Influence of Genotype, Environment, and Crop Management on the Yield and Bread-Making Quality in Spring Wheat Cultivars

Abstract

Obtaining optimal wheat cultivars that balance high productivity and grain processing quality in diverse environmental and crop management conditions requires a comprehensive assessment of the influence of genetic and environmental factors and their interactions. This study investigated the influence of these factors on yield, grain quality, and bread-making traits in spring wheat (Triticum aestivum L.) cultivars. The study was conducted at four trial locations in the temperate climate area over two consecutive growing seasons, each with two different crop management approaches (moderate and high input). We observed a strong influence of genotype on grain quality (e.g., protein content, test weight) and farinographic in spring wheat. Environmental factors strongly influenced the variability of dough softening and quality number among the studied rheological traits. However, we observed that crop management significantly impacted dough stability. The strength of the relationships between yield, grain quality, and bread-making traits depended on the specific crop management used. The multi-trait stability of genotypes in yield, grain quality, and bread-making traits also varied, depending on the crop management method.

1. Introduction

Wheat is one of the most important cultivated crops globally, with winter wheat being the dominant form in the temperate climate zones of Europe. However, spring wheat (Triticum aestivum L.) is grown less due to its lower grain yield and smaller grain size than winter wheat. However, it offers a higher protein content, which translates into better baking quality [1,2,3]. In the face of climate change and decreasing snow cover in Europe, the importance of spring wheat is increasing, attracting more attention from farmers, breeders, and researchers [4,5,6,7]. This crop is often used to replace winter crops, such as rapeseed or winter wheat, which may suffer from frost damage after harsh winters.

The grain yield and bread-making quality of different wheat forms depend on the interaction between genotype (G), environment (E), and their interaction (G × E). This interaction focuses on the variability in genotype performance under different environmental conditions, influencing the selection of optimal genotypes for high-quality grain. In addition to environmental and genetic factors, crop management practices such as sowing time, fertilization, and pesticide use are increasingly crucial for yield and bread-making quality [2,7,8,9].

Bread-making quality in wheat depends primarily on protein content, wet gluten content, gluten quality, and amylolytic enzyme activity [3,9,10,11]. Certain traits, like protein content and gluten characteristics, are mainly determined by genotype and environmental factors, influencing grain yield, thousand kernel weight, and sedimentation volume [6,12]. The relationship between environmental conditions, cultivation practices, and bread-making quality is complex, and the effects on various quality traits still need to be understood. Environmental factors like temperature and precipitation primarily influence certain parameters, such as yield and test weight, grain hardness, and Zeleny sedimentation volume in different forms of wheat and different climatic zones [13,14,15]. Rainfall, for example, in durum wheat, impacts yield and protein content, with more potent effects in favorable years than during drought conditions [7,16]. On the other hand, the stronger impact of genotype on the traits of protein content and wet gluten content may be related to nitrogen use efficiency (NUE) [17]. Stronger pressure of fungal diseases, e.g., Fusarium head blight, affects yield and other important features [18,19], which may be revealed especially in the case of improper or incomplete plant protection as part of crop management. While numerous studies have examined the influence of genotype, environment, and crop management on wheat yield, few have comprehensively assessed their effects on bread-making traits. Furthermore, most research has focused on individual aspects of crop management, such as nitrogen fertilization, without considering a more holistic approach. This study aims to fill this literature gap by emphasizing how crop management influences bread-making quality, thus providing valuable insights into optimizing spring wheat cultivation to produce high-quality grains for bread production.

In the context of climate change, the selection of stable spring wheat genotypes becomes increasingly important. However, the interaction between genotype and environment on yield and quality makes it difficult for farmers to recommend cultivars. Therefore, before introducing new hybrids with high stability, it is necessary to study the genotype-environment interaction to evaluate new lines in different environments. So far, the stability assessment has mainly focused on grain yield or selected single traits [20]. Stability should be evaluated across multiple traits, not just individual characteristics. The multi-trait stability index (MTSI) proposed by Olivoto et al. [21] provides a framework for assessing stability across various bread-making traits. This method combines simultaneous selection for stability of several traits into a single, easy-to-interpret index. Enhancing grain quality and selecting stable genotypes with superior characteristics is critical for producing high-quality, market-preferred products.

This study has three main objectives: (i) to identify spring wheat genotypes with stable performance in yield and bread-making quality across different environments, (ii) to assess the impact of genotype, environment, and crop management on yield, flour, dough, and bread-making quality traits, and (iii) to evaluate the relationships between yield and various quality traits in spring wheat.

