Data accuracy in Ecological Footprint’s carbon footprint

https://doi.org/10.1016/j.ecolind.2019.105983Get rights and content

Highlights

  • EF’s CF data is largely based on estimates.

  • Uncertainty in data makes it impossible to assert that CF is any specific number.

  • This is rarely made clear in EF’s dissemination.

  • Institutions equally focused on research and politics run the risk of bias.

Abstract

Since the UNCED‘s call for the creation of sustainability indicators many such have been put forth in the literature. One of the more successful ones, in terms of popularity, is the Ecological Footprint (EF). Much criticism has been directed at the EF, not least the carbon uptake component (CF). The CF typically makes up around 50% of global EF and is the sole cause for its overshot – i.e. results indicating unsustainable consumption. The aim of this study was to assess the accuracy of the data used for the calculation of CF. The study finds that the data is lacking in accuracy to the point that stating that CF or EF is any given number at any given time is misleading. The reasons for this uncertainty are identified as use of estimates and averages for the calculations as well as discrepancy between data collected locally and data from international databanks. CF or EF results should thus always be prefaced with caveats regarding the uncertainty involved in the estimation. The lack of caveats in EF dissemination is worrying and has led to the most serious criticism of the method to date, that of it fulfilling the criteria for pseudo-science for failing to disclose uncertainties in calculations and results. This study suggests that the reason for this failure may be traced to the Global Footprint Network (GFN) being both a think tank actively promoting the use of EF, and the world’s largest research unit into the methodology. This can lead to uncertainties being down played in dissemination not to confuse current users of the method or dissuade new ones. The study further raises questions about the accuracy of GHG estimates in general since they are often based on the same IPCC default emission factors and activity data as used by the GFN.

Introduction

Human endeavour invariably impacts the natural environment in which we live. Examples from history suggest that when this impact exceeds nature’s inherent regenerative capacity it can lead to substantial changes to the complex web of life that makes up earth’s ecosystems, which is turn can negatively impact people’s quality of life and the robustness of the natural systems (Barnosky et al., 2012, Motesharrei et al., 2014, Roman et al., 2018). This has led to a call for ways to measure human impact on the natural world through so called sustainability indicators (UNCED, 1992).

Quantifying this impact on the environment is not a simple task and many indicators have been created in the past decades (Cobb and Cobb, 1994, Wackernagel and Rees, 1996, Lawn and Sanders, 1999, Hanley, 2000). One of these indicators is the Ecological Footprint (EF) (Rees and Wackernagel, 1994, Wackernagel and Rees, 1996).

The EF falls under a category of aggregate sustainability indicators (van den Bergh and Grazi, 2013). This means that large sets of data are aggregated into one final figure which should indicate the level of sustainability or unsustainability as the case may be. EF has been studied and used extensively in a wide variety of settings from a product level (Frey et al., 2006, Limnios et al., 2009, Hanafiah et al., 2012) to national levels (Haberl et al., 2001, Medved, 2006, Galli et al., 2012, Wang et al., 2012, Salvo et al., 2015, Solarin et al., 2019) and the whole Earth (WWF 2018). EF is supported by a think tank, the Global Footprint Network (GFN), dedicated to its development and distribution. GFN publishes annual accounts of global footprints and that of humanity as a whole in the National Footprint Accounts (NFA).

Since its conception in the early nineties (Rees and Wackernagel, 1994, Wackernagel and Rees, 1996), EF has become an established sustainability indicator and, according to some researchers, one of the most used in recent years (Venetoulis and Talberth, 2008, van den Bergh and Grazi, 2015). The popularity of the EF (or in the words of Giampietro and Saltelli (2014a, p. 620): “The extraordinary success enjoyed by the Ecological Footprint…”) has not been realized without criticism to the method (Gordon and Richardson, 1998, Gordon and Richardson, 1998, van den Bergh and Verbruggen, 1999, Gordon and Richardson, 1998, VROMraad, 1999, Gordon and Richardson, 1998, Ayres, 2000, Gordon and Richardson, 1998, Moffatt, 2000, Gordon and Richardson, 1998, Opschoor, 2000, Gordon and Richardson, 1998, van Kooten and Bulte, 2000, Gordon and Richardson, 1998, EAI, 2002, Gordon and Richardson, 1998, Grazi et al., 2007, Gordon and Richardson, 1998, Lenzen et al., 2007, Gordon and Richardson, 1998, Fiala, 2008, Gordon and Richardson, 1998, van den Bergh and Grazi, 2010, Blomqvist et al., 2013a, Blomqvist et al., 2013b, van den Bergh and Grazi, 2014a, van den Bergh and Grazi, 2014b, Giampietro and Saltelli, 2014a, Giampietro and Salteli, 2014b, van den Bergh and Grazi, 2015).

