Can food have a zero carbon footprint?
This post breaks down reduction strategies by product lifecycle phase, seeks to identify truly residual emissions and explores the role of carbon removal in the food supply chain.
Many talk about the need to decarbonise agriculture and the food industry but so far few have outlined what this looks like practically. This post aims to go through specifics, at least at a summary level, so that we can hopefully start to have a more productive conversation about ‘how can we do it’, rather than ‘what do we need to do’.
A product carbon footprint is broken down into various life cycle phases from from cradle to grave, as depicted below. Similar to a life cycle analysis (LCA), a product carbon footprint focuses only on greenhouse gases (GHG) emissions and doesn’t include other environmental indicators such as water use. Let’s have a look if each individual life cycle phase can possibly emit zero carbon emissions.
Raw Materials
Materials are initially either extracted or grown. Most materials or ingredients involve some sort of agricultural process. Food, cotton, wool, leather, rubber and wood are all agricultural products. Is it possible to grow any of these materials without emitting greenhouse gases?
The short answer is yes if you are only considering fossil fuel emissions and no if you consider livestock and biogenic emissions (those which originate from living organisms e.g. trees, plants and organic matter).
To understand the emissions of raw materials we need to analyse agricultural inputs, machinery, livestock, land use change and the natural carbon cycle.
Agricultural inputs
The most important farming inputs include seeds, feed, fertilisers and pesticides. Let’s assume you run a closed-loop farm and grow your own seeds. That’s one input sorted as long as you can figure out the next two.
Could we do without pesticides? Although research argues that we can reduce our consumption substantially without incurring too many losses, eliminating them completely doesn’t seem viable at scale today. Even organic agriculture uses pesticides, they are simply regulated and limited in scope — originating from natural sources. Indoor, vertical farming offers a partial solution as pesticides are largely unnecessary. This, however, is not quite scalable enough. Crop rotation as well as more resilient crop varieties help reduce disease and pests, but not using pesticides completely reduces yields — an issue for food security amidst a growing population. It seems like we may need these for years to come.
So what about fertilisers? It’s a similar story. Even the healthiest of soils requires occasional fertiliser application. We should start by eliminating carbon-intensive synthetic fertilisers, but then we’ll still need to apply animal manure or compost. This means having animals on farm (which emits GHG emissions) and the manure itself emits methane and nitrous oxide as depicted above. Using plant waste to create compost can be a great way to complement fertilisers. Unfortunately compost releases carbon dioxide, and if anaerobic (i.e. without oxygen) it also emits methane — so there is no perfect solution unfortunately.
The solution showcasing the most promise right now is biochar: a charcoal-like substance produced by pyrolysis of biomass in the absence of oxygen. If we consider the increase in carbon removal of the soil it is applied to, multiple LCA studies have shown that it can be carbon negative on a cradle-to-grave basis. One study shows that 1 tonne of biochar removes 12.25 kg CO2e. So let’s talk about carbon removal on farm and how this works.
Land use change and carbon removal
A substantial % of the carbon footprint of certain foods like palm oil, beef or cacao is attributed to what is referred to as land use change (LUC), in other words: emissions linked to deforestation or land conversion. The good news is that this works both ways! A change in agricultural practices such as no tillage, crop rotation, application of soil probiotic or biochar accelerates soil health, which in turn helps previously depleted soil to remove carbon from the atmosphere.
These measurable changes should also be attributed to the product carbon footprint in that specific timeframe (e.g. 1 year). I would argue that it is only the difference in carbon removal that should be attributed, i.e. the original capacity of the soil to remove carbon should not be accounted for as this would have happened with or without human intervention.
Let’s look at an illustrative example: 1 hectare of farm land currently removes 2 tonnes of CO2 per annum.
As a result of biochar application the same hectare of land now removes 6 tonnes of CO2 per year.
It’s the difference, i.e. 6–2 = 4 tonnes per hectare per year that could be counted as additional carbon removal to be attributed to the individual product carbon footprints by dividing the removal by total yield of that agricultural produce.
The question now becomes for how long should this additional removal capacity be attributed to this action? Once there is no additional carbon removal possible, the soil will reach a new overall annual carbon removal capacity of 6 tonnes a year.
There would come a point where it would be justifiably hard to attribute this removal on an ongoing level to new foods grown as it would do so with or without human intervention. Arguably, it may have a stronger carbon removal capacity if left to re-wild too.
With trees or hedgerows the accounting is a little easier as they will have a reasonably predictable carbon removal schedule which starts slow and typically sees it peak between 15–25 years (depending on the tree species). You could attribute the additional carbon removal of the tree every year by dividing it by the quantity of produce — this is definitely a much need solution much like biochar, but it most likely won’t provide removals forever either.
We then come to natural processes like respiration and decomposition of organic matter. This also releases GHG emissions but are a bit of ‘grey zone’ as some argue this is simply part of the ‘natural carbon cycle’ and therefore doesn’t count.
In summary, it’s a complex matter that warrants more investigation and thankfully the GHG protocol —the leading global GHG accounting standard — will be releasing more guidance on how to account for these types of emissions in the land sector and associated carbon removal in 2022. Getting consensus is proving challenging and it’s a telling sign that the release has been delayed multiple times already.
Livestock
Livestock emits greenhouse gases through respiration and enteric fermentation a.k.a cow burps (not farts as is commonly miscommunicated). Recent announcements point to technology that aims to capture methane directly from cow’s mouths which is promising, but it does seems quite unlikely that we’ll be able to eliminate these emissions completely. It seems that a zero carbon footprint product might be very challenging for ruminants.
