based on L. Merfort et al. Nature Climate Change (2023).

The policy problem

Large fractions of the global forest and natural land cover are not effectively protected. If bioenergy cultivation is not strictly limited to marginal or abandoned land, bioenergy may be grown on agricultural land and displace food production. Shifting food production elsewhere can cause substantial carbon dioxide (CO2) emissions due to forest clearing in regions with weak or no land regulation. This puts policymakers in a difficult situation, because these indirect effects act via a globalized food market and thus are beyond the control of policies of individual nations. Because bioenergy and, in particular, modern biofuels are regarded as a valuable option to reduce emissions from burning fossil fuels, it is thus important to understand if and potentially how locally implementable policies can contribute to reducing bioenergy-induced emissions.

The findings

With an average emission factor (EF) of 92 kgCO2 GJ–1, we find that the production of modern biofuels, if averaged over a 30-year period, causes land-use-change emissions that are higher than those from burning fossil diesel (Fig. 1). If policymakers tax bioenergy according to these average expected emissions, that is, apply a similar carbon price to a litre of biofuels as to a litre of diesel, the total future bioenergy-induced emissions decrease, as the demand is reduced. However, we show that such a policy cannot bring down the high average emissions that are attributed to biofuels. Only strict and globally comprehensive protection of natural land will reduce the EF and hence, only then, will those biofuels that replace fossil fuels effectively reduce CO2 emissions.

Fig. 1: Bioenergy-induced land-use-change (LUC) CO2 emissions, bioenergy production and biofuel EFs under six different policy assumptions.
figure 1

a, Cumulative (2020–2100) global bioenergy-induced LUC emissions (black) and bioenergy production given as the average annual global production (green). White horizontal bars indicate 2020–2050 values. b, Biofuel EFs in different metrics. The red markers show the average 80-year EF, the blue markers the weighted average 30-year EF. The boxplots show the variation over time of the 30-year EF between 2025 and 2070. The minima and maxima of the box confine the interquartile range, the whiskers represent the 1st and 4th quartile (for bioTax30, ‘301’ is the upper bound of the whiskers), and the centre lines are the median value. Figure adapted with permission from L. Merfort et al. Nat. Clim. Change (2023), Springer Nature Ltd.

The study

This study used the integrated assessment modelling framework REMIND-MAgPIE coupling the energy and the land systems to derive alternative transformation scenarios consistent with limiting global warming well below 2 °C. The scenarios differ with respect to assumptions on land-use and energy policies, which have a large influence on CO2 emissions from land-use change and also affect the amount of bioenergy used to fulfil the global energy demand. By deriving a counterfactual scenario, in which bioenergy is not available and hence land-use-change emissions are lower, we can attribute emissions to bioenergy production and derive an EF (emissions per unit of biofuel produced). In comparison to previous studies that analysed biofuel EFs, our approach using future climate change mitigation pathways allows us to compare the EF in the light of different policy frameworks, which we find are the most important to determine the emissions that can be expected from producing biofuels.