Alternative Systems for Cultivating and Side Dressing Specialty Crops for Improved Nitrogen Use Efficiency

1 An ASABE Meeting Presentation DOI: 10.13031/aim.20162456725 Paper Number: 162456725 Alternative Systems for Cultivating and Side Dressing Specialty Crops for Improved Nitrogen Use Efficiency Mark C. Siemens and Ronald R. Gayler Department of Agricultural and Biosystems Engineering, University of Arizona, Tucson, AZ Written for presentation at the 2016 ASABE Annual International Meeting Sponsored by ASABE Orlando, Florida July 17-20, 2016 ABSTRACT. In Arizona raised bed vegetable crop production, crops are conventionally cultivated and fertilized postemergence using tools positioned close to the plant row. These practices can injure plants through excessive soil disturbance and by pruning sensitive feeder roots. Furthermore, conventional side-dress applicators place fertilizer at the edge of bed sidewalls, a location that is not optimal for plant uptake. The goal of this project was to determine if alternative technologies and systems for cultivating and side-dressing that induce minimal soil disturbance and more optimally place fertilizer reduce fertilizer inputs and increase yields in vegetable crops. Field trials with iceberg lettuce, romaine lettuce and broccoli were conducted to compare these alternative systems with conventional methods. Results showed that use of an alternative, point injection fertilizer applicator which placed fertilizer in the root zone increased nutrient uptake efficiency by over 20% and crop yield by 19% in iceberg lettuce when deficient rates of nitrogen were applied. Although nutrient uptake was not improved by using the point injection applicator in romaine lettuce, there was a trend that crop yields were increased by approximately 11% when the device was used and bed sidewalls were shaved (conventional cultivation). Yield improvements were not realized when the point injection applicator was used in conjunction with a cultivating technique that left bed sidewalls intact. This result was not expected since it was hypothesized that bed shaving would prune roots which would limit nutrient uptake, crop growth and yield. In broccoli, use of the point injection in conjunction with cultivation techniques that either shaved sidewalls or left them intact increased nitrogen uptake by more than 27% as compared to the standard applicator. Although not statistically significant, total yield and marketable yields were also higher by >8% and >15% respectively when the beds were not shaved and point injection applicator was used. Combined, the study results imply that fertilizer applicator and cultivation systems that place fertilizer in the root zone and induce minimal soil disturbance improve nutrient use efficiency and/or increase yield in lettuce and broccoli production. Additional trials at the research and field scale levels are needed to confirm this finding. Keywords. applicator, crop production, cultivation, efficiency, fertilizer placement, nitrogen, point injection, soil disturbance, specialty crops. The authors are solely responsible for the content of this meeting presentation. The presentation does not necessarily reflect the official position of the American Society of Agricultural and Biological Engineers (ASABE), and its printing and distribution does not constitute an endorsement of views which may be expressed. Meeting presentations are not subject to the formal peer review process by ASABE editorial committees; therefore, they are not to be presented as refereed publications. Citation of this work should state that it is from an ASABE meeting paper. EXAMPLE: Author s Last Name, Initials. 2016. Title of presentation. ASABE Paper No. ---. St. Joseph, MI.: ASABE. For information about securing permission to reprint or reproduce a meeting presentation, please contact ASABE at http://www.asabe.org/copyright (2950 Niles Road, St. Joseph, MI 49085-9659 USA). 1 ASABE Annual International Meeting 1