We then come to feed. Roughly 40–50% of the emissions linked to chicken for example is due to soya, their main source of food in the UK. This is due to the emissions linked to the land use change, agricultural process and transport. Opting for locally grown wheat, pea and bean-based protein can reduce feed emissions dramatically for chickens for instance. Choosing better feed also means evaluating its impact on enteric fermentation which applies to both beef and lamb. For instance, various studies show that lemongrass or seaweed can help reduce methane emissions from ruminant animals by up to 80%.
Waste outputs such as manure or carcasses also create emissions. Indeed, cow dung generates methane and nitrous oxide as part of the ‘manuring process’. But manure isn’t evil, its impact can be minimised if managed well and is a good source of fertiliser — which we may need for other crops, so it seems like it’ll be a difficult one to eliminate completely. Manure can also be re-purposed as a feedstock for biogas which is often a cleaner source of fuel.
Some industry bodies have argued that ruminant animals have a positive impact on the planet based on the fact that the natural grassland on which they are raised sequesters the same or more carbon than is emitted in the production and distribution of the meat. As outlined above, only quantifiable increases in soil or tree sequestration should be counted if we are to apply credible carbon accounting methods. Otherwise this would be equivalent of saying that humans are carbon neutral by virtue of owning a garden. Gardens would exist, with or without the individual owners, and in fact natural habitats would likely better thrive without humans at all as a number of re-wilding experiments have recently demonstrated.
Machinery, manufacturing and distribution
This part is simple enough in theory: We need to transition to green electric. Not to say electric vans or machines have a zero carbon footprint (across their own lifecycle) in themselves but this is where you need to draw a boundary. Then again the horse and cart wasn’t exactly zero carbon either! One way to reduce the embodied carbon in machinery could be to consider recycled / low carbon materials and for smaller farms to share any expensive machinery.
Technically, direct emissions are zero if the machinery, manufacturing process and distribution are all electric and Scope 2 indirect emissions would be zero if 100% powered by renewable electricity.
Use Phase
Consumer use phase related emissions are mainly linked to cooking (gas-powered hobs, ovens, etc.) and refrigeration (fridges, freezers). With cooking we can transition everything to electric, and thus get to zero relatively easily technologically speaking. The challenge will be financial (subsidies, grants and incentives to move from gas hobs to electric will be key!) and cultural (there is a certain perception issue with induction/electric hobs). This will also mean no more barbecues or wood-fired ovens!
90% of refrigerant emissions happen at end of life, thus effective disposal by purification for reuse or transformation into other chemicals that do not cause warming is essential. Using natural or organic refrigerants such as propane and ammonium can reduce the remaining 10%, but not eliminate them completely. Unless we all become raw vegans, eating food as soon as it is plucked from the earth, it’s unlikely that we will decarbonise refrigeration completely.
Packaging
Packaging is a tricky one. Plastic is often touted as being bad, but the real issue here is single-use. Replacing plastic for cardboard can be worse for upstream carbon emissions. Compostable materials tend to be very water intensive and the composting process in itself is also a major source of carbon emissions.
A relatively easy first step is to move to recycled materials and ensure your packaging is entirely recyclable. This will reduce emissions on both accounts and reduce the amount of waste that ends up in our waterways and oceans. But recycling still requires energy and cannot be repeated an infinite amount of times for all material types (e.g. plastic typically only has 3–4 lives).
So how to get ‘zero carbon’ packaging? Best option: no packaging at all. More practically: we must transition to a re-use model. Unfortunately this is logistically very complex, and for some products near impossible without creating a massively inconvenient experience, or one that has negative side-effects such as increasing food waste — which in turn generates its own emissions! Even re-use has an end-of-life, when the packaging simply cannot be re-used anymore, which in turn causes unavoidable emissions. It seems likely that food product packaging will always see some residual emissions.
End-of-life
Tackling the immense problem of food waste — roughly 1/3 of food produced for human consumption is lost or wasted — will also be crucial so that we only produce the food that we consume. Reducing mindless consumption, designing perfect, individualised food portions, creating further supply chain efficiencies and normalising wonky fruit & veg for everyday consumers will be key. Ensuring we only create 100% recyclable or reusable packaging will also help tackle this final lifecycle phase, and for this to actually be recycled/re-used too.
This said, the realist in me refuses to believe that all humans will always finish their food and/or recycle their packaging. I’d argue that there will likely always be some emissions occurring from landfill, incineration or simply from organic matter rotting in open air. It is very likely for these to be residual, only due to the fact that human behaviour is probably the hardest ‘sector’ to decarbonise.
In summary, when it comes to food products, there are many solutions which can help us decarbonise aggressively. There will however most likely be residual emissions spanning the following: agricultural inputs, livestock, refrigerants, packaging and end-of-life. In the short to medium term, zero carbon food products are therefore unlikely to be possible, hence why the ‘Net’ part is important in ‘Net Zero’.
While we work on implementing all of these initiatives, a good stepping stone can be to opt for Carbon Neutral food whereby the carbon emissions are offset via high quality avoidance, reduction or removal projects. Carbon neutral today, Net Zero tomorrow.
N.B. Although there has been increasing adoption of the Net Zero terminology for climate action commitments which I discuss separately here, I would argue that a ‘Net-Zero Product’ sounds a little odd and that we better stick with simple, more accessible language.
