Review of trade-offs and co-benefits from greenhouse gas mitigation measures in agricultural production

Abstract

Anthropogenic release of greenhouse gasses (GHG) has increased in the last 50 years, contributing to global climate change. Because agriculture is one of the major contributors to the production of non-CO₂ GHG, the opportunities for mitigating GHG emissions from agriculture are often considered by policy makers. However, the implementation of agricultural GHG mitigation policies can have unintended consequences or trade-offs (both negative and positive). A major problem, for policy makers, is that although most of these trade-offs have been described in the past, no overview of them exists; and in many cases, there is no consensus with regard to the impact of the mitigation measures on aspects such as cost effectiveness, social acceptance, environmental impact, etc. The current article gives an overview of the different kinds of trade-off that might occur and their relationships to GHG mitigation and agricultural production. The authors offer policy makers a framework which can be applied to any GHG mitigation measure to determine which trade-offs are the most important and which ones should be taken into consideration. This will help policy makers to create an optimal agricultural GHG mitigation measure.

Keywords:

• greenhouse gas mitigation
• agricultural practices
• trade-offs

1. Introduction

As agriculture accounts for 10–12% of total global anthropogenic emissions of greenhouse gasses (GHG) and contributes 47% and 58%, respectively, to the total anthropogenic emissions of CH₄ and N₂O (Smith et al. 2007b), worldwide research has focused on ways to reduce agricultural emissions. Based on this research, policy makers in different countries have introduced measures to voluntarily (e.g. Canada) or compulsorily (e.g. New Zealand with NZ ETS (only from 2015 onwards)) mitigate GHG in agriculture. Other countries have no specific agricultural GHG mitigation measures, but it can be shown that some measures adopted for reasons not specific to GHG mitigation can also contribute to GHG mitigation. For example, in Europe most carbon sequestration has occurred as the result of non-climate policies (Freibauer et al. 2004). This impact is created by the many interlinkages and trade-offs between GHG mitigation and other practices in industry or society. These non-climate policies will have an ecological, environmental, economic, institutional and societal impact, which might lead to GHG mitigation at a number of levels. In order to find an optimal mitigation measure, whether it was originally designed to limit GHG emissions or not, it is important to take into account these different trade-offs and interlinkages.

A major problem, for policy makers, is that although most of these trade-offs have been described in the past, no overview of them exists and in many cases there is no consensus as to the impact of the mitigation measures on aspects such as cost effectiveness, social acceptance, environmental impact, etc. This paper, therefore, reviews the economic, environmental and social trade-offs of agricultural GHG mitigation strategies to help determine the characteristics of effective GHG mitigation policy measures required to tackle the problem of climate change. First, we review the potential co-benefits and trade-offs from measures to reduce agricultural non-CO₂ GHG emissions. We then discuss the importance of these trade-offs and describe, from a theoretical perspective, which external factors will influence the characteristics of the optimal mitigation measures. Finally, we discuss the associated practical implications for policy makers.

2. Possible co-benefits and trade-offs from GHG mitigation measures

Smith et al. (2005) suggest that actual levels of GHG mitigation are below the technical potential (the maximum level of mitigation) because various barriers exist (see Figure 1) with regard to implementation – including institutional, educational, social and political constraints (Cannell 2003). According to Smith et al. (2005), the barriers can be ranked from the most limiting to the least limiting and this, therefore, offers the potential for mitigation that varies between a minimum and maximum level. These constrained potentials are the points at which negative trade-offs are unacceptable either socially, economically, ecologically or physically. The left hand side of Figure 1 illustrates how GHG mitigation measures will face a theoretically biophysical maximum in terms of GHG reduction, meaning that after this point no further GHG reduction is possible.

Figure 1. Impact of different constraints on reducing GHG mitigation potential from its theoretical biological maximum to lower, realistically achievable potentials. Source: Smith et al. (2005).

However, this sequential representation might cause problems, particularly as the barriers seem to influence each other. Therefore, an alternative representation is given in Figure 2, in which the different trade-offs are shown.

Figure 2. Different categories of trade-offs or co-benefits that have to be identified and assessed to come to a good mitigation measure. Note: Land use includes land use per se, plus direct and indirect land-use change.

