[Note to online version: The report for this project includes tables and attachments that could not be included here. The regional SARE office will mail a hard copy of the entire report at your request. Just contact Northeast SARE at (802)-656-0471 or email@example.com.]
In New York, cabbage diseases have been associated with the presence in the field of susceptible weed hosts. In specific case studies, achieving good weed control resulted in significant disease control that was more efficacious than the use of fungicides. The two weeds that have been documented to promote disease development in cabbage crops are ragweed and velvetleaf. The two diseases that are most likely to occur in weedy cabbage fields are white mold caused by Sclerotinia sclerotiorum, and gray mold caused by Botrytis cinerea. Both diseases can result in significant yield losses, and there are no fungicides registered on cabbage that will control these diseases.
The objectives of this study were to identify crop rotation schemes that suppress weeds and disease development in subsequent cabbage crops, and to identify sustainable weed and disease management practices that result in integrated control of both pests.
Trials were established in three locations, two of which were commercial fields. Each trial was approximately two acres in size. At each location the experimental design was a randomized complete block (RCB) split plot. Crop rotation scheme was the main plot factor (the RCB part), with management practices in cabbage being the sub-plot factor. There were four blocks (replicates) at each location. The trial time period was for three years.
In general, the rotation effect on weed control was consistent throughout all three sites. Fewer weeds were present in the sweet corn plots than in the other rotation treatments. Weed control was poorest in the alfalfa plots, followed by the clover plots. Rye plantings seemed to control most weed species. Disease pressure was low due to hot dry weather. Insignificant levels of white and gray molds (the target diseases) were observed. A new disease that has been observed at low levels in cabbage fields in previous years was the dominant disease in all three locations. The pathogen has been tentatively identified as a Fusarium species. The incidence of cabbage heads with symptoms of infection by Fusarium species was significantly higher following the rye and sweet corn rotational schemes. Total yield was as good in non-chemically managed plots (handweeding, mulch) as in herbicide and fungicide managed plots. Attempting to control foliar diseases while neglecting weed control led to reduced yields.
Approximately 14,000 acres of cabbage are grown in New York annually, and New York is the largest producer of fresh market cabbage grown for winter storage and subsequent production into cole slaw in the United States. The specific diseases that limit cabbage production in New York are white mold caused by Sclerotinia sclerotiorum, gray mold caused by Botrytis cinerea, black rot caused by Xanthomonas campestris pv. campestris, and Alternaria leaf spot caused by Alternaria brassicicola. Sclerotinia sclerotiorum and Botrytis cinerea are cosmopolitan pathogens found on many agronomic crops, ornamentals, and weeds. In New York, cabbage diseases have been associated with the presence of susceptible weed hosts. In specific case studies, achieving good weed control resulted in significant disease control that was more efficacious than the use of fungicides. The two weeds that have been documented to promote disease development in cabbage crops are ragweed and velvetleaf. Herbicides do not always control these two weeds, and hand removal can be labor intensive for producers with large acreages. The ultimate goal of this project was to determine the best way to achieve simultaneous weed and disease management in a cabbage crop.
• Identify crop rotation schemes that suppress weeds and disease development in subsequent cabbage crops.
• Identify sustainable weed and disease management practices that result in integrated control of both pests.
The major activity was the establishment of three large research areas, two of which were in commercial fields. The locations were: Bellona (property of Mr. Paul Roe), Cortland (property of Mr. Don Reed), and Geneva (property of the NYS Agricultural Experiment Station). Each trial was approximately 2 acres in size (Attachment B – plot map). The treatments were arranged in a randomized complete block design with four replications. Each treatment plot was 30 by 70 feet in size and surrounded by a drive alley of 15 or 25 ft. To avoid compaction and herbicide use, the alleys at the Geneva and Bellona locations were planted to a mixture of annual rye grass (71.5%), perennial rye grass (19.3%), and wild white clover (5%) at the rate of 288 lb/A on June 18. After establishment, the grass alleys were mowed about every four to six weeks. At the Bellona location, the trial was placed in a field alongside a commercial planting of cabbage. At the Cortland location, drive alleys consisted of an established alfalfa planting, and were mowed as needed. The trial was located in a commercial field planted to alfalfa.
