In the Columbia Basin, summer rainfall is inadequate for warm-season crops, but the area is well suited for winter annuals and cool-season grasses. Although winter wheat yields in the region have increased over time, soil carbon and organic matter have continued to decline under the winter wheat/summer fallow rotation common in this area--even where soil erosion is minimal. In the 1930’s, several long-term experiments were started at the Pendleton Experiment Station in northeastern Oregon to evaluate the effects of fertilizer, residue management, and tillage on crop productivity in the cereal-producing, dryland regions of the Columbia Basin. A chronosequence of soil carbon determinations from the long-term experiments was evaluated. Soil carbon declined in all conventional management systems, except when substantial organic amendments were applied. Nitrogen fertilization reduced the soil carbon decrease; however, residue burning intensified soil carbon loss. Manure applications increased soil carbon in the upper depths, but even with manure additions, soil carbon below 12 in declined over time. Soil carbon in the upper 12 in increased substantially under a grass pasture management. We concluded that the wheat/fallow system is very detrimental to maintaining soil carbon.
Carbon sequestration, long-term experiments, organic matter, residue management, soil quality
Soil organic matter (SOM) is an important component of soil productivity; carbon (C) comprises about 51 percent of the SOM. It is estimated that the prairie grasslands of the Columbia Plateau required more than 10,000 years to reach a stable SOM content. Large areas of northeastern Oregon and southeastern Washington were first cultivated in the late 1880’s. At the outset, these lands were very fertile; however, immediately following cultivation they lost a substantial amount of their SOM. Frequent cultivation to control weeds removed surface vegetation, accelerated water and wind erosion, and intensified the loss of SOM.
A management system that incorporates wheat/fallow rotations, moldboard plowing, and rod weeding is used on 4.5 million acres in north-central Oregon and south-central Washington. This system is considered economical where rainfall is inadequate to produce a profitable crop every year (Leggett et al. 1974; Bolton and Glen 1983). This system is used primarily to store winter precipitation and control weeds; additional reasons for its use include the accumulation of nutrients from the decomposition of residue and SOM, lower incidence of disease, and fewer problems with residue during tillage or seeding operations. In addition, grain yields are usually less variable over time with this system. The fallow period during summer, however, is detrimental to SOM. The system can increase soil erosion and is not biologically sustainable (Rasmussen and Parton 1994). The development of high-yielding semi-dwarf wheat varieties with high water-use efficiency and disease resistance has not compensated for the decline in biological sustainability of soils in the Pacific Northwest (PNW) (Duff et al. 1995). Since the 1950’s, economic sustainability has also declined in PNW fallow cropping systems because costs continue to rise while wheat prices remain stagnant (Duff et al. 1995).
Long-term experiments are feasibly the only way to determine if agricultural management systems will improve, sustain, or degrade the productive capacity of the soil. They can direct agricultural operations by identifying the effects of management practices on soil quality and crop yield. The Pendleton Experiment Station (PES) was established in 1928 to develop farming systems that improve productivity, sustain soil fertility, and reduce erosion; a short time later several long-term experiments were begun. Two of these, the Crop Residue experiment (CR) and the Grass Pasture experiment (GP) were started in 1931. The CR was designed to assess the sustainability of the conventional wheat/fallow rotation. The objective was to determine the effects of N application, residue burning, and pea vine and manure application on soil properties and productivity in a conventional moldboard plow, winter wheat/summer fallow production system. The GP, simulating perennial grassland, was initiated to enable a comparison with intensively managed areas. The overall goal of the GP was to approximate native productivity without fertilizer, irrigation, and tillage.
The purpose of this investigation was to analyze the changes in soil C concentrations in relation to residue management practices in a summer fallow/winter wheat system. Soil carbon content was evaluated in relation to treatment duration and soil depth.
The PES has a Mediterranean climate. The 50-year annual average precipitation is 16.7 in, and the average annual temperature is 60° F. The soil is a Walla Walla silt loam (coarse, silty, mixed mesic, Typic Haploxerolls). The physical and chemical properties of this soil were reported by Pikul and Allmaras (1986).
The CR rotation is winter wheat/fallow with moldboard plow tillage. The experimental design is an ordered block consisting of nine treatments and two replications. The experiment contains duplicate sets of experiments that are offset by 1 year so that data can be obtained annually. Plot size is 38 by 132 ft. Replicates differ in soil depth, slope, and soil N content. Fall stubble burns are conducted in late September. Spring stubble burns are implemented and organic amendments applied in the spring of the fallow year. Late-winter or early-spring herbicides are used to control vegetative growth in wheat stubble until plots are plowed. Plots are plowed 8 in deep within 3 days after spring burning. The soil is then smoothed with a field cultivator/harrow. Weeds are controlled by tillage during the fallow phase and with herbicides during the crop phase. Nitrogen fertilizer is applied before seeding. The C and N content of the upper 24 in of soil has been determined about every 10 years (1931, 1941, 1951, 1964, 1976, 1986, and 1995).
The GP experiment contains no experimental variables. It is 150 ft wide by 360 ft long, and is divided by a waterway. Slopes range from 0 to 3 percent, and soil depth is about 4 ft. It is periodically reseeded with introduced grass selections, occasionally fertilized, and infrequently irrigated. It was grazed until 1985 but not since; however, the grass is clipped once or twice during summer growth. A profile of the PES long-term experiments, including the CR and GP, was provided by Rasmussen and Smiley (1994).
Soil samples for chemical analysis were air-dried, put through a nine-mesh (Tyler) sieve, and finely ground on a roller mill. Carbon was determined by the loss on ignition procedure until 1976, thereafter by a carbon nitrogen analyzer (Alpkem and Fisions).
