Irrigated Winter Malting Barley Cultivar Performance in Cold Desert and Cold Semiarid Environments

Abstract

Growers in the grain-producing continental cold desert and cold semiarid regions are interested in the local adaptation of winter malting barley (Hordeum vulgare) as a potential alternative crop to winter wheat (Triticum aestivum). Variety selection for specific environments is a critical first step in producing high yields of winter malting barley at the same production costs. Twenty-two winter malting barley entries were planted under irrigation in randomized complete blocks at New Mexico State University’s Agricultural Science Center at Farmington (cold desert; 3 replicates) and Rex E. Kirksey Agricultural Science Center at Tucumcari (cold semiarid; 4 replicates) in September 2023 and harvested for grain in July 2024. All entries at Tucumcari were heavily grazed by wildlife over winter, which may have influenced grain production of some varieties, although there was no site × cultivar interaction for grain yield, which ranged from 2558 to 4157 kg ha⁻¹. Irrigation and N fertilization differences between sites likely influenced (p < 0.0001) grain yield and grain protein (4421 and 2172 kg grain yield ha⁻¹ at Farmington and Tucumcari, respectively; 109 and 93 g grain protein kg⁻¹ at Farmington and Tucumcari, respectively). Future research in cold desert and semiarid regions should evaluate cultivar differences regarding irrigation and nutrient management.

1. Introduction

Barley production has taken place in the Middle East since about 8000 or 10,000 BC [1,2]. Interest in malting barley production is increasing across the USA due to a growing brewery industry [3,4,5,6] and globally [7,8]. This has also resulted in a need to produce grain ingredients for the industry, including malting barley (Hordeum vulgare) to provide local ingredient sources [9] with consistent quality traits [2]. The cool, dry winters of continental cold desert and cold semiarid environments [9] should be conducive to minimize infestations of Fusarium (Fusarium graminearum) head blight (FHB) in barley [10], which produces mycotoxins that are unhealthy for humans and livestock [4,10,11]. Production of spring cultivars has decreased in many areas due to increased pressure by FHB [12]. cultivar selection and fungicide use can be effective in mitigating FHB [4]. Winter cultivars, which are planted in autumn and harvested in late spring or early summer [9] (June–July in the northern hemisphere), have multiple advantages over spring types besides adaptation to environments where FHB risk is minimized, namely, greater yield [12] and earlier maturation that reduces any irrigation requirement [8]. Winter malting barley cultivars have been developed for desert and semiarid rainfed conditions [9] due to a global effort to improve malting qualities of winter types using beneficial traits of spring types [2].

Grain producers in eastern NM, west Texas, and the Intermountain regions of the USA and other arid and semiarid regions with less favorable soil conditions are interested in the local adaptation of winter malting barley as a potential alternative crop to winter wheat (Triticum aestivum) [9]. Including barley in crop rotations can increase crop diversity and economic returns [5,13,14]; however, management for high grain yield and quality, including grain protein, are necessary [2,9,10]. If those standards for malting are not met, the grain is equally suitable to non-malting types for livestock feeding [12,15].

Of particular interest in semiarid regions globally, including eastern NM and west Texas, in the US Southern High Plains, is the use of small grains as dual-purpose crops [13,14,16], particularly winter wheat and triticale [×Triticosecale Wittm. ex A. Camus (Secale × Triticum)], for grazing and grain during the winter period when forage is often limiting [14,16] and grazing is less expensive than feeding hay or other stored feeds [14,17,18,19,20,21]. Grain yields are minimally reduced when grazing is ceased before stem elongation at first hollow stem [14]. Barley can produce greater forage biomass than winter wheat and triticale in late winter and early spring in these regions as an earlier maturing crop [8,17,20,22]. It also provides nutritive value greater than required by grazing livestock [17,19] and is preferred over those species by livestock (Lauriault personal observation). Intensive but controlled grazing can also reduce infestations of foliar diseases in barley [16]. Barley also has a greater harvest index for grain than winter wheat or cereal rye (Secale cereale) [6].