Within Figure 2, the constraints of Smith et al. (2005) are represented as impact categories. The potential of any mitigation measure will not only depend on the constraints that exist at the ecological, environmental, economic, societal or institutional framework level, but will also be heavily defined by the trade-offs that exist between these different categories. We state that a GHG mitigation measure will not only have an impact on the GHG emissions, but also on each of the other interlinked categories. In order to describe this, an overview is given of the trade-offs between each impact category and the GHG emissions.

First, mitigation measures will have an impact on ecological aspects such as biodiversity, landscape and land use. The latter is very important when dealing with climate change issues and includes both direct and indirect land-use change. Direct land-use change describes the introduction of a new crop system at a site, where this form of cultivation has not previously taken place, thereby generating changes in carbon stock (Van Stappen et al. 2011). Often the introduction of biofuel crop production illustrates this direct land-use change. Indirect land-use change occurs when pressure on agriculture exists due to the displacement of a previous activity or the use of biomass which induces land-use changes on other land areas in order to maintain the previous level of (e.g. food) production (Panichelli and Gnansounou 2008). This is also called the “leakage” or “displacement effect”. Discussion and modelling of the impact of this indirect land-use change on GHG emissions is still ongoing (Dumortier et al. 2011).

Second, environmental trade-offs can be observed – with these operating in two directions. On the one hand, mitigation measures designed to decrease the emission of agricultural GHG can lead to other environmental benefits or disbenefits and, on the other hand, measures that tackle other environmental problems can either help or counteract GHG mitigation. For example, balanced N fertiliser application can reduce acidification, nitrate leaching and N₂O emissions, whilst the application of low ammonia manure can increase nitrate leaching and N₂O emissions (Oenema and Velthof 2007). Climate change mitigation through carbon sequestration measures can trigger changes in the fluxes of other GHGs, especially nitrous oxide (N₂O) and methane (Powlson et al. 2011).

The environmental trade-offs can be negative and positive. The negative environmental side-effects of a measure, often referred to as pollution swapping, is the most researched trade-off concerning GHG mitigation strategies (amongst others Smith et al. 2008; Snyder et al. 2009; Stevens and Quinton 2009a; Novak and Fiorelli 2010). The research shows that pollution swapping between gaseous, dissolved and solid particulate physical phases (Dawson and Smith 2010) leads to climate change, acidification, water contamination or soil pollution. However, there are also important win-win solutions, where a particular measure will have a positive effect on different environmental problems, e.g. nutrient management can mitigate GHG emissions and enhance water quality. Many of the measures can also have a positive impact on soil quality.

Third, trade-offs between climate change mitigation and economic activities can occur. In the specific case of agricultural GHG mitigation, a valid objective is agricultural productivity (DeFries et al. 2004). When initiating climate change mitigation instruments, the private economic impacts for the farmer should be examined, particularly as most measures have few positive economic co-benefits for the farmer (Westhoek et al. 2011). These measures can impact on the farm's financial position through a change in yields or input purchases, labour requirements or investments in various types of technology (Freibauer et al. 2004; Beach et al. 2008; Breen 2008; del Prado and Scholefield 2008; MacLeod et al. 2010). The relationship with education and training also needs to be mentioned (Oenema et al. 2009). Equally, as with the environmental trade-offs, no consensus on whether the trade-offs are positive or negative can be found within the relevant literature. While N-accountancy was found to be cost-effective at farm level in the UK by MacLeod et al. (2010), balanced fertilisation was found to have a negative impact on income in the EU-27 by Oenema et al. (2009). Other divergent results were found by Shelton and Dalzell (2007, Australia) and Grainger et al. (2009, New Zealand) regarding the efficiency of forage species with high-condensed tannins. In some countries, the focus on the trade-off between climate change and economic considerations has been introduced into the policy measures (ERM 2011); for example, in Denmark with the ‘agreement for green growth’ and in Japan with the ‘biomass Nippon strategy’.

The above-mentioned economic impacts refer mainly to private economic effects on farmers. However, other stakeholders can also experience the economic impact of mitigation measures. Some of these economic effects are incorporated into the private transaction costs relating to policy measures (see impact five, below) and others can be included in the societal impact of mitigation measures (see impact four, below).

