A First Report on Planting Arrangements for Alfalfa as an Economic Nurse Crop During Kura Clover Establishment

Abstract

Alfalfa (Medicago sativa) persists for several years but must be rotated to another crop before replanting. Kura clover (T. ambiguum M. Bieb) is a perennial legume that can persist indefinitely without replanting; however, establishment is slow, which limits economic returns during the process. Two studies, each with four randomized complete blocks, were planted in two consecutive years at New Mexico State University’s Rex E. Kirksey Agricultural Science Center at Tucumcari, NM, USA, as the first known assessment evaluating alfalfa as an economic nurse crop during kura clover establishment using various kura clover–alfalfa drilled and broadcast planting arrangements. Irrigation termination due to drought limited yield measurements to three years after seeding. In that time, kura clover–alfalfa mixtures generally yielded equally to monoculture alfalfa, except for alternate row planting. After 5 years, the alfalfa stand percentage remained >80%, except for the alternate row treatment (69% stand). Kura clover monocultures attained about 40% stand, and the mixtures had a <25% stand. Alfalfa may persist for more than 5 years before relinquishing dominance to kura clover in mixtures, but the alfalfa would continue to provide economic returns as kura clover continues stand development with minimal production, but develops its root system to maximize production when released from the alfalfa nurse crop.

1. Introduction

Forage legumes are valuable in livestock production and sustainable agriculture due to biological nitrogen fixation [1], from which the N becomes available to associated grasses and increases the crude protein content of the grazed or harvested forage [2]. Alfalfa (Medicago sativa) is the most widely used perennial forage legume globally [3,4,5,6]. Other widely used species in temperate regions include red clover (Trifolium pratense) [6,7], white clover (T. repens) [6], and birdsfoot trefoil (Lotus corniculatus) [7], but these are lower yielding and not as persistent as alfalfa [1,8], although all can be managed to naturally reseed [3] and white clover is stoloniferous [9]. Even alfalfa must be rotated occasionally for stand re-establishment [3].

Kura clover (T. ambiguum M. Bieb) is a widely adapted [6] perennial legume known for its ability to persist under environmental stress [10], except during the seedling stage [1,5,9,11,12,13]. Its deep, branching taproot [14], low crown, and rhizomatous nature allow kura clover to persist indefinitely without replanting [1,5,9,11,12]. Rhizomes can be produced as early as 3 months after planting and can extend out to 1 m, and roots can extend downward to 60 cm in 2 years [6]. Environmental stressors to which kura clover is tolerant include continuous grazing, extreme winter conditions, and many diseases [1], as well as drought and infertile soils [15]. While no herbicides are specifically labeled for kura clover, established kura clover recovers rapidly from broad-spectrum herbicide applications to compete with weeds [15,16]. Perennial forage legumes have two planting seasons, spring or late summer/autumn [17]. Therefore, kura clover and alfalfa can both be safely sown in late summer or autumn in semiarid regions with dry winters to minimize competition with summer weeds.

Kura clover is reported to have similar yield and crude protein content to alfalfa in temperate rainfed environments, but greater digestibility due to the lack of upright stems in growth cycles after the initial spring growth, whereas alfalfa has upright stems for all growth cycles [1]. White clover and kura clover have been shown to be more persistent under two years of continuous grazing from mid-spring through summer than alfalfa, red clover, and birdsfoot trefoil, and maintain >80% stand after two years [14,18]. Kura clover has been evaluated as an alternative to alfalfa for pasture and hay in several environments [1,6], including the semiarid US Southwest [3,19], and found to be second only to alfalfa in productivity once established [1,3,19].

The greatest concern with kura clover, leading to its underutilization [6,9,10,11], is slow establishment [1,8]. This has been associated with what has been called poor seedling vigor [10], which leads to poor competition against weeds and companion species in mixtures and low yields compared to alfalfa and other legumes for 1 to 2 years after seeding [1,11]. Collins et al. [20] reported that daughter plants from the original seeding had lesser root weights, potentially indicating that they still relied on the parent plants for photosynthate. Thus, it could be argued that kura clover is not slow to establish; rather, it is slow to become productive because early stand life is focused on root system development [8], which is its establishment process, after which time it is persistent, extremely competitive, and highly productive, even equaling or exceeding the productivity of alfalfa [3,19].

