The Impact of Planting Dates on the Performance of Soybean Varieties [Glycine max (L.) Merr.] in the Nigerian Savannas

Abstract

Increasing delays in the onset of the rainy season and extended dry spells in the Nigerian savannas are complicating the determination of optimal planting dates for rain-fed crops, which increases risks for farmers. This study evaluated the impact of planting dates on soybean [Glycine max (L.) Merr.] performance to identify optimal planting dates for different soybean varieties in two agroecological zones (AEZs) of Nigeria. The study involved six planting dates (15 June, 22 June, 29 June, 6 July, 13 July, and 20 July) and three soybean varieties (TGX-1835-10E, TGX-1951-3F, TGX-1904-6F). Results showed significant differences in growth and yield parameters based on location, variety, and planting date. In the Sudan savanna (SS), AEZ at BUK-Kano, optimal yields (>1500 kg ha⁻¹) were achieved when planting TGX-1835-10E and TGX-1951-3F from 15 to 29 June and TGX-1904-6F on 15 June. Planting beyond 29 June reduces yields by 12–55% for TGX-1835-10E and 27–63% for TGX-1951-3F. For TGX-1904-6F, planting after 15 June reduces yields by 27–90%. In the Northern Guinea savanna (NGS) AEZ at Zaria, optimal yields (>1500 kg ha⁻¹) were obtained when planting TGX-1835-10E and TGX-1951-3F from 15 June to 6 July, and TGX-1904-6F between 15 to 29 June. Delaying planting beyond these dates significantly reduced yields by 18–31% for TGX-1835-10E and 12–20% for TGX-1951-3F and 10–41% for TGX-1904-6F.

1. Introduction

Soybean [Glycine max (L.) Merr.] has attracted growing interest in Nigerian agriculture due to its rich content of high-quality protein and vegetable oil [1]. It is widely cultivated in Nigeria for food, oil, and feed purposes [2], surpassing other common vegetable or animal feed sources in protein content [3]. Additionally, it contributes to soil fertility improvement and effectively suppresses parasitic weed like Striga [4].

Nigeria and South Africa are the largest soybean producers in Africa, accounting for 29% and 40% of total production, respectively [5]. Soybean ranks the third most important legume crop in Nigeria in terms of production [6], with 650,000 tons of soybeans produced in Nigeria in 2018 on a land area of 696,376 hectares [7]. Despite the rapid increase in soybean production and importance, soybean yield on average is less than 1 t ha⁻¹ compared to the yields of the top world soybean producing countries worldwide, such as the USA, Brazil, and Argentina with yields ranging from 2 to 3.38 t ha⁻¹ and 2 t ha⁻¹ in South Africa [7]. The low yield can be attributed to poor soil fertility, agronomic practices, disease infestation, irregular rainfall, high temperatures, and drought due to climate change [8].

In Nigeria, soybean is mainly grown in the Guinea and Sudan savanna agroecological zones, where rainfall adequately supports its growth and development [9]. However, delays in the onset of the rainy season and long intermittent dry spells experienced during the growing season are becoming more common, even in the wetter southern and Guinea savannas [10]. This significant year-to-year variability in rainfall onset, amount, and duration [11,12] poses challenges for rain-fed crop production, thereby increasing risks for farmers in the Nigerian savannas. For example, drought risk is particularly high in the Sudan savanna zone, where rainfall distribution is unreliable and uneven [13].

The duration of the growing season in the Guinea and Sudan savannas of Nigeria is determined by the onset of the first rainfall, which can vary significantly [12]. Using soybean varieties adapted to these conditions and planting within optimal windows can enhance productivity despite climate variability. It is crucial for farmers to understand the extent to which planting can be delayed and the potential yield losses that may result from late planting.

Over the years, the International Institute of Tropical Agriculture (IITA), in collaboration with national research institutions in Nigeria, has developed soybean varieties that are drought-tolerant, early-maturing, high yielding, resistant to rust, and poor soil, and recommended soil fertility management practices. These varieties are farmers’ preferred choices as they are high yielding, resistant to cercospora leaf spot and bacterial pustule, and nodulate freely with indigenous rhizobia bacteria.

Planting these improved soybean varieties at optimum planting time is considered a hopeful approach to increasing soybean productivity. Selecting the right planting date is critical, as it influences soybean growth stages due to variations in photoperiod [14,15], air temperature [16], and rainfall distribution and amount during the crop cycle [17]. Weber et al. [18] found that over 90% of farmers in Northern Nigeria risk planting earlier in order to take advantage of early crop sales, avoid pest attacks, and increase crop yields through benefiting from the enhanced nutrient availability brought by the first rains. However, early soybean planting when temperatures are typically high may result in poor emergence and reduced growth, while late planting can lead to shorter grain-filling times and lower yields [19]. Since soybean is very sensitive to drought, timely planting is crucial to avoid early season drought and drought at the end of the cropping season. Drought conditions during the seedling stage can significantly impact crop establishment, sometimes necessitating farmers to replant their crops. Additionally, drought coinciding with flowering and grain-filling periods can lead to severe yield instability at farm level, leaving farmers with no opportunity to replant or compensate for yield losses [11].

The planting date is the variable with the largest effect on crop yield [20]. Studies by Ibrahim [21] and Yagoub and Hamed [22] in Sudan emphasized the negative consequences of inappropriate planting dates on grain yield and its components. It has also been observed that early or late planting significantly reduces crop yield [23,24]. Numerous studies have confirmed the significant impact of planting dates on soybean yield and growth. For example, Omoloye et al. [25] examined the impact of planting dates on Hemipteran sucking bugs in soybeans in Nigeria and found that early-planted crops had fewer infestations and less seed damage than late-planted crops. Similarly, Adetayo [26] in Ibadan, Nigeria, reported that plant height, number of pods, and grain yield per hectare all decreased with delayed planting, regardless of the season or variety used. Adetayo emphasized that to achieve optimal grain yield, soybean planting should not be postponed beyond mid-August for late planting.

In another study, Sadiq et al. [27] reported that planting dates had a significant effect on plant height, days to 50% flowering, and grain yield per hectare, with late June planting outperforming early July planting in Samaru Zaria. Additionally, Shegro et al. [28] examined the influence of different planting dates on soybean growth and yield in Ethiopia, demonstrating that earlier planting resulted in significantly higher yields compared to delayed planting. Deciding the appropriate planting date is vital to prevent poor crop establishment and avoid the additional expenses of seed and labor needed for replanting [29].

Implementing crop management techniques in new environments, such as adjusting planting dates [17,30] and selecting suitable maturity groups or varieties [31,32], as well as understanding the implications of these practices can significantly improve soybean yield and resilience under changing climatic conditions. However, information on the optimum planting dates for planting soybean varieties is limited in some major soybean producing zones in Nigeria savannas. This study aims to identify the best improved management practices for soybean production in the Nigerian savannas.