Within the framework proposed by Smith et al. (2005), the last group of tradeoffs concerns the socially and politically constrained potential. Within Figure 2, these two items have been separated. Therefore, as a fourth category of impacts, it is described how mitigation measures in agricultural production might have a societal impact in relation to animal health, food security or public health. For example, better housing conditions with lower GHG emissions will simultaneously enhance animal health (del Prado and Scholefield 2008; Lesschen et al. 2008). In addition, more productive animals, resulting in decreased methane production, will result in higher food security (Smith et al., 2007a). Within the Dutch policy mitigation instrument “supporting cutting-edge equipment”, the focus is not only on environmental issues, but also on animal welfare and GHG abatement (ERM 2011). Although many win-win situations can certainly be identified, more negative trade-offs can also occur. For example, genetic modification could lead to production systems with lower emissions, but such developments are not always acceptable to the public, and therefore not (yet) feasible.

Last but not least, mitigation strategies will face certain institutional or political constraints. The mitigation measures can be in the form of command and control (legal instruments), incentive-based (financial instruments) or information and capacity building (communicative instruments) measures (Prager et al. 2011). All these policy types are embedded in an institutional structure and require design, implementation and monitoring, which will lead to private transaction costs for the farmer and public transaction costs for the government (Mettepenningen et al. 2011). With respect to the private transaction costs, MacLeod et al. (2010) mention that although there is a significant potential for win-win situations (reducing emissions whilst providing a financial saving for farmers), not all farmers opt for such measures. They may argue that some costs have not been taken into account, such as, for example, the administrative costs associated with the adoption of measures. Public transaction costs must also be considered: design, implementation, monitoring and control represent costs for the relevant public administration and will vary for the different measures and instruments implemented.

The trade-offs and co-benefits described above in relation to GHG mitigation and agricultural practices show the complexity of developing an optimal mitigation measure. In Figure 2, all the impact categories are summarised and only the sum of all the trade-offs will indicate whether or not the mitigation measure is successful. It is not so much the sequential occurrence (as suggested by Smith et al. (2005)) that will define the potential of the measure, but much more the total sum of all negative and positive trade-offs. Therefore, a more integrated approach is required, as opposed to the linear representation by Smith et al. (2005).

Furthermore, Figure 2 illustrates the existence of external factors that need to be taken into account when determining the optimal mitigation measure. These external factors include the regional setting, the priorities defined by society or policy and the unit of reference, all of which will have an impact on the acceptability and success of the mitigation measure.

First of all, as no single mitigation option can reduce all trade-offs, the impacts will need to be prioritised. This is described in environmental research, for example, where Stevens and Quinton (2009b) state that no single measure will be able to combat all pollutants. Some authors consider it best to focus on only one impact. Kuosmanen and Laukkanen (2011), for example, believe that focusing on one pollutant may yield greater net benefits than setting uniform abatement targets for all harmful substances. Most of the research on pollution swapping is nonetheless based on two or three pollutants. Dawson and Smith (2010) suggest that mitigation measures for potential pollutants should be integrated, or at least grouped by analogous pollutants, similar physical phases, processes or pathways to prevent pollution swapping. Oenema et al. (2009) state that measures should be effective and beneficial simultaneously for water, air and soil quality (all the environmental impacts included in Figure 2). Stevens and Quinton (2009a) give a good example on how information could be drawn together on the potential for pollution swapping across a range of diverse pollution mitigation options. To enable policy makers to develop an optimal mitigation strategy, they need to have an overview of the potential (environmental) trade-offs, so that they can set priorities on those issues that are the most critical.

The second external factor which influences the choice of optimal mitigation measure concerns the regional setting. On the one hand, possible trade-offs and co-benefits can differ between regions, depending on soil characteristics and climate conditions. On the other hand, the difference between regions can also result in different priority setting (Angus et al. 2003) or a different potential for certain strategies. For example, the potential of animal feed strategies, in terms of low-protein animal feeding, will differ between countries. In some EU Member States, animal feeding is already efficient and the protein content of the animal feed is close to the optimum (Oenema et al. 2009) thereby limiting the potential of animal feed strategies, compared to other regions.

Third, it is important to select the right unit of reference to measure efficacy in terms of GHG reduction. Policy makers need to ask themselves two questions when considering trade-offs and impacts: in which unit are the results reported and what are the system boundaries for the calculations? Only by knowing this will policy makers be able to correctly evaluate the impact of their mitigation policy. This issue is, for example, touched upon by del Prado et al. (2010) who found that most methods of reducing GHG emissions per litre of milk (in this case by increasing the N use efficiency of the farm) increased GHG emissions per ha of land, although thesefarms needed less total forage area. Mondelaers et al. (2009) came to the same conclusion for the environmental benefits of organic farming: due to the lower land use efficiency of organic farming, the positive effect expressed per unit product is less pronounced than the effect per ha. Choosing the right unit of reference will be an important precondition for choosing the best mitigation measure (see Figure 2).

