Return to 2001 Conference Proceedings

How Tillage and Cropping Systems Change Soil Organic Matter. Inland Pacific Northwest Research
Stewart Wuest, Soil Scientist, USDA-ARS, Pendleton, OR; Steve Albrecht, Microbiologist, USDA-ARS, Pendleton OR; Jeff L Smith, Soil Biochemist, USDA-ARS, Pullman, WA; David Bezdicek, Professor, WSU, Pullman, WA.

Soil organic matter is closely associated with soil quality in the minds of most farmers and scientists, and for a good reason. The very definition of soil is linked to the effects of organisms and their residues. It is the accumulation of organic matter that changes deposits of sand, silt and clay into soil.

Every ecosystem has an equilibrium soil organic matter content it will eventually reach if climate, vegetation, and disturbance remain constant. This is true of cropping systems as well as natural systems. I will use an analogy to explain how equilibrium soil carbon levels occur, and how they can be changed by cropping practices. Then I will present data from local experiments that have measured organic carbon changes resulting from different tillage and crop rotations. Tables and figures will present soil organic matter in terms of soil organic carbon, as a percent of total soil weight. (If you want to convert from organic carbon to organic matter, which includes oxygen, hydrogen, nitrogen, sulfur etc. in addition to carbon, multiply soil organic carbon by 1.7.) To conclude, I will address some of the factors which make short-term soil carbon changes difficult to evaluate, and attempt to put soil quality changes caused by reduced tillage in a perspective that includes effects on profitability.

Here is an analogy to help form a picture of how an equilibrium is reached. When building a campfire, you can change the size of the fire by how often and how much wood is added. If you add one log every hour, the fire will not become very large. If you add one log every minute, the fire will grow into a very large campfire indeed. But, it will not grow indefinitely. Eventually the fire will be large enough and hot enough to burn up the equivalent of one log per minute. At that point the fire will stop growing in size, and that is what is meant by equilibrium. The amount added equals the amount consumed, and the quantities of fire, heat, and coals stops changing.

Conversion from native sod to farmland
The soils we farm were originally part of grassland ecosystems that developed over centuries of relatively stable climate and vegetation. At the time we broke them out of native grasses, they were at an equilibrium organic matter content that depended mostly on site productivity (precipitation and temperature, water holding capacity, soil texture, and vegetation types) and conditions for decomposition of organic matter (moisture and temperature of residues).

When we converted the natural systems to cropping systems, we changed the vegetation and therefore the amount of residue returned to the soil. In some places the amount of residue returned to the soil was less than the natural system, and others it may have been more. We also added soil disturbance (tillage) as a factor in the decomposition of residues. Burying residues keeps them moist longer and speeds decomposition. Soil disturbance also has an effect on water infiltration, runoff and erosion, and soil temperature.

Just like the rate that fuel is added to a fire determines is equilibrium size, the rate that residue is returned to the soil helps to determine the soil organic matter level. Different cropping systems add back different amounts of residue to the soil each year. This results in different soil carbon equilibrium levels each cropping system is moving toward. In Table 1 the annual input of above-ground residues is compared for a grass pasture and several plowed cropping systems that have been in place for 40 to 60 years (Rasmussen and Smiley, 1997). The soil carbon levels relate fairly well to the amount of carbon the cropping systems add, but carbon input can not be the whole picture because some large differences in input are not showing up as soil carbon differences.

Table 1. Residue input versus soil carbon in plowed systems top 8 inches
 
Annual carbon
Soil carbon
 
input, lb/a
%
(Grass Pasture)
2240
2.2
Wheat/Pea
2040
1.6
Continuous Winter Wheat
2010
1.5
Wheat/Fallow plus Manure
2150
1.5
Wheat/Fallow, 90 lb N
1500
1.1
Wheat/Fallow, 0 lb N
1010
1.2
Wheat/Fallow, fall burn, 0 lb N
360
1.1

Assuming that the early farmers added no fertilizer or irrigation water, the new equilibrium soil organic matter content would be lower than the natural system. (Farming does not always lower soil carbon levels. In dry areas, irrigation and fertilizers can increase soil organic matter over the original native levels.) Based on long-term Pendleton, OR data, Paul Rasmussen estimates that about 35% of the original soil organic matter was lost in the first 50 years under cultivation. A study in western Nebraska (Peterson et al., 1998) shows how converting from native sod to winter wheat/fallow reduces soil organic matter levels over a twenty-year period (Figure 1). This part of Nebraska has similar annual precipitation to Pendleton (17 inches). Intensity of tillage is a big factor in determining the equilibrium level that each system is headed toward, even if tillage systems produce the same amount of residue. This means that it is not just the quantity of fuel we add to the campfire, but how much we stir the fire.