2. Materials and Methods

2.1. Field Experiments

The seven spring wheat cultivars were evaluated in 4 trial locations and two growing seasons (2019 and 2020). Table S1 presented descriptions of soil and wheatear in study trial locations and growing seasons. The seven spring wheat (Triticum aestivum L.) cultivars included Bombona, Izera, Ostka Smolicka, Radocha, Torridon, Trappe, and Tybalt. The descriptions of study cultivars are shown in Table S2. The spring wheat cultivars were grown at two levels of crop management: moderate-input management (MIM) and conventional, high-input management (HIM). The MIM included fungicide seed treatment before sowing, N fertilization, and the use of herbicides. Depending on the location, the total rate of N for the MIM is about 90 kg ha⁻¹, with 40–60 kg N ha⁻¹ applied before sowing and the rest used at Zadoks Growth Stage (GS) 49. In addition to the MIM level N dose of 40 kg ha⁻¹ at GS 59, foliar microelements fertilizer (MgO 250 g ha⁻¹, Cu 50 g ha⁻¹, Mn 150 g ha⁻¹, Zn 80 g ha⁻¹), two fungicides at GS 31–32 (carbendazim, 625 g ha⁻¹) and GS 49–60 (fenpropidin, 550 g ha⁻¹), and a growth regulator (trinexapac-ethyl, 125 g ha⁻¹) at GS 31 were applied at the HIM level. The sowing density, set at 450 seeds per m², did not depend on crop management levels. The sowing date for individual locations ranged from 23 March to 29 March in 2019 and from 25 March to 1 April in 2020. Harvesting occurred between 28 July and 5 August 2019 and 1 August and 7 August 2020. Individual trials were established as a two-factorial (cultivar and crop management) strip-plot design with two blocks. The crop management levels were assigned in the whole plot, and cultivars were assigned in the subplot. The size of the plot was 15 m². The grain yield and thousand-grain weight TGW were determined from a 1 m² sample collected from the center of the plot.

The quality traits were evaluated from grain samples, including end-use quality traits: test weight (TW), grain ash content (AC), grain protein content (PC), wet gluten content (WG), gluten index (GI), Zeleny sedimentation value (SV), falling number (FN), flour yield (FY); the Farinograph traits: water absorption (WA), dough development (DD), dough stability (DS), dough softening (DSF), and quality number (QN); and the baking properties: loaf volume (LV), and crumb hardness (CH).

2.2. Methods

2.2.1. Properties of Grain

The test weight was determined using AACC Method 55-10. The ash content was determined with the incineration method (AACC Method 08-01.01) using an FCF S muffle furnace (Czylok, Jastrzębie Zdrój, Poland). The protein content (N × 5.7) was determined according to the Kjeldahl method (AACC Method 46-11.02) using a Kjeltec 8200 apparatus (Foss Tecator, Hillerod, Denmark). The wet gluten content and gluten index were determined using the mechanical method (AACC Method 38-12) on the Glutomatic 2200 (Perten Instruments, Stockholm, Sweden). The Zeleny method obtained the sedimentation value (AACC Method 56-61.02). The falling number was determined using the Hagberg-Perten method (AACC Method 56-81B) on the Falling Number test apparatus, type 1400 (Perten Instruments, Stockholm, Sweden).

2.2.2. Grain Grinding

The grain samples were ground in a two-passage laboratory mill Quadrumat Senior (Brabender GmbH & Co. KG, Duisburg, Germany). Before milling, the grains were cleaned on granules (Brabender GmbH & Co. KG, Duisburg, Germany) and conditioned to 14.5% humidity. Based on the milling balance, the total flour yield was calculated.

2.2.3. Properties of Flour and Dough

The water absorption of flour and rheological properties of the dough was determined using the AACC Method 54-21 on the Farinograph-E model 810114 (Brabender GmbH & Co. KG, Duisburg, Germany) [22,23]. The water absorption of flour was determined based on the amount of water added from a burette needed to achieve a dough consistency of 500 FU. The rheological properties of the dough (dough development, dough stability, and dough softening after 12 min of mixing) and quality number were determined from the normal curve graph using the Farinograph v.5 computer program.

2.2.4. Baking Procedure and Properties of Bread

The recipe for the bread dough included 500 g of flour, 15.0 g of fresh compressed yeast, 7.5 g of salt, and water in the amount necessary to achieve a dough consistency of 350 FU. The amount of water added was calculated based on the determined farinographic water absorption. The dough ingredients were mixed in an SP-800A mixer (Spar Food Machinery, Taichung, Taiwan) for 4 min at speed level 2. The dough was fermented in a fermentation chamber DC-32 (Sveba Dahlen, Fristad, Sweden) for 90 min, with punching performed after 60 min. After fermentation, the dough was divided into 250 g portions, shaped by hand, placed in molds, and subjected to proof for 30 min in the fermentation chamber. Baking took place in the DC-32 oven (Sveba Dahlen, Fristad, Sweden) at a temperature of 230 °C for 30 min.

The properties of bread were determined 24 h after baking. The loaf volume was measured using a 3D scanner (NextEngine, West Los Angeles, CA, USA) according to the methodology of Romankiewicz et al. [24] and then converted to 100 g of bread. The crumb hardness was assessed on a texture analyzer TA-XT2i (Stable MicroSystem, Surrey, UK), according to the methodology of Romankiewicz et al. [24]. The assay relied on a dual compression of the crumb sample (thickness of slices—25 mm). A cylindrical mandrel with a 25 mm diameter was used in the measurement. The speed test was 1 mm s⁻¹. 40% penetration of the sample was applied, with a 45 s break between the first and second pressure.

