Introduction
Fresh-market potato prices are often quite volatile among production seasons and even within a season, but the key to economic profitability is to be first to market. The economic advantage of timing the fresh market appropriately is stark and essential to the long-term solvency of fresh-market potato production (M. Copas, personal communication). With market timing in mind, growers must balance rapid crop growth and tuber physiological maturity with tuber physical maturity that enhances skin set and harvest ease.
Potato vine senescence initiates three tuber maturity processes that occur in the final 2 to 3 wk of plant growth. Chemical maturity is particularly important in processing potato, as it affects fry or chip color. Physiological maturity increases crop yield and tuber size. Physical maturity refers to processes by which the tuber skin or phellem, as part of the periderm, binds to underlying phellogen tissue to prevent physical damage during harvest and handling. Additionally, physically immature tubers inhibit machine harvest when the stolons remain connected to the potato vine and take longer to heal wounds that are subject to pathogen infection and disease spread (such as late blight caused by Phytophthora infestans) (Bussan et al. Reference Bussan, Sabba and Drilias2009; Haderlie et al. Reference Haderlie, Halderson, Leino, Petersen and Callihan1989; Rowe Reference Rowe1993; Wiltshire et al. Reference Wiltshire, Milne and Peters2005).
Physical tuber maturation can be hastened by desiccating or cutting potato vines. The current industry practice for pre-harvest vine management in many production regions is to apply diquat herbicide once or twice starting about 3 wk prior to anticipated harvest (Arsenault and Ivany Reference Arsenault and Ivany2001). In many cases, additional herbicides labeled as desiccants are also applied in conjunction with diquat herbicide in a stop-gap approach (Colquhoun et al. Reference Colquhoun, Gevens, Groves, Heider, Jensen, Nice and Ruark2018). Alternatively, mechanical vine management has been utilized to promote tuber maturation and skin set in some production regions and was the mainstay prior to the advent of synthetic herbicides. Mechanical methods include vine flaming, cutting, and pulling. Vine flaming is often used in organic production, but investment and operation costs are prohibitive in larger scale conventional production. Vine pulling induces immediate plant death and thus skin set but is a slow process and can lead to greening of tubers as they are pulled to the soil surface and exposed to sunlight. Vine cutting has been used for potato vines that have begun natural senescence, but plant stems cut when in active vegetative growth can resprout and therefore impede skin set (Kempenaar and Struik Reference Kempenaar and Struik2007).
Potato vine management with herbicides is challenging; adequate coverage is difficult when using contact herbicides that do not move within the plant, as healthy canopies can often exceed 1 m in plant height. Systemic herbicides are rarely considered, as they often end up in the tuber as the resource sink at that time of year, resulting in potential illegal pesticide residue and reduced storability (Colquhoun et al. Reference Colquhoun, Gevens, Groves, Heider, Jensen, Nice and Ruark2018). This conundrum, combined with a lack of new herbicide active ingredients in the age of herbicide-resistant crops, has stymied the ability to register new herbicides as vine desiccants.
Ironically, potato growers have become quite adept at growing very healthy and vigorous potato plants with improved irrigation, fertility, and disease management, but these stress-free plants have in recent years proved to be much more difficult to manage to a level that promotes skin set and allows for harvest of quality, stolon-free tubers (Bussan et al. Reference Bussan, Sabba and Drilias2009). This situation poses a potato perfect storm of sorts: growers have figured out ways to consistently grow healthy plants that support a tuber yield that is economically viable, but the current vine management methods are not allowing harvest in a timely manner.
Keeping in mind the challenges surrounding management of healthy vines with contact herbicides, a few researchers have compared mechanical and chemical methods and their ability to enhance desiccation and skin set. For example, Waterer (Reference Waterer2007) compared vine flail chopping with diquat herbicide application at various treatment timings, concluding that herbicide desiccation required initiation 3 wk prior to tuber harvest, compared to flail-chopped vines that could be harvested immediately––but at the cost of mature skin set.
