat-based approach to measuring impacts and determining offset requirements which
might lend itself to the UK situation. This draws on a version first published in Defra (2009) and
companies. It is based on hectares of habitat of particular type3 or intrinsic
adjusted to account for differences in condition (which in UK habitats is often closely linked to
suitability of management).
system and the requirements of the UK BAP (1995) and should be seen as a starting point for
purposes of discussion and debate rather than a recommended final solution. For purposes of
illustration we present a hypothetical worked example, but rigorous testing would be required to
ensure that the approach At the end of the paper
we identify some potential pitfalls and issues that might need further consideration.
3 IHS is suggested because it encompasses all UK terrestrial, freshwater and marine habitats, including European
and BAP habitats (www.ihs.somerc.co.uk). It is also now widely used at local and regional scales for mapping and
collating habitat data recorded in other classifications (e.g. Butcher, 2008; SERC 2007).
Some key principles and assumptions
There are some circumstances in which use of offsets will never be appropriate: offsets should
not be used for impacts on biodiversity which needs to be conserved in situ for it to survive, for
example. In the UK, offsets might therefore be inappropriate for any BAP habitat with
habitat which takes so long to establish and mature that it is effectively
irreplaceable in any reasonable human timeframe, such as ancient woodland or raised bog.
Another key concept in biodiversity offsetting is additionality, or the requirement for offsets to
deliver conservation outcomes which can be shown to be additional to those that would have
occurred anyway, or which are the responsibility of statutory bodies to deliver. Additionality
might be achieved by protecting lowland deciduous woodland that would otherwise have been
rapidly destroyed, or by creating a diverse habitat on intensive arable cropland. On the other
hand, there would be no additional value in buying woodland that was not under any immediate
threat and then doing nothing with it. Even if the woodland improved in quality over time, for
example through succession, this could not be claimed as an outcome or benefit of the offset
Biodiversity offsets can't be used to solve every conservation challenge and should be used
intelligently in conjunction with other conservation policies and tools. We have therefore
assumed that any UK offset system would have to complement existing requirements for
ecological compensation under the Habitats Directive and would only apply to losses of habitat
occurring outside the Natura 2000 network.
Possible approach/ framework
habitat which will be exposed to an impact and also to land which might be used for an offset.
It is possible to draw on established methods to assess habitat condition, such as those used on
nationally designated sites (Natural England, 2008), but there are no universally agreed
methods for assessing levels of biological distinctiveness. A consultation exercise is currently
underway through the auspices of the Natural Capital Initiative4 to test the extent to which
consensus can be reached if ecologists assign UK habitats to distinctiveness categories a priori
and without in depth assessment on a case-by-case basis, but this is an area where further
research is likely to be necessary.
A potential scoring system is set out in the offset matrix shown in Figure 1. We propose a four
point scale from 0 to 3 in which a score of zero would be assigned
to e.g. as applied by Kyläkorpi et al., 2005) and a score of 3 to
BAP and Annex 1 habitat categories (EU Habitats Directive5). For current purposes we have
doubled the distinctiveness score in the matrix to account for the fact that intrinsic biological
distinctiveness is a more fundamental and less alterable property than current condition. Again,
this requires testing to ensure that reasonable outcomes result in practice. Scores can be
normalised on a scale of 0 to 1 as shown in square brackets. A score of 1 results in cases
where habitat with high distinctiveness and optimum condition is affected.
5 Council Directive 92/43/EEC of 21 May 1992 on the conservation of natural habitats and of wild fauna and flora
Figure 1: Offset scoring matrix
Very Low (0)
The area of habitat to be lost, multiplied by the score from the matrix gives the credits, or
required for the offset. If several habitat types are present, the assessment must
be repeated for each one and the results summed to give the overall offset requirement.
offset must deliver an overall ratio of 1:1 (or better) when offset
gains are compared with the predicted losses due to development. To quantify gains on
potential offset land the final predicted outcome in terms of area x matrix score must be
compared with the baseline or starting condition of the land to be used.
Generating measurable biodiversity gains
In a system such as the one proposed here, the main ways to generate measurable biodiversity
gains are by improving condition of a particular habitat (e.g. by bringing a degraded lowland
heathland into appropriate management) or by elevating distinctiveness category (e.g. by
converting a Category 2 grassland to a Category 3 grassland
Lowland Meadow BAP habitat).