Soil samples were collected from all three locations to assess the populations of weeds and disease organisms. For the disease assessment, the soil was air dried, screened to remove large rocks, and mixed. Organic matter was floated from five 100-ml soil samples (500 ml total/treatment-rep) and placed on sterile sand in glass petri dishes. The sand was watered to above field capacity. The plates were incubated in a growth chamber maintained at a constant temperature of 20-21C, relative humidity of 30%, with cool white fluorescent lights on 12 hours per day (6 am to 6 pm). The plates were periodically scored for the presence of germinating sclerotia of Botrytis cinerea and Sclerotinia sclerotiorum. The soil collected for weed assessments was also air dried and screened. Weed seeds were identified under the microscope.
The following crop treatments were established: sweet corn, alfalfa, clover, and rye. The alfalfa and clover plots were mowed frequently to destroy noxious weeds. Large weeds were removed by hand. The sweet corn grew well and was mowed at the end of the season. A rye cover crop was planted in the corn plots and in the rye plots for the winter months.
Weed counts were taken in each replicate plot at all three locations. Soils samples (approximately one liter) were removed from each plot and weed seed and pathogen populations were assessed. Sweet corn was established in the appropriate areas, and mowed at the end of the season. A rye cover crop was planted in the corn plots and in the rye plots for the winter months.
The cabbage variety “Genesee” was planted at all three trial locations in 2001: Geneva Experiment Station, June 1; Roe Acres, June 14; Reed’s, June 6. Weather conditions were hot and dry throughout the growing season. Total monthly rainfall (in.) was 5.10, 1.98, 2.35, 4.11, and 1.83 for Jun, Jul, Aug, Sep, and Oct, respectively. The only site where irrigation was possible was at the Experiment Station, where a total of 3.2 inches of water was applied. The transplants were commercially produced, and delivered with other shipments to Roe Acres.
Experimental design. The experiment was conducted at three locations in New York (Geneva, Bellona and Cortland). At each location the experimental design was a randomized complete block (RCB) split plot. Crop rotation scheme was the main plot factor (the RCB part), with management practices in cabbage being the sub-plot factor. There were four blocks (replicates) at each location.
The weed seed bank was assessed before rotational crop schemes were established in the spring of 1999, and then again just prior to cabbage transplanting in the spring of 2001. Samples were collected at the main plot level. In each main plot, soil samples were extracted (4″ core ) from 5 locations. The samples were then mixed together in a bucket and bagged. Samples were kept in a cooler to prevent seed germination until processing. Before analysis, the samples were removed from the cooler, and the entire sample was sieved to remove stones and was thoroughly mixed. After mixing, 0.5 kg of soil was removed from the sample for weed seed bank counts and weed species determination.
The incidence of diseased cabbage heads was calculated as the proportion of harvested heads showing symptoms of infection by F. avenaceum. Incidence data were recorded at the sub-plot level. The total weight of cabbage heads harvested per sub-plot was also recorded.
Data analysis. Weed seed bank data had been collected at the main plot level. Two approaches were taken in analyzing weed seed counts. First, the effect of rotational scheme on the total number of seeds (regardless of species) was assessed. Data were analyzed as a randomized complete block, each location separately. Models were implemented using the Glimmix macro in SAS using the log link. This takes into account the Poisson nature of count data.
The second approach took into consideration the fact that seeds were identified to the species level. Because of the multivariate nature of this latter data set, the weed seed counts (by species) were analyzed as a multivariate randomized complete block. Data were first examined for multivariate normality and, if necessary, Box-Cox transformations were used on counts for individual weed species to achieve multivariate normality. Multivariate normality was assessed by tests of Mardia skewness and Mardia kurtosis using the multnorm macro in SAS. Multivariate modeling was done using Proc GLM in SAS. Each location was analyzed separately.