The research presented here was conducted to evaluate the effect of combining mechanical and chemical management strategies on vine management, tuber quality, and fresh-market tuber size distribution and yield. Alternative management strategies included single and split applications of diquat, rolling, flame burning, or mowing the top third of the potato plant, as well as combinations of both chemical and physical methods. We hypothesized that physically removing part of the plant exposes more of the canopy to better diquat coverage in the lower plant canopy and results in better skin set and stolon separation, particularly in growing seasons where potato vines remain vigorous until the desired harvest time.
Materials and Methods
Studies were conducted in 2015 and 2016 at the University of Wisconsin Hancock Agricultural Research Station in Hancock, WI (44.1336° N, 89.5232° W). The soil type was Plainfield loamy sand (sandy, mixed, mesic, Typic Udipsamment) with 0.8% organic matter and a pH of 6.5. The study was arranged in a randomized complete block design with four replications. Individual plots measured 6.1 by 3.7 m with four potato rows spaced 76 cm apart. Plots were planted on April 22 in each season. The field was on a 4-yr rotation that consisted of corn (Zea mays L.), corn, soybean [Glycine max (L.) Merr.], and potato. Potato varieties included ‘Dark Red Norland’ and ‘Yukon Gold’, both commonly grown fresh-market varieties in the production region. All production practices with the exception of vine management were conducted in a manner consistent with local commercial potato production practices (Colquhoun et al. 2018; Westermann Reference Westermann2005).
Treatments consisted of integrated vine management methods starting either 21 or 14 d prior to potato harvest (Table 1). Management timing was based on harvest targeted for fresh-market tuber size distribution and occurred on August 11 in each year. Diquat herbicide (Reglone, Syngenta Crop Protection, Greensboro, NC 27419) was applied with non-ionic surfactant (0.25% vol/vol; Activator 90, Loveland Products, Loveland, CO 80538). The desiccant was applied with a tractor-mounted air-pressure sprayer calibrated to deliver 187 L ha–1 at 186 kPa with Teejet XR8003VS nozzle tips (Spraying Systems Co., P.O. Box 7900, Wheaton, IL 60187). Roller operations were performed with a tractor-mounted four-row plowpacker by Brillion (Landoll Co., 1900 North St, Marysville, KS 66508). Flail-chopping operations were performed with a tractor-mounted customized two-row Brady chopper (Quality Implement Co., 3526 South Treadway St., Abeline, TX 79602). Mowing of the top one-third of the plant was achieved using a gas-powered weed whip with disc blade attachment. Vine flame burning was accomplished with the use of a 9-kg propane cylinder with wand flame attachment designed specifically for plant use.
Potato vine management programs utilized in research conducted at the Hancock Agricultural Research Station, Wisconsin, in 2015 and 2016.
a Diquat applied at rate of 0.52 kg ai ha–1 with non-ionic surfactant (0.25% vol/vol).
Potato stem and leaf reduction were visually assessed on a scale of 0 (no reduction) to 100% (complete reduction via stem or leaf desiccation or removal) at 0 and 8 d prior to harvest (PTH). Vine management programs were compared to nontreated potato vines to determine the overall level of stem and leaf management. Potato tuber harvest was handled in a manner consistent with commercial production practices to simulate realistic tuber skinning and stolon separation. Tubers were then run through a grading and washing machine, with each tuber graded according to USDA-AMS standards (USDA 2011). Fifty tubers were randomly selected from each plot for visual skin set evaluation on a scale of 0 (no skinning) to 100% (complete skin loss) and number of stolons attached to each tuber. Tubers were then placed in cold storage at 13 C and 95% relative humidity for 21 d to simulate commercial storage practices, after which they were once again visually assessed for skinning and stolon presence. Potato leaf and stem management, skin set, tuber yield, and stolon presence were subjected to ANOVA to determine if there was a treatment effect. Interactions were observed between treatments and study year; thus, the data are presented separately by year. Means were separated using Fisher’s Least Significant Differences (LSD) at P=0.05.