2.3. Statistical Methods

The grain yield and bread-making quality traits were analyzed using a two-stage approach. In the first steps, individual trials were analyzed using the linear mixed model (LMM) typical for strip-plot design. In the second stage, we used the combined linear mixed model method, as shown below:

Xijkl = µ + Yi + Lj + YLij + Gk + GYki + GLkj + GLYijk + Ml + MYli + MLjl + MLYijl+ GMkl + GMYkli + GMLlkj + GMLYijkl + eijkl

(1)

where Xijkl is the value of the studied trait; µ is the overall mean; Yi is the random of ith year; Lj is the random effect of jth location; YLij is the random effect of interaction between jth location and ith year; Gk is the fixed effect of kth cultivars; GYki is the random effect of interaction between kth cultivars and ith year; GLkj is the random effect of interaction between kth cultivars and jth location; GLYijk is the random effect of interaction between ith year and jth location and kth cultivars; Ml is the fixed effect of lth crop management; MYli is the random effect of interaction between lth crop management and ith year; MLjl is the random effect of interaction between jth location and lth crop management; MLYijl is the random effect of interaction between ith year and jth location and lth crop management; GMkl is the fixed effect of interaction between kth cultivars and lth crop management; GMYkli is the random effect of interaction between kth cultivars and lth crop management and ith year; GMLlkj is the random effect of interaction between lth crop management and kth cultivars and jth location; GMLYijkl is the random effect of interaction between kth cultivars and lth crop management and jth location and ith year; eijkl is the random residual effect.

Descriptive statistics (the means, minimum, maximum, standard deviations SD, and coefficients of variability CV) were conducted based on adjusted means obtained from model (1) calculated for all study traits. The corrected means calculated this way were also used to assess the varieties’ stability and study their relationships. We evaluated the relationship between all study traits using the Pearson correlation coefficient and principal components analysis (PCA). We used a multi-traits stability parameter (MTSI) to evaluate the stability of cultivars across all study traits [21]. This parameter allows for the simultaneous assessment of genotype stability for many characteristics, which in turn allows for the selection of genotypes with the highest degree of stability for all the traits considered simultaneously. The MTSI parameter is based on factor analysis for the matrix of the means of the standardized study trials, and the standardization of means is performed using the value of genotype-environment interaction effects. The MTSI indicators were assessed separately for crop management, and combinations of year and location were considered environments.

We are also interested in determining the significance and strength of the influence of main effects and their interactions on the variability of the studied traits. For this purpose, in model (1), we have changed the assumptions about the type of effects; all effects are treated as fixed. This allowed for the determination of the sum of squares for study effects. The Wald F test was used to evaluate the significance of the effects.

The statistical analyses were performed using R 4.2.1 software. The MTSI parameters were obtained using the metan package.

3. Results

The variability of the studied traits was assessed separately for genotype and environmental effects and is presented in

Table 1. Description statistics for yield, grain quality, and bread-making traits of genotype (a) and environmental (b) variability in spring wheat cultivars across four trial locations in two growing years (2019 and 2020).

(a) Genotype
	Mean	Minimum	Maximum	Standard Deviation	Coefficient of Variation
GY (t ha−1)	6.20	5.66	6.40	0.26	4.20
TGW (g)	42.21	38.7	45.26	2.26	5.35
TW (kg hL−1)	76.77	73.7	78.88	1.91	2.49
AC (% d.m.)	1.76	1.67	1.82	0.05	2.67
PC (% d.m.)	12.88	11.82	14.00	0.67	5.23
SV (cm3)	33.79	29.42	41.80	4.39	13.00
WG (%)	22.78	19.49	27.23	2.95	12.95
GI (-)	83.36	57.64	91.27	12.32	14.78
FN (s)	338.76	281.96	375.15	32.15	9.49
FY (%)	77.88	77.24	79.15	0.65	0.84
WA (%)	58.28	57.02	59.48	1.01	1.73
DD (min)	2.35	2.06	3.49	0.49	20.75
DS (min)	2.65	1.63	5.59	1.36	51.34
DSF (FU)	71.54	39.13	85.69	17.82	24.92
QN (-)	51.51	39.13	84.88	16.54	32.12
LV (cm3 100 g−1)	380.34	366.34	398.33	10.22	2.65
CH (N)	7.43	5.69	9.21	1.24	16.71
(b) Environmental
	Mean	Minimum	Maximum	Standard Deviation	Coefficient of Variation
GY (t ha−1)	6.20	5.50	6.63	0.49	7.92
TGW (g)	42.21	39.32	44.56	2.42	5.74
TW (kg hL−1)	76.77	73.19	81.17	3.41	4.45
AC (% d.m.)	1.76	1.70	1.82	0.05	2.63
PC (% d.m.)	12.88	11.58	13.78	1.02	7.93
SV (cm3)	33.79	30.35	38.54	3.88	11.50
WG (%)	22.78	20.72	24.96	2.18	9.57
GI (-)	83.36	76.29	92.05	7.45	8.94
FN (s)	338.76	300.40	377.18	35.47	10.47
FY (%)	77.88	77.29	78.69	0.59	0.76
WA (%)	58.28	56.54	60.86	1.88	3.22
DD (min)	2.35	1.79	3.12	0.58	24.68
DS (min)	2.65	1.50	4.38	1.23	46.41
DSF (FU)	71.54	52.79	89.79	15.12	21.13
QN (-)	51.51	39.29	68.82	12.53	24.33
LV (cm3 100 g−1)	380.34	370.67	392.48	9.02	2.37
CH (N)	7.43	6.86	8.62	0.80	10.81

