Global biodiversity impacts of UK economic activity / sustainable consumption

Figure 1: Area of deforestation worldwide associated with UK consumption annually, 2005 to 2022

Biodiversity loss

LIFE metric

The LIFE (Land-cover change Impacts on Future Extinctions) metric provides estimates of changes in the expected number of global extinctions summed across species. This approach integrates information on species richness, rarity, and past habitat loss. A score of one is equivalent to one global extinction – summed across species. Annual UK consumption of crop commodities in 2022 was associated with an estimated ‘LIFE score’ (species extinctions within 100 years) of 4.08 worldwide (Figure 2); this is 28% lower than the annual estimate for the year 2005 (a LIFE score of 5.6). The annual estimated LIFE score associated with UK consumption in 2022 was 15% higher than in the year 2017 (a LIFE score of 3.5), and 1% higher than in the year 2021 (a LIFE score of 4.0). This year is the first time the LIFE score has been included in the indicator suite; as such we would particularly welcome feedback on this metric.

Figure 2: LIFE (Land cover change Impacts on Future Extinctions) score worldwide associated with UK consumption of crop commodities annually, 2005 to 2022

Figure 3: Predicted regional species loss (global total) associated with UK consumption annually, 2005 to 2022

Figure 4: Species richness-weighted crop area worldwide associated with UK consumption annually, 2005 to 2022

Carbon dioxide (CO2) emissions related to deforestation

Annual UK consumption of crop, cattle-related and timber commodities in 2022 was associated with an estimated 12.7 million tonnes of CO₂ emissions linked to deforestation worldwide, including peat drainage (Figure 5); this is 53% lower than the annual estimate for the year 2005 (27.4 million tonnes of CO₂ emissions). The annual estimated CO₂ emissions linked to deforestation (including peat drainage) associated with UK consumption in 2022 was 16% lower than in the year 2017 (15.1 million tonnes of CO₂ emissions), and 11% lower than in the year 2021 (14.2 million tonnes of CO₂ emissions).

Figure 5: Deforestation emissions (including peat drainage) worldwide associated with UK consumption annually, 2005 to 2022

Water consumption and scarcity-weighted water footprint

The surface and groundwater consumed as a result of production (the ‘blue water footprint’) has been multiplied by a weighting factor representing the availability of water regionally after human and aquatic ecosystem demands have been met to estimate ‘scarcity-weighted blue water use’. Annual UK consumption of crop commodities in 2022 was associated with an estimated 758.3 billion cubic metres of scarcity-weighted blue water use worldwide (Figure 6); this is 9% lower than the annual estimate for the year 2005 (836.8 billion cubic metres). The annual estimated scarcity-weighted blue water use associated with UK consumption in 2022 was 18% higher than in the year 2017 (645.2 billion cubic metres), and 8% lower than in the year 2021 (821.2 billion cubic metres). As geographic location of impact is particularly relevant for this metric, users are reminded that they are able to view breakdowns of the results via the associated dashboard.

Figure 6: Scarcity-weighted blue water use worldwide associated with UK consumption annually, 2005 to 2022

Cropland area harvested

Annual UK consumption of crop commodities in 2022 was associated with an estimated total land use footprint of 16.7 million hectares worldwide (Figure 7); this is 19% lower than the annual estimate for the year 2005 (20.7 million ha). The annual estimated total land use footprint associated with UK consumption in 2022 was 12% higher than in the year 2017 (14.9 million ha), and 4% lower than in the year 2021 (17.5 million ha).

Figure 7: Cropland area harvested (land use footprint for crop commodities) worldwide associated with UK consumption annually, 2005 to 2022

Material consumption (tonnes of biomass production)

Annual UK consumption of crop, cattle-related, and timber commodities in 2022 was responsible for an estimated 135.8 million tonnes of material (biomass) production worldwide (Figure 8); this is 17% lower than the annual estimate for the year 2005 (163.5 million tonnes). The annual estimated material production associated with UK consumption in 2022 was 7% higher than in the year 2017 (126.7 million tonnes), and 5% lower than in the year 2021 (143.6 million tonnes).

Figure 8: Crop production (biomass) worldwide associated with UK consumption annually, 2005 to 2022

Relevance

The Government’s 25 Year Environment Plan (25YEP) and its first revision, the Environmental Improvement Plan 2023, recognise the need to measure the environmental risks and impacts associated with UK consumption.

