Management Strategies for Improved Soil Quality with Emphasis on Soil Compaction

1. Vegetable crop response to soil compaction, response to other soil quality parameters, and response to rotation sequence

(a) Beets
Significant yield reductions associated with soil compaction have been observed by our grower cooperator, J. Vincent, L-Brooke farm, and this is one reason this crop species was included in our evaluations. In the one year we attempted to quantify beet yield response to compaction at the L-Brooke farm we found that deep tillage increased yields by 7% compared to plots which received normal shallow tillage (Appdx, Table 1) . In the same year on the same field deep tillage increased snap bean yields by about 15%. Based on qualitative information for other years we suspect yield sensitivity of beets to compaction may frequently be greater than measured in 1995.

Beets responded very positively to addition of a composted chicken manure applied for two years to the same field at a rate of 2 tons/acre at the L-Brooke farm. Marketable table beet yields in composted plots were more than double the yields in non-composted control plots (Appdx, Table 2). This yield benefit was associated in part with reduced incidence of damping off disease of seedlings early in the growing season, and a reduced incidence of beets culled for root rot at final harvest.

A positive beet yield response to composted chicken was also measured at our replicated Geneva field site. Marketable yield of beets in 1995 (the only year beets were grown at this site) was increased by 26% with the addition of the compost at a rate of 5 tons/acre. At this site in this year root disease incidence was low in all treatments, so the compost benefit could not be clearly linked to a reduction in yield losses due to root disease.

(b) Cabbage
Direct-seeded cabbage was the most sensitive to soil compation of all vegetable crops evaluated in our trials. Average yield reduction due to compaction over three years at our replicated Freeville site was 73% (Appdx, Table 3). A key factor in these yield reductions was more severe flea beetle pressure and damage during early growth stages in compacted plots. Cabbage seedlings grew much more slowly on compacted soil, prolonging the period when they are most subject to significant flea beetle damage. Weed competition was also more severe early in the season on compacted compared to non-compacted plots. Another factor causing yield reduction in compacted plots was stunted growth following periods of heavy rain because of poor drainage and a prolonged period of saturated soil conditions. Cabbage and snap beans were more sensitive to wet soil than either sweet corn or cucumber in our Freeville study (beets were not included in the Freeville experiment).

Transplanted cabbage was less sensitive to compaction than direct-seeded cabbage. Average yield reductions of transplanted cabbage (quantified in one year only) were still substantial however-- about 30% lower yields on compacted compared to non-compacted plots (Appdx, Table 3). Transplanted cabbage plants, because they essentially bypassed the early seedling stage in the field, were less negatively affected by flea beetle damage and weed pressure early in the season compared to direct-seeded cabbage. Another factor may be that, in general, transplanted crops tend to have more prolific rooting of fibrous roots in the upper soil profile than direct-seeded crops. Direct-seeded crops tend to be more deep-rooted, but have fewer fibrous roots in the upper soil profile. In our trials, the compacted zone began at about the 5 - 10 inch depth. Transplanted cabbage may have been better able to explore this upper soil zone for water, nutrients etc.. Location of the compacted zone, rainfall and irrigation patterns during the season, and ability of tap roots to penetrate a compacted layer may determine whether the shallow rooting of transplants or deeper rooting of a direct-seeded crop is advantageous on compacted soil. In our trials, transplanting was clearly advantageous compared to direct-seeding, but more research would be required to quantify the role of differences in rooting pattern.

(c) Cucumber
Response of cucumber to soil compaction was highly variable. Within the same compacted plot we sometimes observed one or two plants severely stunted and bearing few fruit alongside plants that were similar in size and fruit load to the controls (plants in non-compacted plots). In general, cucumber appeared less sensitive to soil compaction than direct-seeded cabbage or snap bean, but more sensitive than sweet corn. First harvest was usually delayed in compacted plots, and over three years of trials at our replicated Freeville site, the average total yield reduction in cucumber associated with compaction was 41% (Appdx, Table 3).

(d) Snap beans
Snap beans ranked second to direct-seeded cabbage in yield sensitivity to soil compaction in our three year replicated study at the Freeville field site. The average yield reduction of snap beans associated with compaction at this site was 50% (Appdx, Table 3). One important factor involved in yield reductions on compacted plots was prolonged stunted growth after heavy rains due to poor drainage and extended periods of wet soil conditions, as was also observed for cabbage. Results of leaf tissue testing during the growing season revealed some nitrogen deficiency in compacted plots, which may also have been a yield-determining factor.

