Yield and Yield Components Stability of Winter Wheat and Spring Barley in Long-Term Experiment in Poland

Abstract

The aim of this study was to evaluate the effect of long-term crop rotation and fertilization treatments on the yield and yield component stability of spring barley and winter wheat on the basis of selected data over 30 years. The stability was measured using statistical methods (the Shukla stability index and linear mixed models). The long-term field experiments established in 1955 were conducted in central Poland (Chylice near Warsaw, 52°06′ N, 20°33′ E) and consisted of two crop rotations with the same fertilization regime. The first field experiment (A—NOR) is typical of the Norfolk rotation and consists of the following four crops: 1. sugar beet, 2. spring barley with undersown red clover, 3. red clover, and 4. winter wheat, while the second field experiment (B—NONleg) contains a sequence of crops without legumes: 1. sugar beet, 2. spring barley, 3. winter rapeseed, and 4. winter wheat. The following fertilization regimes are used in both experiments: a control without any fertilization (O), mineral fertilization (NPK), farmyard manure (FM), and mixed mineral and organic fertilization (½ NPK + ½ FM). The average yields of winter wheat and spring barley (in t*ha⁻¹) were 4.34 (a max of 5.48 in NONleg_NPK) and 4.27 (a max of 5.67 in NONleg_NPK) in the NONleg rotation treatment and 5.39 (a max of 6.12 in NOR_NPK) and 4.28 (a max of 5.22 in NOR_½ NPK + ½ FM), respectively, in the legume-based NOR rotation treatment. In the legume-free rotation treatment, the highest yield stability was found in the treatments fertilized only with manure (NONleg_FM) and in a mixed manner (NONleg_½NPK + ½FM), while in the Norfolk rotation treatment, the yield stability was the highest in the treatments with mixed fertilization (NOR_½NPK+½FM) and the treatments in which only mineral fertilizers were used (NOR_NPK).

1. Introduction

In recent years, there has been an increasing appreciation of research findings based on long-term experiments. Long-term experiments are a valuable source of data on the long-term effects of various agronomic factors, weather, and soil practices on yield and yield component stability [1]. Indicators of sustainable agriculture, preventing climate change in crop production, can be determined by long-term experiments. According to Berzsenyi et al. [2], a non-decreasing trend in yield is necessary to call a system sustainable. Long-term research synthesizing the benefits of diverse crop rotations under different land management practices and climatic conditions is essential for developing resilient cropping systems [3].

Cereals are the most important crops in Europe. Wheat and barley are, respectively, the cereals with the highest and third-highest production volumes in European Union countries [4]. Their economic importance has not diminished even in the face of changes in the agricultural market in the last few years due to the extensive use of both species in human and animal nutrition. The role of these species in food security is crucial, hence the need to develop agronomic schemes to ensure the yield stability of these key crop species in Europe and in the world.

In recent years, adapting to climate change has been an important challenge for crop production. Properly selected agrotechnology, especially elements such as crop rotation and fertilization, is a key factor affecting agricultural crop yields. In this context, ecological conditions such as temperature and moisture fluctuations can induce considerable morphological and physiological changes in plants [1], thereby affecting their yield stability and response to agronomic inputs. Cammarano et al. [4] confirm the hypothesis that the impacts of changing agronomic practices might offset the negative impacts of climate change. Research into farming systems and agronomic practices often focuses on agricultural crop yields without considering the temporal stability and reliability of yields. To ensure food security for a growing population, adapt to climate change and meet the requirements of sustainable agriculture, attention must also be paid to yield stability. Yield stability as commonly understood is the constancy of agricultural outputs, especially yield, over long periods of time or across various spatial environments [5]. In the context of agriculture, the concept of stability is mostly used as a criterion to measure the temporal or spatial invariability of specific features [6]. The stability of yield is also an important characteristic to be considered when judging the value of a cropping system relative to others [2].

The most popular research topic in this field is assessing the stability of genotypes in different environments. Piepho [7] emphasizes that methods for determining varietal stability can also be used to compare different agronomic treatments, of which varieties are only one case in point. Macholdt et al. [8] claim that to address yield stability, several agronomic factors need to be considered, which are as follows: the genetic background of cultivars, crop rotation, and crop husbandry (e.g., sowing date/density, mineral/organic fertilization, and pesticide applications). There are still far fewer reports in the literature on which agronomic practices affect the yield stability of agricultural crops and to what extent.

