Previous legume crop influences winter barley yield, N fertilizer response, and malting quality in Missouri

The rapid growth in the craft brewing industry is increasing the demand for winter barley production. Little recent research exists on producing winter barley for malting in the lower Midwestern United States, with a lack of understanding how previous legumes crops grown in rotation with winter barley interact with nitrogen (N) fertilization rates to affect grain yield and malting quality. To investigate these topics, a preliminary, 1‐year study was conducted in Missouri, USA, to test how growing soybean, a common rotational crop, and sunn hemp, a specialty forage crop, prior to winter barley, as well as two fall N fertilization rates (22.4 and 44.8 kg N ha−1), affected barley grain yield. Barley grain yield was unaffected by the previous legume crop choice or the fall N fertilization rate, suggesting that either crop provided adequate residual N to the succeeding barley crop. Additionally, malting quality was analyzed from representative grain samples to reveal adequate levels suitable for use in the craft brewing industry. This study's results reveal preliminary data supporting winter barley production in the lower Midwestern United States, with an indication that previously grown legume crops impact crop yield greater than fall N fertilizer rate.


| INTRODUCTION
Winter barley (Hordeum vulgare) is cultivated on approximately 1.2 million hectares per year in the United States, contributing about 4% of the world's barley production (McMullen et al., 2012).The United States ranks seventh in barley production, and overall, only 22% of US barley is used for feed, whereas about 70% is malted for beer fermentation, resulting in nearly 26 million barrels of craft beer was produced in 2018, an increase of 68% from 2013 (USDA-AMRC, 2018).Therefore, domestic barley production in the United States is critically important to sustain the malting and brewing industry.
In Missouri, there has not been any record of winter barley acreage, yield, and production from 2012 to 2019 (USDA- NASS, 2015NASS, , 2020)).Recent research describing winter barley production in Missouri is absent, as winter cereal grains shifted away from barley production following the dwindling of cow-calf operations and shifted toward winter wheat production, where effort in breeding increased disease resistance and standability in wheat varieties has resulted in an almost complete absence of barley production in Missouri and the lower Midwestern states.Recent demands for local malting and craft brewing have encouraged a renewed interest across the United States in breeding efforts and revisiting how winter barley may fit into current cropping rotations and local agricultural economies.Moreover, craft brewers have preference to locally sourced ingredients that include malting barley grain (Hmielowski, 2017).This creates a renewed opportunity for Missouri malt for craft breweries.One of the earliest tests conducted at the University of Missouri, Columbia, dates between 1978 and1985 when five varieties adapted to Missouri were tested for yields in three counties (Schaffer et al., 1983).This report provides a summary of barley performance test yield ranging from 2004 to 6517 kg ha À1 across years, varieties, and counties.
Those tests reported the incidences of major insect pests including armyworms, aphids, grasshoppers, and several seedling and foliar diseases associated with warm and humid conditions-typical of Missouri.
Modern barley cultivars are reporting greater and more consistent grain yield, as well as disease resistance, with the national average greater than 4708 kg ha À1 (Lazicki et al., 2016).Modern spring barley variety cv."Clearwater" has reported yields of 3677 kg ha À1 in Idaho and cv."Bentley" yielded up to 6235 kg ha À1 in Alberta (Bregitzer et al., 2008;Juskiw et al., 2009), with major barley regions for malt production in the US Pacific Northwest and Northern Rockies and Plains, and US Mid-Atlantic, although there is little winter barley research reported from the lower Midwestern United States (Shrestha & Lindsey, 2019).Likewise, a modern cultivar of winter barley ("Dan"-winter hulless barley) in the Eastern, mid-Atlantic has À1 (Brooks et al., 2011).
Maximum crop recommendations can go up to 135 lb.N acre À1 in a growing season, but exact recommendations vary by growing region.This creates an opportunity for investigating how to achieve through supplemental N fertilization, but not push the barley grain protein above desired malting levels.When managing winter barley for malting quality, optimum digestion may be achieved when barley grain contains at or less than 12% protein.When protein content is between 10% and 12% protein, the optimal ratio of enzymes to starch is achieved to efficiently convert starch to sugars, thus creating a high-quality malt extract (Verhoeven et al., 2019).Therefore, it is commonly recommended to split-apply N fertilizer in the fall and spring to not waste an oversupply of fall-applied N and not create an overabundance of N which the barley plant uptakes and subsequently raises grain protein levels in the spring.Kernel plumpness, which is a good indicator of potential barley malt extract, was found to decrease with increased N fertilizer application (Sainju et al., 2013).In one study, the plump kernel proportion fell below acceptable levels when available N in the soil exceeded 135 kg N ha À1 (Stark & Brown, 1987).
Barley production all but disappeared from most of the lower and central Midwestern United States with the arrival of more profitable cropping rotations, mainly the ubiquitous corn (Zea mays [L.]) and soybean (Glycine max [L.]) summer cropping rotation accepted since the mid-twentieth century (Karlen et al., 2006), with Missouri no exception.Soybean is a grain legume with potential to fix up to 300 kg N ha À1 and up to 95% of its plant-N content through a process called Symbiotic N Fixation (SNF), in which N 2 gas is converted to a usable organic-N form through the mutualistic relationship with Bradyrhizobium japonicum using nitrogenase contained within nodules which form on the root system (Bohlool et al., 1992;Keyser & Li, 1992;Peoples & Craswell, 1992).Very little N fertilizer is applied to soybean fields in the corn-soybean rotation, but it may be expected that as a legume plant, soybean could provide some residual N, likely in organic-N form, for succeeding crops grown in the same field.
Another crop that is being considered for use in Missouri to supplement warm-season grazing forages is sunn hemp (Crotalaria juncea L.), which is a tropical legume reported to produce total nitrogen ranging from 135 to 285 kg N ha À1 (Mansoer et al., 1997;Schomberg et al., 2007).Production value for this legume has mostly focused on green manuring and, more recently, livestock forage (Lepcha et al., 2019;Mansoer et al., 1997;Rotar & Joy, 1983).As only a small percentage of atmospherically-fixed N is removed with harvesting aboveground shoots, there is a potential benefit to grass crops following sunn hemp harvest (Jaramillo et al., 2020).
The effects of differing N fertilizer rates and their influence on yield and barley malting quality in the fall application to winter barley are unknown in Missouri.Both soybean and sunn hemp may serve as valuable summer crops prior to barley production to supplement the nutritional needs of the winter barley as residual organic-N mineralizes to plant-available N to feed the barley throughout the season.
Unfortunately, the lack of dependable quantitative data of actual N contribution at the field level discourages farmers from considering residual N fertility made available through SNF when calculating N fertilizer applications (Bohlool et al., 1992).Previous investigations demonstrated small but appreciable improvements in cereal crop yield following a legume crop, within the 0.5 to 3 t ha À1 range, and it is known that soybean has the potential to increase soil-N balance in some systems (Peoples & Craswell, 1992).It is plausible that the slow mineralization of plant-N from the soybean and sunn hemp residue may be easier to manage grain yields that do not increase grain protein above desirable levels for malting (Peoples & Craswell, 1992), and growing barley following a legume crop in an extended crop rotation is recommended for the benefit of residual N (McFarland et al., 2014).
Previous studies from Croatia (Stipesevic et al., 2007) and Canada (O'Donovan et al., 2014(O'Donovan et al., , 2017;;Williams et al., 2014), which investigated the interaction of previous legume crops and N fertilizer rates on barley grain yield and malting quality, have reported mixed results, with lower N rate effects being overshadowed by the residual N from the previous crops, allowing for reduced N fertilizer inputs.
Therefore, a 1-year study was designed to measure the effects of the preceding summer legume crops, soybean, and sunn hemp, with the interaction of fall-applied N fertilization rates, on the barley grain yield and malting quality of winter barley in central Missouri.

