This overview is intended for those in Canadian agriculture interested in using proven, or partially proven, technologies to reduce net greenhouse gas (GHG) emissions through changes in farming practices.

If you don’t want to read further, my conclusions are:  Opportunities look greatest for reductions in nitrous oxide (N2O) emissions. Further reductions in carbon dioxide (CO2) emissions through more usage of no tillage are possible, and the potential to sequester soil organic carbon through increased perennial forage production exists, but it’s questionable that these will lead to sizable futures decreases in national GHG emissions. Cover crops, with a few exceptions, probably represent limited opportunity for soil C sequestration in Canada with our long winter and short off-season growing intervals, but may prove more valuable for reducing spring-time N2O emissions. Future reductions in methane (CH4) emission are dependent on the extent to which promising research on rumen additives can be implemented in practice.

Carbon Dioxide

Let’s start with CO2 because, globally and nationally, it’s the most important of the three major GHG gases.

(The three gases are CO2, N2O and CH4. When GHG emissions are reported as CO2, that means only the one gas, but when reported as ‘CO2 equivalent,’ or CO2e, that means the collective warming potential of all three gases – plus a few more gases unrelated to agriculture that have smaller effects – all weighted for their relative potential to affect global warming. N2O has about 298 times the warming potential of CO2 on a per-kg basis; for methane it’s about 25, calculated as the average 100-year effect.)

According to GHG accounting protocols specified by the International Panel on Climate Change (IPCC), agriculture represents 8.1 % of total Canadian emissions, but with CO2 only representing 4% of that. These stats are as reported by Environment and Climate Change Canada (ECCC) in its latest National Inventory Report (NIR) to the United Nations Framework Convention on Climate Change (UNFCCC). I’ve summarized Canadian data for 2018 here; the numbers for 2019 are virtually the same – ECCC 2019 summary here – entire report available here.

However, CO2 emissions associated with on-farm fossil fuel usage, biofuel manufacture/usage and net CO2 exchanges between farm soils and the atmosphere are not included in the NIR/UNFCCC calculations for agriculture. If these were in, the Canadian net agricultural emissions would represent 8.5% of the Canadian total (calculations here). Even with these additions and subtractions, CO2 still represents a very small percent of total net agricultural emissions.

If we go further and add GHG emissions associated with fertilizer manufacture (mainly N), the 8.5% grows to about 12%, with most of the increase being CO2, but with CO2 still being a relatively small portion of the Canadian agricultural total.

Finally, if the amount of carbon contained in products that leave Canadian farms and are exported were to be included, the total net GHG emission from Canadian agriculture would actually be negative – removal of CO2 from atmosphere exceeding all forms of ag GHG release. But that calculation remains far from IPCC calculation plans, at least as it appears now.

For the following discussion, I’ll focus only on two categories, net GHG exchanges for ‘Agriculture’ as reported by NIR/UNFCCC, and the agricultural soil portion of another category called, ‘Land Use, Land Use Change and Forestry,’ or LULUCF.

The Canadian NIR recognizes three sources/sinks for CO2 in its LULUCF calculations for agricultural soils. These are: land converted to continuous crop production from periodic summer-fallowing, land converted to reduced tillage and no tillage from more intensive tillage, and land converted into perennial forage production from annual crop production, as well as the reverses.

Emission factors for these land-use changes reported in the Canadian NIR are shown in Table 1 (taken from Table A3.5-8 in Part 2 of Canadian 2020 NIR).

Table 1. Effective linear coefficients of soil organic carbon for land management changes (LMC); for the opposite changes – eg., no tillage to intensive tillage – the linear coefficients are negative of the values shown in column 3.

Of the three factors, the largest effect by far comes from a conversion of annual crop to perennial cropping. Unfortunately, the trend in recent years has been more of the reverse – perennial to annual – a result of better prices for annual crops and declining beef production in some provinces.