Results and Discussion
Stem and Leaf Management
In 2015, ‘Dark Red Norland’ leaf management 8 d PTH was poorest where potato vines were rolled or mowed 14 d PTH, but by harvest there were no differences among treatments (Table 2). Stem management was less straightforward, but mechanical methods were generally less effective than herbicide desiccation. Mowing potato vines was particularly ineffective and resulted in the least stem reduction among treatments when measured at harvest. Leaf and stem management were reduced in the mowing treatment, given that the technique removed only the top third of the plant.
Visual estimation of ‘Dark Red Norland’ potato leaf and vine management at the Hancock Agricultural Research Station, Wisconsin, in 2015.
a Means within columns followed by the same letter are not significantly different according to Fisher’s protected LSD test at P≤0.05. No significant differences within a column were observed when no letters are included.
b Diquat applied at rate of 0.52 kg ai ha–1 with non-ionic surfactant (0.25% vol/vol).
Similar to ‘Dark Red Norland’, ‘Yukon Gold’ vine management in 2015 was poor through harvest where rolling alone was used (Table 3). When evaluated 8 d PTH, mowing was also ineffective and controlled only 65% of leaves but was complete by harvest. Additionally, flame burning 14 d PTH was better at leaf management than mowing or rolling but less effective than all other treatments. Stem management through harvest was particularly poor where vines were rolled at 14 d or mowed at 14 d followed by diquat 8 d PTH. Of the mechanical methods, flail chopping was more efficacious. When stem management was quantified 8 d PTH, stem reduction levels were similar where flail chopping 21 d PTH was followed by either diquat or flame burning compared to where diquat was applied twice.
Visual estimation of ‘Yukon Gold’ potato vine leaf and stem management at the Hancock Agricultural Research Station, Wisconsin, 2015.
a Means within columns followed by the same letter are not significantly different according to Fisher’s protected LSD test at P≤0.05. No significant differences within a column were observed when no letters are included.
b Diquat applied at rate of 0.52 kg ai ha–1 with non-ionic surfactant (0.25% vol/vol).
Potato vines for both varieties were more challenging to manage in the 2016 growing season compared to 2015, although the trends among treatments were quite similar (Tables 4 and 5). ‘Dark Red Norland’ potato vine rolling or mowing continued to be rather ineffective in both leaf and stem management. When potato stem reduction was quantified 8 d PTH, all mechanical methods initiated 14 d PTH were less effective than those initiated 21 d PTH, emphasizing the importance of vine management timing, regardless of method, as reported by Waterer (Reference Waterer2007). At harvest, stem management was 100% for all treatments with the exception of where vines were rolled. Although ‘Yukon Gold’ leaf and stem management was variable among methods when quantified 8 d PTH, reduction was complete for all treatments by harvest (Table 5). Stem management at the first evaluation timing was particularly poor where flame burning was used 14 d PTH, but when followed by diquat, stems were later eliminated.
Visual estimation of ‘Dark Red Norland’ potato leaf and vine management at the Hancock Agricultural Research Station, Wisconsin, 2016.
a Means within columns followed by the same letter are not significantly different according to Fisher’s protected LSD test at P≤0.05. No significant differences within a column were observed when no letters are included.
b Diquat applied at rate of 0.52 kg ai ha–1 with non-ionic surfactant (0.25% vol/vol).
Visual estimation of ‘Yukon Gold’ potato vine leaf and stem management at the Hancock Agricultural Research Station, Wisconsin, 2016.
a Means within columns followed by the same letter are not significantly different according to Fisher’s protected LSD test at P≤0.05. No significant differences within a column were observed when no letters are included.
b Diquat applied at rate of 0.52 kg ai ha–1 with non-ionic surfactant (0.25% vol/vol).
Tuber Skinning and Stolon Count
In 2015, ‘Dark Red Norland’ tuber skinning at harvest was greatest in the nontreated potatoes and where vines were rolled, mowed, or flail chopped 14 d PTH (Table 6). Tuber skinning tended to increase from the harvest evaluation timing to 21 d past harvest (DPH). Tuber skinning at both evaluation times was similar to when diquat was applied twice and when plots were flail chopped 21 d PTH followed by flame burning, indicating that similar skin set can be achieved by mechanical means alone. ‘Dark Red Norland’ stolon presence was similar among all treatments both at harvest and 21 DPH.