International/domestic reporting

This indicator feeds into the Outcome Indicator Framework, a set of indicators describing environmental change related to the ten goals within the 25 year Environment Plan. As part of the Outcome Indicator Framework, this data contributes towards the evidence base used to prepare the annual progress report for the Environmental Improvement Plan. This indicator contributes to OIF indicator K1: Global environmental impacts of UK consumption of key commodities.

The UK Biodiversity Indicators have been reviewed in response to the latest goals and targets agreed under the Kunming-Montreal Global Biodiversity Framework (GBF) of the Convention on Biological Diversity (CBD). The indicators are being adapted, with some new indicators being developed, to better align them with the monitoring framework for the GBF. The suite of biodiversity indicators is therefore expected to change ready for forthcoming UK national reports to CBD in 2026 and 2029.

Web links for further information

• Bird species distribution maps of the world Version 2020.1
• Environmental Improvement Plan
• ‘A Green Future: Our 25 Year Plan to Improve the Environment’
• Official Statistics UK Carbon Footprint
• Commodity footprints dashboard
• EXIOBASE (NTNU University of Science and Technology)
• Food and Agriculture Organisation of the United Nations (FAO)
• The IUCN Red List of Threatened Species
• MapSPAM
• Official Statistics UK Material Footprint
• UKRI Trade Hub
• UN Comtrade webpage
• Water Footprint Network
• Water Use in Life Cycle Assessment (WULCA) – Download AWARE Factors

Methodology

The full methodology for this indicator can be found in the technical documentation. This indicator was previously published as an Official Statistic in Development to facilitate user involvement in its development – information on how the underlying data have been obtained and how the indicator has been prepared is available in the technical documentation.

The indicator is based on multi-regional input-output (MRIO) modelling. MRIOs model global trade flows representing the monetary inputs and outputs across different countries and territories and their commercial sectors (for example, oilseeds, cattle farming, and paddy rice). The MRIO data used for this indicator are from EXIOBASE. The EXIOBASE data set has been selected due to its considerable long-term annual coverage, and to increase alignment and consistency with other UK footprint accounts (the UK Carbon and Material Footprints). Additionally, it allows for data to be broken down into a high number of different sectors compared to other MRIO datasets (however, note that EXIOBASE country and territory data are more aggregated than in some MRIO data; therefore a separate dataset available via the associated dashboard which is not part of this UK data release has been published using data from the Global Trade Analysis Project (GTAP) to provide an alternative that may be more appropriate for certain countries and territories).

The MRIO data are hybridised (linked) with physical data (tonnes of each commodity) from the FAO and, in the case of timber, UN Comtrade, using the Stockholm Environment Institute’s IOTA (Input Output Trade Analysis) model (Croft et al., 2018). This step allows for a higher resolution breakdown of commodities (for example, palm oil and soybeans, instead of just oilseeds) and of countries and territories of origin than MRIO data would provide alone. This step also allows trade – in physical units – at a commodity level to be included before these data are integrated within the sector-level MRIO framework. This gives results at a greater level of product-specificity, and accuracy, than a standalone MRIO-based account. The FAO Statistics were chosen for use because they are a comprehensive set of global production statistics, collected from official national statistics of each country and territory, which can be easily incorporated into the modelling framework. For timber, UN Comtrade data were used instead due to the discontinuation of FAO timber trade data. The modelling framework allows for an estimation of the country or territory of origin of a commodity, accounting for cases where commodities are embedded within other products as an ingredient or input, and cases where commodities are re-exported through multiple countries or territories before the point of consumption in the UK.

To determine deforestation rates and carbon dioxide (CO₂) emissions from deforestation, data from Chalmers University of Technology which link deforestation and commodity production (Singh & Persson, 2024) are used to proportionally attribute UK deforestation impacts based on the volumes of each commodity the UK consumes within each production country or territory. For example, if the Singh & Persson (2024) dataset links x hectares of deforestation in a given country or territory with the production of a particular commodity, and the UK consumes y% of that commodity, then it is assumed that the UK is responsible for y% of those x hectares. This deforestation dataset was selected as it provides data on deforestation and its associated agricultural commodity drivers, with comprehensive coverage. Positive change would likely be represented by a reduction in deforestation rates and CO₂ emissions from deforestation but would need to be interpreted alongside additional information (such as remaining forest area, deforestation per tonne of production, and understanding of the drivers behind the change) for a robust assessment.

