COTMAN^™: First Year's Perspective in North Carolina
(CCN-00-1A-January 19,2000)

A. M. Stewart, J. B. Coltrain, K. L. Edmisten, J. S. Bacheler, and D. W. Mott
Departments of Crop Science and Entomology, NC State and NC Coop. Ext. Service, Martin, Co

COTMAN^TM is a relatively new management tool that models crop growth and development. Plant monitoring and weather data inputs are used to aid a grower or consultant in making decisions regarding early season square loss, mid season irrigation and plant growth regulator applications, late season insecticide termination, and defoliation timing. Experience with COTMAN^TM in North Carolina and the northern regions of the cotton belt has been limited. An evaluation, therefore, of the COTMAN^TM system was initiated in North Carolina comparing it to current Extension Service recommendations.

Materials and Methods

Cotton was planted on May 10 at the Upper Coastal Plain Research Station near Rocky Mount, NC. Six treatments were imposed:

Experimental Design: The experimental design was a randomized complete block with four replications. Plots were six 36 inch rows ride and 50 feet long. Data were collected from the inside four rows of each plot. No irrigation was used. Excess nitrogen was applied to some treatments in an effort to delay maturity, but excessively wet conditions prevented any noticeable maturity delay from occurring.

Data Collection: For data collection, the COTMAN^TM users manual recommends sampling 4-8 sites per field (COTMAN^TM Expert System, Version 5.0. 1998, Univ. of Ark.). In our study, each of four replications were considered a 'site', and the four replications together were considered a 'field', providing the minimum four monitoring sites per field. Scouting for the bollworm/budworm complex as well as for other insect pests, with the exception of plant bugs, was conducted for all treatments, and followed published NC Cooperative Extension Service recommendations. In the COTMAN^TM treatments, data was collected as called for by the COTMAN^TM users manual. In the Extension treatments, in addition to normal insect scouting for the bollworm/budworm complex, plant bug scouting according to NCCES guidelines, plant monitoring according to the Modified Early Bloom Strategy for Pix use, and % open bolls and nodes above cracked boll for defoliation data were collected for management decisions. One of the COTMAN^TM decision rules advises the consideration of a "growth enhancing plant growth regulator" when the curve for optimal square falls below a specified level. Therefore PGR-IV was used in treatments 1 and 4. To obtain a measure of the relative scouting costs of the two systems, additional data were also collected on the time required for obtaining both COTMAN ^TM and Extension inputs. Time was recorded by a digital stop watch, beginning at initial entry into the plot, through data collection, and ending upon exiting the plot. We did not include the additional time required to travel from one site to another in the 'field'.

Results and Discussion

Square Retention and Plant Bugs: Square retention, an index of plant bug damage, averaged 98.0% and 98.9% for June 30 and July 7, respectively, averaged over all Extension treatments, while the COTMAN^TM treatments averaged square retention values of 98.6% and 98.4% for the same dates. High square retention values were not unexpected due to plant bugs typically being a minor pest, the absence of boll weevils, and a light second generation budworm flight in North Carolina. Extension guidelines for measuring square retention and the COTMAN^TM square mapping technique both returned very similar values for percent retention. The time required to map first positions squares in the COTMAN^TM system, however, was much greater (data not shown). In a state in which producers treat an average of approximately 0.6% to 6.1% of their cotton acreage for plant bugs (1990 to 1999), scouting for this pest must have modest time requirements.

Growth Regulator Applications: Due to a late planting season and unusually dry conditions, the crop growth curves generated by COTMAN^TM were shifted to the right and flatter than the target development curve (square populations and square production increases were less than optimum). This situation activated SQUAREMAN Decision Rule #5 which states that "a growth enhancing plant growth regulator may help to retain fruit", triggering the application of 4 oz/acre of PGR-IV to treatments 1 and 4. This occurred twice, on July 7 and July 14. Research in North Carolina has consistently shown no response to PGR-IV. These applications did not appear to shift the growth curves in any direction different from treatments that did not receive PGR-IV. Additionally, no yield differences were observed among any of the treatments (Table 2). In cases where Pix may be a consideration, the COTMAN^TM system recommends consulting local Extension advisories. Following the modified early bloom approach, a Pix application at 12 oz/acre was triggered in all treatments on July 21. Plant monitoring according to the modified early bloom strategy was conducted only once in the COTMAN^TM treatments, when it was called for, and the time required for this plant monitoring is not included in the total time requirement for COTMAN^TM treatments shown in Table 1. Extension treatments were monitored twice and the time required for monitoring according to the modified early bloom approach was recorded.