Disease incidence data were analyzed as a multilocation randomized complete block split plot using the Glimmix macro in SAS. This allowed a more natural modeling of the binary nature of incidence data. Total weight of cabbage heads harvested per sub-plot was analyzed using a multilocation randomized complete block split plot model using SAS Proc Mixed.
The dominant weeds at the Bellona location were chenopodium, ragweed, amaranth, purslane, field pennycress, and eastern black nightshade. The dominant weeds at the Geneva location were ragweed, chenopodium, nettle, chickweed, and annual grasses. Weeds found in the Cortland location included ragweed, cruciferous weeds, and chenopodium. The presence of ragweed at all three locations was significant, because ragweed is difficult to control and is a host for the cabbage pathogen Sclerotinia sclerotiorum (Plant Disease 70:26-28). The fungus infects and colonizes ragweed plants, which effectively increases the population of the fungus in the soil.
Weed control was generally much better in 2000 than in 1999. The clover and alfalfa stands were able to outcompete the weeds and beautiful stands were achieved. Weed control in the corn plots was excellent. Cortland location: Weed counts were taken on June 5. The dominant weeds were ragweed, white cockle, chickweed, and bluegrass. These weeds were present in all treatments except the sweet corn treatment. Weed counts were taken again in the corn plots on July 25, and only a few ragweed plants were present. No cabbage pathogens were detected in the soil samples. Geneva and Bellona locations: Weed counts were taken on July 21. The dominant weeds observed were amaranth, ragweed, nettle, and quackgrass. No cabbage pathogens were detected in the soil samples. Seed counts from soil samples show high populations of lambsquarter at the Geneva location, and chickweed at the Bellona location.
Symptoms of black rot were evident on the transplants at the time of planting at Roe Acres (the last field to be planted). However, the hot dry weather inhibited further disease development.
In general, the rotation effect on weed control was consistent throughout all three sites. Field observations in 1999 and 2000 indicated substantially fewer weeds were present in the sweet corn plots than in the other rotation treatments. This is due in part to timely herbicide applications and to planting a rye cover crop after harvest. Suppression of winter annuals and short-lived perennials, such as shepherds purse and campion, was greatest in the sweet corn and rye plots. Extremely high populations of these weeds were found at the Reed site in the clover and alfalfa rotations. These two rotation treatments did not receive herbicide applications at any of the three sites. Weed control was poorest in the alfalfa plots, followed by the clover plots. Rye plantings seemed to control most weed species.
Disease pressure at all three locations was very low due to hot dry weather. Insignificant levels of white and gray molds, caused by Sclerotinia sclerotiorum and Botrytis cinerea, were observed. A new disease that has been observed at low levels in cabbage fields in previous years was the dominant disease in all 3 locations. The pathogen has been tentatively identified as a Fusarium species. Infections result in small circular lesions that develop into a large rotted area on the cabbage head. In mass, the fungal mycelium appears pink to orange in color. Incidence of this disease was highest at the Reed location.
Weed seed bank total counts. There were no differences in the total count of weed seeds among plots prior to the establishment of the rotational crops in spring of 1999 (Table 1). In the spring of 2001, just before cabbage transplanting, there were still no differences among rotational schemes in terms of the total number of weed seeds at the Geneva location. However, there were significant differences in total weed seed counts due to the rotational schemes at the Bellona and Cortland locations (Table 1). Note, however, that total seed bank counts had increased in the plots from 1999 to 2001 (Table 2).
In 1999, 4 weed species accounted for 95% of the seed bank at the Bellona location. Six weed species accounted for 96% of the weed seed bank at Cortland, and at Geneva 4 weed species accounted for 93% of the weed seed bank (Table 3). Multivariate analyses indicated that there were no significant differences among rotational crop scheme in terms of the weed seed bank species distribution (Table 4).
Weed seed bank counts in the spring of 2001 (just before transplanting cabbage) are a measure of the potential weed species likely to be problematic in cabbage. In 2001, 6 weed species accounted for 80% of the weed seed bank at Bellona, 9 species accounted for 95% of the weed seed bank at Cortland, and 7 weed species accounted for 93% of the weed seed bank at Geneva (Table 3). Multivariate analyses showed no significant differences among the rotational crop schemes at Cortland, but there were significant differences in the weed seed bank distribution among the rotational crop schemes at Bellona and at Geneva (Table 4).