Although in the 2021 and 2022 releases of this indicator only tropical and subtropical deforestation were reported, the estimates for area of deforestation in this indicator now refer to deforestation taking place globally, including in temperate and boreal regions.

To estimate biodiversity loss, three separate methods are utilised. The first method, the LIFE score, is based on data and methods from Eyres et al. (in press) and, Ball et al. (in review). It provides quantitative estimates of the marginal changes in the expected number of extinctions (both increases and decreases) caused by converting remaining natural vegetation to agriculture, or by restoring farmland to natural habitat. This approach integrates information on species richness, endemism, and past habitat loss to estimate the impact on extinctions of land cover change. Positive change would be represented by a decrease in the total expected number of global extinctions linked to UK consumption. The second method uses crop- and country and territory-specific characterisation factors, provided by Chaudhary and Kastner (2016), which are used to estimate the impact per tonne of production for 152 crops or crop groups in 171 countries/territories. This gives an estimate of regional species loss (the number of species predicted to be committed to extinction if current conditions continue per ecoregion). Positive change would be represented by a decrease in predicted species loss. The third method to estimate biodiversity loss (providing separate results from the other methods) uses MapSPAM data (a modelled global data set providing information on which crops are grown where) alongside species richness information from the International Union for the Conservation of Nature (IUCN) and BirdLife International to estimate ‘species richness-weighted extent of crop production’. This represents the hectares of crop production multiplied by the number of species ranges overlapping that hectare, and therefore where there is overlap between production and areas of biodiversity importance. Positive change would be represented by a decrease in ‘species richness-weighted extent of crop production’ embedded in consumption.

Water consumption was estimated from the Water Footprint Network. To account for water scarcity in regions of production, blue (irrigated) water consumption was multiplied by a weighting factor representing the water availability in a region after human and aquatic ecosystem demand has been met, using conversion factors sourced from Boulay et al. (2018). Positive change would be represented by a reduction in scarcity-weighted blue water use.

Caveats and limitations

For accurate interpretation of the results presented within this indicator, it is necessary to understand the following caveats:

• Data tracing all commodities exactly back to their countries or territories of origin are not publicly available. Whilst based on empirical statistics, the outputs produced by this indicator are derived from modelling so should be considered as estimates rather than exact values.
• Only the country or territory of origin, and not the exact location of origin, can be obtained from the current version of the indicator as only national scale data were used. This means impacts are based on average production practices per country or territory, not the actual impacts at the exact location the product came from. This could be improved by using sub-national data (where available) in subsequent iterations of the indicator.
• Data linking impacts to trade are compiled at the national level, meaning any action by the UK in specific regions will be ‘averaged’ across the full global supply chain. Therefore, it will be hard to differentiate UK action from the actions of other consumer nations. The indicator will be more responsive to multi-national/-lateral action than to UK action specifically.
• The objective of this dataset is to identify impacts that are geographically specific, so linking consumption to production location is key to understanding these impacts. An MRIO is therefore linked to additional data through a modelling framework to estimate consumption via production location. We use EXIOBASE for this, so that the financial flows (the picture of the whole economy) align with the UK Material and Carbon Footprints. However, each of these is measuring something that does not vary with location and so does not require this additional step (for example, estimates of carbon from consumption are based on factors applied to the production process, not the location of the process). Hence there are limits to how much this set of indicators and the Material and Carbon indicators can be directly compared.
• Lags in the underlying datasets mean that data are only available up to 2022 in the current release. Care should be taken in analysing trends over time which can reflect complex changes in production volume, trade distributions, estimated inter-sectoral demands and final consumption expenditure.
• Estimates of the environmental impacts of consumption reported by other sources could differ from results reported here due to differences in the underlying data and methods. Use of different MRIOs as the dataset underlying this indicator (for example, GTAP rather than EXIOBASE) could also lead to differences in results, due to factors including geographic and sectoral resolution, temporal coverage and lag. Note that a separate dataset (not forming part of the UK data release) using GTAP has been published on the associated dashboard website, which will allow for a wider variety of countries and territories to have access to data on their consumption impacts.
• The presence of oil palm and soy as key sources of UK deforestation risk is common to other assessments, but the data reveal UK linkages to other supply chains (such as beans, cassava, paddy rice) that are not often considered as ‘deforestation risk’ commodities in other publications. These supply chains warrant further investigation to understand which sectors of consumption link to these estimated impacts.