Seedling survival and competition are critical to the stand establishment of any perennial forage species. The seedling survival of kura clover is also influenced by soil moisture status [21]. When broadcast seeding, conventional tillage is preferable over no-tillage to promote seed-to-soil contact for imbibition and radicle penetration [21], although Taylor and Smith [6] reviewed the literature suggesting that broadcast seeding of kura clover was less desirable than drilling in rows. Because of its slow establishment from seed and because of kura clover’s strong rhizomatous nature, vegetative plantings have been evaluated as a potential alternative to direct seeding, with mixed success and 80% field biomass production in the year after planting [9].

Various strategies have been studied to promote establishment by seeding of other slowly establishing legumes [6,7,11], to provide some economic returns during the establishment period [7,9,17]. While considerable information is available in the scientific literature about using grasses for nurse crops, very little research has been reported about using legume nurse crops for slow-establishing legumes [7]. Ates et al. [7] reported that using annual legumes for nurse crops may not provide a long-enough economic crop life while the preferred species is establishing, or they may be too aggressive and outcompete preferred crop seedlings. Cicek et al. [22] reported that Hungarian vetch (Vicia pannonica Crantz) influenced sainfoin less negatively than triticale (×Triticosecale Wittm. ex A.). Other small grains have also been found to be too competitive with perennial legume seedlings [17,23] while annual legumes are not [17]. Otherwise, the nurse crop should be adapted to the same climatic conditions as the preferred species, be competitive with weeds, and meet the intended purpose of the final stand [7]. Seeding rates can be adjusted to reduce nurse crop competition with the preferred crop when sown in close proximity [6,7,13,17,20,22]. Another strategy is to plant a more rapidly establishing shorter-lived perennial legume as a nurse crop to provide harvestable forage until the preferred species becomes established and productive [7,17,24]. Red clover and birdsfoot trefoil have been evaluated as companion/nurse crops during kura clover establishment in the higher precipitation zones of the eastern US [11,24].

As previously mentioned, kura clover has been evaluated as an alternative to alfalfa for pasture and hay in several environments [3,19] and found to be second only to alfalfa in productivity once established [1]. Guldan et al. [3] found that kura clover crowded out tall fescue (Lolium arundinaceum (Schreb.) S.J. Darbyshire) within 4 years. This may have occurred because the fescue could compete early for surface moisture, but once the kura clover developed its root system, it was able to compete for shallow moisture and access deeper moisture as well as for light in early spring due to its upright stems at that time [1].

Alfalfa is broadly adapted globally [4], but it is not considered short-lived [19], and no literature was found reporting its use as a nurse crop for kura clover establishment or any other forage. Additionally, the literature is scant for evaluations of planting arrangements when legumes are used for nurse crops in the establishment of slow-growing legumes. The objective of this study was to evaluate various kura clover–alfalfa planting arrangements when alfalfa is used as an economic nurse crop during kura clover establishment, with the expectation that the alfalfa would decline over time [3] and yield stand dominance to the kura clover.

2. Materials and Methods

2.1. Site Description

Two identical studies with separate randomizations were planted adjacent to each other in two consecutive years (1999 and 2000) at the New Mexico State University Rex E. Kirksey Agricultural Science Center at Tucumcari, NM, USA (35°12′0.5″ N, 103°41′12.0″ W; elev. 1247 m asl). The soil was quay (fine-silty, mixed, superactive, thermic Ustic Haplocalcids) fine sandy loam. Preplant soil testing revealed pH 8.0, moderate to high preplant soil test levels of salts P, K, and B, low NO₃-N, and very high Na at 550 ppm in the surface 15 cm. The climate in the region is categorized as Köppen–Geiger cold semiarid (BSk; http://www.cec.org/north-americanenvironmental-atlas/climate-zones-of-north-america/, accessed on 22 May 2023), being characterized as semiarid, subtropical, and continental, characterized by cool, dry winters and warm, moist summers. Approximately 83% of the precipitation occurs as intermittent, relatively intense rainfall events from April through October. Weather data were collected from a National Weather Service cooperative station located within 1 km of the study area (

Table 1. Monthly mean air temperatures and total precipitation at Tucumcari, NM, USA, during kura clover establishment with alfalfa as an economic nurse crop, and the long-term (1905–2024) means.