Gain on the distinctiveness scale involves a movement to the right in the matrix (e.g. a low level
of distinctiveness to a high level), while gain on the condition scale involves an upwards
movement (e.g. improvement in condition from poor to good). Use of the matrix to determine
offset requirements means that every offset should contribute to requirements under the UK
1. Achieve condition in existing (priority) BAP habitat.
2. Restore relict/degraded habitat to a BAP habitat type.
3. Expand area of BAP habitats through creation effort.
A hypothetical worked example
Figure 2 shows 14 ha of land near Aberdeen which has been mapped using IHS (NESBREC,
2007) and is due to be lost as a result of a hypothetical development proposal. Hypothetical
condition scores have been assigned to habitat parcels.
Figure 2: habitat types and condition on land to be affected by the development
Figure 3 summarises the losses that will occur and the credits required.
Offset credits or
Dry heath (BAP)
Dry heath (BAP)
Total area of
Figure 3: Credits or habitat units to be lost
Figure 4 shows 32 hectares of land, also near Aberdeen. In our hypothetical example this area
has been offered as a potential offset and we need to decide whether it is able to provide the
required credits or habitat units.
Figure 4: habitat types and condition on potential offset site
Figure 5 summarises the habitats on the potential offset area and their baseline levels of
distinctiveness and condition. The offset needs to deliver gains commensurate with the losses
identified in Figure 3. There are various options for achieving this, the most obvious being
through gains in condition.
HE1 (Dry Heath)
EM4 (Purple Moor
Grass and Rush
GA1 (Lowland Acidic
Non BAP habitat
Non BAP habitat
Non BAP habitat
Figure 5: Baseline assessment of potential offset
Impacts should be offset by actions on specific parts of the proposed offset area, with the final
status of the offset land always equal or better on both distinctiveness and condition scales than
the corresponding impacted land. This means that losses of BAP habitat extent or condition
would have to be offset through gains in extent or condition of the same or a different BAP
habitat. On the other hand, loss of extent or condition in a non-BAP habitat could be offset
through gains in extent or condition in a BAP habitat.
Figure 6 summarises potential enhancements, based on the assumption that all land will be
maintained at, restored to or created as BAP habitat (thus scoring high on the distinctiveness
scale); and that the offset must result in achievement of when it is
mature (suggested rules which are open to debate). There may be a time lag involved in
achieving the required condition, hence the likely need to consider using multipliers in any
eventual UK system (beyond the scope of this paper).
Action to deliver gain
(change from poor to
good condition in
(change from poor to
optimum condition in
HE1 (Dry Heath)
EM4 (Purple Moor Grass
and Rush Pastures)
GA1 (Lowland Acidic
GA0 (Acidic Grassland)
Restore BAP habitat
Expansion (create BAP
G10 (Improved Grassland)
Expansion (create BAP
Figure 6: options for delivering gains
Offset options vary. Focusing on one habitat only, a
the Dry Heath or creating BAP habitat in optimum condition on the Improved Grassland could
deliver the required gain of at least 6.73 credits/ habitat units (see Figure 3). Another option
might be to achieve optimum condition on EM4, and GA1 and create BAP habitat in optimum
condition on CR0 and G10, resulting in a composite offset delivering 7.28 habitat units.
Most biodiversity offset systems in use worldwide include rules relating to allowable exchanges
between habitats when determining offset requirements. Most h-assumption
for replacement of lost habitat, but allow offsets based on different types provided that these are
In our example a similar rule might be to
require the offset to deliver the same habitat as the impacted one, unless there is gain on the
distinctiveness scale. This
ation, to offset the loss of 5.8ha. Creation of other BAP habitat (for example additional
Purple Moor Grass and Rush Pasture) would only be possible once losses of Dry Heath had
Possible unintended consequences and issues requiring further consideration
This paper has focused on just one aspect of biodiversity offsets; there are many others which
will require careful consideration and testing in practice before a robust system can be assured.
Some issues requiring further consideration are outlined here, as well as some possible
unintended consequences of the proposed approach.