Crumb hardness (CH), dough development (DD), dough softening (DSF), dough stability (DS), flour yield (FY), gluten index (GI), grain ash content (AC), grain protein content (PC), grain yield (GY), falling number (FN), loaf volume (LV), test weight (TW), thousand-grain weight (TGW), quality number (QN), water absorption (WA), wet gluten content (WG), Zeleny sedimentation value (SV).

. The GY for the genotypes ranged from 5.66 to 6.40 t ha⁻¹; for the environments, it ranged from 5.50 to 6.63 t ha⁻¹. For yield, variability was greater for environmental effects (CV 7.92%) than for genotypic effects (CV 4.20%). The AC for genotypes means ranged from 1.67 to 1.82% d.m.; for the environments, it ranged from 1.70 to 1.82% d.m. For this trait, we observe a similar level of variability for genotypes and environments (CV 2.67 and 2.63%, respectively). PC for genotypes ranged from 11.82 to 14.00% d.m. We observed a range of 11.58 to 13.78% d.m. for environmental effects. As for PC, we observed greater variability for environments (CV 7.93%) than for genotypes (CV 5.23%). For WG, genotypes ranged from 19.49 to 27.23%, and for environments ranged from 20.72 to 24.96%. For this trait, greater variability was observed for genotypes (CV 12.95%) than for environments (9.57%). For the FN, the values for genotypes ranged from 281.96 to 375.15 s, while for environments, ranging from 300.40 to 377.18 s. The variability was greater for environments (CV 10.47%) than for genotypes (CV 9.49%). The FY for the genotypes ranged from 77.24 to 79.15%; for the environments, it ranged from 77.29 to 78.69%. For FY, we observe a similar level of variability for genotypes and environments (CV 0.84 and 0.76%, respectively). WA for genotypes ranged from 57.02% to 59.48% and had low variability (CV 1.73%); similarly, we observed a low level of variability for environments. In contrast, for DS, we observe a high level of variability in both genotype effects and environments (CV 51.34% and 46.41%, respectively). The LV for the genotypes ranged from 366.34 to 398.33 cm³ 100 g⁻¹; for the environments, it ranged from 370.67 to 392.48 cm³ 100 g⁻¹. For LV, we observe a similar level of variability for genotypes and environments (CV 2.65% and 2.37%, respectively).

A higher average GY was observed for HIM crop management (6.57 t ha⁻¹) than for MIM (5.83 t ha⁻¹)—

Table 2. Description statistics for yield, grain quality, and bread-making traits for moderate-input management (MIM) and high-input management (HIM) in spring wheat cultivars across four trial locations in two growing years (2019 and 2020).

	Mean		Minimum		Maximum		Standard Deviation		Coefficient of Variation
	MIM	HIM	MIM	HIM	MIM	HIM	MIM	HIM	MIM	HIM
GY (t ha−1)	5.83	6.57	4.83	5.42	6.84	7.46	0.56	0.55	9.61	8.41
TGW (g)	40.70	43.72	31.87	37.50	46.70	50.64	3.47	2.91	8.53	6.66
TW (kg hL−1)	75.84	77.71	69.10	68.82	81.87	82.97	3.60	3.70	4.75	4.76
AC (% d.m.)	1.76	1.76	1.61	1.63	1.88	1.87	0.07	0.06	4.06	3.63
PC (% d.m.)	12.29	13.46	10.60	10.58	13.86	16.99	0.98	1.55	7.98	11.49
SV (cm3)	31.08	36.50	25.47	25.98	42.17	51.39	4.45	6.78	14.33	18.59
WG (%)	21.33	24.22	14.85	18.11	26.11	37.98	2.96	4.65	13.86	19.19
GI (-)	85.82	80.90	38.31	44.36	99.75	98.70	13.95	16.36	16.26	20.22
FN (s)	338.02	339.51	239.61	233.14	394.18	422.71	44.69	49.66	13.22	14.63
FY (%)	77.59	78.16	75.20	76.55	79.40	80.35	1.18	0.93	1.53	1.19
WA (%)	57.88	58.29	54.30	55.40	62.25	63.35	2.11	2.29	3.65	3.92
DD (min)	2.06	2.56	1.20	1.35	2.65	7.70	0.33	1.28	15.82	49.84
DS (min)	2.14	3.02	0.90	0.95	6.55	10.00	1.36	2.25	63.54	74.51
DSF (FU)	73.96	68.05	31.00	7.50	110.50	107.50	20.51	24.09	27.74	35.39
QN (-)	47.13	54.68	26.50	27.50	125.50	123.00	22.56	25.29	47.86	46.26
LV (cm3 100 g−1)	386.55	374.13	371.98	369.35	410.48	377.14	14.68	2.97	3.80	0.79
CH (N)	7.35	7.52	4.60	4.33	12.04	12.79	1.48	1.91	20.14	25.39

