PROJECT RESULTS

Development of the Agronomical Information Relevant to the Successful Cultivation of Dry Beans in the Manitoba Interlake Region

Applicant: 

Eric Klassen
S.S. Johnson Seeds Limited
Arborg, Manitoba  R0C 0A0  Canada

Table of Contents:

• Background and Objectives
• Procedure and Project Activities
• Results and Discussion
• Conclusions

ARDI Project:

#98-196

Project Status:

Completed December, 2000

Background and Objectives:

The cultivation of dry beans (phaseolus vulgaris L.) is new for the Manitoba Interlake region, as the need to diversify crop production has led producers to experiment with alternative crops. The development of new upright, earlier maturing dry bean varieties has encouraged Johnson Seeds to investigate the cultivation of dry beans in the Manitoba Interlake region.

One of the issues to be resolved with the production of dry beans in the Interlake is field stones. Dry bean plants tend to hold their pods very close to the ground, making it necessary to keep the cutter bar close to the ground when harvesting. For dry pea production, field rollers have been used successfully to press the stones into the ground. The rolling operation in a pea field occurs very shortly after crop emergence as use of the roller prior to emergence increases the risk of soil crusting. Although the method of handling field stones in dry pea production has been successful in the Interlake, it is not certain it will work as well for dry bean production. In dry beans, the hypocotyl elongates during emergence as opposed to the epicotyl, which elongates in peas. The hypocotyl arch in dry beans is quite rigid, while the epicotyl in peas is much more pliable. This increased rigidity in dry beans should assist in penetrating crusted soils, but could also make the young seedlings more susceptible to damage if flattened by a roller after emergence.

Traditionally, dry beans in Manitoba have been grown under a wide 30 inch row spacing regime using a plant population of around 100,000 plants per acre. This allows for good air circulation under the crop canopy, reducing the risk from white mould. It also allows for inter-row cultivation to assist in weed control. Specialized equipment has been developed to both plant and harvest the crop grown under the wide row system. New growers experimenting with dry beans are interested in producing the crop with their existing equipment. Therefore, they are solid seeding beans using air seeders/drills, press drills and disc seeders and either swathing or direct combining the crop at harvest. Data generated from Ontario and North Dakota indicate that at narrow row spacings, the seeding rate should be increased up to 170,000 plants per acre for 7 inch rows. The University of Saskatchewan recommends 90,000 to 110,000 plants per acre depending on the narrowness of the canopy of a particular variety. Increasing the plant population may affect the crop in the following ways: quicker canopy closure, better crop competition with weeds, no inter-row cultivation, increased risk of white mould, taller crop canopy possibly improving pod clearance, reduced days to maturity, higher yield, reduced stem strength and smaller seed size.

Procedure and Project Activities:

Rolling Bean Trials

Two research trials (a small plot and a field scale) were conducted in 1999 and 2000 to compare the effects of various rolling treatments on dry bean production at Arborg, Manitoba. The rolling treatments in the small plot trial included rolling prior to emergence, 3 days after emergence, 12 days after emergence and a control (no rolling). A 0.92 metre wide drum, giving a soil pressure of 1.2 kg/cm of roller width, was used for the rolling treatments in the small plot trial. The pinto bean variety AC Burrito was used in both the small plot and the field scale trial both years, as it is an early maturing, upright growing type that is adapted to the Interlake region. A 12.2 metre Degelman field roller was used in the trial producing a soil pressure of 5.0 kg/cm of roller width. No additional weight was added to the field roller. Unfortunately, the field trial had to be abandoned in 2000 because of a poor stand and a problem with "drowned out" spots.

A trial was conducted at Arborg in 1999 and 2000 to examine the effect of three seeding rates on the response of three dry bean varieties representing three classes of beans, each with a different growth habit. The same three varieties were used both years: AC Skipper (navy) was chosen for its upright growth habit, AC Burrito (pinto) was chosen for its semi-upright plant architecture while NW 63 (small red) was chosen for its fairly prostrate plant type. In 2000, the row spacing was changed from 12 inches (1999) to 6 inches and the target plant population went from 121, 202 and 283 thousand plants per acre to 101, 182 and 263 thousand plants per acre (30, 50, 70 seeds/m² to 25, 45, 65 seeds/m² respectively).

Results and Discussion:

Rolling Bean Trials

Differences in yield; number of plants/m²; number of damaged plants/m²; and soil bulk densities measured from 0 - 5cm, 5 - 10cm and 10 - 15cm were not significant for the small plot trial in the two year summary (Table 1) or the 1999 trial (Table 3). In 2000, the differences in the number of plants damaged by the post emergent rolling treatments were significant (Table 4). Plants rolled three days after emergence (hypocotyl arch stage) sustained more damage than plants rolled 12 days after emergence (unifoliate leaf stage), but the number of damaged plants for either treatment were still too small to affect the number of plants/m² or the seed yield. Significant differences for the number of damaged plants were also observed in the field rolling trial, but again, this did not significantly affect the surviving number of plants/m² (Table 2). In the field rolling trial, yield; plants/m²; and bulk density differences at 0 - 5cm, 5 - 10cm and 10 - 15 cm were not significant.

Table 1.  Two Year Summary of Plant and Soil Responses as Measured in the Dry Bean Small Plot Rolling Trial, Arborg, MB 1999 & 2000

^a the rolling was conducted using a 0.92m wide drum giving a soil pressure of 1.2 kg/cm of roller.

^b Coefficient of Variation - the amount of variation in the data for which the source is unknown.

^c Least Significant Difference. Any two means in the column that differ by more than this value are considered significantly different with a confidence level of 95 percent. "ns" indicates that the differences are not statistically significant.