An important criticism that has been raised is the fact that of the six land types used in EF only one, carbon uptake land (land needed to sequester human CO2 emissions), seems to be utilized unsustainably on a global scale (van den Bergh and Verbruggen, 1999, Lenzen et al., 2007, Giampietro and Saltelli, 2014a). This does not match other indications being reported in the literature (MEA, 2005, Turner, 2008, Rockström et al., 2009, Niccolucci et al., 2012, Barnosky et al., 2019, Kubiszewski et al., 2013, Steffen et al., 2015, WWF, 2018). Hence, if the carbon uptake land type was removed from the EF it would indicate that humans were living within the regenerative capacity of planet earth – or in other words, living sustainably. This makes the carbon component of EF an important area of study.

EF’s carbon footprint (CF) is made up of four components: Production (emissions) data (P), fraction of CO2 sequestered by the ocean (OSFr), a yield factor – made up of global forest average carbon sequestration rate (AFCS) divided by ratio of C and CO2 – and a so called equivalence factor (EQF) intended to normalize the difference in productivity between the six land types that make up EF. These components have been dealt with in previous studies, at least to some extent.

Giampietro and Saltelli (2014a) have pointed out the difficulty with attaining reliable figures for any of these variables making up the CF equation. They particularly highlight the ocean sequestration fraction rate (OSFr) as problematic in this respect and point to McKinley et al., 2011, Wanninkhof et al., 2012 to support their claim. GFN’s calculations of OSFr are based on the findings of Khatiwala et al. (2009) and personal correspondence with the researchers (Guidebook, 2016). In the study in question Khatiwala et al. (2009) claim that the oceans sequester 20–35% of anthropogenic CO2 and state that:

“…considerable uncertainties remain as to the distribution of anthropogenic CO2 in the ocean, its rate of uptake over the industrial era, and the relative roles of the ocean and terrestrial biosphere in anthropogenic CO2 sequestration.” (p. 346)

They go on to explain that…:

“A key challenge for estimating anthropogenic CO2 (Cant) in the ocean is that Cant is not a directly measurable quantity. Existing estimates of the Cant are thus based on indirect techniques, such as so-called ‘back calculation’ methods that attempt to separate the small anthropogenic perturbation of carbon by correcting the measured total dissolved inorganic carbon (DIC) concentration for changes due to biological activity and air-sea disequilibrium.” (p.346)

Global Footprint Network (GFN) uses 30% as the ocean sequestration fraction.

Blomqvist et al. (2013a) have pointed out how the overshot of carbon uptake land rests on a single determinant – the average forest carbon sequestration rate (AFCS). They go on to explain how suspect this is in light of natural variability in sequestration rates and uncertainties in their measurement. In 2016, Mancini et al. – a group of GFN associated researchers – published a study focusing on refining the AFCS estimation. Their findings were that AFCS was lower than assumed by the standard EF methodology and therefore global CF was higher. In their review of this key parameter within CF, Mancini et al. (2016), estimate the average forest carbon sequestration rate at 0.73 t C ha−1 yr−1, with a standard error of ± 0.37 t C ha−1 yr−1. That is a 50% standard error. This study forms the basis of AFCS in current CF calculations of EF.

The equivalence factor (EQF) variable and its basis in the United Nation FAO suitability indexes from the Global Agro-Ecological Zones (GAEZ) model has been criticized by Venetoulis and Talberth (2008) who proposed a new method of estimating EQF based on net primary production (NPP).

Although Giampietro and Saltelli (2014a) pointed out the difficulty of filling the variables of the CF equation with reliable figures, as mentioned above, no exploration into the remaining variable – the production (P) variable – can be found in the literature nor a holistic assessment of all the input parameter’s accuracies as described above. There is thus a gap in the literature regarding this point.