2. Materials and Methods

2.1. Description of Experimental Site

The experiment was conducted during the cropping seasons of 2021 and 2022 at two different locations: the International Institute of Tropical Agriculture (IITA) experimental station in Samaru Zaria (11°11_N, 7°38_E, 686 m above sea level (ASL)) situated in the northern Guinea savanna (NGS), and Bayero University Research Farm Kano (11°58_N, 8°24_E, 470.9 m ASL) located in the Sudan savanna (SS) agroecological zones (AEZ) of Nigeria. The SS is characterized by a prolonged dry season with a monomodal rainfall pattern and a rainy season from May to October. This period is marked by high mean temperatures ranging from 28 to 32 °C, along with a short growing season lasting 90 to 110 days and low rainfall between 600 and 800 mm [33]. The soils in this region are highly weathered and fragile, exhibiting low clay content, and are predominantly classified as Alfisol [34]. In contrast, the NGS has a growing period ranging from 151 to 180 days [35] and receives an annual rainfall ranging from 900 to 1000 mm. The maximum and minimum temperatures are recorded at 40 and 28 °C, respectively, in the NGS [33]. The soils in the NGS are classified as leached ferruginous tropical soils, characterized by high clay content and overlain by drift materials [36].

2.2. Weather and Soil Characteristics at the Experimental Sites

Daily weather conditions at the experimental sites were monitored using the Watch-Dog weather station manufactured by Spectrum Technologies Inc., Haltom City, TX, USA. This station recorded daily rainfall in millimeters (mm) as well as daily minimum and maximum air temperatures in °C and solar radiation in MJ m⁻² day⁻¹ throughout the experimental period. At BUK-Kano, the average annual rainfall varied across the years, measuring 557 mm and 878 mm, respectively [Figure 1]. Similarly, the annual average minimum and maximum temperatures fluctuated, with values of 20.9 °C and 35.3 °C in 2021 [Figure 1a], and 17.9 °C and 31.0 °C in 2022 [Figure 1b]. Meanwhile, at Samaru Zaria, the average annual rainfall also showed variability, recording 994 mm and 972 mm across the years [Figure 2]. The annual average minimum and maximum temperatures exhibited fluctuations as well, with values of 19.0 °C and 32.1 °C in 2021 [Figure 2a] and 17.3 °C and 28.2 °C in 2022 [Figure 2b]. The annual average solar radiation also varied, with average solar radiation of 19.4 MJ m⁻² day⁻¹ and 21.0 MJ m⁻² day⁻¹ in 2021 and 20.5 and 20.0 MJ m⁻² day⁻¹ in 2022 at BUK-Kano and Samaru Zaria, respectively [Figure 1 and Figure 2].

Before planting in the first year, soil samples were collected from a depth of 0–30 cm at five different points within the experimental sites using a soil auger. The soil cores were thoroughly mixed in a bucket to create a representative sample of the area. The composite samples were sent to IITA, Ibadan, for laboratory analysis of the soil texture, pH, organic carbon, total nitrogen, available phosphorus, and exchangeable potassium. The analysis was conducted according to the analytical procedures of IITA [37] [

Table 1. Physical and chemical properties of soils at the experimental sites.

Soil Composition	BUK-Kano (SS)	Samaru Zaria (NGS)
Physical Properties (%)
Sand	74	34
Silt	11	49
Clay	15	17
Textural Class	Sandy Loam	Loam
Chemical Composition
pH (H2O 1:1)	6.6	6.0
Organic Carbon (g kg–1)	2.9	6.0
Total N (g kg–1)	0.2	0.4
Mehlich P (mg kg–1)	9.3	2.8
Exchangeable Cations (cmol + kg−1)
Ca	2.35	1.83
Mg	0.52	0.53
K	0.44	0.38
Na	0.08	0.04

]. The soils at Samaru Zaria predominantly consist of loam, with sand, silt, and clay contents of 34%, 49%, and 17%, respectively. They are slightly acidic with a pH of 6.0 and exhibit low levels of total nitrogen (0.4 g kg⁻¹), available phosphorus (2.8 mg kg⁻¹), and organic carbon content (6.0 g kg⁻¹). The soils at BUK-Kano are sandy loam, with sand, silt, and clay contents of 74%, 11%, and 15%, respectively. They are slightly acidic, with a pH of 6.6, and display low nutrient content, particularly total nitrogen (0.2 g kg⁻¹), available phosphorus (9.3 mg kg⁻¹), and organic carbon content (2.9 g kg⁻¹).

2.3. Experimental Design and Field Layout

The experiment was laid out in a split-plot factorial design replicated three times with planting dates as the main plot and varieties as the subplot treatment. There were six planting dates at 7-day intervals, 15 June, 22 June, 29 June, 6 July, 13 July, and 20 July designated as PD1, PD2, PD3, PD4, PD5, and PD6 at both locations and across the years, respectively. The soybean varieties used for this study were TGX-1835-10E, TGX-1951-3F, and TGX-1904-6F which were bred by IITA Ibadan. These varieties are farmers’ preferred choices in the region. The seeds were sourced from IITA Kano Station. TGX-1835-10E is an extra-early-maturing variety that has a strong resistance to rust, with a potential yield of 1.5–2 t ha⁻¹. TGX-1951-3F is an early-maturing variety with high tolerance to rust diseases and has a potential yield of 2.5 t ha⁻¹. TGX-1904-6F is a medium-maturing and dual-purpose variety which is cultivated for both grain and fodder. It is resistant to lodging, with a potential yield of 2–2.5 t ha⁻¹. All the varieties are resistant to cercospora leaf spot and bacterial pustule and nodulate freely with indigenous rhizobia bacteria. The three varieties and six planting dates were laid out in a split-plot factorial design at Samaru Zaria and BUK-Kano. The experimental field was disc-harrowed and ridged before planting. Each plot consisted of 4 rows of 5 m length, with 4 rows spaced at 0.75 m apart giving a 3 × 5 m² experimental unit [Figure 3]. Intra-row spacing was 10 cm. Six seeds were planted with an intra-row spacing of 10 cm and later thinned to four plants per hill, resulting in a final population of 533,333 plants ha⁻¹. Additionally, 40 kg ha⁻¹ of K₂O and P₂O₅, in the form of muriate of potash and triple super phosphate (TSP), were applied to all plots at the time of planting. The previous crop was soybean, and all the residue was removed from the field before planting.

