Individual and Combined Effects of Planting Date, Seeding Rate, Relative Maturity, and Row Spacing on Soybean Yield

Abstract

Planting date (PD), seeding rate (SR), relative maturity (RM) of cultivars, and row spacing (RS) are primary management factors affecting soybean (Glycine max (L.) Merr.) yield. The individual and synergistic effects of PD, SR, RM, and RS on seed yield and agronomic characteristics in North Dakota were herein investigated. Early and late PD, early and late RM cultivars, two SR (408,000 and 457,000 seed ha⁻¹), and two RS (30.5 and 61 cm) were evaluated in four total environments in 2019 and 2020. Maximizing green canopy cover prior to the beginning of flowering improved seed yield. Individual factors of early PD and narrow RS resulted in yield increase of 311 and 266 kg ha⁻¹, respectively. The combined factors of early PD, late RM, high SR, and narrow RS improved yield by 26% and provided a $350 ha⁻¹ partial profit over conventional practices. Canopy cover and yield had relatively weak relationships with r² of 0.36, 0.23, 0.14, and 0.21 at the two trifoliolate, four trifoliolate, beginning of flowering, and beginning of pod formation soybean growth stages, respectively. Producers in the most northern soybean region of the USA should combine early planting, optimum RM cultivars, 457,000 seed ha⁻¹ SR, and 31 cm RS to improve yield and profit compared to current management practices.

1. Introduction

The gap between genetic potential and the average producer soybean (Glycine max (L.) Merr.) yield at the farm level has previously been investigated in the north-central USA region. In the north-central USA region, the states North Dakota, South Dakota, and Minnesota comprised 19% of USA soybean production in 2020 [1] with an average profit of $230 USD ha⁻¹ in southern North Dakota [2]. A grower survey of 524 soybean fields in North Dakota by Rattalino Edreira et al. [3] found yield differences between the highest and average-yielding (1200 kg ha⁻¹) fields were due to three primary factors, including planting date (PD) [3], relative maturity (RM) [4], and seeding rate (SR) [5]. Mourtzinis et al. [4] stated that RS practices in North Dakota were about equally split between narrow (~25 cm) and wide (~76 cm) spacings in surveyed soybean fields. The effects of PD, RM, SR, and RS on soybean production have been individually well-investigated. Current North Dakota management trends of second-half of May planting, cultivars with RM not best-adapted for the region, and 408,000 seed ha⁻¹ SR can be improved upon [6], with narrow RS providing yield benefits compared to wide RS in northern growing regions [7]. The potential synergistic effects of early planting, cultivars with longer RM [6], and SR higher than 408,000 seed ha⁻¹ require further exploration, especially in northern USA soybean production environments [8,9].

The soybean planting date is considered the most important cultural management factor to production [10]. Delaying soybean planting from 1 May to 15 May decreased yield by 0.5% d⁻¹ in Minnesota [11]. Delaying soybean planting beyond May resulted in declining yields in the north-central USA region [3,12,13,14,15,16]. In addition, Stanley [6] reported a 0.4% d⁻¹ yield loss when delaying planting beyond 1 May (up to 1 June) in North Dakota. However, soybean yield response to PD can vary considerably from year-to-year, especially depending on the magnitude of environmental constraints [15,16].

The length of the cropping season in the north-central USA has increased by 5 to 20 d since the 1950s, according to Kucharik et al. [17], generating uncertain optimal soybean maturity recommendations [18]. Soybean cultivar RM groups in the USA range from 00.0 to 10.0 [19], with the suggested maturity groups for production in North Dakota ranging from 00.0 to 1.0 [20]. Mourtzinis and Conley [21] delineated optimal soybean maturity group zones and noted a range of RM 0.0 to 1.5 is more suited for North Dakota production. To maximize yield, current production recommendations suggest utilizing a cultivar with the longest maturity group suitable for a growing region [22].

With similar SR, the distance between soybean rows has an impact on plant density within the row and soybean seed yield. Narrow rows can cause earlier in-season canopy closure [23], resulting in greater accumulated light interception compared to wider RS [24,25]. In the northern and central soybean growing regions in the USA, narrow RS can improve soybean yields in normal environmental conditions [9,25,26,27,28,29,30]. Narrow rows can increase soybean yield when planted late or under no-till conditions [14]. In contrast, wide soybean RS improves yields under soil water deficit conditions [29,31].