At the Geneva field site, yield reductions on conventionally tilled compared to deep sub-soiled plots in 1995 ranged from 16 - 31% (Appdx, Table 4). At the Geneva site in this same year we found significantly larger roots in sub-soiled plots, and fewer root lesion nematodes (Pratylenchus spp.) on roots of bean plants grown on sub-soiled compared to conventionally tilled plots, although this latter effect was not statistically significant at P < .05 (Table 4). In the second year of the study, there was no statistically significant residual effect on yield associated with the deep sub-soiling treatment of the first year. At the western NY commercial farm site (L-Brooke Farm), deep tillage resulted in a 15% higher snap bean yield compared to normal shallow tillage (Appdx, Table 1). A 20% pod yield increase was reported for plots that received 5 tons/acre chicken manure compost + synthetic fertilizer compared to plots receiving synthetic fertilizer only in 1996 at the Freeville site. At the Geneva site, where a comparison was made between chicken manure compost (5 tons/acre) only vs. synthetic fertilizer (350 lbs/acre 20-20-20) only, pod yields were slightly lower in the compost treatment, suggesting the need for phosphorus supplement to the compost (Appdx, Table 4). It was noted that the total biomass was similar between treatments, however, and root biomass was higher in the compost plots compared to the plots receiving sysnthetic fertilizer. In our replicated rotation experiments at Freeville, we documented significantly lower yields in plots where snap beans were grown within the prior two years in comparison with plots planted to sweet corn or other rotation crops within the prior two years (Appdx, Tables 5a,b). Yield reductions in snap bean-snap bean rotations could in part be attributed to a higher incidence and severity of root rot (Appdx, Tables 6a,b). We have also found that root rot severity has been increasing year to year in the continuous bean rotation at the Geneva field site. Root rot severity ratings at this site averaged 3.8, 6.2, and 5.8 in 1995, 1996, and 1997, respectively (1=no disease symptoms; 9=severe root rot and decay). Results obtained at the Geneva site of the SARE project and those of other projects have generally shown that bean yields are increased and root rot is reduced after the incorporation of a green manure of grain crops such as oat, ryegrass, grain rye, barley, wheat, sudangrass, and others. Recent greenhouse experiments indicate that root rot severity and growth of beans varies considerably depending on the particular cover crop that precedes beans (Appdx, Fig. 1). Of those cover crops that we also had in field trials, grain rye and sudangrass ranked best in the greenhouse experiments with regard to suppression of root rot of a subsequent bean crop. Hairy vetch, in contrast, resulted in relatively severe root rot severity. Obtaining the benefit from cover crops in a bean rotation requires proper management of the cover crop. In particular, it is important to leave enough time (e.g., 3 - 4 weeks) between incorporation of the cover crop and seeding of the beans to allow for decomposition to occur. We found a negative effect on snap bean yield from fall/winter cover crops of rye+vetch and perennial ryegrass at the Freeville site, where the cover crops were incorporated just a week before bean planting. Observations suggested this was due in part to a higher incidence of root rot in these plots, which was in turn associated with greater seed corn maggot infestation early in the growing season in these plots. Also, reduced nitrogen availability due to nitrogen tie-up by the decomposing cover crops could have been a factor. In the rye+vetch plots very little of the hairy vetch legume actually survived the winter and so there was little nitrogen contribution by vetch (see 1997 Annual Report for more discussion). Results of greenhouse tests indicated that when the soil is wet it is particularly important to allow 3 - 4 weeks between incorporation of the cover crop and seeding of the beans to avoid root disease, poor stands, and poor growth of the bean crop (Appdx, Table 7). (e) Sweet corn
On a relative basis (compared to beets, cabbage, snap beans and cucumber), sweet corn was less negatively affected by soil compaction in some sites in some years (Appdx, Tables 1, 3), although yields can still be reduced by 50% or more. Leaf tissue analyses suggest that these yield reductions can be associated with compaction-induced nutrient deficiency, as we also observed for snap beans. At the Freeville site, the year with little yield reduction on compacted corn plots (1993, Table 3 in Appdx) was a year in which relatively high fertilizer rates were used.

Sweet corn may be a particularly effective rotation cash crop with beans for soils with shallow compacted soil layers. Not only did sweet corn sometimes grow better than other cash crops on compacted soils, we also observed decreases in soil penetrometer readings at the 6-inch (30 cm) depth in compacted plots after one season of sweet corn at Freeville (Appdx, Table 8). At our Freeville site, we found highest final year (1997) snap bean yields on plots which included sweet corn in the rotation (Appdx, Table 5a). In separate greenhouse experiments, we measured greater root growth into a compacted soil layer at the 6-inch depth for sweet corn, compared to cabbage, snap beans, and cucumber (Appdx, Table 9).

We found about a 30% increase in sweet corn yield in plots with 5 tons/acre composted chicken manure in one year of observation at the Freeville site, but no beneficial effect, or even a slight negative effect from addition of this compost at the Geneva site over a three-year period. It should be noted that the Freeville results were based on very small observational plots, whereas the Geneva data involved three years of observation on larger plots.