Introduction In Arizona lettuce and broccoli production, cultivation and post-emergence fertilization (side-dressing) are necessary cropping practices for weed control and optimum crop yield. Conventionally, crops are planted on raised beds to facilitate furrow irrigation with two or more rows planted on each bed. Weeds on bed sidewalls are controlled using cultivators equipped with knives that shave the bed sidewalls, near the plant rows, and deposit the cut soil into the furrow. Following cultivation, the crop is side-dressed with roughly 100 lb/ac of nitrogen (N) using knife blade applicators which deposit fertilizer at the outer edges of the bed wall. Subsequent applications of approximately 80-120 lb/ac of N are made through a second side-dress application or through fertigation. These practices induce high soil disturbance close to the plant, prune feeder roots and do not place fertilizer in an optimal location for plant uptake. As a result, crop growth, nitrogen use efficiency (NUE) and crop yield can be negatively affected. An alternative system would be to side-dress and cultivate crops using implements that minimize plant disturbance and more optimally place fertilizer. For example, crops could be side-dressed using a rotary point injection fertilizer applicator. These systems utilize spikes attached to a rotatable wheel to inject liquid fertilizer into the soil at precise intervals and depths with minimal root damage and soil disturbance. As compared conventional knife blade applicators which deliver product at the edge of the root zone, these systems place fertilizers in the root zone, a more optimal location for plant uptake. Research studies have shown that as compared to conventional knife blade application, use of the system in corn increased grain N content by 16% (Timmons and Baker, 1992) and whole plant N content by 7% (Randall et al., 1997). When tested in sugar beets, use of the system improved NUE by 19% (Stevens et al., 2007). In the studies, the authors reported yield increases of 6-11% for corn and 10% for sugar beets. The system also showed good promise for use in iceberg lettuce production in a one year study (Siemens et al., 2011). During the critical head forming period two weeks after the second side-dress, N concentration in lettuce plant leaf tissue was 50% higher as compared to conventional knife blade application. Plant weights after the first and second side-dress applications were also significantly higher by 11% and 14% respectively. Further research is needed to confirm these preliminary findings. Furthermore, as an alternative to conventional cultivation, a rolling tine cultivator (Lilliston) could be used to control weeds on bed sidewalls. Used in conjunction with a point injection fertilizer applicator, the practice would leave beds intact, induce minimal plant disturbance, virtually eliminate root pruning and improve fertilizer placement. The goal of this project was to determine whether alternative technologies and systems for cultivating and side-dressing that induce minimal soil disturbance and more optimally place fertilizer reduce fertilizer inputs and increase yields in lettuce and broccoli crops. Specific objectives were to determine 1) whether uptake of fertilizer is improved through use of a point injection system type applicator as compared to a knife blade applicator, 2) how crop yield is affected through use of a point injection system type applicator as compared to a knife blade applicator, 3) if applied N rates can be reduced through use of a point injection system type applicator and 4) the effect of cultivating techniques on N uptake and crop yield. Materials and Methods Two sets of experiments were conducted. In the first study, point injection and knife blade applicators were compared in iceberg lettuce. In the second experiment, point injection and knife blade applicators in combination with different cultivation techniques were examined in romaine lettuce and broccoli. A description of the materials and methods used for each experiment follows. Point Injection and Knife Blade Applicator Comparison To facilitate conduction of the experiment, a two bed, two rank frame was fabricated. The unit was designed so that it could be equipped with either standard knife blade applicator knives or point injection type applicators (Fig. 1). The point injection system used was manufactured by SpikeWheel (SpikeWheel, Helix, OR). One spike wheel injector was utilized for each plant row and positioned so that the wheel operates on the bed side wall. To facilitate wheel rotation, the wheels were mounted on an angle so that wheel orientation was perpendicular to the bed side wall. Each spike wheel measured 21 inches in diameter and was comprised of 12, ½ inch diameter point injection units that injected liquid fertilizer at a depth of 4 inches below the soil surface. Spike spacing and wheel geometry was such that a point injection unit entered the soil and injected fertilizer every six inches during operation. Configured in this manner, fertilizer was injected below the plants, into the root zone. Each applicator was calibrated to apply 30 gal/ac of liquid. Trials with iceberg lettuce were conducted in the fall of 2011, fall of 2012 and spring of 2013 on furrow irrigated beds at the University of Arizona s Yuma Agricultural Center, Yuma, AZ. Soil type at the site was Gadsden Clay. Two crop rows were planted on each bed top. Experimental design was a randomized complete block design with 2 factors (applicator type and N rate) and 4 replications. Nitrogen in the form of UAN32 was applied in a split application at a standard rate of 200 lb/ac (100% of standard) and N deficient rates of 150 lb/ac (75% of standard), 100 lb/ac (50% of standard), 50 lb/ac (25% of standard) and 0 lb/ac (0% of standard). Experimental unit plot size was 50 feet long by 4 beds wide. Fertilizer uptake in lettuce ASABE Annual International Meeting Page 2