Abbreviation: crumb hardness (CH), dough development (DD), dough softening (DSF), dough stability (DS), flour yield (FY), gluten index (GI), grain ash content (AC), grain protein content (PC), grain yield (GY), falling number (FN), high-input management (HIM), loaf volume (LV), moderate-input management (MIM), test weight (TW), thousand-grain weight (TGW), quality number (QN), water absorption (WA), wet gluten content (WG), Zeleny sedimentation value (SV).

. On the other hand, slightly higher variability was observed for MIM (CV 9.61%) than for HIM (CV 8.41%). We observe the same AC values between MIM and HIM (1.76% d.m. for both crop management). Slightly higher variability is observed for MIM (CV 4.06%), compared to HIM (CV 3.63%). For the PC, higher values were for HIM (13.46% d.m.) than for MIM (12.29% d.m.). The variability of this feature was also higher for HIM (CV 11.49%) than for MIM (7.98%). As for PC, higher values and variability of WG and FN are observed in HIM (WG—average 24.22%, CV 19.19%; FN—average 339.51 s, CV 14.63%) than in MIM (WG—average 21.33%, CV 13.86%; FN—average 338.02 s, CV 13.22%). We observed higher FY values for HIM (78.16%) than MIM (77.59%). For LV, higher values were for MIM crop management (386.55 cm³ 100 g⁻¹) than for HIM (374.13 cm³ 100 g⁻¹). The variability of this variable was also observed to be more significant for MIM (3.80%) than for HIM (0.79%).

Table 3. Percent of the total variance (the sum of all variance components) of yield, grain quality, and bread-making traits for main effects (year, crop management, location, and genotype) and its interaction effects in spring wheat cultivars across four trial locations in two growing years (2019 and 2020).

Effect	Degrees of Freedom	GY		TGW		TW		AC		PC		SV		WG		GI		FN		FY		WA		DD		DS		DSF		QN		LV		CH
		(t ha-1)		(g)		(kg hL-1)		(% d.m.)		(% d.m.)		(cm3)		(%)		(-)		(s)		(%)		(%)		(min)		(min)		(FU)		(-)		(cm3 100 g-1)		(N)
Year (Y)	1	0.01	*	0.01	*	8.21	**	11	**	9.21	**	10.59	*	0.01		22.3	**	0.01		65.21	**	21.39	**	51.23	**	5.62	*	83.49	**	71.96	**	0.01		41.34	**
Menagment (M)	1	26.4	**	20.67	**	4.79	*	0		12.77	**	15.74	**	11.83	**	1.9	*	0.01	*	1.01	*	19.98	**	2.14	*	44.92	**	15.21	**	27.26	**	13.77	**	0.01	*
Location (L)	3	1.01	*	25.12	**	13.2	**	17	**	20.93	**	4.88	*	0.01		5.23	*	0.01		8.12	**	0.08	*	39.45	**	4.21	*	0.05	*	0.11	*	6.21	*	0.01
YxM	1	0.95	*	2.99	*	0.73	*	3.8	*	5.96	*	0.01		0.01		0.01		0.01		1.11	*	0.05	*	0.05	*	3.15	*	0.05	*	0.11	*	0.01	*	0.01
Genotype (G)	6	5.23	**	21.38	**	11.99	**	24	**	10.93	**	25.85	**	24.64	**	27.72	**	16.4	**	8.1	**	1.33		6.1	*	23.51	**	0.01	*	0.01		0.01		7.69	**
YxL	3	42.81	**	3.14	*	51.7	**	9.1	*	3.23	*	16.73	**	28.24	**	9.14	*	64.2	**	2.13	*	17.05	**	0.78	*	2.34	*	0.96	*	0.13	*	0.01		34.87	**
LxM	3	3.27	*	8.13	**	3.9	*	0		8.8	*	5.64	*	3.39		1.28	*	0.01	*	0.01		0.05	*	0.05	*	8.57	*	0.01		0.01		79.86	**	0.01
YxG	6	0.01		9.68	**	0.58	*	0		0.01		0.01		0.01		4.63	*	0.01		4.75	*	19.98	**	0.01		1.23	*	0.01		0.08	*	0.01		0.01
GxM	6	0.01	*	0.01		0.26		0		0.78		1.44		1.72		0.24		0.01		4.51	*	0.05	*	0.05	*	0.01		0.01		0.01		0.01		0.01
YxLxM	3	0.01		0.04		0.74	*	12	**	5.44	*	17.63	**	0.35		0.01		1.46	**	0.1		0.01		0.01		4.25	*	0.01		0.11	*	0.01		0.01
LxG	18	0.51	*	1.46	*	2.14	*	0		0.37		0.01		0.01		6.15	*	0.01		2.78	*	0.01		0.01		2.15	*	0.05	*	0.01		0.04	*	0.33	*
YxGxM	6	3.57	*	0.53	*	0.13		3.6	*	0.01		0.01		2.53		0.01		0.01		0.01		19.98	**	0.01		0.01		0.05	*	0.09	*	0.01		0.62	*
YxLxG	6	9.03	**	3.61	*	0.32	*	5.8	*	1.61		1.45		7.32	*	0.01		11.04	**	2.14	*	0.01		0.05	*	0.01		0.01		0.01		0.01		0.54	*
LxGxM	18	7.18	*	0.01		0.01		5.2	*	0.01		0.01		1.22		0.01		0.63	*	0.01		0.01		0.05	*	0.01		0.01	*	0.01		0.01		0.01
YxMxLxG	18	0.01		3.21	*	1.3	*	9.1	*	19.95	**	0.01		18.72	**	21.37	**	6.17	**	0.01		0.01		0.01		0.01		0.05	*	0.11		0.01		14.52	**