2. Summer cover crop evaluation for: growth on compacted/poor quality soils; impact on soil compaction and quality; impact on pest/disease/weed complex; integration into rotation scheme; impact on subsequent cash crop yield and quality

(a) Bahiagrass (Paspalum notatum) This cover crop is well known throughout the southern U.S. as one that produces a large deep root system. We experimented with it here to determine if it could be grown in our environment, and if it could be effective at remediating compaction within a single growing season. Unfortunately, we were not successful at finding a planting date or cultural practices that would lead to a good stand of this crop under New York State conditions. In greenhouse pot experiments under controlled environments we were able to grow it and confirmed that it has a relatively large and deep root system. However, in the short-term greenhouse experiments it did not out-perform sudangrass or perennial ryegrass in terms of root growth into a compacted soil layer at the 6 - 12 inch depth.

(b) Berseem sweet clover (Trifolium alexandrium)
Cultural practices: Recommended seeding rate is 15 - 20 lbs/acre broadcast and rolled, or drilled 1/4 to 1/2 inches deep. The higher seeding rate will reduce weed competition. Planting between May 1 - 30 may be appropriate for many regions of New York, but this should be adjusted to local microclimate. Incorporate in late fall (will not overwinter in most Northeast regions).

In milder climates this is a biennial legume that is reported to contribute about 85 lbs nitrogen/acre when grown as a winter annual (planted in late summer and incorporated the next spring/summer). However, it does not overwinter in NY (can survive only light frosts), and so it has been evaluated as a summer annual. It is not a crop that establishes quickly, and weeds can sometimes get ahead of it. We had trouble obtaining good stands, and in general it did not look promising in our 1995 trials in eastern NY and at Freeville (central NY). A trial conducted in 1996 in Genesee county (western) NY, however, measured a higher above-ground dry weight biomass (2.1 tons/acre) in a fall harvest of berseem clover compared to hubam sweet clover, yellow blossom sweet clover, buckwheat, and japanese millet (Appdx, Table 10).

We did not evaluate this crop specifically for growth on compacted, poor quality soils, or ability of roots to penetrate compacted soil layers.

(c) Buckwheat (Fagopyrum sagittatum)
Cultural practices: Typical seeding rate is 60 - 70 lbs/acre, but rates of 100 - 130 lbs/acre are used by some growers. It should be drilled 1/2 - 1 inch deep. Planting date can be between mid May through July. Can reseed and become a weed problem if allowed to flower and set seed before mowing and incorporating in the fall. The crop should be allowed to grow for six weeks from planting to plow down to get the desired beneficial effects on soil quality without producing mature seed. It decomposes relatively quickly after incorporation. Some growers plant two crops of buckwheat in a single year to smother weeds and improve weed control in subsequent seasons. In one trial in eastern NY we tried a buckwheat (30 lbs/acre) and white clover (7 lbs/acre) mix that appeared promising, particularly in the later (June 2) planting date.

Although this crop usually established reasonably well and was a fast grower, it did not result in a particularly high organic matter residue, and it was not particularly effective at root growth into compacted soil layers and remediation of compacted soils. However, it provided some weed suppression, and others have reported that it improves phosphorous availability, and improves soil tilth by improving soil aggregation. It is of value for these purposes. Also, the short time requirement in the field makes it an ideal niche cover crop for some situations.

(d) "Hubam" annual sweetclover (Melilotus alba)
Cultural practices: Seeding rate of 20 - 25 lbs/acre is recommended. Seed can be broadcast and rolled or drilled at 1/4 to 1/2 inches deep. Can be planted late April to June, with optimum planting date probably early to mid May. Produces abundant growth, sometimes up to 6 ft. high, but is relatively easy to incorporate in the fall.

In most of our trials, this performed best of all legumes examined in terms of ease and reliability of establishment, organic matter production, growth of roots into compacted soil layers, and overall beneficial effect on subsequent cash crop yields. Weeds can be a bit of a problem early in the season, but the Hubam crop usually outcompetes and completely smothers the weeds by the end of the growing season. In addition to being a nitrogen-providing legume, this crop is reported to improve phosphorous availability. Hubam ranked third (behind sudangrass and overwintering perennial ryegrass) in terms of root biomass and root growth into compacted soil layers. Its root system is not nearly as deep as sudangrass, but it can improve soil quality of upper soil layers.

(e) "Nitro" non-dormant alfalfa (Medicago sativa)
Cultural practices: Typical seeding rate is 20 - 25 lbs/acre, broadcast and rolled, or drilled 1/4 to 1/2 inch deep. In eastern NY trials, planting dates of May 15 and June 2 (1995) both resulted in a crop that established quickly and smothered weeds. Nitro is a non winter-hardy alfalfa that can be used as a one-year hay source and as a fall plow-down green manure crop. The nondormant characteristic allows for an additional 4 - 6 weeks of fall growth, and therefore more nitrogen fixation within one growing season compared to the perennial dormant cultivars. The seed is relatively expensive.