In both of the management systems examined--winter wheat/summer fallow, and moldboard plow with rod-weeding--soil C has consistently declined from 1931 to 1995 (Figs. 1– 4). When residue was burned and no N fertilizer was applied, the loss rate of soil C was most severe (Fig. 1).
Figure 1. Changes in soil carbon in the top 24 in of soil with fall residue
burn and no N treatment from 1931 to 1995 at the Pendleton Experiment Station,
Oregon.
The addition of N fertilizer diminished the rate of decline in soil C; however, it did not prevent soil C loss over time (Figs. 2 and 3). Also, the addition of pea vines, at a rate of 1 ton of dry vines per acre before planting over a 65-year period, did not preclude soil C loss or substantially alter the rate of loss (Fig. 4). The only system that showed even a slight increase in the soil C level over the 65-year period was when animal manure was applied, at a rate of 10 tons (wet) per acre before seeding.
Figure 2. Changes in soil carbon in the top 24 in of soil with 40 lb N per
acre from 1931 to 1995 at the Pendleton Experiment Station, Oregon.
The rate of soil C loss was similar at all depths, except in the 0- to 12-in depth there was a slight increase in C with manure application. Even with manure additions, the soil C at the 12- to 24-in depth declined at a rate similar to other management systems.
Figure 3. Changes in soil carbon in the top 24 in of soil with 80 lb N per
acre from 1931 to 1995 at the Pendleton Experiment Station, Oregon.
While the amount of C data collected for the GP experiment is not as extensive as for the CR experiment, the long-term trends in soil C are evident. The GP showed a substantial increase in soil C in the upper soil depths from 1931 to 1995 (Fig. 6). However, at the 12- to 24-in depth the soil C remained essentially the same. This indicates that C is not being produced or transported to the lower depths. This suggests that summer fallow and tillage can impact soil C at depths below 12 in and intensify loss of soil C and SOM throughout the profile. While the grass pasture is not an economically viable system, it is included to indicate what is possible in a system without summer/fallow and tillage.
Figure 4. Changes in soil carbon in the top 24 in of soil with pea vine additions
from 1931 to 1995 at the Pendleton Experiment Station, Oregon. Pea vines added
at a rate of 1 ton per acre each crop year.
Figure 5. Changes in soil carbon in the top 24 in of soil with manure addition
from 1931 to 1995 at the Pendleton Experiment Station, Oregon. Manure added
at a rate of 10 tons (wet) per acre each crop year.
Loss of soil C, and also SOM, is generally attributed to microbial action. Burning residue depletes C input into the soil and precludes residue conversion to SOM. Plowing inverts the residue, making it more accessible to soil-dwelling microorganisms. Plowing will also change the gas composition in the soil by reducing carbon dioxide and increasing oxygen. Increasing the oxygen concentration will stimulate aerobic microorganisms by increasing their metabolic activity with the concomitant conversion of residue and SOM to carbon dioxide. Summer fallow is detrimental to
Figure 6. Changes in soil carbon in the top 24 in of soil under grass pasture
from 1931 to 1995 at the Pendleton Experiment Station, Oregon.
SOM because it will generally reduce residue production, thereby reducing the material that can eventually be converted to SOM. Fallow ground also provides an environment with good soil moisture during the warm summer months. This condition favors microbial activity that will reduce residue and also SOM.
After 65 years, the soil C content, although dynamic, varies for different cropping systems. In contrast to almost all cultivated systems, the continuous grass-pasture soil showed an increase in C content. A conventional management system, with moldboard plowing and residue burning, results in more soil C loss than no-till or reduced tillage management. Soil C content can be increased with amendments such as manure.
The wheat/fallow system is very detrimental to soil C content. Residue additions are decreased and, although the fallow system reduces evapotranspiration and saves water for the upcoming crop, the soil moisture in the fallow phase of the rotation promotes microbial oxidation of SOM during the hot summer months. Simply put, continued removal of C and associated nutrients will result in further deterioration in soil quality and productivity.
We thank Roger Goller, Katherine Skirvin, Amy Baker, and Chris Roager for sample collection and technical assistance, and Bob Correa and Karl Rhinhart for farm operations. We also thank Katherine Skirvin and Steve Petrie for helpful discussions, and Paul Rasmussen, Richard Smiley, and Steve Petrie for management or operational support.
Bolton, F.E., and D.M. Glen. 1983. Fallow-cropping systems in the Pacific Northwest. Technical Bulletin 146, Oregon State University, Agricultural Experiment Station, Corvallis, OR.
Duff, B., P.E. Rasmussen, and R.W. Smiley. 1995. Wheat/fallow systems in semi-arid regions of the Pacific NW America. Pages 85-109 in V. Barnett, R. Payne, and R. Steiner (editors). Agricultural Sustainability: Economic, Environmental and Statistical Consideration. John Wiley and Sons, Chichester.
Leggett, G.E., R.E. Ramig, L.C. Johnson, and T.W. Masse. 1974. Summer fallow in the Northwest. Pages 110-135 in Summer Fallow in the Western United States. Conservation Research Report No. 17, ARS-USDA, Washington, DC.
Pikul, J.L., Jr., and R.R. Allmaras. 1986. Physical and chemical properties of a Haploxeroll after fifty years of residue management. Soil Science Society of America Journal. 50:214-219.
Rasmussen, P.E., and W.J. Parton. 1994. Long-term effects of residue management in wheat/fallow. I. Inputs, yield, soil organic matter. Soil Science Society of America Journal 58:523-530.
Rassmussen, P.E., and R.W. Smiley. 1994. Long-term experiments at the Pendleton Agricultural Research Center. 1994 Columbia Basin Agricultural Research Annual Report 933:14-21. [an error occurred while processing this directive]