If we look at some of the long-term studies at Pendleton, we can see the combined effect of tillage and rotation on the rate of carbon change (Table 2). The studies have run for different lengths of time so the numbers are not precisely comparable, but there are clearly some relationships between carbon change and cropping system. Wheat/fallow has lost the most carbon annually, and reduced tillage reduces the rate of loss some. The wheat/pea rotation loss rate is lower than wheat/fallow under plow tillage, and carbon levels are pretty much holding their own under a wheat/pea minimum till. Minimum tillage in this study means sweep four to six inches deep and rodweed at one to two inches. There has been very little change in soil carbon levels under annual winter wheat in a plowed system. Under no-till winter wheat/spring wheat, there was a small gain during ten years of study. A grass pasture, started on ground that had been farmed to wheat for 50 years, gained about one third of one percent soil carbon in 60 years.

Table 2. Effect of tillage on soil carbon (top 8 inches), Pendleton.
 
Years of
C change
 
study
per 60 years
 
%
Wheat/Fallow, plow
60
-0.355
Wheat/Fallow, plow
40
-0.345
Wheat/Fallow, stubble mulch
40
-0.282
Wheat/Pea, plow
28
-0.268
Wheat/Pea, minimum till
28
-0.009
Wheat/Wheat, plow
60
-0.099
Wheat/Wheat, no-till
10
0.060
Grass pasture
60
0.338

Difficulties in measuring soil organic matter So far I have been presenting basic factors that affect soil carbon levels. Making measurements of soil carbon, especially over the short term, is a lot more difficult than it might seem. It is difficult to accurately sample carbon from different cropping systems for a number of reasons. First, there are forms of carbon in the soil which are not soil organic matter. Mineral carbonates, of which caliche is one form, will be included as a part of total soil carbon unless they are removed before analysis. If a soil testing laboratory has removed carbonates before doing carbon analysis, it will probably label the results as organic carbon, or they will use the conversion factor and report the results as soil organic matter.

A second reason it is difficult to accurately measure soil organic carbon levels is spatial variability. Samples taken from different places in a field will produce different results. Repeatedly sampling the same location over a period of decades will help give a consistent picture of changes taking place, but this does not give an accurate measure of what is happening over the rest of the field. To investigate spatial variability Jeff Smith, ARS Soil Biochemist at Pullman, WA took soil samples from a 60-square mile county, and also from a 1.25 acre field plot. The county data is from Whitman County, Washington and the field data from Pullman, Washington. The county ranges in rainfall from 7 to 22 inches and the field samples were from a 20 inch rainfall zone. The county soils range from Shano (0.6 average % carbon) to Palouse silt loam (1.6 average % carbon). Soil samples were taken on a grid pattern from 220 locations across the county, and likewise 220 locations in the field plot. Figure 2 shows the distribution of results from analyses of percent soil carbon and pH for individual samples. Within the 1.25 acre field plot the soil carbon measurements varied from about 0.5 % to 3%. Countywide the variation was similar. In comparison, pH measurements were less varied in the field and more varied over the county. The variation in soil carbon within a 1.25 acre field illustrated in Figure 2 should make it clear that carefully planned sampling strategies are necessary if we want to detect true changes in a field's soil carbon over time.

Organic matter is a mixture of substances
A third reason it is difficult to accurately measure soil organic carbon levels is the mixture of substances that make up soil organic matter. We define soil organic matter as any substance derived from an organism (plant or animal) that can no longer be recognized as part of that original organism. In practical terms, however, anything that contains carbon and makes it through a fine sieve gets analyzed as soil organic matter carbon, even if it is just finely ground chaff from a recent harvest. What we call soil organic matter, therefore, includes everything from un-decomposed fresh residue to recalcitrant, tar-like substances that are no more digestible by bacteria than highway asphalt. Steve Albrecht, ARS, Pendleton, OR, and David Bezdicek, WSU, Pullman, WA, have been studying different types of soil organic matter. They used a mild acid treatment to measure stable carbon versus total carbon. Figure 3 shows the relative amounts of stable carbon to total carbon at 0 to 2, 2 to 4 and 4 to 8 inch depths. The black bars are samples from the 60-year grass pasture, and the gray bars are from a 60-year, plowed, wheat/fallow plot. Notice that most of the increase in soil carbon under grass pasture is not in the stable carbon fraction, but in the more readily decomposable, less stable part of total soil carbon. Also notice that most of the difference is very near the soil surface.