In the eastern NY trial mentioned above, although the Nitro alfalfa established quickly and out-competed weeds, the plantings were devestated by leafhopper damage in late June, and did not grow to maturity. It was clear that leafhoppers had a strong preference for the Nitro alfalfa compared to other cover crops in the trial, and the crop was very sensitve to the damage. In a trial at Freeville (central) NY, conducted on a low quality soil with some compaction problems and poor water holding capacity, the Nitro alfalfa did not establish as well as most other cover crops evaluated. Because of these problems, we cannot recommend Nitro alfalfa at this time, but it would be worthy of further evaluation.

(f) Sudangrass (Sorghum sudanense)
Cultural practices: Sudangrass should be planted into warm soils (e.g. > 65 F), which for many regions of the Northeast will mean seeding after June 1. Seeding later than July 15 will probably not allow adequate time for growth before incorporation in the fall. A seeding rate of about 40 - 50 lbs/acre is typical. On low fertility soils sudangrass will benefit by addition of nitrogen feritilizer (broadcasting at a rate of 75 - 100 lbs. nitrogen/acre). Mowing to a height of about 6 - 12 inches once during the season when the crop is about 3 - 4 ft high is highly advisable as it stimulates prolific tillering, prevents the development of large woody-stemmed plants that can be difficult to cut down in the fall, and may also stimulate a deeper root system. Ideal time to incorporate is before September 15 to avoid problems with excessive surface debris and nitrogen tie-up by residues the following spring. The sudangrass is first cut or chopped and then incorporated to a depth of 4 - 8 inches by several passes with a large disk.

Based on grower observations and our direct quantitative measurements in both greenhouse and field experiments, this cover crop ranked best at ability of the plant to grow on compacted soils (Appdx Tables 11, 12), root penetration into compacted soils (Appdx, Table 12), partial remediation of compaction after only one growing season (Appdx, Table 8), and beneficial impact on subsequent cash crop yields (e.g., Appdx, Table 5b). This crop also ranked highest in most trials at biomass production for organic matter incorporation (Appdx, Tables 10,11,12), and highest for weed suppression (Appdx, Table 11). We know from prior studies and recent greenhouse experiments that this cover crop can suppress root disease (Appdx, Fig. 1) and pathogenic nematodes (Appdx, Fig. 2) in snap beans and other crops (Viaene and Abawi 1998. Plant Disease 80: in press) . In a separate project, sorghum-sudangrass has been found to increase onion yields by about 15-25% on muck soils (personal communication, J. Mishanek, IPM specialist, Orange county, NY).

Note that in most of our trials we used "Trudan 8", a sudangrass (S. sudanense) hybrid. "Sudex", a hybrid cross between sorghum (S. bicolor) and sudangrass (S. sudanense) is a very similar cover crop that has been evaluated in other grower and extension staff trials, and shows very similar traits to the pure sudangrass hybrid that we used. These are likely interchangeable in most circumstances, but we have yet to evaluate the two types in a comprehensive side-by-side comparison to verify this.

(g) Yellow blossom sweet clover (Melilotus officinalis)
This is a biennial legume that is capable of over-wintering in most of New England. It can be planted in late April or May, or August, allowed to over-winter, and then plowed down the following summer before full bloom, usually before early July. Seeding rate is 15 lbs/acre, broadcast and rolled, or drilled 1/4 to 1/2 inch deep. The nitrogen content in foliage is reported to level off when crop is 12 - 16 inches tall, and it is recommended to plow it down early in the second summer to conserve soil moisture.

This crop is reported to have a deep tap root that could potentially break up compacted soil layers. Unfortunately, in the compaction trial we established in 1994 at our Calwell field site (Ithaca, NY), this cover crop, like most of the others included in that study, did not establish well because it was a very dry spring. We did have success establishing a two-year planting in eastern NY, but this was not a compacted field and we did not evaluate root growth in any quantitative fashion. In the eastern NY plantings (May 15 and June 2, 1995) yellow blossom sweetclover did not produce biomass or smother weeds as well as Hubam sweet clover in the first summer comparison, but it produced outstanding growth and was very effective at smothering weeds the following spring and during the second summer.

This is a promising cover crop, but we still need to verify the ability of it to produce deep roots capable of remediating compacted soils in the second year of growth. Also, we need additional studies to develop a planting and harvest schedule recommendation that allows for maximum use of land for cash crops (e.g., August planting rather than May/June planting the first year).