Various SR have been reported as optimal for yield and typically differ between growing regions. Seeding rates below 560,197 seeds ha⁻¹ in Iowa [32], 284,050 to 573,000 seeds ha⁻¹ in high yield conditions in Kansas [29], and 444,600 to 494,000 seeds ha⁻¹ in North Dakota [9], all of which have been found to produce the greatest yield. Soybean iron deficiency chlorosis (IDC) in IDC-prone North Dakota soils can be reduced by increasing plant density from 300,000 to 600,000 plants ha⁻¹, resulting in increased seed yield [33]. To produce similar yields around 3600 kg ha⁻¹, Lee et al. [34] reported optimal SR for planting in May in Kentucky was as low as 171,000 seed ha⁻¹, where SR in Wisconsin was as high as 741,000 seeds ha⁻¹ [14]. Gaspar et al. [5] found SR below the agronomical optimal SR exponentially increased risk and potential yield loss. Reducing risk for soybean yield loss may be circumvented by optimizing canopy cover through management, which would require a greater understanding of the effect of management practices on canopy cover.

Canopy cover is a useful proxy measurement for light interception and crop productivity [35]. Maximum photosynthesis is achieved when plants maximize light interception and utilization of photosynthetic radiation [36,37]. Light interception can be quantified with methods such as quantum line sensors [38], approximated by fractional green canopy cover (FGCC) from pictures using the Canopeo app [39], and leaf area index.

Estimating and predicting crop yields using canopy cover measurements is of high relevance to producers. Crop growth stage [40], RS [41], and canopy structure [42] can affect FGCC. Green canopy cover may allow for better yield prediction and a useful application in soybean production. To our knowledge, no previous research in the northern USA soybean growing region has evaluated the combined effect of management factors and their potential for synergism to increase soybean yields, and their use in yield prediction based on canopy cover.

The objectives of the research were to determine how PD, SR, RM, and RS, as individual factors and when combined, affect seed yield, and agronomic characteristics, if combining factors are more economical than current practices, and to evaluate if FGCC can predict soybean yield.

2. Materials and Methods

Field experiments were conducted in 2019 and 2020 in Fargo, North Dakota, USA (46.932° N, 96.513° W). Each year the location had two experiments: one tile-drained, as described by Kandel et al. [43], and one non-tile-drained. Each experiment and year combination were called an environment, for a total of four environments.

The experimental design was a randomized complete block (RCB) with a split-plot arrangement with four replicates. The whole plot was PD, and the sub-plots were a factorial combination of RM of cultivars, SR, and RS, with the combination of all factors described as a package. Planting dates were at an optimal time, which is no earlier than 5 d before the last projected spring frost in mid-May and a late PD two weeks thereafter (

Table 1. Important dates and environment information for 2019 and 2020 soybean environments.

			Planting Date
	SCN 1		2019		2020		Harvest Date
Environment	2019	2020	1	2	1	2	2019	2020
	eggs 100 cc−1		DOY
Fargo	0	0	137	154	133	149	302	275

¹ SCN, Soybean cyst nematode; DOY, Day of year. DOY 135 is 15 May and DOY 280 is 7 October.

). Cultivars were from the same company and differed by 0.3 RM units and were described as early or late RM relative to the cultivars grown in that environment (

Table 2. Soybean cultivars used and descriptive features.

		NDSU	Company
Cultivar	Maturity	IDC 1	IDC	SCN 2	Canopy	Height
AG 05x9	0.5	2.4	3	R	Medium	Medium
AG 08x8	0.8	-	4	R	Medium	Medium Tall

¹ IDC, iron deficiency chlorosis. NDSU IDC scored on 1–5 scale (1 = green, 5 = dead) [33]. Company IDC scored on 1–9 scale (1 = green 9 = dead). ² SCN, soybean cyst nematode; R, resistant.

). Seeding rates were 408,000 and 457,000 pure live seeds (germinable seeds) ha⁻¹ and adjusted using a 98% germination average. The RS widths were narrow (30.5 cm) and wide (61 cm). The four factors of PD, RM, SR, and RS were combined over 2019 and 2020, with the main effects analyzed individually. To answer if a combination of factors provided higher yield compared to conventional practices, individual factors were combined into ‘packages’, combined across environments, and analyzed by packages using an RCB. The ‘improved package’ consisted of an early PD, late RM, 457,000 seeds ha⁻¹, and narrow RS factors, whereas the ‘conventional package’ was a late PD, early RM, 408,000 seeds ha⁻¹, and wide RS.