Cutout and Insecticide Termination: COTMAN^TM defines cutout as occurring when nodes above white flower equals 5. Late season insecticide termination is recommended when NAWF=5 plus 350 DD60s. Cutout was reached in all of the COTMAN^TM treatments on either August 6 or August 8. For late season insecticide termination, 350 DD60s were calculated beginning on August 6, because historical weather data for Raleigh, NC in the COTMAN^TM program says that August 6 is the latest possible cutout date. 350 DD60s were accumulated on August 22. During that time period, and later in the fall, essentially no insect damage was found in either the Extension or COTMAN^TM plots. At the time of the 350DD60's accumulation, less than one non-terminal square and less than one bloom were found in 200 row feet in the Extension treatments, indicating that these plots were probably at least 4 to 7 days beyond the need for possible protection from bollworms under Extension guidelines (see 2000 Cotton Information booklet for recommendations for terminating insecticide recommendations).

It would appear that NAWF=5 plus 350 DD60s should be adequate to protect bolls from bollworms under normal cutout. However, bollworms and European corn borers could pose a potential threat after 350 DD60s have accumulated in late, rank crops expressing either regrowth, or with prolonged square, bloom and boll production following 'cutout'. Additionally, most cotton fields will cut out on different dates. Further research in the area of late season insecticide termination for the northern regions of the cotton belt needs to be conducted.

Defoliation: COTMAN^TM initiates defoliation at NAWF=5 + 850 DD60s. Extension recommendations in North Carolina suggest defoliation should begin at 40-60% open bolls and/or 3 or 4 nodes above a first position cracked boll, depending upon the boll distribution on the plant. Due to high overall fruit retention on lower and middle nodes, we determined to defoliate the Extension treatments at 50% open and/or NACB=4. Due to Hurricanes Dennis and Floyd, this was not possible. Extension treatments were therefore defoliated on October 1 at 58.5% open and NACB=3.4. The accumulation of 850 DD60s in the COTMAN^TM treatments also fell on October 1, and the entire study was defoliated on the same day. The NAWF=5 + 850 DD60s COTMAN^TM rule coincided almost exactly with 60% open bolls. Research in North Carolina, however, has shown that defoliation can often be initiated much earlier than 60% with no detrimental effects on yield or fiber quality. Due to generally high early season square retention, the majority of the crop in North Carolina is often set over an 8-10 node horizon on the plant. This leads to an overall boll population that is closer in maturity than a crop that is set over a 12-14 node horizon, and often allows earlier defoliation in terms of % open and NACB. It should be remembered, however, that this study is one year's data on defoliation timing, and the impact of two major hurricanes cannot be diminished.

Time Requirements for COTMAN^TM vs. Conventional Cotton Monitoring: As shown in Table 1, the time required for the COTMAN^TM data collection was extremely high compared to the Extension treatments. The majority of the time, 53.8%, was spent mapping squares in the early season (data not shown). There were no significant lint yield differences between any of the treatments (Table 2). The COTMAN^TM system does provide a wealth of information on crop progress and may aid in scheduling harvest. It is, however, a very time consuming method of monitoring crop development. Given that the average field size in North Carolina is 14.2 acres according to the 1999 Boll Weevil Containment Program field reports to ASCS offices, the investment in time to sample and track individual fields would be excessive for a grower or a consultant under our present field size and pest pressure. The COTMAN^TM system also relies heavily on cutout being defined as NAWF=5. Very little of North Carolina's cotton is irrigated, and can often begin flowering at NAWF=6. Lack of irrigation and variable weather patterns often result in 'temporary' cutout, with growth resuming with rainfall. This would add another level of complication to interpreting COTMAN^TM outputs.