3. Fall/winter cover crop evaluation for: growth on compacted/poor quality soils; impact on soil compaction and quality; impact on pest/disease/weed complex; integration into rotation scheme; impact on subsequent cash crop yield and quality

(a) Austrian winter pea (Pisum sativum, var. arvense)
Cultural practices: Seeding rate of 80 - 120 lbs/acre, drilled 1 - 2 inches deep in August. Seed is relatively expensive. It may not survive in severe winters. (It can also be seeded in April and plowed down by late May or early June to provide N for a subsequent cash crop in the same summer).

We did not find any evidence that this cover crop is particularly effective at remediation of compaction, but our evaluations were limited. It is adapted to Northeast weather conditions, and can make a significant nitrogen contribution (up to 100 lbs nitrogen/acre) for subsequent cash crops. It can also help to prevent erosion.

(c) Grain rye (Secale cereale)
Cultural practices: Seeding rates vary widely, typical rates are between 100 - 140 lbs/acre (about 2 bushels/acre), usually broadcast applied and, ideally, rolled. Optimum planting date ranges from late August to mid or late September. Planting too early (e.g., before mid-August) can result in seedheads forming before winter in mild years. It is tolerant of a wide range of soil and other environmental conditions, and can be seeded later in the fall (through October, or even early November in some regions in some years) than any other cover crop grown in the Northeast. It is somewhat tolerant of atrazine so can follow a corn crop. For these reasons it is widely used and has an important niche in the rotations of many growers. To avoid nitrogen tie-up the following year that would negatively affect the summer cash crop, it is best to mechanically kill the crop in the spring (e.g., with a sickle bar mower) before it heads or exceeds 18 inches in height, and allow a few weeks between incorporation and seeding of the cash crop.

The primary benefits of grain rye are winter erosion control, serves as a nitrogen catch crop, and the crop residue in the spring and summer provides significant weed suppression and organic matter. Recent greenhouse results also suggest that grain rye may help to reduce root disease in a subsequent bean crop (Appdx, Fig. 1), as long as there is a sufficient time interval (e.g., 3 -4 weeks) between incorporation of the rye and seeding of the beans. Compared to sudangrass and perennial ryegrass, it does not have a particularly deep or abundant root system for remediating compacted soils, but it is a very hardy crop that may be able to grow on compacted, poor quality soils better than many other cover crops.

Grain rye is well-know to have allelopathic effects that are a factor in its weed suppression abilities. Pigweed, ragweed, crabgrass, and lambsquarters are weeds that may be particularly sensitive to grain rye residues. Direct-seeded small-seeded cash crops, such as cabbage, lettuce, tomato, and carrots can also be negatively affected by the allelopathic properties of grain rye. Transplanting can minimize this problem. In some cases even large-seeded crops such as corn can be affected. Beans are not generally considered to be particularly sensitive to allelopathic effects of grain rye, but in our Freeville trial, where we had allowed only about a week between incorporation of the rye and seeding of the beans we may have observed some negative allelopathic effect. In general, grain rye allelopathic effects are short-lived, and the risks of negative impacts on cash crops can be minimized by allowing sufficient time (e.g., 3 - 4 weeks) between incorporation and cash crop seeding.

(d) Grain rye (Secale cereale) +hairy vetch (Vicia villosa)
Cultural practices: Ideal seeding rates will vary with soil type and purposes of using this cover crop mixture. Typical rates are 50 - 60 lbs/acre of rye and about 15 - 20 lbs/acre of vetch. Ideal planting date probably between Aug 15 and Sept 1 for most Northeast regions. The crop would typically be mowed (flail or sickle-bar mower) in late May and incorporated. Growers will probably need to experiment with rates and dates to best suit their conditions and needs.

This mixture has proven successful in many trials and for some growers in the Northeast. The advantages of the mixture include: quicker establishment and better winter survival of the vetch than when grown as a monocrop; more overall biomass (in some trials) than either crop grown alone; easier mowing in spring than vetch alone because less "matting" of the crop; reduced risk of nitrogen tie-up by the rye in the spring because of the nitrogen contribution by vetch, a legume. Of course, total nitrogen contribution for the subsequent cash crop will be less with this mixture than when vetch alone is grown. In our trials at a relatively cool microclimate site in central NY (Freeville) we had rather poor establishment and very little winter survival of the vetch in late plantings (mid September). This cover crop mixture is not particularly effective at producing deep roots for remediation of compaction.
See also comments for each individual cover crop.

(e) Hairy vetch (Vicia villosa)
Cultural practices: Some sources recommend a seeding rate of 20 - 40 or even 50 lbs/acre, but seed is relatively expensive, and many growers use rates in the 15 - 20 lbs/acre range and still have a good healthy stand. Drilling the seed 1/2 - 1 inch deep works best, particularly if using low rates. The seed can be broadcast and rolled (good soil contact is essential), and in that case higher seeding rates might be more appropriate. Typical planting dates are between Aug 15 and Sept 1. The crop can be sickle- or flail-mowed and incorporated in mid to late May. The crop is quite succulent, and even just disking, without first mowing, is usually sufficient to kill the crop before plowing. There can be problems with winter survival (possibly associated with frost heaving) in some locations in the Northeast in some years.

