This is a personal analysis of our own farm. It’s posted here because the thought process and information might be of interest to other farmers asking the same question.

I’ll start with the conclusions. The explanation of how I reached them is provided in detail below.

I’d best call these interim conclusions as they are very likely to change; this is a subject on which I have much to learn. But based on what I’ve garnered thus far, these seem like the essentials:

1. A rotation of corn and soybeans is inferior to corn-soybeans-wheat for soil organic matter maintenance/enhancement, but the difference is likely small, indeed non-existent if wheat straw is marketed off the farm, and cover crops are not used.
2. The best rotational crop for increasing soil organic matter may well be corn, not wheat. Wheat plus red clover is better, but only if you get decent red clover stands. The worst is soybeans.
3. Control of soil erosion using no-till soybeans and targeted seeding of winter rye into soybeans, or after soybean harvest, is the most important long-term consideration. I’m not keen on no-till corn for our farm, based on past experiences, but strip tillage would likely be better than our present once-over spring cultivation – for improving soil cover, reducing oxidation of soil organic matter, and reducing springtime earthworm mortality.
4. Winter rye’s effectiveness in controlling glyphosate-tolerant broadleaf weeds should prove advantageous in years ahead – as the number of these species increases (at present, we have only fleabane) and herbicide-based solutions become less effective.
5. Higher crop yields – coupled with less tillage – seem the best route for increasing soil organic matter content.
6. Cover crops such as late-summer/early-autumn-seeded oats should also improve soil organic matter content but very slowly – if this can be done at minimal cost to make it economically justifiable – say, $15/acre, or less.
7. Cover crop usage will not likely reduce our annual fertilizer bill.
8. Economically, the weakest part of a corn-soybean rotation may be depressed soybean yields because of too-frequent planting of soybeans.
9. I emphasize that these conclusions apply only to our farm, with no assumption that they are right for others.

—-

For more than 30 years, our small cash-crop farm near Guelph had a four-crop, five-year rotation of corn, soybeans, corn, white (navy) beans and winter wheat, but then we switched about three years ago to the simpler corn-soybeans. The reasons were several, but the main one was us getting older. We wanted something simpler.

That’s led to two questions:  Are we treating the soil properly? And if not, what can we do about it? That’s what this column’s about.

Soil erosion

The first priority in good soil management is eliminating soil erosion, or at least reducing it to the minimum possible.

We started farming in 1972 with no-till corn but a combination of the inadequate design of early no-till planters, inability to incorporate N and K fertilizer, and unsatisfactory yields meant a change to shallow, once-over springtime cultivation before corn. This is quick and cheap, and does a reasonable job of maintaining soil residue cover, though not as good as with no-tillage. With our no-till soybeans, much of the stalk residue from the previous corn crop remains on the soil surface to retard rainwater flow even after soybean harvest.

Despite the corn-soybean residue cover with no-till soys, I still saw some soil movement in areas of the soybean stubble where the rainfall runoff concentrated in early spring. Winter rye broadcast-seeded in those strips in late August before soybean leaf drop should stop that. Indeed, with that change, soil erosion should actually diminish with our shift to corn-soybeans.

Calculations of soil organic matter additions

That brings us to the issue of soil organic matter management, which is far tougher. Organic matter is probably the most important soil-health criterion, as it has such a huge effect on water-holding capacity and drought avoidance. But increasing it is very difficult.

How does corn-soybeans compare to our former four-crop rotation, or to a corn-soybeans-wheat rotation which we could have chosen to use but didn’t?

In seeking answers, I found an excellent analysis done by Kludze et al (Bill Deen’s team) in 2010 at the University of Guelph. (A summary is here.) They used data from the Elora Research Station and across Ontario to calculate the amount of crop organic residue needed to maintain soil organic matter. In Table 1, I’ve used some of their numbers, combined with our 10-year crop-insurance yield records, to compare the three choices for cropping sequence.

Table 1. Calculation of organic matter return to the soil with three cropping sequences on the Daynard farm.

Crop or rotation	Grain yield, 10-yr ave.	Above-ground residue; assume harvest index of 0.5	Root:shoot ratio (including exudates)	Roots and exudates	Above-ground plus roots and exudates	‘Shoot-equivalent’ organic matter residue (shoot + 2.4 times root+exudates)
	bu/acre	dry t/ha		dry t/ha
Corn (C)	166	8810	0.5	8810	17620	29950
Soybean(S)	49	2870	0.6	3440	6310	11130
Wheat ( W)	83	4780	0.8	7640	12410	23110
W. Beans (B)	36	1990	0.6	2350	4370	7710
Average per year
C-S-C-W-B		5450		6210	11670	20370
C-S		5840		6120	11960	20540
C-S-W		5490		6630	12110	21400
C-S-W (75% straw removed)		4290		6630	10920	20200

Notes: Harvest index = grain dry weight/(grain + above-ground crop residue). Values for harvest index and root:shoot from Kludze et al, 2010. White bean ratios assumed to be the same as for soybeans.