2.4. Data Collection

The data were collected from two middle rows representing the net plot (1.5 × 4.5 m²), the two outer rows and a 25 cm distance at the end of each row were left as borders in each replicate. Observations were recorded on growth components (leaf area index and total dry matter) and reproductive and grain yield components (flowering, maturity, 1000-seed weight, number of pods m⁻², number of seed m⁻², grain yield, and harvest index). The number of days to 50% flowering was recorded when half of the plants in a plot had flower. The number of days to physiological maturity was recorded when about 85–90 percent of the pods had dried. Leaf area and intercepted PAR were measured at full flowering (R2) stage using AccuPAR model LP-80 PAR/LAI Ceptometer (Decagon Devices, Inc., Pullman, WA, USA). At harvest, all plants from the quadrat measuring 1 m⁻² were used to estimate the number of pods, seeds, and dry matter. The plants were harvested and separated into pods, stems, and leaves. The pods containing seeds were threshed to separate the seeds. All plant parts, including the leaves collected from the quadrat, were dried in an oven at 60 °C for 76 h until a constant weight was achieved. The dried materials were then weighed and combined to determine the total dry matter. For grain yield determination, the pods from plants in the two middle rows were harvested, sundried, and hand threshed. The grains collected from the quadrat were then combined with those from the two middle rows, and the final grain yield was expressed in kg ha⁻¹. The moisture content of grain samples from each plot was determined using a portable moisture meter (Farmex MT-16, Voluntari, Romania). Final grain yield was adjusted to 12% moisture content, calculated using the relationship below:

Grain yield = grain weight per net plot (kg) × ((100 − MC)⁄88) × (10,000 m²)⁄7.5 m²

MC = Moisture content.

2.5. Data Analysis

The data collected were subjected to an analysis of variance (ANOVA) using SAS for Windows Release 9.4 [38], according to the split-plot factorial design. The SAS mixed model procedure was used for the analysis. Year and replication were treated as random effects and the planting date and variety as fixed effects in determining the expected mean square and appropriate F-tests in the ANOVA. The differences among individual treatments were separated using the least significant difference (LSD) test at the 0.05 probability level.

3. Results

3.1. Analysis of Variance

In both locations, the year had a highly significant effect on almost all parameters, except for days to 50% flowering at BUK-Kano and IPAR and LAI at Samaru Zaria [

Table 2. The ANOVA results for the effects of year, planting dates, and soybean variety on soybean parameters conducted for two years at Bayero University BUK-Kano and Samaru Zaria.

Effect	D50flw	Mat	IPAR	LAI	1000sdwt	Seedno_m−2	TDM	GY
	Bayero university BUK-Kano
Year	0.5592	<0.0001	<0.0001	<0.0001	<0.0001	<0.0001	0.0002	<0.0001
Variety	<0.0001	<0.0001	0.4209	0.0142	<0.0001	<0.0001	0.0733	<0.0001
Year × Variety	0.0043	<0.0001	0.015	<0.0001	0.0003	<0.0001	0.1607	<0.0001
Pdate	0.0778	<0.0001	<0.0001	<0.0001	<0.0001	<0.0001	<0.0001	<0.0001
Year × Pdate	0.8502	0.0008	0.0042	0.0053	0.0614	0.0437	0.5518	<0.0001
Variety × Pdate	0.4281	0.0014	0.0901	0.024	0.0329	0.0003	0.1017	<0.0001
Year × Variety × Pdate	0.2002	0.4288	0.1563	0.1073	0.0083	0.1664	0.85	<0.0001
	Samaru Zaria
Year	0.2828	<0.0001	0.5817	0.2188	<0.0001	<0.0001	<0.0001	<0.0001
Variety	<0.0001	<0.0001	<0.0001	0.0003	<0.0001	0.0004	<0.0001	<0.0001
Year × Variety	0.0007	0.0106	<0.0001	0.1935	0.0004	0.5172	0.789	0.9236
Pdate	<0.0001	<0.0001	<0.0001	<0.0001	<0.0001	<0.0001	<0.0001	<0.0001
Year × Pdate	0.055	<0.0001	0.4534	0.0507	0.8412	0.6499	0.4756	0.6324
Variety × Pdate	0.0638	<0.0001	0.0048	0.1034	0.002	<0.0001	0.4333	<0.0001
Year × Variety × Pdate	0.0797	0.0032	0.3345	0.1179	0.507	0.9119	0.9947	0.4723

D50flw = days to 50% flowering, Mat = days to 95% maturity, IPAR = intercepted photosynthetically active radiation, LAI = leaf area index, TDM = total dry matter, GY = grain yield, 1000sdwt = 1000-seed weight (g), and seedno_m⁻² = Seed number (m⁻²).

]. Variety significantly influenced all parameters in both locations except for IPAR at BUK-Kano. The Year × Variety interaction was significant for most parameters except total dry matter (TDM) at both locations, and for seed number m⁻² and LAI at Samaru Zaria. Similarly, the planting date had a significant effect on all parameters across both locations except days to 50% flowering at BUK-Kano. The Year × Planting Date interaction was only significant for days to physiological maturity at Samaru Zaria, while LAI, seed number m⁻², and grain yield were significant at BUK-Kano. The Variety × Planting Date interaction was significant for the number of days to physiological maturity, LAI, 1000-seed weight, seed number m⁻², and grain yield at both locations. The Year × Variety × Planting Date interaction showed less significance, affecting mainly grain yield at BUK-Kano and the number of days to physiological maturity at Samaru Zaria.

3.2. Phenology

The effect of year on number of days to 50% flowering was not significant at both locations [

Table 3. Year effect on soybean parameters conducted for two years at Bayero University BUK-Kano and Samaru Zaria.

Effect	D50flw	Mat	IPAR	LAI	1000sdwt	Seedno_m−2	TDM	GY
Bayero university BUK-Kano
YEAR
2021	50.2	105.6	71.6	4.5	73.0	1113.5	4866.0	1099.4
2022	50.6	107.9	82.3	5.0	115.0	2051.9	5978.6	1438.9
LSD (5%)	1.62 ns	0.49 **	3.00 **	0.22 **	0.57 **	165.47 **	479.84 **	55.34 **
Samaru Zaria
2021	51.6	103.9	87.1	5.4	11.3	1467.5	5153.6	1672.6
2022	51.3	107.5	86.0	5.2	12.1	1279.5	6129.7	1761.6
LSD (5%)	0.59 ns	0.95 **	1.89 ns	0.23 ns	0.32 **	132.68 **	226.05 **	64.39 **

D50flw = days to 50% flowering, Mat = days to 95% maturity, IPAR = intercepted photosynthetically active radiation, LAI = leaf area index, TDM = total dry matter, GY = grain yield, 1000sdwt = 1000-seed weight (g), and seedno_m⁻² = Seed number (m⁻²), ** = Highly significant at 5% level probability, ns = not significant

]. The number of days to 50% flowering varied among varieties at both locations [

Table 4. Interactive effects of variety and planting dates on days to 50% flowering and physiological maturity in soybeans at two locations in the Sudan and northern Guinea savannas of Nigeria.