plants was assessed by measuring leaf midrib nitrate-n levels (NO 3-N) at four times during the growing period prior to the first side-dress (~ eight leaf stage of growth), two weeks after the first side-dress, two weeks after the second sidedress and at maturity. At maturity, 4 above ground whole plant samples from the middle two beds of each 4 bed plot were collected and weighed. The samples were then dried, weighed and analyzed for total N content to determine crop N uptake in terms of lb/ac. Crop yield was determined by harvesting, trimming and weighing individual heads from 10 ft of row from the middle two beds of each 4 bed plot. A marketable head was considered to be a head that weighed 1.5 lb or more. Data collected from the three experiments were combined for analysis. An ANOVA was performed using SAS (SAS Institute Inc., 2010) to determine statistically significant differences between treatment means. In the analysis, data were blocked by year, the 2 experimental factors and by replication. Point Injection, Knife Blade Applicator and Cultivator Technique Comparison In this experiment, a four bed fertilizer applicator unit was fabricated and used. Applicator units, point injection system wheel orientation and applicator flow rates were as previously described. Trials were conducted in the fall of 2015 on furrow irrigated romaine lettuce at the Yuma Agricultural Center, Yuma, AZ. Soil type was Gadsden Clay. Three crop rows were planted on each bed top as is typical for romaine crops raised for hearts. This crop type was selected as it was thought that the effects of root pruning would be prevalent since the outer crop rows are positioned close to the edges of the bed. Broccoli trials were conducted in the fall of 2014 at the same site with two crop rows planted to each bed. Soil type at this trial location was a Holtville Clay. For both experiments, experimental design was a randomized complete block design with 3 factors and 4 replications. Factors included four N application rates, two applicator types and two cultivation methods. Applied N rates during the first side-dress were 0% (check), 50%, 75% and 100% of the standard rate of N (100 lb/acre). Approximately 100 lb/ac of additional N in the form of UN 32 was applied by furrow irrigation approximately 4 weeks after the first side-dress. Applicator types examined were the point injection system and a conventional knife-blade applicator. Cultivation methods were controlling weeds on bed sidewalls using 1) a cultivator equipped with side knives that shaved the sides of the bed walls (beds shaved) and deposited the cut soil into the furrow and 2) a rolling cultivator (beds not shaved). The side knife cultivation method was used with the knife blade (knife - shaved) and the point injection applicators (point injection - shaved). The rolling cultivator was used only with the point injection applicator (point injection - non-shaved). Experimental unit plot size was 50 feet long by 4 beds wide. Plant N uptake was assessed by measuring leaf midrib nitrate-n levels prior to the first side-dress, two weeks after the first sidedress application and at maturity. At maturity, 6 romaine and 4 broccoli above ground, whole plant samples were collected and weighed from the center 2 beds of each 4 bed plot. The samples were then dried, weighed and analyzed for total N content to determine crop N uptake in terms of lb/ac. For the romaine plots, crop yield was assessed by harvesting, trimming and weighing individual heads from 10 ft of row from the middle two beds of each 4 bed plot. A marketable head was taken to be a heart that weighed more than 0.35 lb. For the broccoli plots, crop yield was determined in the same manner except heads were harvested from 7.5 ft of row and a marketable head was considered to be a crown that weighed more than 0.30 lb. An ANOVA was performed using SAS (SAS Institute Inc., 2010) to determine statistically significant differences between treatment means for each experiment. In the analysis, data were blocked by the 3 experimental factors and by replication. Point Injection and Knife Blade Applicator