An increase in perennial crop acres, ideally as part of crop rotations with annual crops, would be very beneficial for soil organic carbon sequestration, but it is difficult to see how this will happen – given increasing public interest in plant-based (that generally means annual-crop-based) alternatives to ruminant meat. A fledgling cash-crop industry is developing in Ontario and perhaps other provinces for the production and export of high-quality alfalfa hay, but the scale is still too tiny to permit conclusions on whether this will grow to represent a significant means for increasing SOC on a national scale.

The shift from intensive to no tillage provides the smallest per-ha annual change. Table 1 may over-estimate the no-till benefit in Atlantic Canada, Ontario and Quebec. Available research data are inconsistent as to whether there is a net positive no-till effect on soil organic carbon in this part of Canada. Research data show a more consistent increase in soil organic carbon with tillage elimination in the Prairie Provinces. But the biggest shift in Prairie crop acreage to no tillage may already have occurred.

A reduction in summer fallowing causes a larger per-ha annual effect than no tillage, but this is largely limited to the semi-arid Prairies where summer fallowing is still practiced to a significant extent – and this benefit declines as the portion of Prairie farm land available for summer-fallow elimination continues to shrink.

It must also be noted that the annual SOC sequestration provided by these changes in land management declines each year as shown below in Fig 1, extracted from the 2020 Canadian NIR. After 20-25 years, about 50% of the total GHG sequestration potential from a change in land management practice has been achieved.

Fig I. Changes in SOC sequestration with time after implementing changes in land management practices (from Fig A3.5-14 in Part 2 of Canadian2020 NIR). The dashed line denotes an annual sequestration of 25 kg/ha of SOC; once the estimated sequestration falls below this, the values are not included in Canadian LULUCF totals.

A popular means of sequestering soil organic carbon may be cover crops – at least as expressed at farm meetings and as featured in ag media of late. A number of global research publications also make this claim. However, in a recent review of the published evidence, I concluded that the likelihood for such is very limited in short-growing-season environments like Canada. There are exceptions – for example, red clover spring seeded into winter wheat, or for some horticultural crops where the off-season available for cover crop growth may be unusually long. Though I’d love to be proven wrong, I personally don’t see cover crops as a nationally significant option for CO2 sequestration. That may be the reason why cover crops are not included in the listing shown above in Table 1 taken directly from the Canadian NIR.

There is a strong alternative opinion to what I’ve expressed above. For one example, see this backgrounder written by the respected Dr. Brian McConkey (formerly with Agriculture and Agriculture Canada – AAFC – and now chief scientist with Viresco Solutions) for Farmers for Climate Solutions (available here, check Chapter 2). Dr. McConkey’s view is that a portion of any organic matter addition to the soil, no matter how small, contributes to SOC enhancement. That includes cover crops which may only produce a few hundred kg/ha of organic matter per season. His perspective may stem from this article written by Dr. Fan and a number of AAFC researchers including Dr. McConkey who conclude, based on computer models, that more organic matter addition to soil automatically means increased SOC content. However, the Fan article includes a series of graphs (Fig 5 in the paper, if you are looking) that predict, based on this perspective and these models, that the SOC percentage in farm soils of Southern Ontario should increase anytime the annual organic carbon addition exceeds about 3-4 tonne/ha.

But the annual organic matter addition to soil with a typical corn-soybean or corn-soybean-wheat rotation is about 12 t/ha – which equates to 5 t/ha (assuming organic matter is 42% C) and yet the SOC content of Ontario farm soils in principal cash crop areas of the province is in decline. See following graph courtesy of Christine Brown, Ontario Ministry of Agriculture, Food and Rural Affairs (OMAFRA), that is based on thousands of soil samples tested in Ontario labs.

There has to be more to SOC enhancement than just more added organic matter (not that the latter is not also important).

This paper containing results of a 10-year, multi-site study in Iowa provides an explanation. While the addition of organic matter including cover crops to soil will add additional organic matter, it can also increase the rate at which soil organic matter is respired – often resulting in no net gain in SOC.

McClelland et al 2020 in a meta-review found that SOC enhancement with cover crops was generally only significant when the annual cover crop dry matter accumulation was 7 t/ha or higher. Many cover crops in Canada produce only about a tenth of that.