	Bayero University BUK-Kano				Samaru Zaria
	Days to 50% Flowering
Pdate	TGX-1835-10E	TGX-1904-6F	TGX-1951-3F	Mean	TGX-1835-10E	TGX-1904-6F	TGX-1951-3F	Mean
PD1	46.7	56.3	51.5	51.5	47.2	57.8	54.9	53.3
PD2	47.5	56.2	51.8	51.8	47.0	57.8	54.1	53.0
PD3	47.2	56.0	51.3	51.5	46.5	55.7	52.1	51.4
PD4	45.3	55.5	50.0	50.3	45.0	55.7	51.8	50.8
PD5	45.3	50.7	50.0	48.7	44.8	54.9	52.3	50.7
PD6	40.3	55.8	49.8	48.7	45.3	53.7	49.7	49.6
Mean	45.4	55.1	50.7		46.0	55.9	52.5
LSD(5%)pd	2.81 ns				0.89 **
LSD(5%)var	1.99 **				0.68 **
LSD(5%)pdxvar	2.07 ns				1.47 ns
	Days to 95% Maturity
PD1	97.8	119.7	111.0	109.5	97.7	120.1	117.3	111.7
PD2	97.7	118.5	110.8	109.0	95.7	114.2	111.4	107.1
PD3	96.5	114.3	109.3	106.7	94.8	112.0	111.4	106.1
PD4	95.7	113.7	108.2	105.8	94.5	110.9	109.9	105.1
PD5	95.3	113.8	107.3	105.5	93.7	108.1	104.8	102.2
PD6	93.8	113.0	105.0	103.9	93.5	107.5	105.4	102.1
Mean	96.1	115.5	108.6		95.0	112.1	110.0
LSD(5%)pd	0.84 **				1.21 **
LSD(5%)var	0.60 **				1.02 **
LSD(5%)pdxvar	1.46 **				1.78 **

PD1 = 15 June, PD2 = 22 June, PD3 = 29 June, PD4 = 6 July, PD5 = 13 July, PD6 = 22 June. ** = Highly significant at 5% level probability, ns = not significant.

]. TGX-1904-6F had a higher mean of 55 days at BUK-Kano and 60 days at Samaru Zaria than other varieties. Across varieties, the number of days to 50% flowering was significantly affected by planting dates in Samaru Zaria, with early planting window 15–22 June resulting in a higher number of days to flowering than later planting dates (13–20 July) across varieties. Beyond this planting date (22 June), the number of days to 50% flowering was reduced by 2–3 days across varieties. At BUK-Kano, planting dates did not have significant effects on the number of days to flowering. Across planting dates, the number of days to 50% flowering varied from 45–55 days at BUK-Kano and 46–56 days at Samaru Zaria. The interaction between the planting date and variety was not significant at both locations, suggesting that the varieties responded similarly to the different planting dates.

The number of days to physiological maturity significantly varied between the two years at both locations [Table 3]. In 2022, the number of days to physiological maturity was significantly higher than that of 2021 at both locations, with an increase of 2 days at BUK-Kano and 3 days at Samaru Zaria. The number of days to physiological maturity was higher for TGX-1904-6F (mean of 116 days at BUK-Kano and mean of 112 days at Samaru Zaria) compared to other varieties at both locations [Table 4]. The number of days to maturity was also significantly higher for TGX-1951-3F (109 days at BUK-Kano and 110 days at Samaru Zaria) than that of TGX-1835-10E (96 days at BUK-Kano and 95 days at Samaru Zaria). Across planting dates, the number of days to maturity was significantly influenced by the planting date in both locations, with early planting from 15–22 June at BUK-Kano and 15 June at Samaru Zaria, resulting in a higher number of days to physiological maturity than other planting dates. Beyond these dates, the number of days to maturity decreased by 2–5 days at BUK-Kano and 4–9 days at Samaru Zaria. The interaction between variety and planting dates was significant at both locations. At BUK-Kano, the number of days to physiological maturity for TGX-1835-10E, TGX-1904-6F, and TGX-1951-3F were 98, 120, and 111 days when planted on 15–22 June (PD1-PD2). Delayed planting (29 June–20 July) reduced the number of days to maturity by 1–4 for TGX-1835-10E, 4–5 for TGX-1904-6F, and 1–5 for TGX-1951-3F. A similar trend was observed at Samaru Zaria, with early planting recording a higher number of days to maturity and late planting recording the lowest across all varieties. At Samaru Zaria, the number of days to maturity were 98, 120, and 117 for TGX-1835-10E, TGX-1904-6F, and TGX-1951-3F when planted on 15 June (PD1). Delayed planting reduced the number of days to maturity by 2–4 days for TGX-1835-10E, 5–11 days for TGX-1904-6F, and 5–10 days for TGX-1951-3F.

3.3. Intercepted Phtosynthetic Active Radiation and Leaf Area Index

The impact of year on intercepted photosynthetic active radiation (IPAR) was significant at BUK-Kano but not at Samaru Zaria [Table 3]. At BUK-Kano, IPAR was 15% higher in 2022 than 2021. IPAR did not significantly differ among varieties at BUK-Kano, but differences were significant at Samaru Zaria [

Table 5. Interactive effects of variety and planting dates on intercepted photosynthetically active radiation and leaf area index in soybeans at two locations in the Sudan and northern Guinea savannas of Nigeria.

	Bayero University BUK-Kano				Samaru Zaria
	IPAR (%)
Pdate	TGX-1835-10E	TGX-1904-6F	TGX-1951-3F	Mean	TGX-1835-10E	TGX-1904-6F	TGX-1951-3F	Mean
PD1	86.9	89.7	90.5	89.0	87.9	94.1	95.9	92.6
PD2	84.0	89.5	86.8	86.8	84.3	92.4	92.8	89.8
PD3	80.4	79.5	82.5	80.8	77.6	87.5	88.2	84.4
PD4	74.4	75.6	72.7	74.2	74.1	77.4	83.0	78.2
PD5	69.2	68.3	65.4	67.6	70.4	67.5	73.7	70.5
PD6	66.7	64.5	62.1	64.4	68.5	68.5	68.8	68.6
Mean	76.9	77.8	76.7		77.1	81.2	83.7
LSD(5%)pd	2.70 **				2.66 **
LSD(5%)var	1.97 ns				1.88 **
LSD(5%)pdxvar	4.57 ns				4.61 **
	Leaf Area Index
PD1	5.7	5.7	6.2	5.8	5.9	6.8	6.7	6.5
PD2	5.7	5.4	5.7	5.6	5.8	6.5	7.1	6.5
PD3	5.3	4.7	4.1	4.7	5.3	5.6	6.1	5.6
PD4	4.8	4.8	4.1	4.6	5.1	5.4	5.1	5.2
PD5	4.1	4.5	3.7	4.1	4.0	4.2	4.9	4.4
PD6	3.8	4.1	3.4	3.7	3.8	3.5	3.7	3.7
Mean	4.9	4.9	4.5		5.0	5.3	5.6
LSD(5%)pd	0.38 *				0.41 **
LSD(5%)var	0.27 **				0.29 **
LSD(5%)pdxvar	0.65 *				0.70 ns

PD1 = 15 June, PD2 = 22 June, PD3 = 29 June, PD4 = 6 July, PD5 = 13 July, PD6 = 22 June. ** = Highly significant at 5% probability, * = Significant at 5% level probability, ns = not significant.