Considering only aboveground crop residues (column three), the corn-soybean rotation is about 9% better than both our former rotation and corn-soybeans-wheat. But when you include root-plus-root exudates (column six), corn-soybeans is slightly worse than corn-soybeans-wheat if wheat straw is not removed. In truth, given the crudeness of the calculations, all three rotations in column six, except for wheat straw removal, are about the same.

Many researchers have found root and root exudates to be a much better contributor to longer-term soil organic matter than above ground residues – 2.4 times better concluded Rasse et al (2004) in an extensive review. When I take this into account and calculate values for what I’ve termed, ‘shoot-equivalent’ organic matter (final column, Table 1), corn-soybeans-wheat turns out to be 4% better than corn-soybeans.

With cover crops

A common practice is to frost-seed red clover into winter wheat fields in March. This rarely worked satisfactorily on our farm, but other farmers have had more success. Another option – indeed, group of options – is to plant a cover crop into the stubble after wheat, into standing soybeans before leaf drop, or after soybean harvest. The calculations are more speculative than for corn, soybeans and wheat, given the huge differences that have been reported for cover crop yields and the sparsity of data on root:shoot ratios.

Using a variety of sources, including this, I’ve made some guesstimates for aboveground cover crop yield and assumed a cover crop root:shoot ratio of 1:1 (Table 2).

Table 2. Calculation of organic matter return to the soil with several cover crop options following winter wheat and soybeans on the Daynard farm, including a 2.4 weighting for roots + exudates.

Rotation	Cover crop, above ground	‘Shoot-equivalent’ organic matter residue
	dry kg/ha	dry kg/ha
C-S-C-W-B		20,370
C-S		20,540
C-S-W		21,400
C-S-W with spring red clover	3000	24,800
C-S-W with cover crop planted in August	2000	23,670
C-S with cover crop	1000	22,240

 

The ‘shoot equivalent’ organic matter addition increases by 11% with cover crop planted after wheat and 8% with cover crop after soybeans. That’s assuming the cover crop grows adequately when planted after wheat or soybeans. That will not always occur and, hence, the 11 and 5% estimates are likely a bit high. Of course, the calculated value is lower if wheat straw is removal.

Rotational effects and cover crop effects on crop yield

The yields, and notably corn yield, shown in Table 1 are based mostly on a history of a four-crop rotation, but they might be different with other rotations.

Corn has yielded about 2% more in corn-corn-soybeans-wheat than in corn-corn-soybeans-soybeans on silt-loam soil at the Elora Research Station. At Ridgetown on clay loam soil, corn has yielded 13% more in corn-soybean-wheat in tilled plots but only 3% with no-tillage (data are in slide 14 here).

Conclusion: our corn would likely yield slightly more, though not dramatically so, with wheat in the rotation.

With wheat plus red clover cover crop, yield increases have been larger at both Elora and Ridgetown.

Data from both Ridgetown and Elora (see slides 23 and 25 here) show a significantly positive effect on soybean yields with the inclusion of wheat in the rotation, even when the soybeans aren’t planted directly after wheat. I don’t know whether this effect is caused by more wheat or less soybeans in the rotation.

Data showing any effect on corn and other field crop yields using cover crops seeded after wheat and soybean harvest are harder to find. Some reports show a yield increase while others show a decline. This report from Iowa State University explores the yield effects in some detail. The safest assumption seems to be no yield effect.

Measurements of soil organic matter (SOM)

Any reasonable practice that increases SOM, consistent with the need for the farm to generate reasonable profits, is to be encouraged. Unfortunately that’s hard to do, as discussed by Poulton et al (2018) based on experience of more than 150 years of research at Rothamsted Research, United Kingdom.

Data collected after 20 years of crop rotational research at the Elora Research Station show a 5% (statistically non-significant) higher organic soil carbon content for corn-corn-soybeans-wheat compared to corn-corn-soybeans-soybeans, and another 1% increase if red clover is seeded into wheat. These are for the upper approximately 34 cm of soil depth. (The sampling depth was varied to adjust for differences in bulk density; that adjustment is critical as discussed by Poulton et al).

At Ridgetown, Van Eerd et al (2014) measured a 4% (statistically non-significant; averaged across two tillage regimes) increase in the organic carbon content for the upper 120 cm of soil after 11 years of corn-soybeans-wheat compared to corn-soybeans.