In brief, while cover crops in certain situations may increase SOC, it is doubtful that collectively these represent an important opportunity to increase the carbon dioxide sequestration in Canadian farm soils. This is not to detract, at all, from other cover crop benefits, including important reductions in soil erosion.

Nitrous Oxide

The quantity of nitrous oxide (N2O) emitted each year, globally or nationally, is small compared to CO2, but the warming potential of a tonne of N2O gas is about 298 times greater. N2O represents about 5.2% of total Canadian GHG emissions measured as CO2 equivalents, but agriculture represents 76% of that according to the Canadian NIR.

Although some of the agricultural N2O comes from manure storage before field application, about 86% comes from the application of N fertilizers to farm soils, both synthetic N fertilizers and manure. There are some excellent reviews including this one done by a group of researchers at the University of Guelph as well as  this, this, this and this.

The research results are extensive and quite consistent:

• Increased N application rates mean more N2O loss. Indeed, the simplest IPCC calculation protocol for estimating national N2O soil losses expresses loss as a fixed percentage of N applied. Research suggests that N2O loss actually increases exponentially with N application rate, meaning the last 10 kg/ha of applied N causes the biggest N2O loss on a per-kg basis. There have been numerous agronomic/economic studies of the most economical rate – complicated by large annual variations in both actual N losses and crop yields caused by differences in amount and intensity of rainfall. Rapidly increasing yields over time for some crops – corn and canola are examples – also complicate these calculations. How to provide the additional N without increasing N2O losses?
• Fertilizer additives that slow the transformation of urea and ammonium forms of N fertilizer will reduce N2O emissions. The literature suggests an average of about 30% reduction with the combined usage of a urease inhibitor (slows urea transformation into ammonium) and a nitrogenase inhibitor (slows ammonium transformation into nitrates) for treatment of urea or UAN (urea plus ammonium nitrate) fertilizers. The reduction is also sizable, though less, with the usage of polymer-coated fertilizers.
• Partial side-dress application of fertilizer N, as compared to full application of N at or before planting, does not appear to reduce N2O losses consistently as a percent of N applied. But because side-dressing reduces the total N application need, this means a reduction in N2O loss. A lot depends on when intense rainfall events occur relative to time of application.
• In a recent paper, Eagle and several co-authors from across the North American Corn Belt concluded that N2O losses are more closely related to the difference between N applied minus N removed in harvested crops, than to N application rate per se. This means that variable application rates based on within-field differences in measured late-spring soil nitrate levels or historic differences in within-field yield represent another good option for reducing N2O losses.
• For a given rate of soil N addition, any practice that reduces soil nitrogen losses for other reasons, eg., reduced ammonia volatilization with fertilizer incorporation or reduced leaching loss of nitrate will often tend to increase N2O losses, so there are management tradeoffs.
• N2O loss also tends to increase with Increases in soil organic matter content, so a tradeoff once again.

So what does all this mean? With the use of N fertilizer inhibitors, and adding combinations of partial side-dress N application and within-field variable N applications, it should be possible to reduce N2O losses by 30-50% or more with synthetic N fertilizer usage – with the cost for added technology being offset, at least in part, by lower average N fertilizer application rates.

Manure N seems more complicated. IPCC Tier 1 guidelines (the approach countries are to use in calculating GHG emissions when they lack the data or expertise to permit more sophisticated calculations) calculate a lower percent conversion of available N into N2O with manure compared to synthetic fertilizers. However, that is not well supported by North American research results that show that, on average, the two loss ratios are about equivalent. See references cited above.

To that, add the complication that manure is often applied to fields in the late summer or autumn – after wheat harvest, for example – to provide fertility for the crop to be planted the following spring. In provinces/states/countries – like Ontario – with substantial autumn and early spring rainfall, that can mean major N off-season losses via leaching, and N2O emissions.

Data summarized by OMAFRA indicate that 50 to 80% of the N content in late-summer/autumn–applied manure may be lost in off-season months. (The percentage varies across types of manure and whether incorporated or not.)

One solution is to apply manure in spring before or after crop planting but that can be logistically difficult in years with wet springs. Manure additives may reduce the problem though that’s a subject on which I have no knowledge.