Visual estimation of ‘Dark Red Norland’ skinning and stolon presence at the Hancock Agricultural Research Station, Wisconsin, 2015.
a Abbreviation: DPH, d past harvest.
b Means within columns followed by the same letter are not significantly different according to Fisher’s protected LSD test at P≤0.05. No significant differences within a column were observed when no letters are included.
c Diquat applied at rate of 0.52 kg ai ha–1 with non-ionic surfactant (0.25% vol/vol).
‘Yukon Gold’ tuber skinning at harvest in 2015 ranged from 2.6% where diquat was applied twice to 8.8% for the nontreated potatoes (Table 7). Tuber skinning at harvest was least in treatments initiated 21 d PTH but increased slightly after 21 d of storage and ranged from 3.4% in the split application of diquat to 13.9% for the nontreated potatoes. Similar to ‘Dark Red Norland’, stolon presence was similar among management programs and the nontreated potato plants in 2015.
Visual estimation of ‘Yukon Gold’ skinning and stolon presence at the Hancock Agricultural Research Station, Wisconsin, 2015.
a Abbreviation: DPH, d past harvest.
b Means within columns followed by the same letter are not significantly different according to Fisher’s protected LSD test at P≤0.05. No significant differences within a column were observed when no letters are included.
c Diquat applied at rate of 0.52 kg ai ha–1 with non-ionic surfactant (0.25% vol/vol).
In 2016, ‘Dark Red Norland’ tuber skinning ranged from 2.9% when potato vines were mowed 21 d PTH followed by diquat application to 9.0% for the nontreated potatoes (Table 8). Tuber skinning was similar to the industry standard of two diquat applications where mowing was followed by one or two diquat applications, where flail chopping was followed by two diquat applications, and where flail chopping was followed by flame burning. No differences in stolon presence were observed among management programs for ‘Dark Red Norland’ in the 2016 season.
Visual estimation of ‘Dark Red Norland’ skinning and stolon presence at the Hancock Agricultural Research Station, Wisconsin, 2016.
a Abbreviation: DPH, d past harvest.
b Means within columns followed by the same letter are not significantly different according to Fisher’s protected LSD test at P≤0.05. No significant differences within a column were observed when no letters are included.
c Diquat applied at rate of 0.52 kg ai ha–1 with non-ionic surfactant (0.25% vol/vol).
Tuber skinning in the ‘Yukon Gold’ variety in the 2016 season was generally greater than in 2015 (Table 9). Drastically lower skin losses both at harvest and 21 DPH were observed in treatments initiated 21 d PTH when compared to those initiated 14 d PTH. Tuber skinning at harvest ranged from 3.0% for the split application of diquat starting 21 d PTH to 22.3% for the nontreated potato vines. Interestingly, stolon number at harvest was greater than in the nontreated vines in the same management programs mentioned above where skinning was least. By 21 DPH, stolon number was only greater than in the nontreated potato vines where mowing or flail chopping were followed by two diquat applications.
Visual estimation of ‘Yukon Gold’ skinning and stolon presence at the Hancock Agricultural Research Station, Wisconsin, 2016.
a Abbreviation: DPH, d past harvest.
b Means within columns followed by the same letter are not significantly different according to Fisher’s protected LSD test at P≤0.05. No significant differences within a column were observed when no letters are included.
c Diquat applied at rate of 0.52 kg ai ha–1 with non-ionic surfactant (0.25% vol/vol).
Tuber Yield
Tuber yield for ‘Dark Red Norland’ potatoes in the 2015 season was similar for all treatments and in all tuber size categories except for the 369- to 454-g tuber range (Table 10). We hypothesize that this result was due to the relatively greater and faster leaf and stem management during the 2015 growing season. This suggests that in some years any of the evaluated management programs will result in similar potato yields across tuber grades, regardless of differences in leaf and stem management. In 2015, total ‘Yukon Gold’ tuber yield was generally greater in treatments that were started 14 d PTH when compared to those started 21 d PTH (Table 11). We propose that this result is due to an increase in living leaf and stem tissue during tuber bulking stage compared to the earlier vine management treatments. However, cull tuber yield was less where treatments were started 21 d PTH. B-sized and 57- to 112-g tuber yield, which are the most desirable for fresh-market production for these varieties, were similar among treatments in the 2015 growing season.