3. Dealing with trade-offs and co-benefits in mitigation strategies

Figure 2 shows that if policy makers want to build an optimal mitigation measure for agricultural GHG emissions, they need to understand the trade-offs and to weigh these against certain external factors. This creates a need for review and overview studies that describe all possible trade-offs for a particular mitigation measure. Although the literature on the topic of pollution swapping (negative environmental trade-offs) is expanding, there are very few studies incorporating economic trade-offs and even less studies incorporate all the potential trade-offs. DeFries et al. (2004) found a similar lack of trade-off research regarding the ecosystem impact of land use.

One issue that emerges from the relevant literature is the impression that there is a clear link between economic and environmental impact, on the one hand, and societal impact, on the other. For example, Smith et al. (2008) directly link the social issue of food security with the economic issue of agricultural productivity. Others have linked societal issues such as animal welfare to the economic characteristics of the farm, including feeding or housing conditions (del Prado and Scholefield 2008; Lesschen et al. 2008) or replacement rate (Lesschen et al. 2008). del Prado et al. (2010) give an example of how human health is related to economic farming aspects such as animal feed and the associated milk quality. Freibauer et al. (2004) have linked the societal value of amenities and aesthetic issues to the environmental issue of biodiversity. Because of the existence of this link, it might be possible to focus less on the societal issues and more on the economic and environmental ones. However, at the moment, not all win-win situations and trade-offs are fully understood, thereby making it difficult to say with certainty whether looking at economic and environmental aspects is sufficient to guarantee the best mitigation strategy.

Another issue that arises when dealing with trade-offs and co-benefits in mitigation strategies is that it is not just the different trade-offs for one mitigation measure that require consideration. All of the interactions between the different GHG mitigation measures should be taken into account, as these interactions reduce the abatement rate for subsequent measures (del Prado et al. 2010; MacLeod et al. 2010). In addition to this, interactions with other policies and trends should also be considered. Although the institutional impacts are, in general, quite absent from theliterature, they are nevertheless very important. This is probably the result of the more theoretical approach required to study the institutional aspects involved, as opposed to the more scientific approach that can be followed for the other categories. For example, different authors point to balanced fertilisation as a good agricultural GHG mitigation measure (del Prado and Scholefield 2008; Oenema et al. 2009; MacLeod et al. 2010). However, in some countries this measure has already been implemented in relation to the Nitrate Directive (such as Flanders), and therefore no longer belongs to the group of preferred mitigation measures. In these countries, policy makers should seek to maintain cost-effectiveness by increasing the stringency of this measure.

Finally, another important issue is that uncertainty might exist concerning the potential of a mitigation measure because of the interdependency between mitigation and adaptation. For example, some authors believe that climate change will negatively affect the growth rate of agricultural crops, which means that more land will be needed, and this could offset the impact of other beneficial strategies on mitigation (Smith et al. 2007a; Falloon and Betts 2010; Thornton and Gerber 2010). Other authors believe that climate change will increase agricultural productivity, because of higher temperatures and higher CO₂ concentrations (Reidsma et al. 2009; Hermans et al. 2010), thereby positively influencing the interdependency between mitigation and adaptation. Therefore, robust conclusions are necessary regarding the interactions between carbon sequestration and the mitigation of agricultural emissions (Vermont and De Cara 2010). Smith and Olesen (2010), who mention both positive and negative synergies, endorse the need for further research on the interlinkages between mitigation and adaptation.

4. Practical implications

Retrieving full information on trade-offs and co-benefits is complicated, prioritising appears to be difficult and taking trends into consideration seems almost impossible. Therefore, some practical guidelines are suggested here, to deal with some of the issues policy makers and researchers currently face.

First of all, although there is a need for review and overview studies that describe all possible trade-offs for a specific policy measure, shortcuts can be made in some cases, when win-win situations can be identified in at least two different fields. Forexample, when a mitigation measure leads to a net positive economic effect and a net positive environmental effect (taking into consideration the most important trade-offs within each effect) then applying this measure is expected to achieve a good mitigation outcome, even when not all trade-offs are known.