The objective of the present study was to evaluate the performance of winter malting barley cultivars under irrigation in two diverse environments (cold desert and cold semiarid) with a continental precipitation pattern.

2. Materials and Methods

2.1. Site Descriptions

2.1.1. Farmington

This study was conducted during the 2023–2024 winter growing season at the New Mexico State University Agricultural Science Center in Farmington, NM, USA (36.39°, −108.31°; elev. 1790 m asl). The soil was Shiprock (coarse–loamy, mixed, superactive, mesic Typic Haplargids) fine sandy loam. The climate in the region is Köppen–Geiger cold desert (BWk: http://www.cec.org/north-americanenvironmental-atlas/climate-zones-of-north-america/, accessed on 22 May 2023). Weather data were collected from a National Weather Service cooperative station located within 0.9 km of the study area (

Table 1. Monthly mean temperature and total precipitation during the study period and the long-term averages and irrigation applied during the study period at Farmington and Tucumcari, NM, USA.

Site	Period	Aug.	Sep.	Oct.	Nov.	Dec.	Jan.	Feb.	Mar.	Apr.	May	June	July	Average/Total
Temperature, °C
Farmington	2023–2024	23.3	19.4	12.8	6.1	1.1	0.0	3.9	6.7	10.6	14.4	22.8	24.4	12.2
	1969–2023	23.3	19.4	12.2	5.0	0.0	−1.1	2.2	6.7	10.6	16.1	21.7	24.4	11.7
Tucumcari	2023–2024	27.8	24.4	16.7	10.0	6.7	2.2	8.9	10.6	15.6	20.6	27.2	27.2	16.7
	1905–2023	25.0	21.7	15.0	8.3	3.9	3.3	5.6	9.4	14.4	18.9	24.4	26.1	15.0
Precipitation, mm
Farmington	2023–2024	12	10	1	8	14	18	6	22	7	0	36	22	156
	1969–2023	25	27	23	16	13	13	12	16	14	14	7	21	200
Tucumcari	2023–2024	53	18	5	2	25	15	13	8	9	29	48	47	273
	1905–2023	68	39	34	17	16	10	12	19	28	47	47	67	398
Irrigation, mm
Farmington	2023–2024	0	33	76	0	0	0	0	0	124	239	39	0	511
Tucumcari	2023–2024	0	89	75	50	76	57	0	0	51	0	51	0	449

). Irrigation was applied through a center pivot system with canal water as the source.

2.1.2. Tucumcari

This study was also conducted during the 2023–2024 winter growing season at the New Mexico State University Rex E. Kirksey Agricultural Science Center in Tucumcari, NM, USA (35.20°, −103.68°; elev. 1247 m asl). The soil was Redona (fine–loamy, mixed, superactive, thermic ustic Calciargids) fine sandy loam. The climate in the region is Köppen–Geiger cold semiarid (BSk; http://www.cec.org/north-americanenvironmental-atlas/climate-zones-of-north-america/, accessed on 22 May 2023). Weather data were collected from a National Weather Service cooperative station located within 0.5 km of the study area (Table 1). Irrigation was applied through a center pivot system with treated municipal wastewater as the source. The entire test area was continuously grazed to nearly ground level by Canada geese (Branta canadensis) from about mid-December through late February when northern migration began.

2.2. Test Management

2.2.1. Farmington

The test was planted into three randomized complete blocks. Cultivars were planted on 21 September 2023 into a prepared flat seedbed with plots measuring 1.1 × 6.1 m using a disk drill fitted with a cone seed distribution system. The seeding rate for all cultivars was 88.1 L ha⁻¹ on 20 cm row spacing. A 1.5 m unplanted border was left between plots in the planting direction, but no border existed between plots. On 17 October 2023 and 10 April 2024, 67 and 90 kg N ha⁻¹, respectively, were uniformly applied to the test. On 3 February 2024, 3.5 L ha⁻¹ of Pendimethalin (N-(1-ethylpropyl)-3,4-dimethyl-2,6-dinitrobenzenamine; BASF Agricultural Solutions US LLC, Research Triangle Park, NC, USA) was applied for pre-emergent weed control. On 27 February 2024, 1.2 L ha⁻¹ 2,4-D (2,4-Dichlorophenoxyacetic acid, isooctyl ester; Tenkoz Inc., Alpharetta, GA, USA) was applied, and on 3 April, 1.75 L ha⁻¹ 2,4-D (Dimethylamine Salt of 2,4-Dichlorophenoxyacetic acid; Alligare, LLC, Opelika, AL, USA) was applied, both for broadleaf weed control. The entire test area was irrigated with an overhead system as needed. Monthly total irrigation amounts are shown in Table 1.