In many of the countries and territories where commodities are driving rapid impacts, the UK represents a small proportion of the total demand. However, the underlying dataset allows the proportion of demand to be identified and the other consumers also contributing to the demand, hence providing better opportunities to work with producer countries and territories and to work multi-laterally with other consumer countries and territories to address sustainable production and reduce impacts.

To discuss the data and methodology, caveats, limitations and uncertainties associated with this indicator, the development team at the Stockholm Environment Institute, University of York, can be contacted at info@commodityfootprints.earth.

Additional notes

• All crops with data recorded in FAOSTAT are included (see technical documentation for further detail and exceptions).
• “Cattle-related commodities” refer to meat, offal, fats and hides from cattle and buffalo as reported within FAOSTAT, and aligning with data usages within Singh & Persson, 2024. Note that impacts attributed to cattle result from land used for pasture, whilst impacts from commodities used as feed are presented as impacts associated with the raw commodity (for example, soy).
• Underlying datasets currently limit analysis to the years 2005 to 2022. See the technical documentation for further information about planned data updates for each underlying dataset.
• Characterisation factors are used as estimates of an environmental impact per unit of stressor (for example, the land use per kg of production or the biodiversity loss per unit of pollutant).

Recent developments

This year’s release includes a new biodiversity loss metric – the LIFE score metric. This provides a more nuanced understanding that integrates information on species richness, endemism, and past habitat loss to estimate the impact of land cover change on extinctions, rather than simply relying on estimates from species area curves or overlaying land use with expected species richness. This has built on work undertaken by the UKRI Trade Hub.

Deforestation and associated emissions data are also based on an updated underlying deforestation dataset, which provides greater spatial resolution for a number of crop-specific data layers and improved methods relating to the masking of plantation forestry and the estimation of carbon emissions.

Further detail on recent developments can be found in Appendix 3 of the technical documentation.

Work has also been undertaken on a number of developments that are related to, but planned to be published separately from this official statistic release, due to their more experimental nature. These include an initial dataset on the material footprint of metal and mineral commodities (which could, at a later date, be extended to include environmental impact metrics) and an initial dataset integrating finer resolution trade data for certain commodities.

Development plan

This indicator was previously published as an Official Statistic in Development. In line with our commitment to the Code of Practice for Statistics, we are setting out a plan for developing these statistics further. This plan outlines what we hope to do to improve our publications in future. This plan is informed by responses from users and stakeholders, as well as the UK Biodiversity Indicators Steering Group. We are keen to hear feedback from users of these statistics, please send your feedback to: biodiversity@defra.gov.uk.

We will look to update our development plan at least annually.

Developments planned for the next statistical release to be published in 2025:

• Scoping work for the addition of agri-chemical related metrics, such as nitrogen and phosphorous pollution. The current impact metrics are mainly associated with areas of new and expanding production. However, many impacts come from intensification of current production. Including metrics related to this will be important. Work this year will scope out whether it is possible to add such data in the 2025 release, or further into the future.

• Intercomparison of datasets available. We are working to understand how the use of alternative datasets affects end results, and what this can tell us about uncertainty. We are investigating ways that this could be presented as part of the dataset, such as by incorporating data flags or reporting on a range of possible values instead of having a single number for each data point.

• Metals/minerals data. Before the 2025 release, we plan to publish some initial and interim data on the material footprint of metals and minerals. This will be in the form of a JNCC report, rather than part of the formal official statistic, as it will be based on different methods and so will not be comparable.

• Assessment of value and use. Before the 2025 release, we plan to publish a report summarising the findings of a survey and a series of usability testing sessions that we have run this year, to report on the value and use that people see in the indicator and understand what further improvements could be made.

Longer term development plans:

This work will take slightly longer and will be part of post-2025 statistical releases.

• We are currently in discussions about funding for longer term development plans. Ideas include continuing work on the metals and minerals data to include environmental impact data, producing data about sector-linked demand (e.g., the breakdown of the data by government spend vs non-government spend), scoping work to include social impact metrics, the addition of data about land use change beyond deforestation, and the addition of marine commodities.