Year		Temperature, °C
Test 1	Test 2	Jan.	Feb.	Mar.	Apr.	May	June	July	Aug.	Sep.	Oct.	Nov.	Dec.	Mean
0	----	6.5	9.4	9.9	13.5	18.0	23.4	26.8	26.2	20.8	15.7	12.1	4.6	15.6
1	0	6.6	9.7	10.7	15.7	22.3	24.6	27.7	28.2	24.1	15.1	5.2	2.3	16.0
2	1	1.6	6.9	8.2	17.0	19.1	24.8	28.8	26.1	22.7	16.1	11.2	5.4	15.7
3	2	4.8	5.2	8.4	16.6	20.8	27.3	26.5	27.1	21.6	13.5	8.1	4.4	15.3
4	3	5.9	5.2	10.9	16.0	21.3	22.7	28.6	26.9	21.4	17.6	9.5	6.3	16.1
5	4	5.5	4.0	12.1	13.5	21.6	25.4	25.3	24.1	22.0	15.3	7.2	6.3	15.6
----	5	6.5	3.9	12.1	13.6	21.6	24.8	26.8	24.7	23.0	15.6	11.3	4.3	15.7
1905–2024		3.5	5.6	9.5	14.2	19.1	24.3	26.3	25.3	21.6	15.2	8.6	4.0	14.8
		Precipitation, mm
0	----	4	1	6	24	47	31	51	34	43	35	25	15	316
1	0	23	16	21	3	14	10	58	20	14	31	14	3	227
2	1	17	24	67	3	72	45	17	72	5	4	40	6	371
3	2	14	4	8	13	17	17	111	18	102	30	38	19	391
4	3	0	16	32	18	48	101	13	108	8	24	23	6	397
5	4	2	15	24	95	7	47	58	70	101	71	58	10	557
----	5	34	28	29	59	60	4	76	113	109	14	0	0	526
1905–2024		10	12	19	28	47	47	67	68	39	34	17	16	398

Temperature and precipitation means are the mean annual temperature and mean total annual precipitation.

).

2.2. Treatments (TRTs) and Planting

Treatments included the following: alfalfa, broadcast (ALFMONOBROAD); alfalfa, drilled (ALFMONOROW); kura clover, broadcast (KCMONOBROAD); kura clover, drilled (KCMONOROW); alfalfa–kura clover, drilled in alternate rows (MIXALTROW); alfalfa–kura clover, broadcast together (MIXBROAD); alfalfa–kura clover, drilled together (MIXROW); alfalfa, broadcast kura clover, drilled (MIXKCROW); and alfalfa, drilled kura clover, broadcast (MIXKCBROAD). Each test had four randomized complete blocks.

2.3. Test Management

Tests were sown in September of their Year 0 into conventionally tilled seedbeds formed into beds on 0.91-m centers for furrow irrigation. Each plot was 1.83 m × 5.18 m (across two 0.91-m furrow beds), all of which were harvested. There was a 1-m skip between plots in the harvesting direction to serve as a border. A disk drill fitted with a seed-metering cone was used for planting drilled treatments in eight 15-cm spaced drill rows. The MIXALTROW treatment was planted by making two passes of four 30-cm spaced drill rows. The drill seeding depth was set to be approximately 1.25 cm using depth bands on each drill unit. Packer wheels followed the drill units to promote seed-to-soil contact. Broadcast treatments were hand-sown and raked-in to improve seed-soil contact prior to any drilled component of that treatment. Planting each year was completed in a single day. The seeding rate for monocultures was 22.45 kg ha⁻¹ [6] and half that rate for each species in mixtures, both using seed coated with the appropriate rhizobium [8].

Irrigations were applied through gated pipe using stored surface water, on the dates as shown in

Table 2. Irrigation and harvest dates and first autumn occurrence of −2.8 °C temperatures for established furrow-irrigated alfalfa–kura clover mixture studies at Tucumcari, NM, USA.

	Month								First −2.8 °C
Test/Harvest Year	Apr.	May	June	July	Aug.	Sep.	Oct.	Nov.
Irrigation dates
Test 1/Year 0						22	1 & 21	15	----
Test 1/Year 1; Test 2/Year 0	18	25		3 & 25	24	15			----
Test 1/Year 2; Test 2/Year 1		21	19	18	17	18			----
Test 1/Year 3; Test 2/Year 2		14	7						----
Test 2/Year 3									----
Harvest dates
Test 1/Year 1; Test 2/Year 0		1	21	20	17	11		2	8 Nov.
Test 1/Year 2; Test 2/Year 1		10 1	11	10	7	12	29		27 Nov.
Test 1/Year 3; Test 2/Year 2		9	4	9	6	18	29		16 Nov.
Test 2/Year 3		27	1	1 & 29		3	31		23 Nov.

¹ No harvest was taken in May for Test 1/Year 1 or Test 2/Year 1 because maturity at this location is delayed in the spring after late summer planting of alfalfa.