This study aimed to assess the accuracy of the carbon footprint (CF) component of the EF calculations from the standpoint of the reliability of the data used for the P variable of the CF equation in combination with the recognized uncertainties in AFCS and OSFr. Two previous studies have shown the importance of data accuracy in EF calculations and how inaccuracy in a single data point can have a big impact on the results (Jóhannesson et al., 2018, Jóhannesson et al., 2019) at least in relation to the fisheries component of EF. Since CF is responsible for about 50% of the global total EF it is important to ascertain if similar inaccuracies and sensitivity are found within the data used for the P variable of the CF equation.

In the study, the input data (P) used for the CF was traced to its origins and its accuracy assessed and sensitivity tests were made to assess the impact on the final results. In addition, calculations were made incorporating the standard error for AFCS from Mancini et al., 2016 and the upper and lower levels of uncertainty of OSFr from Khatiwala et al., 2009.

To focus the research, Iceland was used as a case study to highlight issues pertaining to EF’s national and global CF accounting. Iceland makes an interesting case in this respect since very little fossil fuel is used for space heating and electricity generation due to the harnessing of local renewable resources such as water and geothermal heat. 99.9% of all electricity production in the country is powered by renewable energy and about 96% of space heating (Orkustofnun, 2018a, Orkustofnun, 2018b).

The results show that estimates play a major role in GFN’s CF calculations mainly due to the use of IPCC default emission factors. Further, activity data from international databanks rarely match locally sourced data. The change in CF under the data scenarios created range from a 42% decrease in CF to a 147% increase. Relevant caveats regarding estimations in CF calculations are found lacking in GFN’s dissemination of results.

The next section of this paper briefly explains the standard CF method and how the study was conducted, section 3 presents the results, section 4 provides a discussion about the results and section 5 lays out conclusions and considerations.

Section snippets

The Ecological Footprint

The Ecological Footprint aims to measure nature’s annual resource production – referred to as biocapacity (BC) – and measure that against human consumption of those natural resources – named footprint of consumption. The unit of measure is primary production (PP), which is then converted into productivity-adjusted hectares called global hectares (gha). The method’s creators William Rees and Mathis Wackernagel (1996, p. 227) say EF attempts to answer the question:

“How large an area of productive

Results

The study finds that the upper and lower limits of the case scenarios created from local data, AFCS standard error (Mancini et al., 2016) and OSFr upper and lower limits (Khatiwala et al., 2009) results in a change to CF from a 42% decrease for the lower limit to a 147% increase for the upper.

Further, of all the data points in the CF calculations for Iceland only two – industrial processes and product use (IPPU) and electricity production – match locally sourced data. Other data points show a

Discussion

The aim of this study was to assess the accuracy of the data used for calculations of the P variable of the CF equation of EF calculations and reviewing the combined impact of any inaccuracies with the recognized uncertainties in the AFCS and OSFr variables. This was done by tracing the data used for each data point to its origins and testing the sensitivity of the accounts according to the uncertainty involved.

The results suggest that the data used for the CF of EF is lacking in accuracy and

Conclusions

EF is arguably one of the most successful concepts to come out of the field of environmental science in terms of raising public awareness of the pressures that human endeavour is putting on the natural world. This makes the EF a very special thing.

To state categorically that the Ecological Footprint of humanity – or a nation, or any other subset of individuals – is any given number, is inherently misleading and false. The uncertainties in the data behind the carbon footprint – responsible for

Funding

This work has been funded by the University of Iceland Research Fund and Nordforsk through the project GreenMar. The funding bodies had no input in the study design; in the collection, analysis and interpretation of data; in the writing of the report; nor in the decision to submit the article for publication.

CRediT authorship contribution statement

Sigurður E. Jóhannesson: Conceptualization, Validation, Formal analysis, Investigation, Writing - original draft, Project administration. Jukka Heinonen: Resources, Writing - review & editing, Supervision, Funding acquisition. Brynhildur Davíðsdóttir: Resources, Writing - review & editing, Supervision, Funding acquisition.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References (94)

  • H. Haberl et al.

    How to calculate and interpret ecological footprints for long periods of time: the case of Austria 1926–1995

    Ecol. Econ.

    (2001)
  • M.M. Hanafiah et al.

    Comparing the ecological footprint with the biodiversity footprint of products

    J. Clean. Prod.

    (2012)
  • S. Henders et al.

    Accounting methods for international land-related leakage and distant deforestation drivers

    Ecol. Econ.

    (2014)
  • S.E. Jóhannesson et al.

    Standard ecological footprint method for small, highly specialized economies

    Ecol. Econ.