Experimental units were seeded using a Hege 1000 no-till planter (Hege Company, Waldenberg, Germany). Seeds were sown to a depth of approximately 3 cm, with the experimental unit size being 1.52 m by 5.47 m. Soils were tested (

Table 3. Soil test results for soybean environments in 2019 and 2020.

Depth	NO3-N	P	K	pH	OM
cm	kg ha−1	mg kg−1			%
	2019
0–15	8	15	495	8	6
15–61	14	5	300	8	4
	2020
0–15	22	18	489	8	6
15–61	26	6	353	8	4

) for plant essential nutrients before seeding to ensure fertility was not a limiting factor based on North Dakota State University recommendations [20]. Soil fertility levels were sufficient and no additional fertilizer was applied. Soybean seed was pre-treated with Acceleron seed treatment, (Bayer CropScience, Monheim, Germany). Seeds were inoculated with Bradyrhizobium jamponicum using Vault SP (BASF, Ludwigshafen, Germany) at a rate of 40 g per 23 kg of seed.

Two applications of Roundup PowerMAX (Bayer CropScience, Monheim, Germany) at a rate of 1.6 L ha⁻¹ were applied during the season for weed control. Soybean aphid (Aphis glycines Matsumura) and grasshopper (Orthoptera: Acrididae) were controlled when pest densities surpassed the economic threshold [44] using the insecticide Mustang Maxx (FMC Corporation, Philadelphia, PA, USA) at a rate of 1.75 L ha⁻¹.

After planting, a number of established plants per experimental unit were recorded at the V2 or two trifoliolate stage [45]. Established plant density was determined by counting a 0.91 m length from the middle soybean rows, and the final plant density was determined by counting plants from the same length before harvest. During the growing season, soil cover percent was recorded using the Canopeo app (Oklahoma State University, Stillwater, OK, USA). Fractional green canopy cover photos were processed to determine a canopy coverage percentage [39]. Canopy pictures were taken approximately 1.5 m from the soil surface in the center of each plot using an iPad (Apple, Cupertino, CA, USA). Matlab software (MathWorks, Inc., Natick, MA, USA) was used to estimate canopy cover by FGCC. The FGCC measurements were recorded when soybean plants in the early PD were at the V2, V4, R1, R3, R5, or R7 growth stage (two trifoliolate, four trifoliolate, beginning of flowering, beginning of pod formation, beginning of bean development, and pod and leaf yellowing), respectively [45]).

Experimental units were harvested after physiological maturity [45] using a Wintersteiger classic plot combine (Wintersteiger AG, Ried, Austria). Seed samples were cleaned and weighed post-harvest and analyzed for oil and protein content using a Perten Instruments DA 7250 NIR analyzer (Perten Instruments, Inc., Springfield, IL, USA). Moisture and test weight were determined using a GAC 2100 moisture tester (DICKEY-John Corp., Minneapolis, MN, USA) and observations were corrected to 13% moisture content. Seed weights were obtained by counting 1000 seeds using an electric counter and weighing them. Weather data were collected using the North Dakota Agricultural Weather Network [46] providing weekly maximum and minimum air temperature and rainfall using the Fargo NDAWN weather station.

Normality and homogeneity of variance assumptions of ANOVA were met, as determined by residual histograms and the ratio of the highest and lowest error mean square being less than 10 [47]. Analysis of variance was performed for 2019 and 2020 environments using the GLIMMIX procedure in SAS 9.4 (SAS Institute, SAS Circle, Cary, NC, USA). The individual factor analysis used fixed and random effect designations described by Carmer et al. [48], where PD, RM, SR, and RS were considered fixed effects, while environment, and replicate within environment, were considered random. Analysis of variance of dependent variables for packages were performed using the GLIMMIX procedure, where package was considered a fixed effect, and environment a random effect. The FGCC dependent variable was analyzed as a repeated measure over time using the design structure similar to the respective analysis [49]. Simple linear correlations between FGCC and yield were performed using the CORR procedure. Orthogonal contrasts were made using Proc GLIMMIX to compare improved and conventional management packages. Means were separated using a Tukey’s HSD with p = 0.05.