. All irrigations were of sufficient duration to completely soak the beds to the center for their full length, which was anticipated to bring the surface 45 cm to field capacity. Historic irrigation records indicated that approximately 20 cm were applied during each irrigation application using this technique [25]. Shortly after planting in Year 0 for Test 1 (hereinafter designated as Test 1/Year 0, for example), 4.25 cm of precipitation fell, delaying any need for immediate supplemental water. Irrigation water was no longer available after the spring of Test 1/Year 3 (Test 2/Year 2) due to persistent drought in the watershed (Table 1 and Table 2).

On 4 April of Test 1/Year 1, Sethoxydim (Diethyl 2-[(dimethoxyphosphorothioyl)sulfanyl]butanedioate), Bentazon (3-(1-methylethyl)-1H-2,1,3-Benzothiadiazin-4(3-one 2,3-dioxide), and crop oil concentrate were all applied at 2.34 L ha⁻¹ as a tank mix to eliminate existing non-legume plants. Ethalfluralin (N-ethyl-N-(2-methyl-2-propenyl)-2,6-dinitro-4-(trifluoromethyl)benzenamin); 2.34 L ha⁻¹) was applied on 29 June for general pre-emergent weed control, and on 19 July, 2.34 L ha⁻¹ of Malathion 5 (Diethyl 2-[(dimethoxyphosphorothioyl)sulfanyl]butanedioate) was applied for grasshopper (Melanoplus spp.) control. All of these products were applied only to Test 1.

On 11 February of Test 1/Year 2 and Test 2/Year 1, 24.7-116.8-49.4 kg N-P₂O₅-K ha⁻¹ were broadcast based on the expected removal of P by an 18 Mg ha⁻¹ year⁻¹ alfalfa yield. On 3 April of that year, 584 mL ha⁻¹ Pounce 3.2EC (Permethrin) was applied to control cowpea aphids (Aphis craccivora). These applications were applied to both tests. No other fertilizers or pesticides were applied.

2.4. Measurements

Plant counts (plants m⁻²) were taken in June of Year 1 in two locations in each plot and averaged [18]. Portable quadrats (0.4181 m²) were used to include plants across the furrow bed continuum. Whole plot percent ground cover of each sown species (ALFPC and KCPC for alfalfa and kura clover stand percentage, respectively) was rated for each test in Years 2 to 5 [1] as an indication of establishment success [23]. Brummer and Moore [18] reported that plant counts and stand percentages were highly correlated for perennial forage legumes, including alfalfa and kura clover.

A six-harvest management regime was used for test Years 1 to 3, with at least 6 weeks between the two final harvests to allow the legumes to store root carbohydrates, although cutting management did not influence kura clover root architecture or carbohydrates [6], as it does root carbohydrates in alfalfa [25]. Harvests were taken to 7.5 cm stubble [1] using a self-propelled forage plot harvester equipped with a reciprocating sickle and electronic scales. Harvest dates are shown in Table 2 for each test year. At each harvest date, a subsample of the harvested material from each plot was collected, weighed, and dried at 70 °C for 48 h to determine dry matter (DM) concentration, which was used to convert plot fresh weights to DM mass Mg ha⁻¹. Total annual DM yield for each plot was calculated as the sum of the plot’s harvest masses.

2.5. Statistical Description

Plant counts were analyzed using the mixed procedure of SAS version 9.4 [26] to compare Test and TRT and the Test × TRT interaction. Data for the annual DM yield of alfalfa and stand percentage of alfalfa and kura clover were subjected to SAS GLM techniques [26] for repeated measurements of stand age (Age), comparing TRT and Age × TRT. Test and replicates within Test were identified as unique to Test and considered random, while TRT and Age were fixed effects in both analyses. All differences reported are significant at p ≤ 0.05, and when the F-test for Age, TRT, or any Age × TRT interaction was significant (p ≤ 0.05), lsmeans were compared using least significant differences using the PDMIX800 macro [27].

3. Results and Discussion

3.1. Alfalfa and Kura Clover Plants in the First Year After Seeding

The main effect means and results of the statistical analysis for alfalfa and kura clover plant counts in the spring after seeding are presented in

Table 3. Alfalfa and kura clover plants m⁻² in the spring after two consecutive seeding years (Tests 1 and 2) of various planting arrangements using alfalfa as an economic nurse crop during kura clover establishment at Tucumcari, NM, USA. Values are the lsmeans of four replicates within each test.

