Tillage Practices for Improving Nitrogen Cycling and Soil Quality

Immediate responses of soil to tillage events. After discing and rototillage of two Salinas soils, microbial biomass and activity declined within a few hours, but carbon dioxide emission and production of nitrate increased. Carbon dioxide emission appears to be caused by physical degassing and equilibration of carbon dioxide that has accumulated in the soil, and not to an increase in microbial respiration. Thus tillage appears to be a disruptive event with immediate potential for C and N loss. Our cooperators were Triangle Farms and Tanimura & Antle, Inc. Less emphasis was being placed on such short-term studies due to other funding sources (e.g., USDA-NRI, Kearney Foundation). Our focus for the SARE grant has subsequently emphasized longer-term, on-farm projects.

On-farm comparison of three types of reduced tillage. A long-term assessment was conducted on effects of alternative tillage practices on soil properties, vegetable yield, and crop diseases on a large, mainstream farm in the Salinas Valley. Our farmer-cooperator is Israel Morales from American Farms, who has designed tillage practices and equipment for deep minimum tillage on semi-permanent beds for fields under sprinkler/furrow irrigation. His methods for retaining beds consist of five tillage operations totaling 1.5 to 2 hours per acre, and require much less time and fuel than conventional methods that breakdown beds, disk, and re-form beds after every crop.

The five tillage operations used in the deep five-step minimum tillage method are: 1) mow/chop crop residues; 2) minimum-till chisel which simultaneously chisels the furrows and disk-hills the beds; 3) ‘Sundance’ which disks the surface of the beds and clears the furrows; 4) minimum-till rip which breaks the compacted layer at depth in the beds; and 5) surface roto-till/mulch to smooth the surface and prepare a seedbed.

We found that the five-step minimum tillage method can produce less soil compaction than conventional tillage methods. We compared bulk density (g soil/cm³ at 0 to 47 cm deep) on fields farmed by American Farms with neighbors' fields under conventional tillage. Four soil types were compared. The soil types ranged from a sandy loam to a clay loam. In two of the soil types, bulk density was significantly lower on the fields farmed with the five-step minimum tillage method, and on two other soil types, there was no significant difference in bulk density.

We monitored the long-term effects of three reduced tillage methods on a vegetable field on a silty clay loam at American Farms. The three tillage methods were imposed as strip plots in a field that was conventionally tilled with deep chiseling and ripping in winter of 1996. The three treatments are: 1) the five-step minimum tillage method described above; 2) only ‘minimum-till chisel’ (see operation 2 above); and 3) only ‘Sundance’, i.e., shallow reduced tillage. All of the three tillage treatments are designed to minimize time and fuel use, and to retain semi-permanent beds in place over multiple years. A rationale for semi-permanent beds is to restrict compaction from heavy machinery to the furrows between the beds, so that the beds have less compaction in the root zone.

Results. By 1998, pronounced differences had occurred between the different methods for reduced tillage. Lettuce fresh weight was significantly lower in the ‘Sundance’ treatment that shallowly tills the beds (842 g/head in ‘Sundance’ vs. 861 g/head in the ‘minimum-till’ chisel vs. 903 g/head in the five-step reduced tillage treatment). Higher propensity for plant diseases may have contributed to the yield decrease in the shallowly–tilled treatment, where there was a significantly higher percentage of lettuce plants with lettuce drop (Sclerotinia minor) wilt symptoms in the ‘Sundance’ treatment (5.3%) than either of the ‘minimum-till’ treatments that till the soil more deeply (1.2 and 1.9%). Also, the mean percentage of the taproot showing cracking and yellowing characteristic of corky root disease (Rhizomonas suberifaciens) was significantly higher in the ‘Sundance’ treatment. Concentrations of inorganic N in the surface layer (0-10 cm depth) were higher in the ‘Sundance’ treatment at crop harvest. Soil microbial biomass C was higher in the ‘Sundance’ treatment than in the deeper five-step reduced tillage treatment. Assays for N mineralization potential were conducted, but are difficult to interpret due to very high concentrations of ambient inorganic N. No difference in bulk density was found between the treatments.