Based on review of methods in use worldwide, we conclude that it would be possible to develop
a workable method to assess losses and gains as part of an offset system which could deliver
demonstrable gains in BAP habitat extent and condition. Any UK system for biodiversity offsets
would probably apply only in certain prescribed circumstances, however. It is likely that offsets
would apply to losses of habitat occurring outside the Natura 2000 network and that impacts on
as discussed earlier in this paper. The
method suggested here could apply to any impact (however small) on any habitat (whether
considered important or not), with potential benefits in terms of tackling cumulative impacts
(such as those highlighted in a recent article on development creep in The Guardian 6) but
use for small scale impacts on habitats of low distinctiveness and condition would only be
practicable if a straightforward system for requiring and identifying offsets could be established
to ensure an acceptable bureaucratic load for numerous smaller transactions (a system of
It might be necessary to put safeguards in place to avoid an outcome in which impacts are
always offset by condition enhancement on existing habitat, rather than by gains in
distinctiveness (which are likely to be harder to achieve in practice). This could be avoided by a
rule requiring offsets to include an equivalent area of habitat expansion and/or restoration to
ensure that there is no loss of extent of BAP habitat, where impacts will result in deterioration on
the biodiversity distinctiveness scale.
Another necessary safeguard might be a requirement for the final condition of any offset to be
good or optimum. Clearly further work is required to identify suitable criteria for determining
when such a condition has been achieved and to establish a reasonable timeframe for this. It
may be necessary to establish indicators which can be used to determine whether
implementation of appropriate management does indeed deliver demonstrable gains in
condition and/or distinctiveness, for example. Whatever system is introduced, and whoever
delivers the offsets ("habitat banks" or some other agency), it will be essential to have
independent, trusted auditing or verification.
Recognising that habitat expansion, restoration and condition achievement carry varying levels
of outcome uncertainty, it may be necessary to consider use of appropriate multipliers to ensure
a robustly fair offset. Similarly there may be long delays in achievement of prescribed outcomes.
In some cases it will not be appropriate to replace habitat lost now with the same habitat units in
viability could be compromised by temporary loss of habitat. Multipliers could also play a part
here, but this is a controversial area, beyond the scope of this paper.
The proposed metric (area x condition x distinctiveness) is a general metric that can be applied
to all UK sites/habitats and is fungible. In effect it represents a kind of "lowest common
denominator". Supplementary methods would be necessary to ensure that high-priority
biodiversity features for which habitat is not a good surrogate are appropriately measured and
offset. Risk of impacts on habitats with high distinctiveness or the presence of BAP species
might be the trigger for further consideration of this kind.
If a biodiversity offset system allows impacts on one habitat type to be offset through actions to
enhance another (as suggested here), it is important to check that the overall balance of
habitats is maintained with impacts and offsets in place over time and that some habitats do not
gain at the expense of others (perhaps those which are more challenging to restore). This
exchange rules such as the one suggested earlier. A reliable system of strategic spatial
planning might also be important to ensure that offsets are delivered on suitable land or that
opportunities to develop habitat networks are realized. This is an aspect which will require
careful consideration given recent changes in regional planning.
References and web links
Butcher, B (2008) Regional Habitat Data Project, Stage 2. West Midlands Biodiversity
Partnership, unpublished report.
Defra (2009) Scoping study for the design and use of biodiversity offsets in an English Context.
Final Report, Contract NEE 0801,
eftec, IEEP et.al (2010) The use of market-based instruments for biodiversity protection The
case of habitat banking Technical Report to European Commission DG Environment.
http://ec.europa.eu/environment/enveco/index.htm (accessed 08.08.2010).
Kyläkorpi K, Rydgren B, Ellegård A, Miliander S, Grusell E (2005) The Biotope Method 2005. A
method to assess the impact of land use on biodiversity. Vattenfall, Sweden.
Natural England (2008) SSSI condition assessment: A guide for owners and occupiers. NE 61
Natural England, Peterborough.
North East Scotland Biological Records Centre (NESBReC)
http://www.nesbrec.org.uk/modules/biomaps1/index.php (accessed 08.08.2010)
SERC 2007 SERC (2007) Integrated Habitat System Regional Dataset for SE England, South
East England Local Records Centres. Somerset Environmental Records Centre, Wellington.
UK BAP (http://www.ukbap.org.uk (accessed 08.08.2010)
UK BAP (2006). Review of BAP Targets Target Type Definitions, UK BAP 2006