Test	Alfalfa		Kura Clover
1	44		3
2	65		6
SEM 1	15		6
Treatment (TRT) 2
ALFMONOBROAD	95	A 3	0	C
ALFMONOROW	82	AB	0	C
KCMONOBROAD	0	E	13	B
KCMONOROW	0	E	30	A
MIXALTROW	43	D	3	C
MIXBROAD	60	C	0	C
MIXKCBROAD	70	BC	0	C
MIXKCROW	72	BC	1	C
MIXROW	66	BC	0	C
SEM	15		7
p-values
Test	0.3630		0.6479
TRT	<0.0001		<0.0001
Test × TRT	0.0014		0.0003

¹ SEM signifies standard error of the mean. ² ALF, KC, MONO, MIX, BROAD, ROW, and ALT signify alfalfa, kura clover, monoculture, mixture, broadcast planted, drilled in all rows, or drilled in alternate rows, respectively. When no legume species is mentioned in the TRT name, both species were planted the same way. For MIXKCBROAD and MIXKCROW, alfalfa was drilled in rows or broadcast, respectively. Mixture component species were sown at half the seeding rate of the monocultures (22.45 kg ha⁻¹). ³ Treatment means within a column are not significantly different when followed by a similar letter.

. Recall that the seeding rate for mixture components was half that of the monocultures. While there were no differences between tests for either the alfalfa or kura plant counts, the TRT and Test × TRT interactions were significant for both. Regarding the interaction for alfalfa plants m⁻², there was a minor change in rank for MIXKCBROAD that was considered not biologically significant. Otherwise, there were no differences between tests for the remaining treatments, and the differences among treatments within tests were consistent. The results for the main effect of TRT (Table 3) indicated a reduction in alfalfa plants m⁻² whenever it was mixed with kura clover, but there was no difference whether the alfalfa was drilled in rows or broadcast.

The alfalfa plant counts in the present study (Table 3) were similar to the 97 plants m⁻² reported for a fall dormancy 6 variety at two sites in Australia by Stutz et al. [5], but considerably lower than the 236 reported for alfalfa 17 months after a spring seeding [18], and even more so than the >300 plants m⁻² reported for sprinkler-irrigated alfalfa 3 weeks after late summer seeding using the same planting technique and seeding rate but in a different soil type at the location of the present study, for which soil Na and salts were 151 ppm and 0.27 mmho cm⁻¹, respectively [4]. The alfalfa in the latter study [4] had not yet undergone winter conditions.

The interaction for kura clover plants m⁻² (Table 3) occurred as an interaction of magnitude because of a much greater difference between tests for KCMONOROW (13 and 47 plants m⁻² for Test 1 and Test 2, respectively) than for other TRTs, which were not different between tests. Differences for the main effect of TRT in kura clover plants m⁻² occurred because drilling kura clover seed in rows led to more plants m⁻² on average compared to broadcasting the seed (Table 3). Taylor and Smith [6] reported that broadcast seeding was the least successful planting method for kura clover.

The kura clover plant counts for the monocultures were considerably less than the 85 plants m⁻² 4–5 weeks after planting reported by Stutz et al. [5], and the 161 plants m⁻² reported in the year after seeding for one planting year by Seguin et al. [11], but similar to those the latter study reported for another planting year (19 plants m⁻²), also in the year after seeding. The latter study [11] attributed the lower populations to soil crusting. Planting with birdsfoot trefoil led to about half the kura clover plants compared to the monoculture kura clover [11]. Collins et al. [20] reported as many as 200 kura clover plants m⁻² 10 days after planting in the spring, but 365 days later, there were only about 20 plants m⁻² in strip mine spoils vs. 53 plants m⁻² in natural soil. Steiner and Snelling [23], in a slightly more moist environment south of the Equator with a similar summer precipitation pattern (Table 1), reported a decline in kura clover plants m⁻² from shortly after planting in early autumn (May) to spring (Aug.) or early summer (Nov.), with plant counts being similar to those measured at about the same time after winter as in the present study (Table 3).

The pH of the strip mine spoil was 7.9 [20], which is similar to the pH of the soil in the present study, potentially indicating that pH is a factor in kura clover establishment. Although DeHann et al. [10] found that kura clover establishment was improved as pH increased from 4.9 to 6.5, and Guldan et al. [3] successfully established kura clover in a calcareous soil having a pH of 7.4, Sheaffer et al. [12] stated that kura clover was reputed to be adapted to a wide range of soil types.