In 1999, lettuce yield was again significantly lower in the ‘Sundance’ treatment (856 g/head) compared to the deeper tillage treatments (926 g/head in the ‘minimum-till’ chisel treatment and 955 g/head in the five-step minimum tillage treatment). The occurrence of lettuce drop symptoms was higher in the ‘Sundance’ treatment (1.9%) compared to the ‘minimum-till’ chisel treatment (1.0%) and the five-step reduced tillage treatment (0.6%). There was a trend toward higher corky root severity in the ‘Sundance’ treatment, but differences were not significant. Soil microbial biomass C was not significantly different between treatments, but there was a trend towards higher values in the shallow tillage treatment, as was observed in the previous year.

In summary, this on-farm project has shown that after two years of shallow minimum tillage with only a ‘Sundance’ implement, lettuce yield decreased and severity of lettuce drop disease increased. Yields were highest when both chiseling and ripping were included in the operations for maintaining semi-permanent beds. Corky root disease severity was variable from year to year, as was soil microbial biomass C. These results indicate that shallow tillage cannot be recommended for more than one year at a time. Shallow minimum tillage may be satisfactory for short periods, for example between vegetable crops, as a means to reduce labor and fuel use, and hasten the planting of a second crop during the summer.

Whole-system on-farm comparison of reduced and conventional tillage, with and without high inputs of organic matter. In April, 1998, we established a large multidisciplinary project to investigate the combined effects of tillage and OM additions on various aspects of the vegetable cropping system: yield, soil fertility, soil microbial biomass and activity, plant disease, insect pests, weeds, and economic and fuel costs. Our cooperator is Ron Yokota of Tanimura & Antle, Inc., a large and progressive vegetable operation. The project will continue through April, 2000.

The study consists of a 10-hectare field that has been conventionally farmed for vegetables for several decades. The grower has produced three iceberg lettuce crops per year at the site during the course of the study. The soil type is a silt loam. After taking a comprehensive set of measurements for background soil properties after a winter cole crop in April, 1998, the site was divided into four blocks. Each block has an independent system for surface drip irrigation. Each block was divided into four sections: minimum tillage + OM; minimum tillage - OM; conventional tillage + OM; conventional tillage - OM (see definitions below). Thus, each treatment is replicated four times in the field. Each treatment is ~0.63 hectares.

The minimum tillage treatment consists of using the ‘Sundance’ system, a liliston, and rollers. No subsoiling is done, so that the soil is not disturbed below approximately 20 cm depth. The same 1-m wide beds are being utilized in the minimum tillage treatment for the entire study. By contrast, conventional tillage practices consist of disking, cultivating with a liliston, subsoiling, and chiseling. The soil is disturbed to a depth of approximately 50 cm depth. Beds are re-made between every crop. This is the typical tillage method for vegetable production in this area.

In treatments receiving added OM, compost is added two or three times per year, and a cover crop is grown during the fall or winter. The compost is based on yard waste material, and has a low C/N ratio (approximately 13), high NO3- (e.g., 200 µg/g), and low ammonium (NH4+) (e.g., 5 µg/g). In 1998, four tons/acre of compost was added in April, and in late July. A Merced rye cover crop was grown during August and September right after compost addition, and was harvested before anthesis after producing 465 g dry weight/m². In 1999, compost was added in June and September, and a cover crop was grown from September through November, producing 265 g dry weight/m². In the treatments receiving no added OM, only the vegetable crop residue was incorporated into the soil. This is the typical amount of OM that is used for most vegetable production here, except for occasional manure.

The general premise of field operations is that the entire field is managed similarly, except for the four different tillage and OM treatments. ‘Best Management Practices’ were used for application of inorganic fertilizer and irrigation. This required strong cooperation with the grower, who has managed all treatments similarly as far as planting and harvest dates, and fertilizer, pesticide and water inputs. He also has monitored all inputs, including tractor time and labor. Since every management event and input is recorded with the identity of tractors and implements, economic analysis of the costs to produce vegetables with each of the treatments, and to assess labor and fuel costs are now underway.