| Study location and design
This study occurred at the University of Missouri Bradford Research Center near Columbia, MO, in 2016-2017.The study design was a split block design, with two fixed blocks of a summer annual legume crop (sunn hemp and soybean) grown prior to winter barley, with randomized strip plots of N fertilization treatments (22.4 and 44.8 kg N ha À1 ) across the legume blocks in three replications.In the summer of 2016, the field site was disk harrowed then planted in sunn hemp (cv."Tropic Sun") grown for forage in one section of the field and soybean produced for grain planted in a separate section of the field.No N fertilizer was applied.The soil was classified as a Mexico silt loam (fine, smectic, mesic, Aeric Vertic Epiaquafs).The sunn hemp was drilled early June 2016 in 0.178 m rows at a rate of 22.5 kg ha À1 .Soybeans were planted shortly after the sunn hemp in .76in.row spacing for a target population of approximately 140,000 plants acre À1 .Sunn hemp harvest occurred in late September, by cutting, baling, and removing all aboveground biomass.Soybean grain was harvested at maturity.Following harvest of both crops, the field was tilled using a power cultivator and disk harrowed, and then winter barley (cv."Violetta") was drilled into the field October 3, 2016 at approximately 67 kg seed ha À1 with 0.178 m row spacing.Violetta is described as having a winter growth habit and is a hulled, two-row barley variety developed for European production and introduced to US markets.
Representative soil samples were collected from the two blocks of the field in which sunn hemp and soybeans were previously grown, heretofore referred to as blocks, by collecting a composite soil sample collected in a z-pattern within each legume block from the 0-153 mm depth on October 14, 2016.The sample analysis, including total N (ppm), is reported in Table 1.
Immediately following soil sample collection, the fall applications of 22.4 and 44.8 kg N ha À1 in the form of urea (46% N) were hand spread using a disk spreader in three strip plots for each assigned N fertilizer rate at October 14, 2016.Each N fertilizer subplot measured