Martin County Note: Time Requirements for COTMAN^TM Field Data Collection

An additional 1999 evaluation of the time required for gathering COTMAN^TM data was conducted in Martin County in the northeastern coastal plain of North Carolina, in two cotton Farms totaling 37 acres. One 17-acre farm (Farm 1) with varying soil types was composed of 6 fields varying in size from 1.6 to 4.7 acres, with each individual cotton field serving as a COTMAN^TM study field. In the more uniform soil on the second farm (Farm 2), the 20 acres was divided into16 variety strips each with an adjacent check variety. Each variety served as an individual COTMAN^TM study field for the purpose of our study. For all but the initial sampling date, each of the 22 COTMAN^TM 'fields' was subdivided into 4 sampling units, as per COTMAN^TM User's Guide instructions.

The time taken to obtain and record data on stand counts, first fruiting node, SQUAREMAN square number and position data (4 assessments per field), and BOLLMAN data (4 trips) were recorded with a digital stop watch and refers to the data collection time only, not the time taken to walk between 'fields' (as was also the case in the Rocky Mount study). As can be seen in Table 3, The SQUAREMAN portion of the plant data takes the longest (1.5 to 2.0 hours per field) and is the most difficult to justify with our generally low plant bug pressure (also, see PGR-IV comment below). The BOLLMAN portion of the model, though also time-consuming, had approximately half the time requirements and the SQUAREMAN component, and at least provides information about a field's susceptibility to further damage from an annual, treatable pest, the cotton bollworm. Farm 2 had smaller field sizes than Farm 1, and in both cases they are probably smaller than what producers would find on their own farms. Based on one year's data, however, it appears that the time input per acre decreases with increasing field size (Table 3). It would follow that as the user becomes more experienced with the program and field sizes become larger, the time input per acre would go down. Given an average field size of only 14.2 acres in North Carolina, the time input per acre would have to decrease significantly to justify COTMAN^TM use over a large area.

Summary

The COTMAN^TM system, after one year's experience, appears to require a heavy time input for the value of the information received, compared with present recommendations. Additionally, the COTMAN^TM model suggests the application of PGR-IV growth regulator when crop growth does not satisfy certain parameters. Under the conditions of this study, these applications resulted an unneeded expenditure of $16.68/acre ($5.84 for PGR-IV plus $2.25 application. cost x 2 applications. = $16.68).  Parts of the model, such as the square mapping portion, appear to have limited use in North Carolina, while other parts, such as the NAWF measurements, may be useful in tracking the progress of individual fields. In any case, the time needed to collect the required data needs to be reduced. To determine if this model has any utility for North Carolina cotton producers, further research is needed to evaluate the various components of the program for non-irrigated cotton in the more northern cotton producing regions of the Southeast.

TABLE 1. Time expenditure averaged over all COTMAN^TM and Extension treatments¹ Rocky Mount, NC, 1999.

1. Time indicated is in addition to regular insect scouting, primarily for the bollworm/budworm complex.

2. Includes plant bug scouting (twice), Pix plant monitoring (twice), percent open bolls (twice), and NACB measurements (twice).

TABLE 2. Lint yields, Rocky Mount, NC, 1999.

TABLE 3. Time required to obtain various COTMAN^TM plant assessments;
Martin County, North Carolina, 1999 (n = 22  'fields').

1. Each 'field' consisted of 4 data sites during the June 28 through August 15 assessments;  6 sites/=Field= were utilized for the stand counts. Time is in addition to normal scouting for bollworms and other late season pests.
2. 6 fields ranging from 1.6-4.7 acres. Total of 17 acres.
3. 16 fields of 0.6 acre size plus borders equaling 20 total acres.
4. 233.4 = 3hrs and 53 minutes.; 198.6 = 3hrs and 19 minutes.