A partial net profit economic analysis for the individual factors and packages were only included costs associated with seed and no other costs, such as machinery and labor which do not affect the economics of SR. The partial net profit was calculated by subtracting seed cost (price per seed (PPS)⨯ live seeding rate (SR)) from gross revenue (seed yield (SY)⨯ market price (MP)): $\mathrm{Partial} \mathrm{Net} \mathrm{Profit} = (\mathrm{SY} \times \mathrm{MP}) - (\mathrm{PPS} \times \mathrm{SR})$. Price per seed was calculated by dividing a $49.20 seed cost unit⁻¹ by 140,000 seeds [50]. Gross revenue calculations used the 2010–2020 soybean market price average of $0.41 kg⁻¹ from the Chicago Board of Trade. Means were separated using Tukey’s HSD with p = 0.05.

3. Results and Discussion

3.1. Weather Data

Rainfall amounts and temperature conditions differed slightly between the years. Rainfall events were more frequent in 2019 compared to 2020 (Figure 1). Total rainfall between weeks 1 to 25 was greater in 2019 (558 mm) compared to 2020 (404 mm). Weekly rainfall averages post week 15 were considerably drier in 2020 compared to 2019. The two-week period following planting in 2019 was warmer and drier compared to the first two weeks following planting in 2020. In 2020, minimum air temperatures decreased below 0 °C, causing slight frost damage to the longer relative maturity soybean cultivar. On average, average temperatures of 16.2 and 16.4 °C between weeks 1 and 25 were similar in 2019 and 2020, respectively.

3.2. Individual Factor Analysis

Analysis of interactions among individual factors were non-significant, PD, RM, and RS affected most dependent variables (

Table 4. Mean agronomic trait observations for planting date, relative maturity, seeding rate, and row spacing averaged across controlled tile drainage and naturally environments in 2019 and 2020 for a total of 4 environments.

SOV 1	ES	FS	Protein	Oil	TSW	Yield	Partial Net Profit	FGCC
PD	plants ha−1		g kg−1		g	kg ha−1	$ ha−1	%
Early	380,608 b	344,881 b	335 a	173 a	140 a	3884 a	1604 a	56 a
Late	432,649 a	403,757 a	335 a	171 a	137 a	3573 b	1476 b	58 a
RM
Early	409,419 a	362,826 a	339 a	168 b	142 a	3584 a	1480 a	56 a
Late	403,838 a	385,812 a	331 b	176 a	136 b	3873 a	1600 a	58 a
SR
seeds ha−1
408,000	392,081 b	360,607 b	334 a	172 a	139 a	3686 a	1522 a	57 a
457,000	421,176 a	388,031 a	335 a	172 a	139 a	3772 a	1557 a	58 a
RS
cm
30.5	424,782 a	395,411 a	334 a	173 a	138 a	3862 a	1595 a	59 a
61	388,475 a	353,227 a	336 a	171 a	140 a	3596 b	1485 b	55 a

Within columns, for each main factor, means followed by the same letter are not significantly different at p ≤ 0.05. ¹ SOV, source of variation; ES, established stand; FS, final stand; TSW, 1000 seed weight.

). Early planting resulted in reduced established plant density compared to late planting, thereby affecting final plant densities (Table 4). Above-normal rainfall events before the early PD created poor seeding and emergence conditions. The yield of early PD soybean was significantly higher compared with late PD soybean. Delaying planting beyond May in the upper Midwest can reduce yields [32] and consequently resulted in a 9% decrease in partial net profit (Table 4). Although late planting provided better germination and establishment, early planting allows for greater accumulated solar radiation throughout the growing season providing greater yields comparatively [51]. In Japan, planting three weeks later than optimal decreased soybean yield due to reduced node production, slower canopy development [52]. Delaying planting can shorten vegetative and reproduction growth stage periods resulting in less radiation accumulation when planting after 20 May in Iowa [53].

The later RM cultivar yielded significantly higher than the early RM cultivar. Planting longer maturing (2.0) cultivars in Minnesota and Wisconsin tend to increase yields compared to earlier RM (0.5) confirming the trend in our experiment [22]. Cultivars had significant differences in seed protein, oil content, and weight, which can partly be explained by the genetics of the cultivar (Table 4). Other studies have also found that longer-maturing cultivars can accumulate greater amounts of solar radiation throughout the season often resulting in greater seed yield and oil content, and reduced protein content [51,54]. However, autumn frost damage is a risk in the northern growing region therefore selecting a RM which can mature before the average frost date is critical to maximize production.

Seeding rate did not have a significant effect on observations, other than plant stands. Established and final plant stands were greater for the higher SR (Table 4). Both SRs had similar stand loss from established to final density of around 9%. In southern Brazil, optimal SRs were greater in high yielding environments (>5000 kg ha⁻¹) where SRs could be reduced by 18% without a yield penalty compared to low yield environments (<4000 kg ha⁻¹) [55]. Gaspar et al. [56] recommends soybean farmers in Wisconsin to adjust SRs to the economic optimum of 255,000–293,000 seeds ha⁻¹ for maximum yield. For this research, increasing the separation between the two SRs or adding additional rates may have shown an effect of SR.