Table 2.  Plant and Soil Response as Measured in the Dry Bean Field Rolling Trial Arborg, MB 1999

^a the rolling was conducted using a 12.2 metre wide Degelman Industries field roller. The empty Weight was 6,125 kg giving a pressure of 5.0kg/cm of roller.

^b Coefficient of Variation - the amount of variation in the data for which the source is unknown. Expressed as percentage of the overall mean.

^c Least Significant Difference. Any two means in the column that differ by more than this value are considered significantly different with a confidence level of 95 percent. "ns" indicates that the Differences are not statistically significant.

Table 3.  Plant and Soil Responses as Measured in the Dry Bean Small Plot Rolling Trial, Arborg, MB 1999

^a the rolling was conducted using a 0.92m wide lawn roller, giving a soil pressure of 1.22 kg/cm of roller.

^b Coefficient of Variation - the variation in the data for which the source is unknown.  Expressed in terms of a percentage of the overall mean.

^c Least Significant Difference - "ns" indicates that the differences are not statistically significant.

Table 4.  Plant and Soil Responses as Measured in the Dry Bean Small Plot Rolling Trial, Arborg, MB 2000

^a the rolling was conducted using a 0.92m wide drum and rolled two times across each rolled plot giving a soil pressure of 1.20 kg/cm of roller.

^b Coefficient of Variation - the amount of variation in the data for which the source is unknown.

^c Least Significant Difference. Any two means in the column that differ by more than this value are considered significantly different with a confidence level of 95 percent.  "ns" indicates that the Differences are not statistically significant.

Seeding Rate Trials

Growing conditions for the two years were quite different. In 1999, the spring was fairly warm and the bean plant canopies closed by early flowering, although there were a few gaps in the plots with the lowest seeding rate taking a little longer to cover the ground. In 2000, the weather was unusually cool and wet and the canopies, even at the highest seeding rates, did not close. The yields in 2000 were half those of 1999 and AC Skipper did not produce a crop at all in 2000. The agronomic and economic results from 1999 can be seen in Tables 5 and 6 respectively, while the 2000 results can be found in Tables 7 and 8. In 1999, significant differences due to the variety effect were observed for yield, plant population, days to flower, days to maturity, canopy height and pod clearance. Significant linear effects were observed for the yield, plant population and pod clearance responses due to the seeding rate. Other significant effects due to seed rate were observed in days to flower and days to maturity. The thousand kernel weight seemed to have a linear response, but the reduction in seed size with increasing seed rates was small. Similarly, the treatment with the lowest seeding rate only took one extra day to mature than the other two seeding rate treatments. The interaction between the variety and seed rate effects were not significant at the 95 percent confidence level for the factors measured in this trial, with the exception of canopy height which increased linearly for the variety AC Skipper only. In 2000, with AC Skipper treatments removed, the trends were similar to those in 1999. More thousand kernel weight data were recorded in 2000, allowing the statistical analysis of the data. The thousand kernel weight did not drop significantly from the 25 to 45 seeds per m² seeding rate, but did drop significantly from the 45 to 65 seeds per m² seeding rate. No white mould was observed in any of the treatments either year and the increase in plant stand did not cause lodging in any of the plots.

The results of the economic analysis of the trials both years show that as the seed rate increased, the overall gross potential return increased, but at the cost of an increased risk of production due to the higher seeding costs. The incremental cost and thus the increased risk due to seed rate is smallest for the smallest seeded cultivar, AC Skipper. From the 1999 data, at the price levels used in this analysis, an argument could be made that the best seed rate for AC Burrito and NW 63 would be 200,000 seeds per acre. The net incremental revenue for these two varieties either stayed the same or decreased for the highest seed rate. The net incremental revenue for AC Skipper still increased for the highest seed rate, but not by as much as the increase from the 121,000 plants per acre seeding rate to the 202,000 plant per acre seeding rate. In 2000, the highest seeding rate of 263,000 seeds per acre produced the best returns. The net incremental revenue was also the greatest for the highest seeding rate, indicating that, in retrospect, for the 2000 Arborg, Manitoba growing season, the best recommendation would have been to plant dry beans at 263,000 seeds per acre.

Unfortunately, a warm, moist year was not encountered in either 1999 or 2000. Using these data, it could be expected that the recommendation would be to seed at somewhere between 160,000 to 220,000 seeds per acre. Given a warm moist year, however, where white mould could become a factor, these seeding rates may be too high, especially for the more prostrate growing varieties.

Conclusions:

In both 1999 rolling trials, a rain heavy enough to cause significant crusting was experienced. In spite of this, the dry bean seedling emergence did not seem to be affected. Therefore, it is recommended that dry bean fields that need to be rolled, should be rolled prior to emergence. If problems due to equipment availability, grower schedule, or soil moisture conditions prevent rolling prior to emergence, rolling can be conducted from approximately seven to 12 days after emergence without affecting yield. Rolling should occur after the straightening of most of the hypocotyl arches in the field, but prior to the growth of the first trifoliate leaf. All vehicle travel over a wet field should be avoided as wet soil conditions increase the possibility of compacting the soil (Smith, 1996).

Using the data obtained from the 1999 and 2000 seeding rate trials, it could be expected that the recommendation would be to seed at somewhere between 160,000 to 220,000 seeds per acre. Given a warm moist year, however, where white mould could become a factor, these seeding rates may be too high, especially for the more prostrate growing varieties. Further seed rate research should be conducted to gather more data over different growing conditions.

References:

Smith, J. 1996. Soil Compaction - A major problem in dry edible bean production. "The Bean Bag". Volume 14 Number 2.

Acknowledgements:

This work could not have been conducted without the support received from the Agri-Food Research and Development Initiative (ARDI).

ARDI Home - Contact ARDI