Lack of kura clover seedlings in the spring after seeding for mixture treatments in which kura clover was broadcast [6] may be due to low surface soil moisture leading to seedling desiccation and/or crusting [11] that inhibited radicle penetration [21]. The furrow irrigation technique used in the present study applies large amounts of water less frequently (Table 2) than sprinkler irrigation, with deep percolation into the soil profile, but it also leads to a greater likelihood of crusting [11], especially in coarse-textured soils, such as those of the present study. More frequent irrigation to supplement precipitation (Table 1 and Table 2) was used shortly after seeding to prevent problems, but it was often not frequent enough. Taylor and Smith [6] reviewed the literature and found that kura clover can establish under drought conditions, but that drought after germination led to many stand failures.

Awan et al. [21] reported that seedling survival of field-sown kura clover was only 20% of sown seeds 20 days after sowing, and Collins et al. [20] found 33% germination of the kura clover pure live seed they planted. This may also have been a factor in the low number of alfalfa seedlings in this study compared to those reported elsewhere [18], in addition to the high preplant soil Na levels and the calcareous nature of the soil type, because kura clover has been reported to perform best at soil pH < 7 [10].

Poor nodulation has also been implicated in the poor establishment of kura clover, with non-nodulated plants being unthrifty and eventually dying [1,6]. Collins et al. [20] stated that when the application rate of inoculant was 5× the recommended rate, kura clover seedling densities increased from 5 to 30 plants m⁻². The latter number is consistent with the kura clover plants m⁻² in the present study for KCMONOROW, indicating that conditions were conducive for nodulation when the most optimum planting technique was used (drilling in rows vs. broadcast seeding [6]). The high soil pH in the present study should also have enhanced nodule numbers of kura clover [10]. LaBerge et al. [8] waited for two weeks after seeding to inoculate plants in their study and reported a linear increase in nodule number for eight weeks thereafter, although supplied N reduced nodule numbers. Therefore, kura clover does not need immediate rhizobium infection for survival.

LaBerge et al. [8] found that low soil N status during the germination and seedling stages reduced shoot and root growth even when an effective inoculant was used. They [8] also found that kura clover plants that were inoculated and supplied with inorganic N had similar whole plant dry matter to white clover, but greater root dry matter. Using an aggressively establishing, high-yielding, short-lived legume such as red clover [1,7] led to reduced plant numbers of birdsfoot trefoil [7], but such species that nodulate well, including alfalfa, should provide sufficient N to meet kura clover’s needs until it can sufficiently nodulate. The preplant soil N levels in the present study should have been adequate to support the plants through delayed nodulation. Preplant P levels and the P application in the year after planting Test 2 would also have promoted kura clover growth [6,20]. DeHaan et al. [10] reported that low P uptake from soil with limited P levels due to pH would likely not have been a factor in low kura clover yields.

3.2. Alfalfa Total Annual Yield for 3 Years After Seeding

Recall that irrigation was no longer available after the spring of Test 1/Year 3 (Test 2/Year 2) (Table 2). Consequently, only 3 years of yield data are available. The main effect of Test was not significant (p = 0.27); however, the main effects of Age, TRT, and the Test × TRT, Age × TRT, and Test × Age × TRT interactions were significant (p ≤ 0.0001) for total annual yield. The interactions were fairly consistent with the main effect means across tests and stand ages, with age 2 having the greatest yields compared to ages 1 and 3, and KCMONOBROAD and KCMONOROW having no yield, which led to significant (p ≤ 0.0001) linear and quadratic effects for the Age × TRT interaction (Figure 1). Among TRTs including alfalfa, there was no difference at age 1; however, at 2 years of age, ALFMONOROW had greater yields than all the other TRTs having alfalfa, including ALFMONOBROAD, and MIXALTROW had lower yields than all alfalfa-containing TRTs. Furthermore, at age 3, there was no difference among alfalfa-included TRTs, except that MIXTALTROW again yielded less than all the other alfalfa-containing TRTs (Figure 1). Total annual alfalfa yields in the present study were consistent with those measured for a similarly managed alfalfa variety test nearby the present study (L. Lauriault, unpublished data).