]. Higher IPAR was recorded for TGX-1951-3F (83.7%), followed by TGX-1904-6F (81.2%), and the lowest for TGX-1835-10E (77.1%). The IPAR was significantly lower with delayed planting at both locations. Early planting recorded higher IPAR (89% for BUK-Kano and 92.6% for Samaru Zaria) when planted early on 15 June (PD1). Delayed planting beyond PD1 reduced IPAR by 2% for PD2, 9% for PD3, 16% for PD4, 24% for PD5, and 28% for PD6 at BUK-Kano. At Samaru Zaria, IPAR declined by 3, 9, 19, 24, and 26%, respectively, for PD2, PD3, PD4, PD5, and PD6. The interaction between the planting date and variety was significant at Samaru Zaria but not at BUK-Kano. At Samaru Zaria, higher IPAR values were recorded for TGX-1951-3F, TGX-1904-6F, and TGX-1835-10E when planted on 15 June (PD1), with a decline of 3–23% for TGX-1835-10E, 11–28% for TGX-1904-6F, and 4–31% for TGX-1951-3F with delayed planting beyond 15 June.

Year significantly influenced the leaf area index (LAI) at BUK-Kano but not in Samaru Zaria [Table 3]. Higher LAI (5.0) was recorded in 2022 than in 2021 (4.5). LAI varied among varieties at both locations [Table 5]. TGX-1835-10E and TGX-1904-6F exhibited higher LAI values of 4.9 compared to TGX-1951-3F, which had an LAI of 4.5 at BUK-Kano. At Samaru Zaria, TGX-1951-3F had a higher LAI value of 5.6, closely followed by TGX-1904-6F (5.3) and the lowest for TGX-1835-10E (5.0). With delayed planting beyond June 15, LAI declined by 3–36% at BUK-Kano and 13–43% at Samaru Zaria across varieties. The interaction between planting dates and variety was significant in BUK-Kano but not in Samaru Zaria. When planted on 15 June (PD1) at BUK-Kano, LAI was 6.2 for TGX-1951-3F and 5.7 for both TGX-1904-6F and TGX-1835-10E. With a delay in planting, LAI declined by 7–33% for TGX-1835-10E, 5–28% for TGX-1904-6F, and 8–45% for TGX-1951-3F. Delaying planting to July 20 (PD6) gave the least LAI of 3.9, 4.9, and 4.5 for TGX-1835-10E, TGX-1904-6F, and TGX-1951-3F, respectively.

3.4. Seed Weight and Number of Seeds m⁻²

The effect of year on the 1000-seed weight was significant at both locations [Table 3]. In 2022, a higher 1000-seed weight was recorded compared to 2021 at both locations, with an increase of 58% at BUK-Kano and 7% at Samaru Zaria. Seed weight was significantly affected by planting dates and varieties at both locations [

Table 6. Interactive effects of variety and planting dates on 1000-seed weight and number of seeds m⁻² in soybeans at two locations in the Sudan and northern Guinea savannas of Nigeria.

	Bayero University BUK-Kano				Samaru Zaria
	1000-Seed Weight (g)
Pdate	TGX-1835-10E	TGX-1904-6F	TGX-1951-3F	Mean	TGX-1835-10E	TGX-1904-6F	TGX-1951-3F	Mean
PD1	117	108	121	115	132	119	148	133
PD2	113	82	117	104	129	109	129	122
PD3	107	92	92	97	120	104	129	118
PD4	108	89	95	97	119	104	129	117
PD5	88	63	91	81	104	101	121	109
PD6	84	44	85	71	96	100	114	103
Mean	103	79	10		117	10.6	12.8
LSD(5%)pd	9.9 **				5.0 **
LSD(5%)var	7.0 **				3.8 **
LSD(5%)pdxvar	17.1 *				8.2 ns
	Seed number (m−2)
PD1	2128.8	2415.9	2287.8	2277.5	2163	2795.6	2710.1	2556.3
PD2	2148	1723.9	2265.2	2045.7	2131	2349.7	2458	2312.9
PD3	2072.6	1424.9	1542.6	1680	1969.7	1673.2	2082.9	1908.6
PD4	1692.4	1310.6	1376.9	1460	1683.8	1605.8	1578.6	1622.7
PD5	1331.6	925.7	1137.6	1131.6	1353.2	1131.8	1333.8	1272.9
PD6	942.9	678.9	1082.8	901.5	1097.9	697.3	1004.4	933.2
Mean	1719.4	1413.3	1615.5		1733.1	1708.9	1861.3
LSD(5%)pd	178.45 **				118.95 **
LSD(5%)var	165.47 **				90.87 **
LSD(5%)pdxvar	314.36 **				194.21 **

PD1 = 15 June, PD2 = 22 June, PD3 = 29 June, PD4 = 6 July, PD5 = 13 July, PD6 = 22 June. ** = Highly significant at 5% level probability, * = Significant at 5% probability, ns = not significant.

]. A higher 1000-seed weight was recorded for TGX-1835-10E compared to the other varieties at BUK-Kano, while at Samaru Zaria, TGX-1951-3F had a higher 1000-seed weight than that of the other varieties. The least seed weight was recorded for TGX-1904-6F at both locations. Across varieties, seed weight decreased with delayed planting beyond 15 June (PD1) by 10–30% at BUK-Kano and 8–23% at Samaru Zaria. The interaction between the planting dates and variety was significant at BUK-Kano, with TGX-1951-3F producing higher seed weight (121 g) when planted early on 15 June (PD1). When planted beyond this date, seed weight decreased by 3–28% for TGX-1835-10E, 24–59% for TGX-1904-6F, and 3–30% for TGX-1951-3F.