Comparison Results Prior to the first side dress, midrib nitrate-n levels were roughly 3,000 ppm and uniform across all treatments (data not shown). Fertilizer applicator type was found to have a significant effect on midrib nitrate-n levels (Fig. 2a,b,c). At the 25% rate of applied N, use of the point injection system resulted in midrib nitrate-n levels that were significantly higher after the first side-dress (29%), after the second side-dress (23%) and at maturity (29%) as compared to the knife blade applicator. At the 50% rate of applied N, use of the point injection system resulted in midrate nitrate levels that were 23% higher after the first side-dress. The results also showed that when the point injection system was used to apply 25% or more of the standard rate of N, midrib nitrate-n levels at all sample dates were equivalent to or higher than those found when the knife blade applicator was used to apply 50%, 75% or 100% of the standard rate of N. Differences between fertilizer applicator types were not found when applied N rate exceeded 75% of the standard rate. Very similar results were found for total plant N uptake at maturity (Fig. 2d). At the 25% and 50% rates of applied N, total N in above ground plant material for the point injection system were 41% and 20% higher respectively as compared to the knife blade applicator. Also, when the point injection was used to apply 25% or more of the standard rate of N, total plant N values were equivalent to or higher than those applied by the knife blade applicator at any rate, including the full 100% rate. Fertilizer applicator type also had a significant effect on yield parameters (Fig. 2e,f,g). At the 25% and 50% rate of applied N, use of the point injection system resulted in significantly higher head weight (>8%), total yield (>10%) and marketable yield (>19%) as compared to the knife blade applicator. The data also show that when the point injection ASABE Annual International Meeting Page 3

system was used to apply 50% of the standard rate of N, yield parameters were equivalent to, or higher than when 75% or the full rate of N was applied with either applicator type. At applied N rates exceeding 75%, there were no differences between applicator types. Interpretation of these results indicates that only 75% of the standard rate of N was required to raise the crops, due presumably to excessive amounts of N applied for the relatively low yielding crops. It is appropriate then that results at the 75% and lower rates of applied N be used to compare differences in fertilizer applicator type. If one compares the point injection system at the 50% rate with the knife blade system at the 75% rate, the point injection system had midrib nitrate-n levels, total plant N and crop yield parameters that were equivalent to or higher than the knife blade applicator. Consequently, these results suggest that fertilizer application rates can be reduced by at least 25% (50 lb/ac) without negatively affecting fertilizer uptake levels or crop yield. Point Injection, Knife Blade Applicator and Cultivator Technique Comparison In the romaine lettuce trial, applicator type or cultivation method had no significant effect on midrib nitrate-n levels or N uptake (Fig. 3a,b,c). Also, although there was a trend for midrib nitrate-n levels to increase with increasing levels of applied N, there were few significant differences between treatments at any applied rate of N other than the control. This result is not logical and may be the result of the unusually high variability in the data. At the 75% and 100% applied rates of N, mean head weight was generally higher by more than 10% for the point injection - shaved treatment (Fig. 3d). This translated into higher total yields (>12%) and marketable yields (>17%) as compared to the knife blade shaved treatment (Fig. 3e). Similar total and marketable yield increases of >13% and >11% respectively were found as compared to the point injection-non-shaved treatment (Fig. 3f). Marketable yields for the point injection shaved treatment were also higher at the 50% level of applied N as compared to the other two treatments. These results were unexpected as it was hypothesized that shaving the bed side walls would prune roots, induce soil disturbance and consequently hinder crop growth and yield. A possible explanation for these results was that romaine was grown with 3 crop rows per bed. Yield parameters for the knife blade shaved treatment were low due to root pruning on the outer two rows and poor fertilizer placement for the center row. In comparison, although the roots of the outer two rows were also pruned in the point injection shaved treatment, fertilizer