Initial samples in April, 1998, were taken for soil organic C and N (combustion), pH, CEC, soil particle size analysis, microbial biomass (fumigation-extraction; Vance et al., 1987; Wyland et al., 1994), inorganic N (KCl extracts), net mineralizable N (anaerobic incubation; Waring and Bremner, 1964). Bulk density was measured in the surface and at 20 cm depth with brass rings. Two samples were then taken per plot (n=32) at the time of harvest of each crop or cover crop. To monitor indigenous weed populations, micro-plots of 8.1 by 15.3 m were established in all plots. Twenty soil core samples were taken in each micro plot as described in Forcella et al. (1992) in April, 1998, and May, 1999. Soil sampling is planned for April 2000. Each core was partitioned into 0 to 15 and 15 to 30 cm sections. To determine the number of viable seeds in the soil cores, the samples are being processed using the direct emergence methods described by Barberi et al. (1998). Weed density counts are being taken at least once per cropping cycle. It was not possible to take weed density counts in the no-amendment plots in September, 1998 and November, 1999 because these plots were not irrigated and there was no weed emergence.
The key lettuce diseases were monitored and assessed throughout the study. For all treatments, plant stands (number of germinated lettuce seedlings per linear foot of row) were evaluated shortly after plant emergence as a gauge of possible damping-off diseases. Near harvest, the most important foliar disease, downy mildew (caused by the fungus Bremia lactucae), was evaluated. Three important soilborne diseases, corky root (caused by Rhizomonas suberifaciens), lettuce drop (caused by Sclerotinia minor) and big vein disease (caused by the big vein virus-like-agent) were also assessed near crop maturity.

Leafminer populations were evaluated by collecting randomly chosen heads from each of the treatment areas. Whole plant samples, five per treatment replicate in May, 1999, and six per treatment replicate in August, 1999, were placed into a bucket-cage fitted with a yellow sticky trap to allow the leafminers and parasites to complete development. The number of adult flies and parasites that flew into and stuck to the sticky card were counted six weeks after sampling to ensure development of all leafminers and their parasites.

Results. Soil microbial biomass C (0-15 cm depth) was ultimately strongly affected by OM addition. In the first growing season of the experiment, however, when only compost had been added, no differences were observed between treatments. Soil microbial biomass C increased markedly after the fall cover crop + compost additions in late summer of 1998. It remained higher in these treatments throughout the next year. Values were typically 75 to 100 µg C g-¹ dry soil in the -OM treatments vs. 120 to 220 µg C g-¹ dry soil in the +OM treatments. Adding a very labile source of C in the cover crop residue simultaneously with a more recalcitrant C source as compost appears to have supplied microbes with available C for a long period of time.

Soil nitrate in the top 90 cm of the soil profile was typically >200 kg N ha-¹ throughout the study. There was a significant decrease in nitrate after both of the fall cover crops, reducing the potential for leaching loss. Nitrate in the surface soil was made available earlier in the spring with conventional as compared to minimum tillage.

For the lettuce crop that was planted in the first season (April, 1998), yields were high, and we found little difference in yield among treatments. Mean fresh weight ranged from 8351 to 8860 g fresh weight m-², and dry weight ranged from 49 to 53 g/head. In the lettuce crops that followed the fall cover crop + compost addition, yield tended to be significantly higher than in the -OM treatments. In May, 1999, the lettuce yields were 6710, 6418, 6358, and 6045 g fresh weight m-² in the conventional tillage +OM, conventional tillage -OM, minimum tillage -OM and minimum tillage +OM treatments, respectively. For the same treatments in August, 1999, values were 9249, 8889, 8112, and 8313, respectively. The increased lettuce yield appears to be associated with higher soil microbial biomass C. Yield was highest in the conventional tillage +OM treatment in May and in August, 1999. The minimum tillage treatments produced significantly less yield than the conventionally tilled treatments on these two dates. Differences in plant N content were not as pronounced as for yield.