Narrow rows were found to be superior to wider rows regarding yield. Cox and Cherney [28] found a linear increase in yield as RS when decreasing RS from 76 to 19 cm in north-central USA. Row spacing results showed the benefit of narrow RS (30.5 cm) in the northern USA soybean growing region (Table 4). Previous research in the northern USA soybean growing region deemed RS a major yield contributing factor with narrow RS (30.5 cm) providing a 6% greater yield than wide RS (61 cm) [9]. Andrade et al. [7] similarly reported 540 kg ha⁻¹ yield benefit to narrow RS compared to wide RS. The yield increase in narrow RS is likely due to an advantage of increased light interception during the limited growing season [7]. Lee et al. [34] reported complete canopy cover at the R5 growth stage was required for maximum yield potential. Narrow RS of 19 and 38 cm also can provide weed control for multiple-resistant waterhemp (Amaranthus tuberculatus) which is commonly found in the north central growing region of the USA [57].

In general, PD was expected to be the most critical determinant of yield. Moreover, narrow RS improved yield and partial net profit by 7% compared to wider spacings (Table 4). Many previous studies have confirmed the importance of soybean RS particularly due to increased light interception [24,25]. Evaluating these individual effects within an improved management package of early PD, a late-maturing cultivar, high SR (457,000 seed ha⁻¹), and narrow RS (30.5 cm) permits assessment of the potential synergism between four yield benefitting factors.

The fractional green canopy cover (FGCC) was significantly different across the growing season. The greatest FGCC occurred at R3 (89%) and R5 (90%). Canopy measurements at the V2, V4, R1, and R3 growth stages were significantly correlated with yield, with r² values for all growth stages ranging from −0.05 to 0.36 which are not relatively strong relations (

Table 5. Simple correlation coefficients for the relationship between soybean seed yield and fractional green canopy cover (FGCC) measured at different growth stages throughout the growing season for the combined environments in 2019 and 2020.

	Correlations
Growth Stage	r2	p	FGCC
			%
V2	0.36	<0.001	11 e
V4	0.23	<0.001	46 c
R1	0.14	0.033	75 b
R3	0.21	0.002	89 a
R5	0.10	0.138	90 a
R7	−0.05	0.442	33 d

Within column, means followed by the same letter are not significantly different at p ≤ 0.05.

). The FGCC measurement at the V2 growth stage was the most closely correlated with yield, suggesting management practices that maximize FGCC may improve soybean yields. Edreira et al. [51] found that the average producer can improve their accumulated photosynthetically active radiation, and therefore seed yield by 14%, through earlier canopy closure. Our data reinforces that maximization of FGCC earlier in the season likely improves light interception and seed yield. Although significant, the associations between V2 and R3 growth stages and yield were relatively poor (on average r² = 0.24). Additionally, subsequent environmental events (e.g., drought, hail, early frost) can alter seed yield after early canopy measurements making these measurements unreliable for end-season seed yield predictions.

3.3. Treatment Package Analysis

Analysis by packages allowed for a more comprehensive depiction of a producer’s management system as producers typically compare yield differences through a systematic approach. The package analysis followed similar trends as the individual factor analysis (PD, RM, and RS). Although factors like early planting, late RM, and narrow RS have yield advantages alone, consolidating these factors into a singular package can improve yield when compared to other potential packages farmers may use.

Established and final plant stand results did not show any definite trends related to the packages (

Table 6. Treatment package means and contrasts for soybean agronomic characteristics and partial net averaged across environments in 2019 and 2020.