At Bellona, the weed seed bank distribution following the rye rotational scheme was the most different compared to the distributions following the other rotational schemes (Table 5). Seed density of C. bursa-pastoris, Galinsoga cillata, P. major, P. oleracea were increased much more than with the other rotational crops (Table 6).
At Geneva, the only contrast of significance of rotational crop on the weed seed bank distribution was alfalfa compared to rye (Table 7). The density of seeds of Amaranthus spp., A. artemisiifolia, and S. sarrachoides were higher following the rye rotation compared to the other rotational crops (Table 8). However, the density of S. media seeds was higher following the alfalfa rotation compared to the other rotational schemes.
Incidence of cabbage heads infected by F. avenaceum. The interactions of location with rotational crop scheme and with management practices in cabbage were insignificant (determined by likelihood ratio tests). This meant that results could be pooled across locations. There were significant effects of rotational crop scheme (P = 0.0004) and management practices in cabbage (P = 0.0001), but the interaction between rotational crop scheme and cabbage management practices was not significant (P = 0.2641). The effects of rotational crop scheme were compared using simple contrasts (Table 9). The incidence of cabbage heads with symptoms of infection by F. avenaceum was significantly higher following the rye and sweet corn rotational schemes (Table 10).
Simple contrasts were also done among management practices in cabbage. The main result was that the incidence of cabbage heads infected by F. avenaceum was significantly lower in sub-plots under management practice #4 (Table 11).
Total weight of cabbage harvested. There was a significant effect of management practice in cabbage (P < 0.0001) on total weight of cabbage harvested (Table 12). Simple contrasts of the effect of management practices indicated that there was no benefit of applying foliar fungicides in addition to herbicide use (Table 13). Total yield was as good in non-chemically managed plots (handweeding, mulch) as in herbicide and fungicide managed plots. Attempting to control foliar diseases while neglecting weed control led to reduced yields. The overall nonsignificant F-test (P = 0.1859) for the effect of rotational crop masked underlying significant effects that were site-specific. Each location must be interpreted separately. Yields appeared to be higher following the sweet corn rotation, in general (Table 14). Contrasts showed that yields following the sweet corn rotational scheme were consistently higher than those following the rye rotation (Table 15). Yields were higher following sweet corn compared to the clover and alfalfa rotations, but not at the Geneva location (Table 15).
The results and philosophy behind this trial have been discussed at cabbage advisory committee meetings, and at informal gatherings of cabbage producers. We are in the planning stages towards development of a publication that will encompass the entire trial. We are also in the planning stages towards development of a publication and fact sheets on the identification of the new cabbage rot caused by Fusarium sp.
Impacts of Results/Outcomes
Hot dry weather in the final and most crucial year of this trial resulted in reduced disease incidence in cabbage at all 3 locations. Disease caused by the target organisms did not materialize. Thus, the impact of sustainable simultaneous weed and disease control could not be demonstrated.
The outcome was complicated by the presence of a new pathogen in the system, and the lack of traditional pathogens due to unfavorable environmental conditions. The new pathogen (Fusarium species) has been previously observed but disease incidence has always been low and not considered a significant threat to most cabbage crops. The early symptoms of this new disease resemble those of white mold, caused by Sclerotinia sclerotiorum. It is possible that growers have mistakenly identified diseased caused by Fusarium sp., as white mold, caused by S. sclerotiorum.
Growers have shown continued interest in ways to control weeds and disease in a sustainable manner. An organic producer worked with us on a separate trial in 2002 looking at different mulches that might provide simultaneous weed and Alternaria leaf spot control in several crucifer species produced on his farm. The use of mulches and compost for weed and disease management continues to be an area of considerable interest among growers.
Areas needing additional study
Additional research is needed on the use of mulches for disease and weed control in vegetable cropping systems. Additional research is also needed to define the environmental conditions that promote infection of cabbage by species of Fusarium, resulting in the head rot symptoms documented in this trial.