Kura clover yields were never measurable during this study period due to small plants [1]. Little growth has also been reported in the humid southeastern US at a similar latitude to the present study [6,20]; however, kura clover has also been reported to succeed in a drier climate [6]. Brink et al. [2] reported that kura clover aboveground production > 7 cm was not sufficient for nutritive value analysis by near infrared spectroscopy. Laberge et al. [8] reported that early growth of kura clover was comparable to white clover when soil N was not limited; however, Andrzejewska et al. [1] reported that kura clover yielded the least among red and white clover and alfalfa in the first year after spring seeding, but kura clover yields were second only to alfalfa in the second and third year after seeding, although yields declined over the three-year study period. DeHaan et al. [10] found that kura clover whole plant mass at 56 and 91 days after planting was about 50% that of birdsfoot trefoil, which is also considered to be a slow-establishing legume [3,7,10]. Guldan et al. [3] reported that it was not until the fourth year after seeding that kura clover yields equaled alfalfa yields when both were mixed with tall fescue, and Lauriault et al. [19] found that kura clover exceeded alfalfa yields through the eighth year of that same study [3]. Andrzejewska et al. [1] stated that kura clover yields were greater at a location with less precipitation and a light soil, where moisture deficit was observed, and alfalfa, red clover, and white clover yields were reduced compared to a location with greater precipitation during the three years after seeding.

3.3. Alfalfa and Kura Clover Stand Percentages for Years 2 to 5 After Seeding

The main effects of Age and TRT were significant for alfalfa stand percentage (p < 0.0085 and p < 0.0001, respectively), but the interaction was not significant. Repeated measurements analysis revealed that the Age effect was linear over time (p < 0.0033) because of an increase from an average of 85 to 88% from years 2 to 5 for TRTs including alfalfa. Because alfalfa has no mechanism for vegetative reproduction and was not allowed to produce seed, this increase was likely due to crown expansion. For the TRT effect, all mixtures, except MIXALTROW, had the greater ALFPC, with some differences within that grouping and the KC monocultures having the least ALFPC because alfalfa was not planted in those plots (95, 95, 0, 0, 69, 87, 89, 84, and 91 ALFPC for ALFMONOBROAD, ALFMONOROW, KCMONOBROAD, KCMONOROW, MIXALTROW, MIXBROAD, MIXKCBROAD, MIXKCROW, and MIXROW, respectively, 5% LSD = 5). Sulc and Rhodes [28] reported about 80% stands for 1-year-old alfalfa, and Andrzejewska et al. [1] reported that monoculture alfalfa maintained 100% stand for 3 years after the establishment year. These results are consistent with the results of the present study when differences between observers among studies are taken into account. Cupina et al. [17] found that alfalfa was generally not influenced by companion cropping if its light demand was met. Lauriault et al. [19] found that alfalfa-tall fescue yield remained equilibrated at 53% alfalfa in the sward for 5 to 8 years after seeding, yielding an average of about 7.1 Mg ha⁻¹ of alfalfa in the mixture with total mixture yields of 12.7 Mg ha⁻¹. Therefore, it may take more than 8 years for alfalfa to relinquish dominance to kura clover in alfalfa–kura clover mixtures, but the alfalfa would still provide economic returns in the meantime.

For KCPC, the repeated measurements analysis revealed significant Age, TRT, Test × TRT, and Test × Age × TRT effects (p < 0.0111, 0.0001, 0.0048, and 0.0026, respectively). The three-way interaction, which had a linear trend (p < 0.0079), is shown in Figure 2. The interaction was largely caused by the difference in KCMONOROW, which fairly steadily increased over the years in Test 1, but declined after Year 3 and then increased again slightly from Year 4 to Year 5 in Test 2. The trend in Test 2 is not understood because the environmental conditions were the same as for Years 4 to 5 in Test 1 (Table 1), when the KC monocultures experienced the greatest increase in KCPC (Figure 2). In both tests, MIXALTROW increased slightly from Year 2, but remained around 10% stand through Year 5. When the kura clover was in closer proximity to the alfalfa, as in MIXBROAD, MIXROW, MIXKCBROAD, and MIXKCROW, KCPC remained low or exhibited erratic changes over time (Figure 2). This said, the KCPC of MIXROW was also different across tests and stand ages such that it remained at or near 0 in Test 1 across years but increased from 0 in Test 2 to about 25% in Test 1/Year 5. These results are likely consistent and environmentally driven since Test 1/Years 2–5 are the same calendar years as Test 1/Years 1–4 (Table 1 and Table 2). Note from Table 1 that Test 2/Year 4 received over 150 mm greater precipitation than all of the years for Test 1 and all previous years for Test 2.