No significant differences for the various lettuce diseases were observed for either spring and summer lettuce crops. Tillage, compost and cover crops did not result in increases or decreases of the plant diseases we evaluated. But disease severity at the study site was very low. In the May, 1999, lettuce crop, mean corky root severity was 2.2 to 2.7 on a scale of 1(low severity) to 12 (high severity). In August, 1999, means ranged from 3.1 to 3.5. Plants with lettuce drop symptoms (Sclerotinia minor) composed 1.3 to 1.6% of the stand in May, 1999, and 0.3 to 0.4% in August, 1999. Big vein disease was present in 3.0 to 3.3% of the plants in May, 1999, and was not evident in the second lettuce crop. Downy mildew disease and damping-off disease were absent from both lettuce crops.

The major weeds found thus far in the study have been burning nettle (Urtica urens) and shepherdspurse (Capsella bursa-pastoris), after a total of nine weed density counts to date. No effects of tillage on shepherdspurse or burning nettle densities have been observed. This is not surprising since in a similar study Schreiber et al. (1992) found that tillage effects on weed population dynamics were only evident two to three years after the adoption of no-till farming practices. The addition of organic amendments did result in reduced shepherdspurse and burning nettle densities. Mean shepherdspurse densities ranged from 3 to 61 plants per 0.56 m-² in the -OM treatments, as compared to 1.0 to 21 plants per 0.56 m-² in the +OM treatments. Burning nettle densities were 0.6 to 4.0 and 0.3 to 3.2 plants per 0.56 m-², respectively. Organic amendments may favor increased biological activity that in turn results in weed seed degradation. We are currently analyzing the soil seedbank samples to determine if there are differences between the treatments in terms of the numbers of weed seeds in the seedbank.

Leafminers, a pest found in other reduced-till situations, were not significantly different between treatments, although means were lower in the conventionally tilled treatments. Insect pressure on the field was limited to the Pea Leafminer. All of the leafminers found were Liromyza huidobrensis and only a few parasites were found, mostly Diglyphus intermedius. There were no significant treatment effects. An initial block effect may have been due to the impact of flies migrating in from a neighboring field.
In summary, this multidisciplinary study showed that additions of compost and cover crops significantly increased soil microbial biomass for almost a year. Lettuce yield also generally increased in the following spring and summer after these OM additions. Yield, however, was lower with shallow minimum tillage (‘Sundance’) than conventional tillage (disking, subsoiling, and bed shaping). A single application of compost alone without a prior cover crop (April, 1998), however, had no effect on yield. Nitrate decreased after cover cropping, reducing the potential for leaching loss. Weed density was reduced in the +OM treatments, but was not affected by tillage treatment. Insect pests and diseases were not affected by either OM or tillage treatment. Economic analysis is now underway.

References Cited

Barberi, P., M. Macchia, E. Bonari. 1998. Comparison between the seed extraction and seedling emergence methods for weed seedbank evaluation. In Weed Seedbanks: Determination Dynamics and Manipulation. C.T. Champion, A.C. Grundy, N.E. Jones, E.J.P. Marshall, R.J. Froud-Williams eds. Aspects of Applied Biology Vol. 51, pp. 9-14.

Forcella, F., R.G. Wilson, K.A. Renner, J. Dekker, R.G. Harvey, D.A. Alm, D.D. Buhler, and J.A. Cardina. 1992. Weed seedbanks of the U.S. Cornbelt: magnitude, variation, emergence and application. Weed Sci. 40:636:644.

Schreiber, M.M. 1992. Influence of tillage, crop rotation, and weed management on giant foxtail (Setaria faberi) population dynamics and corn yield. Weed Sci. 40:645-653.

Vance, E.D., P.D. Brookes and D.S. Jenkinson. 1987. An extraction method to estimate soil microbial biomass C. Soil Biology and Biochemistry 19: 703-707.

Waring, S.A. and Bremner, J.A. 1964. Ammonium production in soil under waterlogged conditions as an index of nitrogen availability. Nature 201:951-952.

Wyland, L.J., L.E. Jackson and P.D. Brooks. 1994. Eliminating nitrate interference during Kjeldahl digestion of soil extracts for microbial biomass determination. Soil Science Society of America Journal 58: 357-360.