*, **: significant at the 0.05 and 0.001 probability levels, respectively; abbreviation: crumb hardness (CH), dough development (DD), dough softening (DSF), dough stability (DS), flour yield (FY), gluten index (GI), grain ash content (AC), grain protein content (PC), grain yield (GY), falling number (FN), loaf volume (LV), test weight (TW), thousand-grain weight (TGW), quality number (QN), water absorption (WA), wet gluten content (WG), Zeleny sedimentation value (SV).

shows the influence of the individual main and interaction effects for the studied features, determined using the percent of total variance based on the sum of squares. To compare the variance of factors and their interactions, the sum of squares was presented as percentages, the total sum of squares for considerate effects, and it is named as a percentage of the total variance. This form of presenting the results allows us to determine the strength of the influence of the main and interactive effects on the studied traits. For agronomic traits, such as GY and TGW, the values of these traits were most strongly conditioned by the leading environmental effects (year or location), as well as crop management or interaction effects between them. On the other hand, traits associated with the quality of grain, flour, dough, and bread are more often conditioned by genotypic effects and interactions with genotypes. However, we observed a few exceptions; for example, FY and DSF were most strongly determined by the main effect of the year (belonging to the group of environmental impacts).

We conducted separate MTSI analyses for both management approaches (MIM and HIM), focusing on grain quality and the resulting flour and bread of seven spring wheat cultivars (

Table 4. The multi-trait stability index for spring wheat cultivars in moderate-input management (MIM) and high-input management (HIM) across four trial locations in two growing years (2019 and 2020).

Multi Trait Stability Index (MTSI)
Cultivars	MIM	Stability Ranking MIM	HIM	Stability Ranking HIM
Bombona	2.45	1	3.04	2
Izera	3.71	4	2.73	1
Ostka Smolicka	4.10	6	3.79	3
Radocha	3.39	2	3.99	4
Torridon	3.52	3	4.75	7
Trappe	5.15	7	4.74	6
Tybalt	4.04	5	4.44	5

High-input management (HIM); moderate-input management (MIM).

). Its parameter interpretation is the same as the commonly used stability index, e.g., Shukla stability variance (lower is better). In the case of MIM, the top two genotypes with the highest quality MTSI were selected using a selection pressure of 15%. This approach helped to determine the selection differentials for genotypes, such as Bombona (2.45) and Radocha (3.39), which hold the first and second positions in the stability ranking, respectively. In HIM, the genotypes Izera (2.73) and Bombona (3.04) hold first and second positions in the stability ranking, respectively. The correlation coefficient between MTSI in MIM and HIM crop management was equal to 0.56.

The correlation matrix Figure 1 in the upper triangle reveals the associations among the parameters for MIM. Significant positive correlations of the PC with WG (r = 0.61), DD (r = 0.56), and QN (r = 0.46). The positive correlations of the WG with SV (r = 0.56), DD (r = 0.45), and DS (r = 0.43) were observed. Conversely, negative correlations were observed between TGW and AC (r = −0.61) and WA and FN (r = −0.76). On the other hand, we observe weaker significant negative correlations between PC and DSF (r = −0.61) and between WG and DSF (r = −0.59). Moreover, no correlation was observed between DD and CH for management HIM (the lower triangle). Significant positive correlations were identified between DS and DD (r = 0.82). We also observed positive correlations of the PC with WG (r = 0.54), SV (r= 0.61), WA (r = 0.57), DD (r = 0.63), and DS (r = 0.66). On the other hand, negative correlations were observed between DS and DSF (r = −0.74) with statistical significance at p < 0.05. Also, we observed a negative correlation between PC and DSF (r = −0.53). Generally, we observe weaker correlations of WG and FN for HIM crop management with farinographic parameters than for MIM.