2.2.2. Tucumcari

The test was planted into four randomized complete blocks. Cultivars were planted on 21 September 2023 into a prepared flat seedbed with plots measuring 1.5 × 6.1 m using a disk drill fitted with a cone seed distribution system. The seeding rate for all cultivars was 70.5 L ha⁻¹ on 15 cm row spacing. A 1.5 m unplanted border was left between plots in the planting direction, but no border existed between plots. No fertilizer or herbicides were applied; however, the study was sprinkler-irrigated as needed with treated municipal wastewater, which applied 47 kg N ha⁻¹, of which about 56% was applied within a month after seeding due to high N levels in the wastewater that returned to minimal levels shortly thereafter.

2.3. Measurements

Heading was observed every other day beginning on 1 April 2024 at both locations and continued until all cultivars had reached 50% heading. On 9 and 12 July 2024, all plots were combined at Farmington and Tucumcari, respectively. At Farmington, a Classic Plus small plot combine was used (Wintersteiger, Ried im Innkreis, Austria), while at Tucumcari, a model SPC40 small plot combine (Almaco, Nevada, IA, USA) was used. Grain was not deawned or dehulled. After combining, grain was weighed and a subsample from each plot at each site was shipped to the Navajo Agricultural Products Industries Lab (Farmington, NM, USA) for grain moisture and grain protein analysis. Grain moisture was used to adjust grain yields to 14.5% moisture content.

2.4. Statistical Analysis

Data were combined across sites and subjected to SAS MIXED version 9.4 [23] procedures to compare the site, cultivar, and site × cultivar interaction effects for heading date, grain yield, and grain protein content. Replicates were defined as unique within site and considered random. When the effects of cultivar or the site × cultivar interaction were significant (p ≤ 0.05), lsmeans were separated by LSD. The LSD value was calculated using the standard error of the difference provided by the PDMIX800 macro [24] for the effect multiplied by the t-value (p ≤ 0.05) returned by an online source (https://goodcalculators.com/student-t-value-calculator/, accessed on 11 November 2024).

3. Results and Discussion

Main effect means and results of statistical analysis are reported in

Table 2. Main effect means and results of statistical analyses for winter malting barley cultivars grown in 2023–2024 at Farmington and Tucumcari, NM, USA. Values are the lsmeans of three replicates at Farmington and four replicates at Tucumcari.

Effect	50% Head Date	Grain Yield 1	Grain Protein
	DOY 2	kg ha−1	g kg−1
Site
Farmington	112	4421	109
Tucumcari	103	2172	93
Cultivar
11ARS652-7	107	3815	102
13ARS526-8	103	3090	99
16ARS622-248	109	3452	89
16ARS627-037	106	2943	98
2MW19_3013-004	107	2983	115
CHARLES	106	3384	93
DH141947	107	2880	99
DH162310	102	2940	111
DH170472	109	2859	99
DH171854	113	3613	90
FLAVIA	110	3712	98
GHRIL0201-103	109	3253	118
GHRIL02JY-077	110	3548	96
GHRIL02OPL-190	109	3165	102
GHRIL02SCL-010	109	3037	103
HIRONDELLA	109	3719	97
LCSVIOLETTA	109	3121	108
MAROUETTA	109	4157	90
W2M001	108	3513	102
W2M002	96	2558	107
W6M003	107	3481	100
WINTMALT	113	3307	100
LSD, 0.05 3	3	749	NS 4
p-values
Site	<0.0001	<0.0001	<0.0001
Cultivar	<0.0001	0.0020	0.1440
Site × Cultivar	0.0126	0.1610	0.0216

¹ Grain yield was adjusted to 14.5% moisture. ² DOY signifies day of year. ³ Cultivar values within a column that differ by the LSD are significantly different at p < 0.05. ⁴ Not significantly different at p < 0.05.