    (2018)
  • S.E. Jóhannesson et al.

    Increasing the accuracy of marine footprint calculations

    Ecol. Indic.

    (2019)
  • G.W. Kim et al.

    Assessment of direct carbon dioxide emission factor from urea fertilizer in temperate upland soil during warm and cold cropping season

    Eur. J. Soil Biol.

    (2017)
  • G.W. Kim et al.

    Evaluation of carbon dioxide emission factor from urea during rice cropping season: a case study in Korean paddy soil

    Atmos. Environ.

    (2016)
  • J. Kitzes et al.

    A research agenda for improving national Ecological Footprint accounts

    Ecol. Econ.

    (2009)
  • I. Kubiszewski et al.

    Beyond GDP: measuring and achieving global genuine progress

    Ecol. Econ.

    (2013)
  • P.A. Lawn et al.

    Has Australia surpassed its optimal macroeconomic scale? Finding out with the aid of ‘benefit' and ‘cost' accounts and a sustainable net benefit index

    Ecol. Econ.

    (1999)
  • M. Lenzen et al.

    On the bioproductivity and land disturbance metrics of the Ecological Footprint

    Ecol. Econ.

    (2007)
  • M.S. Mancini et al.

    Ecological Footprint: refining the carbon footprint calculation

    Ecol. Ind.

    (2016)
  • S. Medved

    Present and future ecological footprint of Slovenia—The influence of energy demand scenarios

    Ecol. Model.

    (2006)
  • C. Monfreda et al.

    Establishing national natural capital accounts based on detailed Ecological Footprint and biological capacity assessments

    Land Use Policy

    (2004)
  • S. Motesharrei et al.

    Human and nature dynamics (HANDY): modeling inequality and use of resources in the collapse or sustainability of societies

    Ecol. Econ.

    (2014)
  • V. Niccolucci et al.

    Biocapacity vs Ecological Footprint of world regions: a geopolitical interpretation

    Ecol. Ind.

    (2012)
  • J. Prakash et al.

    A technology-based mass emission factors of gases and aerosol precursor and spatial distribution of emissions from on-road transport sector in India

    Atmos. Environ.

    (2018)
  • J. Prakash et al.

    A technology-based mass emission factors of gases and aerosol precursor and spatial distribution of emissions from on-road transport sector in India

    Atmos. Environ.

    (2018)
  • W. Rees et al.

    Urban ecological footprints: why cities cannot be sustainable—And why they are a key to sustainability

    Environ. Impact Assess. Rev.

    (1996)
  • S. Roman et al.

    The dynamics of human–environment interactions in the collapse of the classic Maya

    Ecol. Econ.

    (2018)
  • P. Sadavarte et al.

    Trends in multi-pollutant emissions from a technology-linked inventory for India: I. Industry and transport sectors

    Atmos. Environ.

    (2014)
  • G. Salvo et al.

    Estimating the human appropriation of land in Brazil by means of an Input-Output Economic Model and Ecological Footprint analysis

    Ecol. Ind.

    (2015)
  • Y. Shan et al.

    New provincial CO2 emission inventories in China based on apparent energy consumption data and updated emission factors

    Appl. Energy

    (2016)
  • L. Shen et al.

    Factory-level measurements on CO2 emission factors of cement production in China

    Renew. Sustain. Energy Rev.

    (2014)
  • S.A. Solarin et al.

    A multi-country convergence analysis of ecological footprint and its components

    Sustain. Cities Soc.

    (2019)
  • G.M. Turner

    A comparison of The Limits to Growth with 30 years of reality

    Global Environ. Change

    (2008)
  • J.C.J.M. van den Bergh et al.

    Spatial sustainability, trade and indicators: an evaluation of the ‘ecological footprint’

    Ecol. Econ.

    (1999)
  • J.C.J.M. van den Bergh et al.

    Reply to the first systematic response by the Global Footprint Network to criticism: a real debate finally?

    Ecol. Ind.

    (2015)
  • B.C. Wang et al.

    Ecological footprint of Taiwan: a discussion of its implications for urban and rural sustainable development

    Comput. Environ. Urban Syst.

    (2012)
  • AYRES

    Commentary on the utility of the ecological footprint concept

    Ecol. Econ.

    (2000)
  • A.D. Barnosky et al.

    Approaching a state shift in Earth’s biosphere

    Nature

    (2012)
  • L. Blomqvist et al.