The impact of the year on number of seeds m⁻² was significant at both locations [Table 2]. The number of seeds m⁻² was 84% higher in 2022 than in 2021 at BUK-Kano, while at Samaru Zaria, the number of seeds was 13% lower in 2022. The number of seeds m⁻² was significantly affected by the planting dates and variety at both locations [Table 6]. TGX-1835-10E had a higher number of seeds m⁻² than those of the other varieties at BUK-Kano. While at Samaru Zaria, TGX-1951-3F had a higher number of seeds m⁻² than that of the other varieties. TGX-1904-6F produced the least number of seeds at both locations. Across varieties, the number of seeds m⁻² decreased by 10–60% at BUK-Kano and 10–63% at Samaru Zaria with delayed planting beyond 15 June. The interaction between the planting dates and variety was significant at both locations, with TGX-1904-6F producing higher seeds m⁻² than that of the other varieties when planted early on 15 June (PD1) at both locations. Delayed planting beyond PD1 reduced the number of seeds m⁻² by 4–56% for TGX-1835-10E, 29–72% for TGX-1904-6F, and 32–52% for TGX-1951-3F at BUK-Kano, and 8–48% for TGX-1835-10E, 16–75% for TGX-1904-6F, and 9–63% for TGX-1951-3F at Samaru Zaria.

3.5. Total Dry Matter and Grain Yield

The impact of year on total dry matter (TDM) was significant at both locations [Table 2]. TDM was 23% and 19% higher in 2022 than in 2021 at BUK-Kano and at Samaru Zaria, respectively. The TGX-1951-3F variety produced a total dry matter (TDM) that was higher than that of the other varieties at Samaru Zaria, and there was no significant difference among varieties at BUK-Kano [

Table 7. Interactive effects of variety and planting dates on total dry matter and grain yield in soybeans at two locations in the Sudan and northern Guinea savannas of Nigeria.

	Bayero University BUK-Kano				Samaru Zaria
	Total Dry Matter (kg ha−1)
Pdate	TGX-1835-10E	TGX-1904-6F	TGX-1951-3F	Mean	TGX-1835-10E	TGX-1904-6F	TGX-1951-3F	Mean
PD1	5870.1	6977.9	6453.4	6433.8	6042.4	7073	7133.5	6749.6
PD2	5582.8	5895.3	6260.3	5912.8	5675.9	6612	6537	6275
PD3	5245.3	5406.8	5449.3	5367.1	4885.9	5806.9	5588	5426.9
PD4	4648.7	4072.4	4605.2	4442.1	4433.9	4761.8	5021	4738.9
PD5	3795.8	3370.9	3706.5	3624.4	3768.7	4118.7	4457.4	4114.9
PD6	2837	3021.7	3531.6	3130.1	3393.5	3625.1	3847.3	3622.0
Mean	4663.3	4790.8	5001.1		4700.1	5332.9	5430.7
LSD(5%)pd	410.26 **				318.28 **
LSD(5%)var	290.11 ns				229.24 **
LSD(5%)pdxvar	710.61 ns				544.29 ns
	Grain Yield (kg ha−1)
PD1	1938.9	1838.9	2041.8	1939.9	1947.2	2378.4	2351.5	2225.7
PD2	1996.0	1341.9	2084.3	1807.4	1885.2	2054.9	2145.6	2028.6
PD3	1626.2	1113.3	1539.8	1426.5	1787.9	1599.9	1880.2	1756.0
PD4	1423.6	877.0	1127.2	1142.6	1609.2	1446.1	1525.3	1526.9
PD5	971.5	325.8	662.9	653.4	1312.2	1176.2	1334.3	1274.2
PD6	736.5	172.3	565.5	491.4	1103.8	943.1	1218.5	1088.5
Mean	1448.8	944.9	1336.9		1607.6	1599.8	1742.6
LSD(5%)pd	84.63 **				93.50 **
LSD(5%)var	62.95 **				68.97 **
LSD(5%)pdxvar	141.29 **				157.12 **

PD1 = 15 June, PD2 = 22 June, PD3 = 29 June, PD4 = 6 July, PD5 = 13 July, PD6 = 22 June. ** = Highly significant at 5% level probability, ns = not significant.

]. Planting dates significantly affected TDM at both locations. When planted beyond 15 June (PD1), TDM declined by 8–51% at BUK-Kano and 7–46% at Samaru Zaria as the planting dates progressed from 15 June to 20 July (PD1-PD6). There was no interaction between the planting dates and variety at either location. However, the TGX-1904-6F variety had the highest decline and the TGX-1951-3F variety had the least decline with delay in planting.

The year had a significant impact on grain yield at both locations [Table 2]. Specifically, grain yield was 31% higher in 2022 than in 2021 at BUK-Kano and 5% higher in 2022 than in 2021 at Samaru Zaria. Grain yield varied across planting dates and varieties at both locations [Table 7]. Higher grain yield was recorded for TGX-1835-10E (1448.8 kg ha⁻¹) than that of other varieties at BUK-Kano with least yield recorded for TGX-1904-6F (944.9 kg ha⁻¹). At Samaru Zaria, grain yield was TGX-1951-3f (1743 kg ha⁻¹) > TGX-1835-10E (1608 kg ha⁻¹) > TGX-1904-6F (1600 kg ha⁻¹). Grain yield decreased by 9–75% at BUK-Kano and 9–51% at Samaru Zaria with delayed planting for all the varieties as the planting dates progressed from 15 June to 20 July (PD1-PD6). At BUK-Kano, TGX-1951-3F had a higher grain yield when planted early, between 15 and June 22 (PD1-PD2), compared to other varieties. Delayed planting beyond 22 June (PD2) decreased yields by 19–63% for TGX-1835-10E, 27–90% for TGX-1904-6F, and 26–73% for TGX-1951-3F. At Samaru Zaria, TGX-1904-6F had the highest grain yield which was closely followed by TGX-1951-3F when planted early on 15 June (PD1), with yield declines of 3–43% for TGX-1835-10E, 13–60% for TGX-1904-6F, and 9–48% for TGX-1951-3F with delayed planting beyond this date.

Generally, days to 50% flowering, days to maturity, IPAR, LAI, 1000-seed weight, seed number, and total dry matter exhibited positive and significant correlations with grain yield across various planting dates in both locations [

Table 8. Correlation coefficients of characters with grain yield for each variety across the planting date.