While farm publications often contain testimonials of major increases in soil organic matter occurring after a few years of cover crop use, supportive documentation in scientific literature is sparse.

Poeplau and Don (2014), in a meta-review of published research on effects of cover crops on soil organic carbon, concluded that “The time since introduction of cover crops in crop rotations was linearly correlated with [soil organic carbon] stock change (R² = 0.19) with an annual change rate of 0.32 + 0.08 [t/ha/year] in a mean soil depth of 22cm.” Unfortunately, their analysis has critical flaws. For one, selected research papers  include several (and including most from Canada) that involve dedicated ‘green manure’ type crops grown for the whole season. Secondly, 70% of the selected studies include no measurement of bulk density; the authors attempt to estimate missing bulk density values using a single negative linear relationship between percent soil organic matter and soil density – for studies in 11 countries on four continents. There are also some weaknesses in statistical analyses. I’m very skeptical about accepting the conclusion from this review.

Calculating changes in SOM

A review of scientific literature suggests it’s difficult to predict changes in SOM content based on annual additions. So many other factors are important.

Kludze et al (2010) assumed, for the Elora Research Station, that 15.1% of crop residues are converted into soil organic matter and that, on average, 2.5% of resident soil organic matter is mineralized (i.e., respired, broken down, lost) per year. They calculated that a minimal annual addition of about 11 t/ha/year of crop residues, roots and root exudates is needed for SOM to remain constant. They assumed the same rate of decomposition for shoot and root residues/exudates.

I did similar calculations for a corn-soybean on our farm, using a range in rates of annual soil OM mineralization and decomposition of added crop residues, and including a 2.4 root/exudate weighting (Table 3).

Table 3. Calculations of soil organic matter losses and gains with various assumptions about rate of soil OM mineralization and annual decomposition of added residues, roots and exudates.

	Rate of SOM mineralization (loss) per year (%)
	2.0	2.25	2.5
SOM loss, kg/ha/year	-1740	-1960	-2180
	Annual rate of conversion of crop residues, roots, exudates into SOM (%)
	10%	12.5%	15%
SOM gain, kg/ha/year	+1490	+1860	+2230

Note: It is assumed a corn-soybean rotation adds 20,540 kg/ha of ‘shoot-equivalent’ organic dry matter per year (root and exudates weighted by 2.4), that soil is 4% organic matter in upper 15cm of soil and that soil bulk density is 1.45 g/cm3. Assume the added organic matter is 42% carbon and soil organic matter is 58% carbon.

Table 3 shows that by using different combinations of rates of annual SOM mineralization and addition, I can calculate either an annual gain or loss in SOM for our farm. If I use the Kludze et al choice of 2.5% mineralization and 15% annual addition, our farm soil breaks even.

Importantly, there seems no good reason to believe that our SOM is increasing, and it may be slowly declining.

Note, I am using the range in rates of mineralization and decomposition for a silt loam soil at Elora, which should be applicable to our Guelph Loam soil. These values would be different for a soil higher in clay (slower mineralization or residue decomposition because organic compounds are bonded to clay minerals), and the reverse for a sandy soil. (See Rasse et al.)

Effects of cover crops and higher crop yields

Any increase in annual addition through cover crops or other means will increase the likelihood of a gain in SOM.

Using the same calculation procedure used in Table 3, a cover crop seeded into or after soybean harvest would mean about a 250 kg/ha/year increase in SOM, assuming 15% annual conversion into SOM.

The 250 kg/ha/year equates to an annual gain in percent soil organic matter of about 0.01% with cover crop usage. That’s tiny though it does equate to nearly 0.5% gain over my 47-year farming career.

One other option is increased crop yields. What if our average yields were 20% higher – say to 200 bu/acre for corn and 60 bu/acre with soybeans? That’s a reasonable projection given expected genetic improvement. That would mean about a 620 kg/ha/year increase in SOM using the same assumptions – or about 2.5 times that which might be achieved using cover crops. Of course, it’s reasonable to consider both higher yielding crops and cover crops.

 

What about tillage/no-tillage?

More soil tillage means greater destruction of soil micro-aggregates and faster microbial decay of the organic matter contained within. Organic matter left on the surface with reduced or no tillage also deteriorates more slowly (reference here).

The research data on changes in soil organic matter content with no tillage in Ontario are conflicting. Some comparisons show increased in soil organic matter with no tillage, some the reverse (reference here).

I’m of the opinion that lack of a decrease in soil organic matter may reflect lower crop yields with no-tillage – and with the lower annual organic addition more than offsetting a lower rate of organic matter oxidation in the absence of tillage.

Van Eerd et al (2014) at Ridgetown measured substantially more (>14%) SOM for no-tillage versus annual tillage. Corn generally yielded about 10% more with no-tillage in a corn-soybean rotation.