Cover crops are an opportunity to reduce emissions but this is a source of uncertainty. Research by Dr. Wagner-Riddle and colleagues at the University of Guelph (see publication here, and personal communication) has shown that early spring N2O emissions are significantly lower with a cover crop like winter rye. A substantial portion of the year’s total annual N2O emissions can occur during this short period; the cover crop effect seems to be one of reducing the depth of frost penetration, or maybe the number of early-spring freeze-thaw cycles.

However, other researchers at the University of Guelph hold a different opinion. “Conservation tillage, [and] cover cropping … increase N₂O emissions,” state Ashiq et al (2021) as a highlighted conclusion in a recent review. My read of the literature indicates such an unqualified conclusion is probably inaccurate, though it likely reflects any effect of cover crops, if present during the main growing season, in preserving soil moisture and, hence, increasing the likelihood of anaerobic soil conditions for N2O formation.

Non-legume cover crops are well recognized for their value in reducing soil nitrate levels in autumn, and the expectation is that this will reduce the potential for autumn-winter-spring leaching losses. Cover crops may thus reduce the amount of nitrate available for N2O formation in early spring. But if the cover crop dies quickly following late-fall-early-winter frosts, thus releasing nitrates back into the soil profile in weeks to follow, this may mean more nitrate available for N2O formation in early spring – maybe even more than with no cover crop at all. Legume cover crops mean more soil nitrate availability and the potential for more N2O formation, especially at time of cover-crop death.

A very informative meta-review was published in 2014 by Basche et al. They found 29 relevant studies of paired comparisons – cover crop versus no cover crop. Of those, 40% found a decrease in N2O emissions with cover crop usage and 60% found an increase. In general, there was more likely to be an increase in N2O with cover crops when: the cover crop was an N-fixing legume versus grass/cereal or other species; the residues from killed cover crops were incorporated into the soil versus left on the surface; and when N2O emissions were measured during the initial time of cover-crop residue decomposition versus for longer periods of time. The reviewers also state that, to the extent that cover crops reduce nitrate losses from leaching, they will also reduce off-site N2O losses which can be substantial though not measured in any of the 26 studies.

A best option seems to be cover crops that remain fully viable over winter followed by no tillage so as to not incorporate residues. But getting them killed in time to not affect the growth and yield of the subsequent spring-planted crop can be a challenge – again especially in years with above-average spring rainfall. Dr. Eileen Kladivko of Purdue University provides a good discussion on these options here.

Cover crops may ultimately prove to be of much more value in reducing GHG emissions by suppressing N2O formation, than by sequestering soil carbon, but it is difficult to be certain at this time.

Methane

The greenhouse gas, methane (CH4), represents a much smaller quantity of global emissions than CO2, but provides a substantially higher ‘warming potential’ per tonne. Unlike N2O and CO2, CH4 has a relatively short half-life in the atmosphere (about 10 years) before being broken down into water and CO2 molecules.  Methane represents about 13% of total CO2e emissions for Canada, according to IPCC-based calculations, of which about 31% comes from agriculture – and 86% of the 31% comes from ruminant digestion. There are a couple of key issues though I’ll spend limited time on them here as they are addressed well elsewhere, including here and here.

One involves the fact that IPCC calculations, based on average warming effects of GHG gases over 100 years do not work well for short-lived gases like methane, with its half-life of about 10 years. And based on an improved calculation procedure developed at Oxford University, the true long-term GHG effect of methane should be much smaller than in the numbers submitted by Canada and others in their annual NIR. This is especially true for Canada where methane emissions from ruminant livestock have been in decline for about the last 15 years. However, a related blog with analysis that I posted in September 2020 has generated about zero interest, and I’m not seeing much greater interest in related articles written by those with much stronger credentials and connections. I’m not optimistic that the work from Oxford U will result in any change in calculation protocol specified by IPCC for annual NIR submissions.

One problem is that the improved calculation methodology also applies to methane emissions associated with fossil hydrocarbons. That’s most of the 69% in Canada not represented by agriculture. This includes natural gas pipeline leaks. Politically, few in the environmental/climate-change lobby want to make changes that would be seen to benefit ‘Big Oil and Gas.’