Graded yield of ‘Dark Red Norland’ tubers at the Hancock Agricultural Research Station, Wisconsin, 2015.
a Means within columns followed by the same letter are not significantly different according to Fisher’s protected LSD test at P≤0.05. No significant differences within a column were observed when no letters are present.
b B size potatoes include those with a diameter of 4.4 cm or less.
c Diquat applied at rate of 0.52 kg ai ha–1 with non-ionic surfactant (0.25% vol/vol).
Graded yield of ‘Yukon Gold’ tubers at the Hancock Agricultural Research Station, Wisconsin, 2015.
a Means within columns followed by the same letter are not significantly different according to Fisher’s protected LSD test at P≤0.05. No significant differences within a column were observed when no letters are present.
b B size potatoes include those with a diameter of 4.4 cm or less.
c Diquat applied at rate of 0.52 kg ai ha–1 with non-ionic surfactant (0.25% vol/vol).
In 2016, ‘Dark Red Norland’ cull tuber yield was greatest where vines were flail chopped or rolled (Table 12). We propose that heavy rainfall soon after the implementation of these treatments eroded soil from around tubers near the surface and resulted in tuber greening, which, in turn, causes those tubers to be designated as “cull.” In 2016, ‘Yukon Gold’ tuber yield differed among vine management programs in the 57- to 112-g, 170- to 282-g and greater than 454-g categories, but not for total tuber yield (Table 13). However, no discernible pattern among treatments was observed for grade categories in which differences were observed.
Graded yield of ‘Dark Red Norland’ tubers at the Hancock Agricultural Research Station, Wisconsin, 2016.
a Means within columns followed by the same letter are not significantly different according to Fisher’s protected LSD test at P≤0.05. No significant differences within a column were observed when no letters are present.
b B size potatoes include those with a diameter of 4.4 cm or less.
c Diquat applied at rate of 0.52 kg ai ha–1 with non-ionic surfactant (0.25% vol/vol).
Graded yield of ‘Yukon Gold’ tubers at the Hancock Agricultural Research Station, Wisconsin, 2016.
a Means within columns followed by the same letter are not significantly different according to Fisher’s protected LSD test at P≤0.05. No significant differences within a column were observed when no letters are present.
b B size potatoes include those with a diameter of 4.4 cm or less.
c Diquat applied at rate of 0.52 kg ai ha–1 with non-ionic surfactant (0.25% vol/vol).
Vine management results for both varieties and years suggested that treatment timing and not method have a greater effect on overall plant senescence. Additionally, skin set improved when potato vine management was initiated 21 d PTH compared to 14 d PTH. Previous research suggests that longer time intervals between date of vine management initiation and date of harvest result in a decrease of tuber skinning (Terman et al. Reference Terman, Cunningham and Goven1952; Bethke and Busse Reference Bethke and Busse2010). The use of flail chopping and flame burning can be substituted for diquat with similar results for skin set and overall tuber yield, particularly when vine management programs are initiated 21 d PTH. These results are useful to growers in the organic industry, who are unable to use synthetic pesticides such as diquat for their desiccation needs.
Vine management cost is an important consideration. The cost to flail chop is double that of a single application of diquat and also more time-consuming. Flaming requires 112 L ha–1 propane when used in conjunction with a chemical desiccant (Murphy Reference Murphy1968). Flail chopping, mowing, and rolling machinery manage fewer rows when compared to pesticide spray booms, requiring more passes through fields and greater time and fuel consumption (Murphy Reference Murphy1968). These costs should be weighed against crop value, particularly relative to optimizing the grower return on investment for early fresh-potato production in volatile markets.
Acknowledgments
Partial funding for this project was provided by the USDA Specialty Crop Block Grant Program and the Wisconsin Potato Industry Board. No conflicts of interest have been declared.