. The site effect was significant for all variables, while cultivar was significant only for 50% heading date and grain yield. The site × cultivar interaction was significant for 50% head date and grain protein. Variables will be reported and generally discussed in the order shown in Table 2.

Regarding the site × cultivar interaction for heading date (

Table 3. The site x cultivar interactions for heading date and grain protein between Farmington (FAR) and Tucumcari (TUC), NM, USA, in 2024.

Cultivar	Heading Date		Grain Protein
	FAR	TUC	FAR	TUC
	DOY 1		g kg−1
11ARS652-7	112	103	123	80
13ARS526-8	110	96	120	78
16ARS622-248	113	105	100	78
16ARS627-037	111	100	108	87
2MW19_3013-004	110	103	130	99
CHARLES	111	100	97	89
DH141947	111	103	103	96
DH162310	108	96	124	97
DH170472	114	105	99	99
DH171854	119	107	89	91
FLAVIA	115	105	115	82
GHRIL0201-103	112	105	133	102
GHRIL02JY-077	116	105	87	106
GHRIL02OPL-190	113	104	111	93
GHRIL02SCL-010	113	105	110	96
HIRONDELLA	114	105	94	101
LCSVIOLETTA	113	104	113	104
MAROUETTA	112	106	88	91
W2M001	111	105	121	82
W2M002	98	93	118	95
W6M003	110	105	106	93
WINTMALT	119	107	105	96
LSD, 0.05 2	4		28

¹ DOY signifies day of year. ² Interaction values within a variable that differ by the LSD are significantly different at p < 0.05.

), all cultivars matured later at Farmington. The greatest gap between sites was within 13ARS526-8, DH162310, and W2M002 in the earlier maturing group, with a 14, 12, and 5 d difference across sites, respectively, and DH171854, Flavia, GHRIL02JY-077, and WINTMALT as the latest maturing group with 12, 10, 11, and 12 d differences across sites, respectively. All other cultivars matured within 5–9 d across sites (Table 3). Cultivar W2M002 matured earliest at both locations, but not significantly earlier than DH162310 or 13ARS526-8 at Tucumcari, which matured earlier than all other cultivars at that location. Cultivars DH171854 and WINTMALT were the latest maturing at both locations, but they were not different from ten other cultivars at Tucumcari. Spunar et al. [12], at a latitude farther north (49 N latitude) than the present sites (about 35.9 N latitude), reported a 4–8 d difference across years in heading date.

Cowger et al. [4] reported a 10 to 13 d range of relative maturities within years and across year among four locally adapted cultivars at a similar latitude (35.7 N) as the present study sites. They [4] also reported a lack of uniformity in maturity within plots, attributing it to temperature effects on different plants within the cultivar. Nonuniformity in heading date was also observed at both sites in the present study, but the differences were between the centers and ends of the plots and could have been induced by soil temperature, moisture, or nutrient availability (none of those measured) as an effect of the unplanted border between plots and lesser irrigation and N applied than at Farmington. In the present study, non-uniformity of heading within plots did not appear to have had a direct influence on grain yield since all cultivars were harvested on the same day, well after grain maturity, with dry down in the field to approximately 8% moisture content, which was consistent among cultivars at both locations. Otherwise, heading date does not influence other traits, which allows for development of cultivars with high yield and malting quality across a spectrum of maturity and climate [1]. That said, heading date is associated with maturity and first hollow stem (Zadoks 29 [14,17,25]), before which grazing should be ceased to avoid yield reductions in dual-purpose small grain species, such as barley [17].