Package 1	PD 2	RM	SR	RS	ES	FS	Protein	Oil	TSW	Yield	Partial Net Profit 3	FGCC
			seeds ha−1		plants ha−1		g kg−1		g	kg ha−1	$ ha−1	%
I	E	L	457,000	N	326,306 a	399,705 bcd	330 f	177 ab	136bcd	4162a	1719 a	58 a
C	L	E	408,000	W	396,754 a	536,117 abc	339 abcd	165 e	143abc	3316g	1369 g	57 a
1	E	E	408,000	N	289,801 a	331,818 d	338 abcde	170 bcde	139 abcd	3867 abcde	1597 abcde	55 a
2	E	E	457,000	N	321,823 a	365,121 cd	339 abcd	170 bcde	142 abcd	3843 abcde	1587 abcde	53 a
3	E	E	408,000	W	335,272 a	457,984 abcd	340 ab	170 bcde	147 a	3639 bcdefg	1503 bcdefg	58 a
4	E	E	457,000	W	369,215 a	526,511 abcd	340 ab	168 de	147 a	3581 cdefg	1478 cdefg	60 a
5	E	L	408,000	N	348,081 a	425,322 abcd	331 def	180 a	141 abcd	4085 ab	1687 ab	54 a
6	E	L	408,000	W	334,631 a	470,793 abcd	330 f	174 abcd	136 bcd	3831 abcde	1582 abcde	55 a
7	E	L	457,000	W	353,204 a	489,366 abcd	333 bcdef	178 ab	138 abcd	4013 abc	1657 abc	56 a
8	L	E	408,000	N	355,125 a	424,041 abcd	339 abcd	167 de	140 abcd	3489 defg	1441 defg	60 a
9	L	E	457,000	N	376,260 a	458,625 abcd	337 abcdef	171 bcde	136 bcd	3670 bcdefg	1515 bcdefg	54 a
10	L	E	457,000	W	421,731 a	593,116 ab	341 a	167 de	145 ab	3356 fg	1385 fg	56 a
11	L	L	408,000	N	304,531 a	413,794 abcd	330 f	176 abc	136 bcd	3783 abdef	1562 abcdef	62 a
12	L	L	457,000	N	361,831 a	471,787 abcd	332 cdef	174 abcd	133 d	3944 abcd	1629 abc	61 a
13	L	L	408,000	W	398,675 a	579,027 ab	331 def	174 abcd	133 d	3437 efg	1419 efg	56 a
14	L	L	457,000	W	421,090 a	604,644 a	333 bcdef	174 abcd	134 cd	3542 defg	1462 defg	57 a
P					ns	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001	ns
Improved vs. Conventional P					0.572	0.809	0.962	0.510	0.956	0.004	0.004	0.518

Within the column, means for each treatment followed by the same letter are not significantly different at p ≤ 0.05. ¹ I, suggested improved management for growers; C, conventional management currently used by growers. ² PD, planting date; RM, relative maturity; SR, seeding rate; RS, row spacing; E, Early; L, Late; N, narrow (30.5 cm) row spacing; W, Wide (61 cm) row spacing; TSW, 1000 seed weight; FGCC, fractional green canopy cover. ³ Partial net profit accounts for $49.20 per 140 000 seeds and $0.41 kg⁻¹ market price.

). In general, packages with the early RM cultivar had higher protein content and lower oil content, with the inverse true for the late RM, however, this is mostly due to the genetics of the cultivars used. Conventional compared to the improved package did not differ for plant stand, protein content, oil content, and seed weight. Producers are currently not directly compensated for relatively high protein or oil content and should not consider them when investigating profitability.

The improved management strategy provided for higher yield (26% increase) and partial net profit (+$350 ha⁻¹) compared to conventional practices (Table 6). Comparing the conventional to treatment 8 (Table 6), the difference in RS had a non-significant yield increase, suggesting that a change in RS does not impact yield or partial net profit if a late RM soybean is planted late at a lower SR. In this case, changing the SR of the improved treatment to the lower SR (Treatment 5) results in similar profits, but would likely buffer plant density-reducing events (e.g., disease, hail, early frost). Schmitz et al. [9] found increased yields associated with narrow RS compared to wide RS in the same growing region. In North Dakota, in the absence of ideal temperature and rainfall, wide RS can result in incomplete canopy closure, whereas narrow RS improves the potential to maximize accumulated photosynthetically active radiation, therefore, increasing grain yield [24,51].

4. Conclusions

Our research benefits soybean producers in the north-central USA region, as other research on these management strategies has been limited in this region. Planting as early as appropriate and using narrow RS can increase soybean yield. Although an individual management strategy may not improve yields alone, combining yield-promoting management practices provided greater yield and partial net profit compared to conventional practices. These findings display a synergistic yield effect when combining favorable management strategies. The relationship between canopy cover at early growth stages and yield was significant; however, it is important to consider that various environmental stressors can influence seed yield beyond early season measurements. It is recommended that northern USA soybean producers combine favorable management practices of early planting, optimal maturity cultivars adapted to the area, and narrow RSs to maximize yield potential.