This winter annual legume can supply substantial quantities of nitrogen (up to 150+ lbs nitrogen/acre), about half of which is available the first year, the rest within the next two years. It is also reported to help make phosphorous and some micronutrients more available. It has a viny growth habit and has the potential to produce substantial amounts of biomass. Sometimes it becomes matted and is difficult to mow and incorporate in the spring. When well-established it is a good weed suppresor, and some suspect it may have some allelopathic properties. It has the potential to become a "weed" problem itself, however, in some rotations if it is allowed to go to seed. We have found higher populations of root lesion nematode in soils in which hairy vetch has been grown. Therefore, hairy vetch in rotation with snap beans or other crops susceptible to this nematode could be problematic, and should be carefully considered.

Vetch does not perform well on poorly drained soil, so is probably not a viable option for remediation of soil compaction in most situations. However, it may be a good rejuvenating cover crop to follow a deep tillage (chisel plowing) operation.

(f) Oats (Avena sativa)
Cultural practices: Seeding rates often between about 80 - 120 lbs/acre, but can be much higher. Lower rates are used if drilling the seed (1/2 - 1 inch deep) and higher rates for broadcasting. Sometimes it is interseeded with legumes. In the Northeast it must be seeded before August 15 to produce sufficient biomass before winter kill. It does not survive NY winters, but residue mulch left on surface can suppress weeds the following spring. It can also be grown as a summer cover or cash crop.

This cover crop can produce considerable biomass in the fall that can prevent winter erosion, suppress weeds in the spring, and provide organic matter for the soil. There is some evidence that it has allelopathic properties affecting some weed species. It is not, however, particularly effective at producing deep roots capable of breaking up compacted soil layers.

(g) Oilseed radish (Raphanus sativus)
Cultural practices: Seeding rate of 10-15 lbs/acre. Plant between August 15 and September 10 to avoid flowering and seed set before winter. Shallow drilling of seed is optimal, but broadcasting into a fine seedbed followed by cultipacker or harrow is also effective.

A good stand will provide rapid groundcover, abundant biomass, and take up significant amounts of residual nitrogen before winter killing, usually in December. If left to winter kill, the crop residue deteriorates considerably by the following spring. However, spring weed control is usually excellent following oilseed radish since very few weed seedlings become established in the cover crop during the fall.

(h) Perennial ryegrass (Lolium perenne)
Cultural practices: Seeding rate of 20 - 40 lbs/acre, broadcast and rolled, or drilled 1/4 - 1/2 inch deep. Plant between August 1 - September 1. It generally will overwinter, and produce abundant biomass the following spring, with substantial root growth as well as above-ground biomass. It is usually disked and incorporated in late May or early June. Some sweet corn growers are establishing perennial ryegrass early by broadcasting the seed into an existing corn crop during cultivation, and allowing the ryegrass to continue to grow as a cover crop after the sweet corn is harvested in the fall.

Of those cover crops we evaluated, this crop was one of the best for producing a strong, large root system and abundant biomass (Appdx, Table 12) under a wide range of conditions for remediation of compacted soils. The root system is not as deep as sudangrass, however.

(i) Yellow blossom sweetclover (Melilotus officinalis)
This crop can be planted in August, allowed to overwinter, and be plowed down in early summer (June) the following year. However, we evaluated it planted the first year in May/June, grown as a summer cover crop, allowed to overwinter, and then grown into early summer the second year. See discussion of it under summer cover crops (part 2, above).

This crop has the potential to produce deep roots that can help in the remediation of compacted soils.

(j) Yellow mustard (Brassica hirta)
This cover crop does not over-winter in most regions of NY, but it can be planted in August, and will grow and produce biomass well into the fall since, as a Brassica, it is well-adapted to cool temperatures. A typical seeding rate is 10 -12 lbs/acre, usually drilled 1/4 - 1/2 inch deep. Broadcasting into a tilled bed and cultipacked or harrowed (less than 1 inch deep) is also possible. Avoid allowing to go to seed, because this can become a weed problem, particularly in subsequent Brassica cash crop (e.g., cabbage) plantings. In our trials, with planting dates in mid to late August, it usually was winter-killed before going to seed.

Yellow mustard produces a deep tap root that can potentially break up shallow compacted soil layers. We had variable success at establishment of this crop. In some years and locations it established quickly, competed well against weeds during the fall, and produced significant biomass (e.g., over 2 tons/acre dry weight in first two months of growth). In other locations and years it did not compete effectively against weeds and biomass production was small. In most trials there was little crop residue remaining on the soil surface in the spring following a yellow mustard planting, so it was not effective as a weed suppressor in the spring.