	Bayero University BUK-Kano			Samaru Zaria
Parameters	TGX-1951-3F	TGX-1835-10F	TGX-1904-6F	TGX-1951-3F	TGX-1835-10F	TGX-1904-6F
D50flw	0.34 *	0.22 ns	0.34 *	0.55 **	0.44 **	0.82 **
Mat	0.85 **	0.64 **	0.73 **	0.81 **	0.64 **	0.82 **
IPAR	0.86 **	0.74 **	0.90 **	0.83 **	0.75 **	0.83 **
LAI	0.84 **	0.71 **	0.87 **	0.85 **	0.80 **	0.86 **
1000sdwt	0.74 **	0.63 **	0.65 **	0.73 **	0.88 **	0.62 **
Seedno_m−2	0.79 **	0.84 **	0.79 **	0.98 **	0.97 **	0.98 **
TDM	0.86 **	0.80 **	0.88 **	0.90 **	0.86 **	0.89 **

D50flw = days to 50% flowering, Mat = days to 95% maturity, IPAR = intercepted photosynthetically active radiation, LAI = leaf area index, TDM = total dry matter, 1000sdwt = 1000-seed weight (g), seedno_m⁻² = seed number (m⁻²), * = significant at 5% level probability, ** = highly significant at 5% level probability, ns = not significant.

]. At BUK-Kano, the days to 50% flowering (D50flw), days to 95% maturity (Mat), intercepted photosynthetically active radiation (IPAR), leaf area index (LAI), total dry matter (TDM), and 1000-seed weight positively and significantly correlated (R value) with grain yield. Among these traits, the strongest correlations were observed for days to maturity, IPAR, LAI, and TDM, suggesting their potential as reliable indicators for influencing grain yield. Similarly, for the TGX-1835-10F and TGX-1904-6F at BUK-Kano, positive and significant correlations were observed for most traits with grain yield, suggesting that these traits had a significant impact on grain yield formation across different planting dates. At Samaru Zaria, consistent positive and significant correlations were observed between all examined traits (D50flw, days to maturity, IPAR, LAI, 1000-seed weight, number of seeds m⁻², and TDM) and grain yield across the planting dates for all varieties. High significant correlations were observed for D50flw, days to maturity, and TDM, indicating their crucial roles in influencing grain yield formation in this variety.

4. Discussion

Soybean is progressively emerging as a key economic crop in the savannas of Northern Nigeria. However, its yield is constrained by inadequate soil fertility and unpredictable rainfall patterns caused by climate change. Identifying the best crop management practices, including optimal planting times, can offer valuable insights for strategic planning in the region.

Our results reveal the significant impact of planting dates on the performance of soybean varieties in the Nigerian savannas. The growth and yield of soybeans were influenced by location and year, which could be attributed to soil types, rainfall patterns, temperature and solar radiation across AEZs [39]. There were significant differences in soybean performance between the two years. Soybean performance was generally better in 2022 than in 2021. The differences observed between the two years are attributed to the differences in rainfall amount and distribution. For example, BUK-Kano received 557 mm of rainfall in 2021 and 878 mm in 2022, with a higher amount of rainfall in October in 2022 than in 2021. Though rainfall amount was almost similar for both years (994 mm in 2021 and 972 mm in 2022) at Samaru Zaria, the distribution was better in 2022. Rainfall was significantly higher in October 2022 than in October 2021 which shows that there was enough rainfall in October 2022 to support soybean growth. The growth and yield of soybeans were also affected by location. For instance, flowering was hastened in BUK-Kano then in Samaru Zaria, probably due to the higher temperatures in this AEZ than those of the NGS AEZ.

Grain and yield components such as number of seeds m⁻² and 1000-seed weight also showed significant variation between the two locations. For example, number of seeds m⁻², 1000-seed weight, TDM, and grain yield were generally higher at Samaru Zaria than those at BUK-Kano. This may also be attributed to better soil and climatic conditions at Samaru Zaria than BUK-Kano. Rainfall at Samaru Zaria in the NGS AEZ is higher than at BUK-Kano in the SS AEZ. Also, the soils at Samaru Zaria have higher clay, organic carbon, and N content than BUK-Kano soils [Table 1]. These findings are consistent with those of previous studies by Bebeley et al. [11], who simulated higher soybean grain yields in Samaru Zaria than in other locations because of its suitable weather conditions and the good soil fertility. Similarly, Tofa et al. [10] also simulated higher maize yields in Samaru Zaria than in other regions. Varietal differences affected the growth and yield parameters, indicating genetic variability among the soybean varieties studied. For example, TGX-1904-6F generally took longer to flower and mature compared to TGX-1835-10E and TGX-1951-3F in both locations. In contrast, TGX-1835-10E consistently flowered earlier than the others across all planting dates (15 June to 20 July). TGX-1904-6F is a medium-maturing variety and takes longer to flower than the early-maturing TGX-1951-3F and TGX-1835-10E. The LAI and IPAR were significantly influenced by variety, with TGX-1951-3F having higher values in both locations consistent with the findings of Shegro et al. [28].

The TGX-1951-3F variety consistently showed higher TDM with early planting (15 June–22 June) at both locations. TGX-1904-6F also performed well in earlier planting dates, while TGX-1835-10E generally had the lowest TDM at early planting. The lower total dry matter of TGX-1835-10E may be attributed to its early maturity, which limits its biomass production [8]. The higher TDM observed in TGX-1951-3F could be due to its genetic potential for greater biomass production and better adaptation to local conditions. TGX-1835-10E had the highest grain yield at BUK-Kano, followed by TGX-1951-3F and TGX-1904-6F. The higher yield of TGX-1835-10E may be due to the shorter growing season at BUK-Kano which lies in the SS AEZ. This could have impacted the performance of the medium/late-maturing varieties. The early-maturing TGX-1835-10E variety had already completed the pod-filling stage before the cessation of rain, resulting in a higher yield than those of TGX-1951-3F and TGX-1904-6F. This agrees with the previous studies of Kamara et al. [12], which showed that early-maturing variety yielded higher than medium and late varieties in the SS AEZ of Nigeria. At Samaru Zaria, TGX-1951-3F consistently produced the highest yield, followed by TGX-1904-6F and TGX-1835-10E. This suggests that the early-maturing TGX-1835-10E variety is best suited to the SS AEZ, where the growing season is short and the rainfall is erratic. The results also show that the medium-maturing TGX-1904-6F variety is best suited in AEZs such as the NGS with a longer growing season and higher rainfall for optimum productivity. From our findings, TGX-1951–3F outperformed the other soybean varieties because it is a robust, early-maturing variety that is less photosensitive and well-suited to both the NGS and SS AEZ of West Africa, where the rainy season typically begins in late June and ends in early October. This is consistent with previous findings of Bebeley et al. [11] who simulated higher yield for TGX-1951–3F over a range of planting windows than for other soybean varieties in Northwest Nigeria. Similarly, Kamara et al. [12] also simulated higher yield of TGX-1951-3F than TGX-1448-2E in Northeast Nigeria. In a field experiment, Sadiq et al. [27] also obtained high yield of TGX-1951-3F in Kubwa and Samaru Zaria respectively. The TGX-1951-3F variety is, therefore, well suited to the savannas of West Africa, where the growing season is becoming shorter due to climate change. In periods of early rainfall cessation, this variety is expected to outperform late-maturing varieties, as reported by Kamara et al. [12], in Northeast Nigeria.