The other issue involves the potential use of seaweed extracts and other rumen additives to reduce methane production. These technologies look very promising though are largely still at the ‘proof of concept’ stage of development. More R and D work will be required before they can be truly classed as a principal solution for reducing GHG/CH4 emissions associated with ruminant agriculture.

From a broader agricultural perspective, any technology that encourages beef/sheep production will also encourage greater use of perennial forage crops in Canadian farm crop rotations – with major benefits to soil quality and C sequestration opportunities.

Scope for Zero Emissions?

The focus of international discussion, at least in high-profile circles of late, is not so much strategies for partial reductions, but rather, for zero net emissions. Nations and corporation are both competing for media attention in announcing their new strategies for achieving zero emissions by 2050 or even sooner. The cynic in me doubts that most of these mean much as they depend on major future use of technologies that are far from proven, and/or calculations based on avoiding future GHG-producing practices (e.g., credits for not cutting down trees or draining swamps). They also depend on a level of political and public support which is not yet apparent.

So what’s the prospect for zero emissions in agriculture? I’m not aware of many attempts by main-stream agriculture to tackle this question but there are a few. I give credit below to two farm/agricultural groups that have attempted to develop/define strategies, even though I think that both have major flaws.

One is the National Farmers Union for England and Wales which, in this document, describes a strategy for achieving zero emissions from agriculture by 2040. The strategy is based in part on a national strategy developed by the Royal Society. In the NFU’s plan to eliminate the current emission total of about 45 Mt CO2e, it proposes to achieve about 25% of that by reductions in direct emissions of CO2, N2O and CH4 from farming; 20% by additional C sequestration in farm soils, fence rows, farm woodlots and wetlands; and about 55% by the  combination of biofuel production/consumption coupled with the storage of CO2 released in ethanol manufacture using  ‘carbon capture and storage’ (CCS) technologies (essentially underground storage of CO2, though the NFU leaves the door open to other CCS options).

It’s a big stretch to assume that either the sequestration or biofuel-CCS goals are achievable, but full marks to the NFU for offering something.

Another report is one done by the Aspen Institute in Colorado for the US Farmers and Ranchers Alliance (USFRA). The report, based in part on an analysis of negative emission and sequestration technologies by the National Academies of Sciences, Medicine and Engineering (it, in turn, based on an earlier report by Eagle et al), projects that US agriculture could reduce net GHG emissions by 147% by expanded use of technologies such as no tillage, cover crops, better grazing methods and use “frontier” technologies now in early stages of development. (USFRA, in a more recent report with the Foundation for Food & Agriculture Research and based on an analysis done by a team led by the Nature Conservancy, states that the potential offset is actually more than 200%, though with much of this coming from the management of new and existing forests on farmer-owned land.)

One problem with both the NFU and the USFRA proposals is that, even if the projected reductions were to be attained, most of them would not be credited to ‘Agriculture.’ Rather, the credits would go to other sectors such as ‘Energy’ (for biofuels) and Land Use and Land Use Changes (for C sequestration).

 Using the present IPCC calculation methodology, one could offset 100% or more of agricultural emissions with soil carbon storage – as in the Aspen-USFRA calculations – and still have agriculture identified as the source of about 8-10% of national GHG emissions as at present.  With the NFU strategy, 75% of the projected GHG reductions/offsets would not be credited to Agriculture.

In brief, while it may be important politically for agriculture to portray that it has plans to achieve zero emissions by mid-century (as the NFU has done), my conclusion is that this is not achievable using even semi-proven technology known today, and especially with current accounting procedures.

Bottom line:

Sizable opportunities exist now to reduce nitrous oxide emissions in agriculture using proven technologies, and with others looking promising – eg., selective use of cover crops to reduce N2O spring-time emissions. The potential for methane emission reduction is promising if new rumen-additive technologies prove feasible. The opportunities for C sequestration appear more limited unless means can be found for more greatly expanded use of perennial forages in farm crop rotations.