Crop rotation is of great importance in achieving high and stable yields. It is a foundational component of sustainability and long-term profitability, without requiring additional financial investments. The design of stable cropping systems that are better adapted to a changing climate is crucial for sustainable agricultural production [6,9]. Many authors emphasize that a properly planned crop rotation reduces the use of crop protection products, reduces the infestation of weeds, pests, and diseases, and also improves the physical properties of the soil, soil fertility, and biological life [10,11]. According to Berzsenyi et al. [2], no amount of chemical fertilizer or pesticide can fully compensate for the effects of crop rotation.

Cereal-dominated rotations show a higher risk of yield losses, due to increased disease and weed pressure [2,12]. Kuś et al. [13] indicated that in a monoculture, the average wheat yield for 10 years can be as much as 30% lower than in an integrated system with correct crop rotation. The introduction of legume crops into crop rotation has a beneficial effect on soil properties and nutrient availability. Studies show that the introduction of a 3–4 year perennial alfalfa crop in the rotation has a significant effect in this respect, supplementing the soil with nitrogen and contributing to soil improvement [10]. Including red clover or alfalfa in the rotation improves aggregate stability compared to continuous maize cultivation [14]. The Norfolk crop rotation is a classic, biologically documented pattern of plant sequence with the most significant yield-increasing effect (Grzebisz et al. [15]) [2]. Yang et al. [16] observed that the addition of a red clover seedling in barley (Hordeum vulgare L.) crops increased the SOC (soil organic carbon), and the addition of red clover to winter wheat increased the rate of carbon mineralization compared to continuous winter wheat. The increase in SOC implies the role of crop diversity and red clover in improving soil carbon accumulation and potentially soil health.

Balanced mineral and organic fertilization can influence yield stability and offset the impact of climate change. Manure applications significantly influence soil aggregation, SOC, and yield stability [17]. Macholdt et al. [18] found that well-planned cropping systems with additional organic matter inputs reduced the yield vulnerability of winter barley, as indicated by a higher yield stability, better environmental adaptability, and lower production risks than those of cereal-cropping systems without organic inputs. Additionally, Macholdt et al. and Qin et al. [8,19] indicated that NPK with manure increased crop yields more than NPK alone or NPK plus straw. Chen et al. [1] reported that yield was higher and more stable with combined fertilisation (½NPK + ½ FM), due to improved soil fertility and reduced variability due to climate change in this fertilization combination.

The aim of the present paper was to evaluate the effects of crop rotation and fertilization treatments on spring barley and winter wheat yield and yield component stability on the basis of 30 years of data. We answer the following questions:

Q1: Which fertilization and crop rotation combination ensures stable yields of winter wheat and spring barley?

Q2: Are there correlations between the soil organic carbon content and yield stability of winter wheat and spring barley?

Q3: What are the effects of additional organic inputs (farmyard manure) on the yield stability of winter wheat and spring barley?

2. Materials and Methods

2.1. Case Study

Two long-term field experiments (LTEs) in this study were established in 1955 in central Poland (Chylice near Warsaw, 52°06′ N, 20°33′ E) on Phaeozem (according to WRB) derived from sandy loam [20]. The vegetation period in this region lasted about 190–200 days, with a short period of snow cover (36–65 days), early springs, and a relatively long period (about 250 days) without winter [20]. The multi-year average (1991–2022) air temperature was 8.4 °C, and the average total annual precipitation was 645.7 mm. The annual precipitation totals and average annual temperatures for the individual years of the study are shown in

Table 1. Sum of precipitation in Chylice in 1991–2021 (mm).