was more optimally placed for the center row. A possible reason this hypothesized improved placement of fertilizer placement did not result in improved midrib nitrate-n levels was the sampling method used. Here, an equal number of samples were taken from each crop row, so only 1/3 rd of the samples were from the center row. As a consequence, no significant differences in N uptake were found. The point injection non-shaved treatments had low yield parameters because without bed shaving, fertilizer was placed too far from the center row to be effectively utilized. In future studies, outer and center row plant samples should be analyzed separately to correctly determine the effects of fertilizer placement, root pruning and soil disturbance on nutrient uptake. In the broccoli trial, applicator type had a significant effect on midrib nitrate-n levels after the first side-dress (Fig. 4a). Use of the point injection applicator with or without bed shaving resulted in midrib nitrate-n levels that were higher than the knife blade shaved treatment at the 50% rate (>27%), 75% rate (>36%), and 100% rate (>60%) of applied N. No significant differences between the point injection shaved and non-shaved treatments were found. A logical explanation for why these differences did not also translate into higher midrib nitrate-n levels at maturity or total N uptake by the crop could not be formulated (Fig. 4b,c). Perhaps excess fertilizer was accidently applied during the post side-dressing fertigation operation where experimental protocol was to apply 80 lb/ac of N. This hypothesis is supported by the result that N levels at maturity for the control treatment (0% applied N rate) were generally equivalent to treatments where the full complement of fertilizer was applied. Although there were few statistically significant differences in yield parameters (Fig. 4d,e,f), there was a trend that root pruning and soil disturbance results in lower yields. This is supported by the result that at the 50% rate of applied N, the point injection non-shaved treatment had numerically higher total yields (>8%) and marketable yields (>15%) as compared to either shaved treatment. Similar results were found at the 75% rate of applied N where total and marketable yields were numerically higher by >9% and >16% respectively. Conclusions Results of the study showed that applicator type, fertilizer placement and cultivation technique had significant effects on nutrient uptake and crop yield in raised bed lettuce and broccoli production. In iceberg lettuce trials, nutrient uptake was generally improved by more than 20% when fertilizer was placed in the root zone (point injection applicator) as compared to at its edge (knife blade applicator). Improved nutrient uptake translated into higher head weights (>8%), total yield (>10%) and marketable yield (>19%). The trials also showed that applied N rates could be reduced by at least 50 lb/ac through use of the point injection system without negatively affecting crop yield. Additional trials at the field scale level are needed to confirm this result. In trials with romaine lettuce, use of the point injection applicator did not improve nutrient uptake as compared to the conventional knife blade applicator. This result conflicts with the findings of the iceberg lettuce experiment. Although the obvious significant difference between the two trials was that iceberg lettuce was planted with two crop rows per bed while romaine was planted with three crop rows per bed, a logical explanation for this finding could not be formulated. ASABE Annual International Meeting Page 4

Although not statistically significant, increases in head weight (>10%), total crop yield (>12%) and marketable yield (>11%) were found when the point injection applicator was used in conjunction with the standard cultivation technique where bed walls were shaved, but not when beds were left intact. These results were not expected as it was hypothesized that shaving bed side walls prunes plant roots and consequently retards nutrient uptake, crop growth and yield. A possible explanation for this was that since injection points were relatively short compared to the width of the bed top, the point injection applicator placed fertilizer more equidistantly between the outer and center rows when bed sidewalls were shaved. Further research utilizing the point injection applicator with different length injection points is needed to better understand how fertilizer placement affects nutrient uptake and crop yield in crops planted with 3 rows on each bed top. In broccoli, use of the point injection applicator in conjunction with cultivation techniques that either shaved