Changes in planting dates are known to influence soybean yield. Our study reveals that the planting dates significantly affected the growth and yield of soybeans. The findings show that early planting between 15–29 June in SS and between 15 June–6 July in NGS results in higher yield and yield components due to better utilization of the growing season. This is consistent with the findings of Shegro et al. [28], who reported that early planting dates produced a higher seed yield than late planting in Ethiopia. Conversely, delays in planting beyond the optimal period led to significant declines in performance at both locations. This corroborates the findings of Yagoub and Hamed [22], who reported that delayed planting resulted in a reduction in yield and its components of soybean in Sudan. Our study shows that planting early in 15–22 June increased the number of days to 50% flowering and maturity for all varieties at both locations. This may be as a result of variations in solar radiation, which is critical in influencing the growth and development of the soybean varieties because they are photosensitive. Early planting in West Africa, when the day lengths are longer until July, prolongs the number of days to flowering of photosensitive plants like soybean and cowpea. Setiyono et al. [40] demonstrated that soybeans are sensitive to changes in day length, which significantly influences the timing of flowering and maturity.

Delayed flowering and maturity associated with early planting also led to increased higher leaf area index (LAI), higher IPAR, and biomass accumulation, ultimately enhancing yield potential. In this study, early planting (15–22 June) resulted in higher LAI at both locations. This agrees with previous studies by Shegro et al. [28] who reported that early planting dates produced a higher LAI for soybeans than late planting in Ethiopia. This may be as a result of exposing the plant to a longer growing season, which promotes greater leaf area development and canopy coverage. A similar trend was observed for intercepted photosynthetic active radiation (IPAR), as early planting increased the IPAR of the soybean varieties compared to late planting (13–20 July) at both locations. These findings may be due to reduced cloud cover and longer day length, allowing more sunlight to reach the crops and resulting in increased IPAR. There was a significant decline in total dry matter (TDM) as planting was delayed beyond 22 June across varieties at both locations. Delayed planting may shorten the vegetative phase due to the shortened growing season, thereby limiting the time available for vegetative growth and biomass accumulation [41]. This result is in agreement with previous studies by Ennin et al. [42] who reported a significant decline in total dry matter as a result of a delay in planting in Ghana.

The timing of soybean planting also significantly influences yield components like 1000-seed weight and seed number m⁻², which together contribute to overall grain yield. Our results show that early planting (15–22 June) enhances these parameters, leading to higher yields. Increase in the number of seeds m⁻² and 1000-seed weight for the early-planted soybean may likely be due to the soybean varieties taking full advantage of the growing season which can result in a longer period for vegetative and reproductive growth, leading to better seed development and higher seed weight. This result is in agreement with previous studies by Ibrahim [21], who reported that early planting increases soybean yield components in Sudan.

Our results revealed that the optimal planting dates were dependent on location and soybean variety. In the SS AEZ, planting the early-maturing TGX-1835-10E and TGX-1951-3F soybean varieties between 15–29 June can achieve a desirable yield of >1500 kg ha⁻¹. Delaying planting beyond 29 June reduces grain yield of TGX-1835-10E by 27–63% and TGX1951-3F by 12–55%. For TGX-1904-6F, planting after 15 June reduces yields by 27–90%. This finding corroborates the results of Bebeley et al. [11] who simulated 11–70% yield reduction when planting TGX-1835-10E beyond 28 June and 8–69% yield reduction in TGX 1951-3F when planting beyond 21 June. Due to the short planting window and the unpredictability of rainfall at the start of the season, planting the medium-maturing TGX-1904-6F variety in the SS AEZ is not advisable. This is in agreement with the long-term simulation results of Bebeley et al. [11]. Farmers should focus on early-maturing varieties to avoid significant yield reductions. Meanwhile, at Samaru Zaria in the NGS AEZ, a desirable yield of >1500 kg ha⁻¹ are achievable when TGX-1835-10E and TGX-1951-3F are planted between 15 June–6 July and when TGX1904-6F is planted in 15–29 June. The yield of TGX-1835-10E and TGX-1951-3F will decline by 18–31% and 12–20%, respectively, when planting is carried out beyond 6 July. Planting TGX1904-6F beyond 29 June will reduce yield by 10–41%. There was less yield reduction in TGX-1835-10E and TGX-1951-3F compared to TGX-1904-6F when planted beyond 6 July due to moderate rainfall and a longer growing season. Samaru Zaria was found to be more favorable for soybean production due to the longer length of growing period and its higher organic matter and clay contents, resulting in high water retention and nutrient content. Other authors [10,11] reported that Samaru Zaria has a long growing season and favorable soil conditions for grain crop production. This reduction in yield when planting is conducted beyond the optimum planting date may be due to drought stress resulting from insufficient moisture before the crop completes its life cycle in October. In the Nigerian savannas, late-season rainfall is unpredictable [10] and may stop in September or October before the crops mature, leading to substantial yield losses [11,12]. Providing farmers with information on optimal planting dates to avoid end-of-season drought would be beneficial. While our results provide short-term information on optimum planting dates for these locations, the results cannot be extrapolated to other locations or regions in the Nigeria savannas. This is more due to heterogeneity of soils and differences in microclimates in the face of climate change. To overcome these limitations, we recommend long-term simulation studies using biophysical and multiple-year weather information from other regions.

5. Conclusions

Our study reveals that the performance of the soybeans was dependent on year, AEZ, variety, and planting dates. The growth and yield of all the soybean varieties were better in 2022 than in 2021 in both locations because of better rainfall distribution in 2022. Soybean varieties generally yielded higher in the NGS AEZ compared to those in the SS AEZ due to better soil fertility, higher rainfall, and longer growing seasons compared to the SS. Among the varieties tested, TGX-1951-3F consistently outperformed TGX-1904-6F and TGX-1835-10E and was suitable for cultivation at both AEZs, while the extra-early-maturing TGX-1835-10E is more suited for cultivation in the SS AEZ. Because of medium-maturity, TGX-1904-6F is not recommended for planting in the SS AEZ. Farmers should be advised to plant TGX-1835-10E and TGX-1951-3F soybean varieties on 15–29 June in the SS AEZ and 15 June–6 July in the NGS AEZ. The results show that the medium-maturing TGX-1904-6F variety should be planted on 15–29 June in the NGS AEZ, beyond which there will be substantial reduction in grain yield. Because of the unstable nature of rainfall in the region and an inability to extrapolate short-term experimental results to other locations or regions, long-term simulation studies under changing climates are required to determine the optimal planting windows for these varieties for wider areas in Northern Nigeria.