According to the PCA depicted in Figure 2a (MIM crop management), PC1 accounts for 27.7% of the total variance of the parameters, and PC2 explains approximately 24.4% of the variance. When combined, PC1 and PC2 collectively account for 52.1% of the variance observed in all the analyzed parameters. In the PCA of management HIM, as shown in Figure 2b, PC1 accounts for 38.1% of the total variance of the parameters, while PC2 explains approximately 16% of the variance. PC1 and PC2 contribute to 54.1% of the total variance observed in all the analyzed parameters.

Figure 2a represents the PCA of MIM crop management; we observed a negative relationship in agronomic traits (GY and TGW). The quality traits (AC, CH, and GI) showed a positive relationship with each other, and SV with DS and FY with WA also showed a positive correlation. A strong negative relationship was observed between DS and DSF and between BS and DSF, and no correlation was observed between SV and CH. In grain quality traits, PC and FN displayed a strong positive correlation. In contrast, TW and SV displayed a negative correlation. Generally, a strong negative correlation was shown between TGW and FA, TW and AC, DS and DSF, and FN and WA. At the same time, a strong positive correlation was observed between GI and FA.

In the PCA of management HIM (Figure 2b), agronomic traits (GY and TGW) showed a negative relationship. A positive correlation was observed between AC, GI, and CH in flour quality traits. DS, FN, FY, and DSF showed a significant positive correlation. In comparison, a significant negative correlation was observed between CH and LV. A significant negative correlation was found between PC and GY and between GY and TW. In contrast, a significant positive correlation was found between PC and TW, SV and DD, and CH and FA. In grain quality traits, TW and PC showed a strong positive correlation.

The color-coded indicator values (cos2) in Figure 2a,b reflect how well the variables are represented on the primary component. In the present study, the variables LV (Figure 2a) and GY (Figure 2b) had cos2 values below 0.2, suggesting that additional interpretation of these parameters may not be necessary.

4. Discussion

Our study explored genotype, environmental, and crop management effects on seven spring wheat cultivars across four trial locations under two farming conditions over two growing seasons, emphasizing yield size and quality traits. We found that grain yield spring wheat variations strongly stemmed from location and growing seasons interaction effects, explaining about 50% of the yield variability. Such a large impact of this interactive effect makes it difficult to predict the yield and justify its value. The main effect of crop management also had a significant share in the yield variability, at approximately 25%. Wheat research, including winter and durum forms, shows a powerful influence of environmental effects and their interactions [2,9,25,26,27]. However, in our study, we observe a relatively high impact of crop management effects on yield, which has yet to be fully explored before.

PC is one of the most important parameters used to assess the suitability of wheat grain for selecting raw material for milling into baking flour [2,5]. The PC depends on environmental effects (location and year) and crop management, but only 10% were influenced by genetic factors. Among wheat proteins, particular importance is attributed to storage proteins (gliadins, glutenins), which, when combined with water, form a viscoelastic gluten structure [28]. In our study, WG was more dependent on genotype than environmental influence. On the other hand, the GI, reflecting the quality of the protein, was already influenced by the genotype by over 25% and, to a minimal extent (about 2%), by crop management. Unfortunately, PC, WG, and GI values were significantly conditioned by the four-way interaction effect among year x location x genotype x crop management. Unfortunately, the substantial contribution of this interaction effect in shaping these characteristics makes it challenging to formulate appropriate recommendations and guidelines.

Scarce studies include the assessment of the impact of genotypic, environmental, and crop management effects on grain quality and bread-making quality traits. We detected a significant impact of crop management on DS, accounting for approximately 45% of the variability. However, we also observe a relatively large effect of crop management on features such as WA and QN. The WA is an essential parameter on which the productivity of dough and bread depends, which translates into the economic effects of the bakery [3]. The knowledge that these traits are determined by crop management may contribute to protection from the impact of unfavorable weather during the growing season and, as a result, will allow the delivery of the appropriate quality product, for example, by increasing the intensification of plant protection or fertilization.