Adjusted grain yield was significant only for the site and cultivar effects (Table 2). Differences among sites are likely attributable to management and environmental differences. In this study, grain yield at Farmington was about twice that at Tucumcari (Table 2). Cultivar differences were expected (Table 2). The greatest yielding cultivar was Marouetta, which was not different from eight other cultivars, while the least yielding cultivar was W2M002, which was not different from 10 other cultivars, and there was no overlap between the greatest and least yielding groups. The lack of any site × cultivar interaction (Table 2) indicates the broadness of adaptation in this region of the cultivars evaluated despite environmental and management differences between sites and that the Tucumcari site was heavily grazed.

Winter malting barley has been studied in more northern semiarid regions of the USA (average 45.7 N latitude) under rainfed conditions in a semiarid, cold Mediterranean climate having warm, dry summers and cold, wet winters and found to have had grain yields ranging from 3369 to 6513 kg ha⁻¹, which was about 90% of yields for barley feed grain [3]. Barley grown under these conditions can equal or exceed irrigated spring barley yield [3,7]. Grain yields at Farmington in this study were intermediate to those measured in various, rainfed, humid continental locations in Ohio, USA (average 40.6 N latitude) [10]. Frischke et al. [17] stated that dual-purpose crops are feasible in regions with higher precipitation that is more uniformly distributed during the growing season, which would also represent fully irrigated production, such as the Farmington site in the present study (Table 2).

It is possible that the intense grazing by wildlife at Tucumcari led to the reduced grain yields [16], particularly for earlier maturing cultivars [20] (13ARS526-8, DH162310, and W2M002), which produced an average of 2862 kg ha⁻¹ (calculated from Table 2). The later maturing cultivars DH171854, Flavia, GHRIL02JY-077, and WINTMALT yielded an average of 3545 kg ha⁻¹, although there was no site × cultivar interaction for grain yield. So, in the present study, grazing may not have influenced the grain yield of the early maturing cultivars, as reported elsewhere for winter wheat, but not for barley, at p < 0.06 [17]. Frischke et al. [17] reported that defoliation delayed maturity of most small grain cultivars tested [16], which may have limited the grazing influence in the present study since the plots were uniformly continuously grazed until late February. They [17] also suggested that yield penalties under grazing may not be as frequent in N limited conditions, such as those at the Tucumcari site, that are frequent under producer management. Latta [20] also reported that little grain yield reduction occurred due to grazing during low-precipitation growing seasons when lesser yields should be anticipated [14]. Otherwise, later-maturing cultivars allow for a longer grazing period before first hollow stem [14,16] while earlier maturing cultivars would allow for double cropping [6], especially where irrigation water is available year-round in cold desert and cold semiarid regions. That said, earlier maturity at higher elevations or latitudes where late freezes can occur can lead to winter damage and yield reductions [2].

Nitrogen application rates influence grain yields [9] without regard to whether the applications were made in the autumn or spring, such that rates above 34 kg N ha⁻¹ with 304 mm winter precipitation or 112 kg N ha⁻¹ with 430 mm winter precipitation led to decreased yields, with no explanation as to the cause. Although lodging has been mentioned as a concern when N is applied [3], that can be mitigated with a growth regulator [12]. Otherwise, winter barley has a lower N requirement to produce grain than winter wheat [6].

The site × cultivar interaction existed for grain protein, which included a strong site effect but no cultivar main effect (Table 2). Habshied et al. [26] and Spunar et al. [12] both reported a year effect in barley grain protein, particularly associated with precipitation amounts, which would be consistent with the strong site effect in the present study because of differences in irrigation levels (Table 1). Otherwise, while most cultivars had significantly equal grain protein between sites, 11ARS652-7, 13ARS526-8, 16ARS622-248, 2MW19_3013-004, DH162310, Flavia, GHRIL0201-103, and W2M001 all had lesser protein at Tucumcari (Table 3). There also were changes in rank such that, while GHRIL02JY-077 was not different across sites, it had the least grain protein at Farmington but the greatest at Tucumcari. The lack of a difference between sites within most of the cultivars indicates a more broad adaptation across cold desert and cold semiarid environments for those cultivars (Table 3).