    Does the shoe fit? Real versus imagined ecological footprints

    PLoS Biol. Ogy

    (2013)
  • L. Blomqvist et al.

    The ecological footprint remains a misleading metric of global sustainability

    PLoS Biol. Ogy

    (2013)
  • Boden, T.A., Marland, G., Andres, R.J., 2013. Global, Regional, and National Fossil-Fuel CO2...
  • C. Cobb et al.

    The Green National Product: A Proposed Index of Sustainable Economic Welfare

    (1994)
  • E. Demir et al.

    A comparative analysis of several vehicle emission models for road freight transportation

    Transport. Res. Part D

    (2011)
  • EAI

    Assessing the Ecological Footprint: A Look at the WWF's Living Planet Report 2002

    (2002)
  • Cited by (13)

    • The impact of space development structure on the level of ecological footprint - Shift share analysis for European Union countries

      2022, Science of the Total Environment
      Citation Excerpt :

      This directs the understanding of the concept of the ecological footprint as a necessary tool in assessing the degree of balancing the consumption of natural resources with the pace of their renewal. In addition, in the context of sustainable development, the concept of ecological footprint occurs in relation to other concepts such as: biocapitalization (De Pascale et al., 2021; Hu et al., 2015; Toderoiu, 2010; Wackernagel et al., 1997), decarbonization (Sinha et al., 2022; Hammond et al., 2019; Pulselli et al., 2019; Mazzanti and Rizzo, 2017; Szopik-Depczyńska et al., 2017; Alderson et al., 2012), carbon footprint (Doğan et al., 2022; Jiang et al., 2022; Jóhannesson et al., 2020; Bello et al., 2018; Mancini et al., 2015; Galli et al., 2011), water footprint (Feng and Zhao, 2020; Bello et al., 2018; Galli et al., 2011), lean production (Taddeo et al., 2019; Aldieri et al., 2019; Lenzo et al., 2017). It should be emphasized that the concept of ecological footprint is not the same as the cited concepts.

    • Bracketing sustainability: Carbon footprinting March Madness to rethink sustainable tourism approaches and measurements

      2021, Journal of Cleaner Production
      Citation Excerpt :

      It can help direct management strategies to increase operational efficiencies and reduce those emissions (Pandey et al., 2011). This approach is especially welcome in new contexts considering disaggregated industries into aggregated products such as sport (Jóhannesson et al., 2020). Sport and event managers ought to be ecocentric leaders (Sartore-Baldwin and McCullough, 2018) by being more aware of the environmental impacts of sporting events to address the opportunities to improve environmental performance.

    • Deindustrialization, reindustrialization and environmental degradation: Evidence from ecological footprint of Turkey

      2021, Journal of Cleaner Production
      Citation Excerpt :

      In addition, the most important limitation of the study can be stated as an ongoing debate exists about the ecological footprint calculation that’s used as an indicator of environmental quality in the study. According to Jóhannesson et al. (2020), small changes to a data point in the EF computation matrix can result in large changes in the computed value. Additionally, it is written that the emission values presented by the IPCC were derived using various estimates and averages.

    • Dynamic assessment of ecological sustainability and the associated driving factors in Tibet and its cities

      2021, Science of the Total Environment
      Citation Excerpt :

      For example, the ecological security risks of the Guanzhong urban agglomeration were analyzed for the period 2005–2017(Yang and Cai, 2020). The carbon footprint data of the EF is lacking accuracy in existing studies because the non-fossil energy-related emissions from industrial processes are usually ignored in the calculation (Jóhannesson et al., 2020). Carbon footprint or energy EF, as a sub-footprint of EF, refers to the required forest area to absorb the man-made carbon dioxide (Wackernagel and Rees, 1996; Lin et al., 2018; Mancini et al., 2016).

    • An overview of mercury emissions in the energy industry - A step to mercury footprint assessment

      2020, Journal of Cleaner Production
      Citation Excerpt :

      Land, water, and carbon footprints are important indicators in the implementation of the circular economy (Holmatov et al., 2019). However, the results of footprint assessment tools are only as good and accurate as their input data (Jóhannesson et al., 2020). Gathering of reliable and accurate input data as well as the exploration of the connections (nexus) between various footprints and of their implications (ecological, economical, societal, etc.) is very important in the development of footprint assessment tools.

    View all citing articles on Scopus
    View full text