Our study revealed wider genotypic variation in WG and GI. These results are consistent with the studies on spring wheat by Sułek et al. [7] and Feledyn-Szewczyk et al. [29] and on winter wheat by Ma et al. [30] and Rozbicki et al. [9], emphasizing the dominant role of genetics in wet gluten content in wheat grain. Genotypic effects also impact FN, reflect wheat grain development, and are too strongly influenced by genotypic effects. It was also found for spring wheat in Poland by Sułek et al. [7] and in India by Farhad et al. [31] that genotypic factors exert a greater influence on falling number variation than environmental factors. Moreover, for FN, we observe a significant contribution to the interaction of environmental factors (year and location), explaining about 60% of the variability of this trait. A considerable impact of environmental effects on shaping this trait was noted for winter wheat in a temperate climate in the studies by Rozbicki et al. [9] and Mitura et al. [2] in the case of spring wheat. For LV, the decisive effects were crop management and location. However, the most influence is observed in the interaction between location and crop management, which explains almost 80% of the variability. Similar dependencies were demonstrated by Rozbicki et al. [9] for winter wheat and by Wysocka et al. [3] for spring wheat.

We conducted separate Multi Trait Stability Index (MTSI) analyses for MIM and HIM approaches, assessing seven spring wheat cultivars across agronomic, grain, flour, dough, and bread traits. It turns out that entirely different cultivars can be considered stable in MIM and different in HIM crop management. Unfortunately, our study cultivars do not observe consistency in cultivar evaluation between MIM and HIM, as indicated by a relatively low correlation coefficient. Indirectly, we can conclude that multi-trait stability depended on the crop management type used. The recommendation of stable cultivars in terms of many traits will allow for maintaining the standards of the final product even in unfavorable weather conditions during the growing season. Selecting such multi-trait stable cultivars becomes essential in breeding crops grown in uncertain, changing climate conditions. Cultivars with consistent performance across traits and practices are valuable for stable wheat production; however, identifying such cultivars is often impossible or difficult. The Bombona cultivar is worth noting as it turned out to be the most stable for MIM and the second most stable for HIM. It may be worth introducing to breeding programs to obtain more stable multi-traits new cultivars.

Under MIM, negative correlations between WA and SV suggest wheat cultivars with lower softening and water absorption have higher baking scores [28]. A solid positive DS-DD correlation in HIM suggests that improved dough development contributes to better dough strength and baking scores. However, in MIM, this relationship between these features was much weaker but still positive. Lower DSF often signifies a balanced ratio of glutenin and gliadin proteins, enhancing dough strength [28]. A positive relationship among FA, CH, and GI suggests higher protein and gluten strength correlate with increased ash content, associated with more minerals, similar to durum wheat in the study of Ficco et al. [32], and improved dough strength [33]. Positive correlations were found between FN and DS, as well as FY and WA. A higher supply of nitrogen in high input management (HIM) contributes to an increase in PC and WG, which translates into higher WA and better rheological properties of the dough [34].

A strong positive correlation between TGW and PC in HIM suggests larger grains can store more proteins, leading to increased PC [28]. Negative correlations between agronomic traits (GY and TGW) and grain/flour quality suggest a trade-off between maximizing yield and optimizing bread quality. Selecting management and cultivars balancing yield and quality is crucial in plant breeding.

The positive correlation between FY and WA indicates that cultivars with higher flour yield may possess increased water absorption capacity, impacting dough hydration and final bread quality [35]. This results from a higher dietary fiber content with high water absorbency [36].

Between MIM and HIM, we observe differences in the strength of the relationships between characteristics. We observe a significantly stronger negative correlation between GY and protein content PC or between PC and FN in HIM compared to MIM. This makes the protein content more sensitive to spring wheat productivity in HIM. Generally, in HIM, we observe more substantial dependencies of PC on other considered characteristics. We also observe a change in the type of relationship between levels of crop management. We see this situation for PC and WA; for MIM, this relationship is positive, and for HIM, it is negative. These crop management methods are not optimal and contribute to biotic and abiotic stresses in plants, unlike in HIM, where optimal fertilization and plant protection are applied. Unfortunately, a different field experiment would have to be conducted to indicate a single element of applied crop management to demonstrate this. These elements would be studied as separate factors, which would significantly complicate the statistical analysis of this type of multi-environment trial.

This complicates the formulation of clear recommendations and obtaining a homogeneous raw material. These results are specific to spring wheat cultivars and MIM/HIM crop managements; outcomes may differ for other wheat types, regions, or approaches.

5. Conclusions

This study assessed the influence of genotype, environment, and crop management practices on yield and bread-making quality traits in spring wheat cultivars. Our findings revealed significant genotype-environment interactions, particularly emphasizing the role of location and crop management in shaping both yield and grain quality. While high-input management systems generally enhanced yield and protein content, moderate-input systems showed improved loaf volume and bread-making traits, indicating that no single management practice universally optimizes all desired characteristics. The multi-trait stability index (MTSI) analysis emphasized the variability in cultivar stability across environments, with specific genotypes excelling in different management systems. This underlines the importance of considering crop management and the environment when selecting genotypes for stable wheat production in a changing climate.

Future breeding programs should focus on multi-trait stability to identify genotypes that maintain high grain quality and yield under varying environmental conditions. Such efforts will be crucial to ensure stable and high-quality wheat production under the pressures of climate change.