Second, this also means that whenever a specific measure leads to a very negative net benefit for one category of trade-off, it might be pointless to research the other trade-offs. For example, because the economic trade-off is so important, a measure will only be adopted when it creates a viable situation for all stakeholders (with or without compensation payments). Therefore, once a net negative result is identified for one of the most important categories, policy makers might immediately decide to seek another more cost-effective strategy.

The interaction between scientists and policy makers has an important role in determining good mitigation measures. Policy makers should have full knowledge of as many trade-offs as possible, so that they can define priorities and develop the optimal mitigation measure. Therefore, policy makers should indicate what kind of hiatus in information they perceive and in which direction they would wish to steer future policy. At the same time, it is very important for scientists to make their results known to policy makers.

5. Discussion and conclusion

When considering mitigation, measures should be designed with a very broad scope. There are a number of different trade-offs (environmental, economic, social, ecological and institutional) all of which have to be taken into account. Knowledge concerning the co-benefits and trade-offs is essential as “one can only take the full range of consequences into account if consequences are identified and quantified” (DeFries et al. 2004). It is a challenge for science to address the knowledge gaps and to reach a consensus on impact measures. More research is needed to identify the exact interactions between the different environmental phases, the social context and the ecological and economic impact. The uncertainty around the abatement potential of mitigation measures is significant, making monitoring and accounting for the measures even more challenging (McKinsey and Company 2009; Vermont and De Cara 2010).

From the review, it can be concluded that the importance of pollution swapping is evident and should be accounted for. At the same time, policy makers should understand the social and ecological impacts of proposed GHG mitigation measures, but from the literature review there would appear to be a link between economic or environmental impact and societal impact. Therefore, it might be possible to focus less on the societal issues and more on the economic and environmental ones. Above all, more research is needed on the economic impact of agricultural mitigation measures; and motivation and other aspects that affect the uptake of such measures by farmers. As farmers will need to apply the measures, policy makers should be aware of their likely financial impact and design the policy accordingly, so that farmers will be willing to implement the required measures. Verspecht et al. (2011) believe that stimulating awareness amongst farmers about existing and new mitigation policies, as well as their probable impact, is a key success factor for creating an efficient policy. In Canada, awareness raising is taken up explicitly within their GHG mitigation programme (ERM 2011). Awareness raising is crucial, particularly during the early stages of introducing a new measure. During the later stages of implementation, communication and capacity building will become more important in motivating farmers to understand and adopt the required GHG mitigation measures (Smith et al. 2008). In Canada and the US capacity building instruments were developed for farmers (ERM 2011). At the same time, policy makers will need to refine the mitigation measures they use so that they become (or remain) cost-effective and viable.

The prioritising of pollutants seems to be situation and place specific, but as climate change is a global problem the geographical scale requires consideration – for example, deforestation to enable long-distance trade in animal feed (Herrero et al. 2009) or the displacement of production when emissions are reduced by lowering production (Golub et al. 2009; MacLeod et al. 2010). This displacement effect has also been mentioned under the ecological impact. The time scale is also important, highlighting the need for a long-term outlook. Many of the strategies require 5–10 years before they result in measurable evidence of payoff (McKinsey and Company 2009). This will make mitigation potentials in the agricultural sector uncertain, making a consensus difficult and hindering policy makers (Smith et al. 2007a).

The proposed framework seeks to clarify the different links and trade-offs that can occur when introducing mitigation measures. The introduction of a mitigation measure depends on the overall impact of all negative and positive co-benefits. Moreover, the success and acceptability of a measure is also dependent on the regional setting and policy priorities.

An integrated policy approach is needed and co-impacts (dis-benefits or benefits) of policy measures have to be taken into account. Even when all possible trade-offs are known, it is important that policy makers set priorities in relation to the pollutants and environmental problems they wish to tackle. Policy makers should focus on those measures that have a good mitigation performance and preferably with some important co-benefits and only small trade-offs. Therefore, policymakers, as well as researchers and private actors, should identify good mitigation measures that are time and location specific with minimal trade-offs and have a sustainable abatement potential. It is likely that a combination of measures will be required to overcome the many constraints to GHG mitigation in agriculture, and may include incentives, regulations and education.

Acknowledgements

The authors want to thank the NL Agency who, on behalf of the Dutch Ministry of Economic Affairs, Agriculture and Innovation and the Ministry of Infrastructure and the Environment, commissioned the research on which this article is based.