Much of the precipitation fell after the stand percentage ratings were taken in June of Test 1/Year 5 when that test’s final stand rating was taken. Precipitation was considerably less for the period of June through May, after which the stand ratings for Test 1/Year 5 and Test 2/Year 4 were taken, likely contributing to the slight decreases in stand percentage for MIXALTROW from Years 4 to 5 in Test 1 and for several TRTs from Years 3 to 4 in Test 2 (Figure 2). Interestingly, although few if any plants were observed in the first year after seeding for the kura clover–alfalfa mixtures (MIXALTROW, MIXBROAD, MIXROW, MIXKCBROAD, and MIXKCROW) (Table 3), when the kura clover was drilled in rows, 10–15% stand percentage was attainable (Figure 2), and alfalfa yields in the mixtures remained equal to the alfalfa monocultures, except for MIXALTROW (Figure 1).

Although kura clover is known to not have a high percentage of hard seed [20], it is possible that some hard-seededness existed in the kura clover seed lot used in the present study [11] that broke dormancy over time, generating new plants as indicated by stand percentage, which increases as kura clover becomes established [15] (Figure 3a).

Sheaffer and Martin [12] reported that kura clover monoculture stand percentages in May were >90% after three production years of a spring seeding, and Steiner and Snelling [23] reported 10% kura clover stands after two years, and Peterson et al. [14] started their study with 6-year-old kura clover having 95% stands.

Walker and King [13] found that moisture was a limiting factor during three of four growing seasons and that poor moisture conditions resulted in insufficient growth for harvest of kura clover. Kura clover is known to recover from non-selective herbicide applications [15,16]. Therefore, perhaps it can also recover from periods of drought [1,5,6,9,13,23]. Because the kura clover in the present study was slow at increasing stand percentage (Figure 2) and based on the results of other studies in the semiarid US Southwest [3,19], when kura clover is planted with a companion species, it may take 3 or more years for it to gain dominance and maximize stand percentage, especially when precipitation or irrigation are inadequate. In the present study, because irrigation water was no longer available after Test 1/Year 3 (Test 2/Year 2) (Table 2), kura clover did not attain full stand and could not produce harvestable forage, either as hay or pasture, but there was an increase when greater precipitation occurred. The alfalfa was also not productive after irrigation was terminated, but it could recover and be as productive as before irrigation was terminated [29], if it had not been too long and stands were not depleted.

It is unfortunate that irrigation water became unavailable and kura clover yields were not measurable during the current study because Ates et al. [7] reported that even growing an aggressive legume, such as red clover, did not negatively influence the productivity of birdsfoot trefoil after the red clover stands declined. Because of the competition for light and soil moisture by alfalfa in the present study, the kura clover may have continued as a minimally photosynthetically active species with small aboveground structures, as reported for birdsfoot trefoil by Ates et al. [7], but was developing its root system, and so it could be ready to be fully productive when released from the alfalfa nurse crop, possibly after 6 to 7 years. Lauriault et al. [3,19] reported that kura clover had not attained full production status until 4 years after being sown with tall fescue, which was found to be extremely competitive when sown with kura clover [6,24]. In an ongoing fully irrigated study at another location in New Mexico, similar effects have been observed in an alfalfa-kura clover mixture where kura clover did not increase in stand percentage to any significant degree until three years after seeding with alfalfa (Marsalis, personal observation; Figure 3b). Sheaffer et al. [12] reported that kura clover became one of the highest-yielding legumes, second only to alfalfa. Lauriault et al. [19], at another location in the semiarid US Southwest, found that kura clover had greater average yields from the fifth to eighth year after planting.

4. Conclusions

This study is the first known report to begin filling the knowledge gap for using alfalfa as an economic nurse crop during kura clover establishment, as demonstrated by kura clover’s ability to survive and increase stand percentage in monoculture and when planted with alfalfa. The uncontrollable termination of irrigation water led to a slower stand development of the kura clover and eventual production potential. This was a limitation of this study in providing sufficient information to conduct an economic analysis of establishment through optimum production potential by the kura clover. Under typical irrigation management in semiarid regions, it is expected that the kura clover stand would develop more rapidly, and the alfalfa would eventually yield dominance in the stand. Alfalfa may persist for more than 5 years before relinquishing dominance to kura clover in mixtures, but the alfalfa would continue to provide economic returns as kura clover continued stand development with minimal production, but developed its root system to maximize production when released from the alfalfa nurse crop. Future research should be conducted in an area with a more consistently available irrigation water source so this study can be completed under typical irrigation management until the kura clover attains stand dominance and maximizes production potential, and an economic analysis can be applied.