4. Summary of result highlights

(a) Evaluation of vegetable crops for their yield sensitivity to soil compaction

A three-year field study conducted on an Eel silt loam soil at the Cornell Vegetable Research Farm in Freeville, NY compared cabbage, snap bean, cucumber, and sweet corn for their growth and yield response to an artificially compacted soil layer beginning at about the 6 inch depth. Slower growing cabbage seedlings in compacted plots were more subject to flea beetle damage than the uncompacted controls. This was more of a problem in direct-seeded cabbage than in transplanted cabbage. Prolonged flooding after heavy rainfall events in compacted areas had a more adverse effect on cabbage and snap bean than on cucumber or sweet corn. Separate field experiments at the Geneva site, and greenhouse studies, have indicated that snap bean yield reductions on compacted soils can be associated in part with increased root disease severity. Symptoms of nutrient deficiencies were observed in compacted plots in some years, particularly in snap beans and sweet corn. Maturity of cabbage, snap bean, and cucumber was delayed by compaction in all years. The average reduction in total marketable yield in (direct-seeded) compacted plots was 73, 50, 41, and 39% for cabbage, snap bean, cucumber and sweet corn, respectively. Yield reduction in transplanted cabbage (evaluated in 1993 only) was 29%. In separate field trials in western New York we observed significant yield reductions in table beets due to soil compaction, and this may have been in part due to increased incidence of root disease. Our results from all of these trials indicate that yield response to compaction in the field is often associated with crop sensitivity to secondary effects of compaction, such as prolonged flooding after rainfall events, reduced nutrient availability or uptake, and prolonged or more severe insect, disease, or weed pressure.

Our results suggest that of the five cash crops we evaluated, sweet corn may be a good option to include in rotation for fields where remediation of soil compaction is an issue. Although sweet corn yields were significantly reduced when grown on compacted soils, it was not as sensitive to compaction as snap beans and cabbage. Also, of the cash crops we evaluated, sweet corn was best at producing roots that could penetrate into shallow compacted soil layers.

(b) Evaluation of cover crops and rotation sequences for bioremediation of compacted soils, impact on other soil quality parameters, and impact on subsequent cash crop yield

Overall, considering the focus of this study-- remediation of soil compaction-- as well as considering ease of crop establishment and year-to-year and site-to-site stability of performance, the two cover crops which ranked highest in our trials were sudangrass (as a summer crop) and perennial ryegrass (as a fall/winter crop). Of these two, sudangrass has the deepest root system and generally ranked highest with regard to root growth into compacted soil layers and organic matter production. Sudangrass also was very effective at suppressing weeds and parasitic soil nematode populations. Hubam sweet clover was another cover crop that consistently performed well, including growth on compacted soils. However, Hubam did not produce as much below-ground biomass as sudangrass or perennial ryegrass. Yellow blossom sweet clover is reported to produce deep roots capable of breaking up compacted soil layers, but we did not have an opportunity to observe its performance on a compacted soil in our trials. We did find it well-adapted to our region in a trial where we grew it for two consecutive summers.

In years with low rainfall during the spring, it was particularly difficult to establish most of the summer cover crops, and the plantings were often completely overrun by weeds.

Grain rye, hairy vetch, and grain rye + vetch mixtures are fall cover crops that performed well in most of our trials, but we did not find any clear evidence that they were effective at compaction remediation. They can produce sufficient biomass in the fall to minimize winter soil erosion, and produce additional biomass in the spring. Grain rye can negatively affect subsequent cash crop yields due to nitrogen tie-up and/or allelopathic effects if you do not allow 3-4 weeks between incorporation and cash crop planting. The allelopathic effects of grain rye are more likely to be a problem in small-seeded crops (and corn is also reported to be susceptible) when direct-seeded too soon after incorporation of the rye. Interseeding rye with hairy vetch (a legume) can alleviate the nitrogen tie-up problem. This crop mixture was successful in many of our trials, although when planted late (e.g., mid September) very little of the vetch survived the winter. Also, with regard to use on compacted soils, vetch does not grow well on poorly drained sites, so may not be a good option for most compacted soils, unless they have first been deep tilled.

Yellow mustard, and other cover crops in the Brassica genus, are potential soil compaction remediators because they produce deep, penetrating tap roots. However, we have encountered some problems in establishing a good stand in some sites in some years. Although the Brassicas aid in weed control by suppressing weed establishment in the fall, most of the Brassicas winter kill and do not leave a significant amount of residue the following spring.