There are three implications of measuring a mixture of organic carbon substances in making accurate estimates of soil organic matter gains. First, test results will vary depending upon what season the samples are taken. If they are taken soon after harvest, there will be more un-decomposed, fresh residues included in the sample. If the samples are taken late in the spring, much of the easily decomposed residue will be gone, especially in tilled systems. Second, soil organic matter follows the easy-come, easy-go principle. Organic matter that accumulates quickly can also be lost quickly. In a particular soil type in a particular climate, a cropping system that produces a higher soil carbon level will also have a higher proportion of less stable organic substances.

A third implication of organic matter being measured as the total of all organic substances that pass through a fine sieve is that untilled systems accumulate significant amounts of organic material which does not pass through the sieve and is therefore not counted. Dave Bezdicek and Mary Fauci measured the amount of material that did not pass through a 2 millimeter sieve from the top two inches of a 25-year no-till field near Pullman, WA. In a two-inch deep soil sample, six percent did not pass through the screen, and about half of that material was organic residues. This means that there was additional carbon in partially decomposed residue in or on the surface soil that would not get measured in a soil test. This accumulation of surface residue has important effects on the soil environment. No-till systems which accumulate residues on or very near the soil surface are undergoing some very significant changes in soil quality which are not due to measured soil organic matter increases.

Organic matter increases are not the whole picture
Cropping systems that avoid tillage not only reduce organic matter loss (by not "stirring the campfire"), they accumulate a larger fraction of soil organic matter near the surface. Since the soil surface is a critical interface for air and water movement, high organic matter contents near the surface have a greater effect on soil quality than if an equivalent amount of organic matter was distributed throughout the plow layer. Figure 4 shows organic carbon levels for four wheat/pea tillage systems. Both of the plowed systems (circles) have less carbon in the surface 0 to 4 inches, and more in the 4 to 8 inch layer than the non-inversion tillage systems (triangles) because plowing mixes the top 8 inches of soil every year. The minimum till system has the greatest carbon level in the surface layer, and the lowest at 4 to 8 inches. Therefore, even if a reduction in tillage does not result in a substantial increase in the soil organic matter equilibrium, stratification of organic matter on or near the soil surface will have an impact on soil quality. The minimum-till system also accumulates un-decomposed residues on the soil surface.

Measuring the value of increased soil quality in terms of profitable production is difficult. Having a soil which crusts less often, infiltrates water better, drains more rapidly after a rain, erodes less, is more mellow when tilled or seeded, or provides easier root growth should provide greater profits in the long run, even if maximum yields under normal conditions are not higher. Fortunately, the need to reduce production costs and increase the use of available precipitation is leading to a reduction in tillage. This should result in soil quality increases in terms of soil organic matter accumulation near the soil surface.

Substantial changes in stable soil organic matter take decades to develop and are difficult to measure accurately, even though residue inputs can vary widely on a yearly basis. Differences in soil quality attributed to minimum tillage cropping systems are probably due to a combination of tillage related factors, of which soil organic matter increase is only one part.

References

Peterson, G.A., A.D. Halvorson, J.L. Havlin, O.R. Jones, D.J. Lyon, and D.L. Tanaka. 1998. Reduced tillage and increasing cropping intensity in the Great Plains conserves soil C. Soil and Tillage Research 47:207-218.

Rasmussen, P.E., and R.W. Smiley. 1977. Soil carbon and nitrogen changes in long-term agricultural experiments at Pendleton, Oregon. In Soil Organic Matter in Temperate Agroecosystems, Long Term Experiments in North America. ed. E.A. Paul, K. Paustain, E.T. Elliot, and C.V. Cole, pp. 353-360. CRC Press, Boca Raton.