| Statistical methods
Yield of winter barley grain analysis of variance (ANOVA) was conducted using PROC MIXED in SAS 9.4 (SAS Institute, 2013) with legume crop block as a fixed variable and N treatment as a random variable.Dependent variable was barley grain yield (kg ha À1 ).Means comparisons were conducted using Fisher's protected LSD and significant differences at p = 0.05 significance levels.Proc CORR was used to test for any relationship between either the legume crop and N treatment to the grain yield with significant correlations identified at p < 0.05.

| Crop yields
Barley grain yield across treatments averaged 2929 kg ha À1 , but neither the previous legume crop grown in rotation nor the level of fall N fertilization affected grain yield (Table 3).Although raw values appear greater when sunn hemp was grown prior to barley (3552 kg ha À1 ) compared to soybean (2449 kg ha À1 ), statistically there was no benefit (p = 0.1958) because of high variability among the plot grain yields with the statistical limitation of no replication in the soybean and sunn hemp treatment area.
Nitrogen fertilization treatments were overlaid across the crop blocks in the field in three replications of two fertilization treatments: 22.4 and 44.8 kg N ha À1 applied in the fall.Grain yield did not differ according to N fertilizer rates across the study (p = 0.2541) or when examined within each crop block of soybean or sunn hemp (p = 0.5018).
Correlation analysis of the variables revealed a strong, positive relationship between the previously grown legume crop to barley grain yield (r = 0.84, p < 0.01).Nitrogen fertilizer rate was not significantly correlated to yield, reflecting the ANOVA previously mentioned.Thus, despite the lack of significant differences between the two legume crops in grain yield, there does appear to be a correlative trend between the choice of legume crop and subsequent barley grain yield.

| Malting quality
One grain sample harvested from each legume block was sent to Hartwick College (Oneonta, New York), for malting quality analysis.The descriptive results of 20 quality characteristics are presented in Fall fertilization treatment Grain yield following preceding legume crop that the digestion enzymes had sufficient starches to convert.No appreciable differences were present in the quality characteristics between the two samples in grain characteristics.Each sample has comparable percent protein but was at or slightly above the 12%-12.5% threshold for optimum enzyme: starch ratio.Slight differences did exist in fermentation characteristics.For example, percent germination energy, 8 ml, decreased considerably in the barley grown following sunn hemp, whereas the color, measured in degrees SRM, and percent soluble protein were slightly higher in the same sample.This concurs with Nedel et al. (1993), who reported a deleterious increase in malting protein in grain produced following pea.This suggests that the sunn hemp's greater N residual, which supplied N to the barley grain was perhaps too high and translated into some reduced malting quality characteristics.The lack of sample duplication detracts from further interpretation but does indicate that grain quality, regardless of the previous legume crop grown, may be suitable for malting.