Other studies reported that grazing [17] or defoliation [20] had little effect on grain protein, except in a few cases where there was a dilution effect with increasing yield. This may have been a factor in grain protein content in the present study at Farmington (Table 3), as indicated by a comparison of the earlier maturing, lower yielding cultivars with the later maturing, higher yielding cultivars that were previously mentioned (Table 2). Lesser grain protein at Tucumcari (Table 2) is likely an indication of N deficiency, although the difference may also be attributable to potentially excessive N at Farmington to be described next.

The nitrogen in barley reduces grain quality [9], making the beer cloudy [1]. Optimum grain protein levels reported in the literature were 95 to 105 g kg⁻¹ [7], 115 to 135 g kg⁻¹ [3,10], and 95 to 125 g kg⁻¹ [9]. Munoz-Amatriain et al. [8] stated that it should be ≤135 g kg⁻¹. Several studies [3,7,10,26] reported barley grain protein in a range of 105 g kg⁻¹ or greater. While all cultivars had <135 g kg⁻¹ grain protein (Table 3), only six of the cultivars at Farmington had grain protein within the optimum level stated by Belcar et al. [7], with five below and 11 exceeding this level (Table 3). At Tucumcari, half of the cultivars contained grain protein within the Belcar et al. [7] optimum, with one (GHRIL02JY-077) potentially exceeding it and 10 below that optimum. Both low and high protein content can have a negative effect during the malting process [7]. Brewers also expect consistent grain protein content for optimum ethanol production [7]. The observation of differences in grain protein at Tucumcari (Table 3) regarding the optimums may indicate differences in N uptake and utilization among cultivars [6] that could be exploited in breeding programs and by producers to reduce inputs for economic and environmental safety benefits. Otherwise, even malting barley containing excessive protein content can be used for non-malted distillation, which is acceptable for grain protein between 115 and 140 g kg⁻¹ [27].

Castro et al. [3] evaluated cultivar and N levels in more northern semiarid Mediterranean rainfed conditions and found that N fertilization increased both grain yield and grain protein differently among cultivars in three site-years. Generally, the application of 34 kg ha⁻¹ with 304 mm winter precipitation or 56 kg ha⁻¹ with 430 mm winter precipitation increased grain protein [3] to exceed the upper optimum level of 105 g kg⁻¹ reported by Belcar et al. [7] but below the 135 g kg⁻¹ stated by Munoz-Amatriain et al. [8]. In a recent review of barley agronomic management, Shrestha and Lindsey [9] reported that N applications > 100 kg ha⁻¹ reduced barley grain malting quality in the US Southwest near the sites of the present study. Barley continues to take up soil N after yield requirements are met, which increases grain protein and decreases grain quality [9]. Nitrogen application timing also affects grain protein with greater protein resulting from later N applications, although fall applications were not as beneficial to grain yield as spring applications [9]. The amount and timing of N fertilization at both locations of the present study (high and low N fertilization at Farmington and Tucumcari, respectively) likely led to the differences observed in grain protein in the present study (Table 3). Hence, growers should recognize the need for appropriate fertilization by cultivar to optimize grain protein content. Grain protein tends to be greater in cold semiarid and cold desert rainfed systems than under irrigation [9], possibly due to yield dilution, which should also be considered in continental cold semiarid rainfed systems.

4. Conclusions

The data from the present study showed a grain yield range of 2558 to 4157 kg ha⁻¹ among cultivars having 78 to 133 g grain protein kg⁻¹ and returned strong site and cultivar effects with some site × cultivar interactions pertaining to irrigation, N fertilization, and grazing, indicating the need for additional research regarding winter malting barley opportunities and management in cold desert and cold semiarid environments with a continental precipitation pattern, particularly as genetic improvement continues. In addition to agronomic management regarding irrigation, nutrient management, and other factors, future research should evaluate cultivar differences in malting quality due to soil and weather conditions in cold desert and cold semiarid regions. In the meantime, since dual-purpose winter cereal production is commonly practiced in these regions globally, those producers are encouraged to follow the same recommendation as for dual-purpose wheat to remove livestock at first hollow stem for the particular cultivar to protect grain production potential and receive the added value and environmental benefits from livestock production.