Dr. Bill Deen (slide 15 here) found corn yield at Elora to be higher with reduced/no tillage when corn followed soybeans, wheat or barley in rotation, but lower for corn after corn or wheat-red clover.

I prefer strip tillage for corn as it provides an easier opportunity for soil-incorporating P and K fertilizer, and perhaps nitrogen too, and counters the tendency for no-tilled soils to warm up more slowly in spring. I wish I knew more about relative effects of no-tillage, strip tillage and shallow-once-over-spring-tillage on earthworm mortality. I suspect that shallow spring tillage is bad, no-tillage is good, and that strip-till is somewhere in between. Some great references are here, here and here.

The ‘sleeper,’ continuous corn

According to Table 1, the best rotation for building soil organic matter would be monoculture corn.

But data from crop rotation studies in Ontario rarely show the highest soil organic matter levels with continuous corn. Perhaps that’s because corn mostly yields less in monoculture than in rotations.

That said, many Ontario farmers have grown corn successfully several years in succession – especially on sandy soils that need all the organic matter they can get.

Christine Brown of the Ontario Ministry of Agriculture, Food and Rural Affairs calculated in a 2017 presentation that a corn-corn-soybean rotation could be more beneficial to SOM than corn-soybeans-wheat. A corn-corn-soybean rotation would also address concerns about too frequent inclusion of soybeans in a corn-soybean rotation – though it would mean a yield reduction for second-year corn.

The calculations above are based on an assumption that all crop organic matter is equivalent for SOM enhancement, with the condition that root+exudate organic matter is 2.4 times more valuable. But farm experience indicates a faster rate of breakdown for some plant material, especially vegetative material that is lusher and with a lower C:N ratio. Highly suberized residue breaks down more slowly. (See Rasse et al for discussion.)

The answers are far from ‘all-in’ on relative merits of different cropping-tillage combinations for SOM preservation and enhancement.

Other cover crop benefits for our farm?

Deep-rooted cover crops can reduce soil compaction where this has occurred because of heavy field equipment travel, especially on wet soil. That’s not a priority on our farm, to the best of my knowledge, but it may be critical for others.

Water infiltration is a closely related issue with infiltration increasing in response to any technique that slows water flow across the soil surface or opens pores to deeper soil and tile drains below. Maintenance of surface residues with low/no-tillage is very effective. Earthworm-channel formation promotes infiltration as do dead-root channels. I’ve no doubt that cover crops would improve infiltration even more though it is difficult to quantify the benefit.

Weed control is another known benefit. Winter rye has been found very effective in controlling problem weeds like glyphosate-resistant fleabane. Should our existing weed control program prove insufficient for that purpose, rye will be the obvious choice.

A problem comes when the cover is incomplete (eg, red clover into wheat) and weeds grow abundantly in those patches if corrective measures are not taken. There is also need to ensure that the cover crop species itself does not become a weed of concern – and that new weeds aren’t introduced in cheap cover crop seed.

Finally, cover crops have been considered as a means of reducing nitrogen fertilizer need for corn. Research data show, consistently, the benefit of spring-seeded red cover. The same applies to a much lower extent for legume species planted after wheat harvest.

Research commonly shows lower soil nitrate levels in late autumn after cover crop establishment – at least for non-legumes species. However, the same plots often show little difference in soil nitrate levels in spring compared to soils without cover crops. Studies in Ontario have generally shown no reduction in N requirement for the subsequent corn crop after cover crops (legumes excepted, especially red clover). (See here and here, and also unpublished data from Greg Stewart, former OMAFRA corn lead.) Ontario crop consultants generally report no reduction in amount of N recommended for corn following a cover crop, unless the cover crop is a legume – and some consultants recommend more.

Acknowledgments

This column represents the culmination of a personal project of more than a year. I thank the following individuals for providing advice, insight and information (though none should be blamed for what’s written above): Drs. Laura Van Eerd, Bill Deen and Dave Hooker, and Ken Janovicek, University of Guelph; Andrew McGuire, Washington State University; Anne Verhallen, Christine Brown, Sebastian Belliard, Jake Munroe and Ian McDonald, Ontario Ministry of Agriculture, Food and Rural Affairs; Dr. Angela Straathof and Harold Rudy, Ontario Soil and Crop Improvement Association; Patrick Lynch, Peter Johnson (Real Agriculture), Greg Stewart (Maizex Seeds), Ken Nixon and several Ontario crop consultants (whom, for fear of missing someone, I’ll not list here); the many knowledgeable and experienced farmers I’ve interacted with on Twitter; and finally, thanks to you – all five of you – who have read this far!