the sidewall or left it intact increased nitrogen uptake by more than 27% as compared to the standard applicator. This was true for all applied rates of nitrogen examined. This finding indicates that for efficient N uptake in broccoli, fertilizer placement is more important than cultivation method. Although these differences did not result in higher levels of plant N at maturity, total yield and marketable yield were numerically higher by >8% and >15% respectively when the beds were not shaved and the point injection system was used. Combined, these results imply that fertilizer applicator and cultivation systems that place fertilizer in the root zone and induce minimal soil disturbance improve nutrient use efficiency and/or increase yield in lettuce and broccoli production. Additional trials are needed to confirm this finding. References Randall, G.W., Iragavarapu, T.K. & B.R. Bock. 1997. Corn responds best to spokewheel injected UAN in ridge-till trials. The Fluid J. Online 19(4): 16-19. SAS Institute Inc. (2010). SAS Enterprise Guide. Ver. 4.3. Cary, NC: SAS Institute, Inc. Siemens, M.C., Nolte, K.D. & Gayler, R.R. Improving lettuce production through utilization of spike wheel liquid injections systems. ASABE Paper No. 1111245. St. Joseph, Mich.: ASABE. Stevens, W.B., Blaylock, A.D., Krall, J.M., Hopkins, B.G. & Ellsworth, J.W. 2007. Sugarbeet yield and nitrogen use efficiency with preplant broadcast, banded or point-injected nitrogen application. Agron. J. 99(5): 1252-1259. Timmons, D.R. & Baker, J.L. 1992. Fertilizer management effect on recovery of labeled nitrogen by continuous no-till. Agron. J. 84(3): 490-496. Acknowledgements The Arizona Department of Agriculture, Agricultural Consultation and Training provided partial project funding using Specialty Crop Block Grant funds provided by the USDA, Agricultural Marketing Service. The views or findings presented are the authors and do not necessarily represent those of the Arizona Department of Agriculture, the State of Arizona or the USDA. The authors thank them for their support. ASABE Annual International Meeting Page 5

Fig. 1. Two bed knife blade applicator operating in iceberg lettuce (upper left) and close up view of bed after application. Two bed point injection fertilizer applicator operating in iceberg lettuce (upper right) and close up view of bed after point injection (lower right). ASABE Annual International Meeting Page 6

Fig. 2. Effect of fertilizer applicator type at various application rates of N on iceberg lettuce (a) midrib nitrate-n content after first side-dress, (b) midrib nitrate-n content after second side-dress, (c) midrib nitrate-n content at maturity, (d) total N uptake by crop at maturity, (e) head weight, (f) total yield and (g) marketable yield in trials conducted at the Yuma Agricultural Center, Yuma, AZ in 2011, 2012 and 2013. Applicator types were a point injection system (Point Injection), and a conventional knife blade applicator (Knife Blade). Standard rate of N was 200 lb/ac, split equally between the first side-dress and the second side-dress operations. ASABE Annual International Meeting Page 7

Fig. 3. Effect of fertilizer applicator type and cultivation method at various application rates of N on romaine lettuce (a) midrib nitrate-n content after first side-dress, (b) midrib nitrate-n content at maturity, (c) total N uptake by crop at maturity, (d) head weight, (e) total yield and (f) marketable yield in trials conducted at the Yuma Agricultural Center, Yuma, AZ in 2015. Applicator types were a conventional knife blade applicator (Knife) and a point injection system (PI). Cultivation methods were conventional cultivator equipped with side knives that shaved bed sidewalls (Shaved), and a conventional cultivator with side knives removed (Non-Shaved). Standard rate of N was 100 lb/ac, applied during the first side-dress operation. ASABE Annual International Meeting Page 8

Fig. 4. Effect of fertilizer applicator type and cultivation method at various application rates of N on broccoli (a) midrib nitrate-n content after first side-dress, (b) midrib nitrate-n content at maturity, (c) total N uptake by crop at maturity, (d) head weight, (e) total yield and (f) marketable yield in trials conducted at the Yuma Agricultural Center, Yuma, AZ in 2014. Applicator types were a conventional knife blade applicator (Knife) and a point injection system (PI). Cultivation methods were conventional cultivator equipped with side knives that shaved bed sidewalls (Shaved), and a conventional cultivator with side knives removed (Non-Shaved). Standard rate of N was 100 lb/ac, applied during the first side-dress. ASABE Annual International Meeting Page 9