| DISCUSSION
This 1-year pilot study provides some fascinating, albeit limited data, investigating the effects of two legume crops grown in the summer growing season prior to winter barley on the resulting grain yield and its malting quality in Missouri, USA.This is the first known pilot study of its kinds in Missouri.Despite its limited scope and applicability there are valuable outcomes of this pilot study, which elicit further investigation.
Grain yield was not affected by either legume crop or N rate in this 1-year study, which is surprising because barley is a heavy N user and not suitable for poorer fertility fields.Thus, our results force a rejection of the study's hypothesis that supplemental N, especially during fall seedling emergence or through residual organic-N sources, is helpful in establishing a plant crown that can survive the winter freezing temperatures.This suggests that residual soil-N was adequately available and sufficient to supply the barley crop regardless of preceding legume crop, and thus, the levels of N fertilization were either too close in amount or too low to induce any differences in yield response to N rate.Although N recommendations can go up to 134.5 kg ha À1 for winter barley, recommendations in Michigan for 67 to 90 kg N ha À1 of available N are adequate for optimum yields of >2690 kg ha À1 (McFarland et al., 2014).Our grain yields results are comparable, suggesting adequate soil-N.Thus, any future study should include a replicated control or "zero" N-added to account for the residual N in this silt loam soil.
One study in Croatia helps to explain our data, wherein there was no increase in barley yield in a disk harrowed field following soybean production when applying approximately 30 kg N ha À1 (2752 kg grain ha À1 ) compared to 0 kg N ha À1 applied (2571 kg ha À1 ).Yet, when greater rates of N fertilizer was applied, there was a significant increase in yield of about 26% when increasing N fertilization from 30 to 60 kg N ha À1 (3479 kg ha).Furthermore, the greatest rate of 60 kg N ha À1 was identified as producing the greatest crop profit compared to greater N fertilization rates (Stipesevic et al., 2007).Our Soil samples collected 7 days after drilling the barley provide an explanation for the effects measured by the previous legume crop, as samples were analyzed for soil total N in each crop block.In the block in which soybean was grown total soil-N concentration was 1500 ppm, whereas it was almost double that in the block with sunn hemp grown, which contained 2800 ppm (Table 1).Although pH was slightly higher and kg P ha À1 was almost twofold greater in the sunn hemp block, this would support the potential for yield differences between the two crop blocks.Soybean can result in greater N removal if grown in succeeding years, especially with supplemental N fertilizer, which subdues SNF, as reported in Nebraska, with average N removal of 136 kg ha À1 , much greater than other cropping rotations, which included corn or oats (Varvel & Peterson, 1990).Soybean is an effective scavenging plant for remaining soil-N available from the previous corn crop prior to nodulation, but once nodulation has begun, then adequate N is expected to fix through SNF for soybean grain production.Although, nodulation is greatly determined by environmental factors that can limit nodulation and trigger the uptake of soil-N to meet the soybean's large N requirement.
These results are understandable considering that much of the residual N remaining following soybean grain harvest is contained within the desiccated leaf tissue following the harvest of grain, the dominate plant sink.Little residual N is known to remain in crop residue following soybean production.Indeed, previous studies in the Midwest are mixed in demonstrating the value to supplemental N fertilization to increase soybean grain yield.Yet, there is a wellestablished soybean N credit when grown prior to corn in the Midwest because of increased soil inorganic-N available for corn (Gentry et al., 2001).A 15-year study in Illinois reported about a 20% yield increase (+3 Mg ha À1 ) including soybean in rotation with corn and reduced N 2 O emissions by about 35% (Behnke et al., 2018).Soybeans essentially act as more or less efficient N users but can be an N source or sink according to soil fertility levels and thus may not provide much readily available inorganic-N for the succeeding crop, such as a winter cereal grain.
The evident benefit to growing sunn hemp prior to winter barley is demonstrated in the greater grain yield in this study.Sunn hemp is a tropical legume, which can fix up to 163 kg N ha À1 in aboveground biomass and is known to leave residual N ranging from 58 to 84 kg N ha À1 depending upon the amount of residue left on the field, especially in the well-developed root system (Cherr et al., 2006;Rotar & Joy, 1983).This then suggests that although some or most of the residual N may be in organic form, unavailable to barley plant uptake, there is a greater total N pool within the top 0.15 m sampled soil depth for uptake by the barley.Sunn hemp has a potentially untapped value in Missouri and across the Midwest for use as a high protein, annual, warm-season forage to livestock (Lepcha et al., 2019).
Interpreting the malting quality is limited as only single samples representing each preceding legume crop were analyzed.The analyzed samples do reflect barley, which is suitable for malting quality.
Malting quality is known to decrease at high N fertilizer rates with the increase in grain protein.A previous study from the Western Plains of Canada reported no effects of growing faba bean, field pea, or lentil preceding spring barley on the grain malt quality, contrasting with mostly deleterious malting quality impacts by applying N fertilizer from rates of 0-120 kg N ha À1 (O' Donovan et al., 2017).Sample analysis in this present study did not include any of the N fertilizer plots following no statistical difference by grain yield, but perhaps further investigation would have revealed malting quality differences by N rate that were not reflected in the grain yield.Therefore, it is suggested that future studies more thoroughly evaluate the preliminary findings of this study to understand how the production of legume crops and specifically a rotational crop as important to Midwestern US agriculture as soybean, as well as alternatives such as sunn hemp, a forage crop, affect malting quality.
This research holds potential value for the craft brewing industry as this study's findings suggest it is possible to produce above-average grain yields in Missouri, USA, granting the limited scope of this study.
As well, this study elicits the need to further investigate if growing summer annual legume crops, either for grain (soybean) or forage (sunn hemp), can add value to the cropping rotation system in reducing the need to intensively manage N fertilizer applications for optimum barley grain yield.Indeed, after further study, sunn hemp may demonstrate its ability to supply greater residual N supply in the soil for the following winter barley crop, compared to soybean, which is the typical grain legume grown in the Midwest United States.