(c) The soil pest complex as affected by tillage, cover crops, and rotation sequences

Root diseases caused by fungal and nematodal pathogens frequently reduce yields of many economically important vegetable crops of New York and the Northeast region. It is know that damage by root diseases is most severe on poor quality soils, such as compacted soils with poor structure and inadequate drainage, soils low in organic matter, and soils with low nutrient availability. Results of this project demonstrated that part of the reason for increased snap bean yeids on deep-tilled (subsoiled) plots was associated with a significant reduction in root diseases, and in some cases there was some evidence of reduced populations of root lesion nematodes. However, no residual effect of the subsoiling was observed on the second bean crop planted a year later, suggesting the need for annual subsoiling in some soils, particularly if bean is to follow bean in rotation.

Increasing soil organic matter through the use of cover crops and green manures is well documented to improve soil physical and chemical properties, and also is known to increase the number and diversity of the total soil microbial community. The latter results in direct suppression of root pathogens and/or the production of bigger and more vigorous root systems that are tolerant to damage by root pathogens.

Our results have shown that the use of cover crops is generally beneficial in increasing yield and reducing root disease severity and crop damage. Not surprisingly, in both replicated field trials we found that bean monoculture without rotation to sweet corn or other cash or cover crops led to a decline in yield associated with an increase in root disease severity. Our results also documented that the various rotation cover crops are not equal in their impact in suppressing root diseasees or the population of individual pathogens. For example, rapeseed, crown vetch, wheat and sudangrass were the most effective at reducing root rot and increasing bean growth as compared to a fallow check, hairy vetch or white clover. Also, sudangrass, rapeseed, ryegrass, and white mustard suppressed the number of lesion nematodes in bean roots, whereas hairy vetch, and to a lesser extent, alfalfa, white clover, grain rye and alsike clover contributed to higher root populations of this nematode. These results clearly demonstrate the importance in selecting the appropriate cover crop if soil pest suppression or avoidance is an issue.

(d) Deep tillage effects on specific soil quality parameters in relation to crop yields and rotation sequence

In our study we quantified the beneficial effects of mechanical deep tillage on compacted soils with regard to soil quality parameters such as penetration resistance (Appdx, Table 8), and bulk density, time to ponding, water infiltration rate, and soil water holding capacity (Appdx, Table 13). Not shown in Table 13 are our data on soil pore size distribution, which were correlated with the other data, and indicated a higher percentage of large air pores (> 300 ìm diameter) in soils which had been deep tilled or subsoiled. The improvement in soil quality parameters due to deep tillage had significant beneficial effects on crop yields in the first year (Appdx, Tables 1, 3, 4, 5, and 11).

Deep tillage in the first year of the study did not have a statistically significant impact on penetrometer resistance measured three years later at the Geneva or Freeville sites. However, at the Freeville site, bulk density was slightly lower and water holding capacity slightly higher at the 12 - 15 inch depth in the third year on plots which had been deep tilled compared to those not deep tilled in the first year (Appdx, Table 14). Organic matter content measured the third year at Freeville was slightly higher in the deep-tilled compared to non deep-tilled plots (statistical contrast significant at P=.12, Appdx, Table 14). Also, there was evidence to suggest that deep tillage the first year reduced the incidence and severity of bean root rot in the third year of the study in those plots which had also been planted to beans in the first year (Appdx, Table 6b). These longer term, residual effects from deep tillage may be associated with the substantial beneficial effect that it had on crop growth and the pest/disease complex in the first year.

The specific rotation sequence used in combination with deep tillage may affect the long-term effects on soil quality parameters. Those plots at our Freeville site which had a continuous bean rotation had a faster ?time-to-ponding@ (i.e., more prone to flooding, runoff, and erosion) in the third year compared to other rotation sequences (Appdx, Table 14; this was significant at P < .05). In general, the combined use of deep tillage and cover crops appears to provide additive soil quality improvements. We are still analyzing the data of Table 14, but results from this site in combination with other data suggest that including sweet corn and sudangrass in the rotation sequence may enhance or prolong the beneficial effects of a deep tillage operation on soil quality parameters, and make the soil less susceptible to compaction. (e) Frost tillage as a soil management option Frost tillage is a primary tillage method which may be performed when a thin (1 4 inches) frozen layer exists at the soil surface. When frost enters initially unfrozen soil, the process of freezing induced water redistribution causes soil drying below the frost layer and may therefore allow for tillage. A multi year analysis of frost tillage demonstrated that this may be an attractive management alternative for vegetable growers in the Northeast, especially for early season crops on medium to fine textured soils. Frost tillage allows spring field work to be performed during the winter and facilitates soil drying in the spring, thereby potentially reducing soil compaction from early field work. Using model simulations based on climate data from the Northeast, frost tillage opportunities occur most often (4 5 days on the average per winter season) at the 40 to 43 degree latitude, with generally lower number of frost tillable days to the north and south of this belt.