| CONCLUSION
This 1-year study examined the effects of producing two legumes, soybean, and sunn hemp, prior to the production of winter barley in 3.6 Â 47.7 m in the soybean block and 3.6 Â 82.0 m in the sunn hemp blockdichlorophenoxyacetic acid (2,4-D) controlled winter broadleaf weeds.Then during green up of the winter barley, a foliar application of N fertilizer (28% N) at a rate of 22.4 kg N ha À1 was distributed over all plots on March 6, 2017.Additionally, pyraclostrobin fungicide was sprayed at labeled rate to control and protect from Fusarium head blight (Fusarium graminearum), or smut, on April 24, 2017.Weather data were automatically collected by a weather station located approximately 0.4 km from the field site.Temperature and rainfall data from October 2016 through June 2017 was accessed and downloaded in July 2017.A summary of the weather data for the winter barley growing season as well as the 30-year averages for the research site are presented in grain yields were comparable to those in the Croatian study, with no N response between our two N fertilizer rates, suggesting that future research should include greater N fertilizer application to truly investigate fertilizer response in winter malting barley to elucidate the effects of the likely dominant residual N, which shadowed any fertilizer impact in the Croatian study as in this pilot study.Previous work in Canada on spring barley reported positive benefits in carryover N from faba bean (Vicia faba L.), lupin (Lupinus angustifolius L.), and pea (Pisum sativum L.) crops to support barley yield in the subsequent crop year without the addition of fertilizer (Williams et al., 2014).A separate Canadian article reported faba bean, field pea, and lentil (Lens culinaris Medik.)increasing barley grain yield by 6%-14% 2 years following the legume crops and consistently responded to N fertilizer with maximum yield between 90 and 120 kg N ha À1 (O'Donovan et al., 2014).These comparative studies further support the need for further research in the lower Midwest United States into the impacts of and possible rotational benefits of previous legume crops in the ability to trade off applied N fertilizer on winter barley.Despite no significant difference with N treatment present in either legume crop block in our study, in part explained by low degrees of freedom and high standard errors in our data to explain the error in the statistical model, the mean values within each crop do suggest a possible separation of the residual impacts of soybean and sunn hemp that demands future research.
Missouri, USA.The rapid growth in the craft brewing industry is increasing the demand for locally grown winter barley production, to which this study's preliminary results may indicate that adequate residual soil-N can be supplied through legume crop production to service the barley production without further supplemental application in the fall months after seeding, indicated by the lack of N fertilizer response in this study.Additionally, either legume crop resulted in similar grain yields, despite almost double residual soil-N concentration at barley drilling in the sunn hemp block compared to the soybean block.Additionally, malting quality was analyzed from representative grain samples to reveal adequate levels suitable for use in the craft brewing industry.This study's results reveal preliminary data supporting winter barley production produced in rotation with legume crops in the lower Midwestern United States, with the need for expanded investigation to verify these study results.

Table 2 .
Weather data for the 2016-2017 growing season and 30-year average near Columbia, MO 2.2 | Data collectionGrain was at physiological maturity by late May, and in early June, a Wintersteiger plot combine (Wintersteiger Ag, Austria) was used to harvest the strip plots into sample bags, and then samples were processed through a sample grain cleaner to remove the remaining chaff.Whole plot samples were weighed, and subsamples tested for percentage grain moisture using a DICKEY-john Grain Analysis Computer 2100b (DICKEY-john, Auburn, Illinois).Grain yield was calculated as kg ha À1 adjusted to 14.5% standard moisture.Two representative grain samples, one from the soybean block and one from the sunn hemp block, were analyzed for full malting quality by the Hartwick College Center for Craft Food & Beverage T A B L E 1 Soil sample analysis following the harvest of soybeans and sunn hemp taken from the 0-153 mm depth(Hartwick College, Oneonta, New York).This analysis includes 20 quality factors for a full panel of malting quality characteristics, which were then used as descriptive analysis to evaluate the crop block treatments to grain malting, including moisture, assortment, friability, fine extract, coarse extract, F/C difference, β-glucan, FAN, soluble protein, S/T, DP, α-amylase, color, filtration time, and clarity.