Year	Sum	Month
		I	II	III	IV	V	VI	VII	VIII	IX	X	XI	XII
1991	511.3	24.1	33.6	23.6	27.4	65.3	67.8	51.5	62.6	22.9	27.8	62.9	41.8
1993	513.3	56.4	29.9	38.2	22.7	35.0	71.4	62.8	34.4	45.9	32.5	22.3	61.6
1997	669.7	7.3	29.8	28.4	41.9	59.2	56.9	220.2	33.4	36.7	59.1	53.2	43.5
1999	643.7	25.7	47.5	31.1	81.4	49.1	133.0	68.5	43.5	33.9	54.2	37.5	38.5
2003	483.0	41.4	18.7	17.7	43.8	1.8	34.6	75.3	49.0	50.3	58.9	36.6	54.8
2005	535.0	44.7	42.8	44.8	27.8	86.1	46.5	62.4	27.9	23.0	6.8	33.5	88.7
2009	727.9	25.7	39.7	68.5	5.3	83.6	121.4	89.4	66.3	25.9	95.6	52.2	54.3
2011	627.5	38.8	31.6	16.6	47.1	69.2	49.5	239.1	54.6	12.8	22.4	4.4	41.7
2013	656.1	55.2	38.0	48.4	52.1	146.2	106.0	27.7	35.3	64.9	23.8	35.6	22.8
2015	509.9	56.9	10.4	42.6	39.3	78.9	36.8	49.5	6.6	74.0	39.3	46.5	29.2
2017	698.2	22.1	41.4	51.1	73.6	63.3	56.4	2.7	46.3	149.5	90.4	53.2	48.2
2019	510.3	44.6	28.0	40.2	16.8	94.6	24.0	45.2	52.7	59.2	23.3	31.3	50.4
2021	739.4	47.8	34.4	25.6	57.8	76.1	51.2	100.9	167.6	53.6	12.2	40.3	72.1

and

Table 2. Mean temperature in 1991–2021 in Chylice.

Year	Average	Month
		I	II	III	IV	V	VI	VII	VIII	IX	X	XI	XII
1991	7.7	−1.8	−5.0	3.3	7.4	10.9	16.3	19.8	19.0	15.0	7.8	2.6	−2.7
1993	7.5	−1.8	−2.7	0.7	8.2	16.5	16.0	17.6	17.8	12.7	7.8	−3.2	0.0
1997	7.2	−7.2	0.3	2.4	4.5	14.0	17.4	18.5	19.3	13.5	5.7	1.3	−2.9
1999	8.6	−1.9	−2.4	3.5	9.2	12.7	18.7	20.7	18.6	16.2	8.0	0.4	−1.2
2003	7.8	−4.9	−7.3	0.7	6.4	15.8	18.6	21.5	19.8	14.4	5.1	4.3	−0.2
2005	8.2	−0.2	−4.6	−0.8	8.7	13.7	16.7	21.5	18.6	16.6	9.4	2.2	−3.3
2009	8.2	−4.4	−2.1	1.6	10.3	13.6	16.5	20.2	18.9	15.7	6.2	4.1	−2.3
2011	8.3	−3.0	−4.5	2.0	9.8	14.3	18.8	18.5	18.8	15.3	7.8	1.7	0.7
2013	8.0	−5.7	−2.1	−2.8	6.8	15.2	18.4	19.9	19.7	12.3	9.6	4.3	0.4
2015	9.5	−0.6	−1.0	4.2	7.9	13.3	17.4	20.6	23.3	15.7	6.7	4.2	2.8
2017	8.3	−6.2	−3.2	4.7	6.8	14.0	18.2	18.9	20.0	13.8	9.1	3.6	0.5
2019	10.2	−3.7	1.1	4.9	9.6	13.4	22.6	19.8	21.3	15.1	10.7	5.5	1.6
2021	8.0	−3.9	−4.8	2.2	6.3	12.2	20.1	22.4	17.6	13.7	8.6	4.1	−2.9

. The years with the most favorable precipitation distribution for cereals were 1991, 2005, 2009, 2013, 2019, and 2021, when a large amount of precipitation was recorded during critical periods for cereals, e.g., in May. However, taking into account the total rainfall, 1991 and 2019 were dry years. Particularly in 2003, when a spring drought occurred, the humidity conditions worsened. Additionally, the year 2019 stood out in terms of average temperature, which was very high in comparison to multi-year average.

These studies are univariate, arranged using a randomized block design with four replications. Each plot measures 12.5 by 4 m and contains a harvested area of 50 square meters. The experimental treatments consist of four fertilizer types: mineral fertilizers (NPK); farmyard manure (FM); a combination of mineral and organic fertilizers (½ NPK + ½ FM); and a control with no fertilization (O). Manure is incorporated into the soil in autumn. Mineral fertilizers are applied in the form of ammonium nitrate, granulated superphosphate (containing 18–19% available phosphorus), and potassium salt (60%). The fertilization procedures are detailed in

Table 3. Fertilization in Norfolk rotation (A).

Plants	Fertilization
	NPK			FM	½ NPK + ½ FM			FM	Control
	N	P	K		N	P	K
	kg ha−1			t ha−1	kg ha−1			t ha−1
Sugar beet	200	56.0	200	40	100	28.0	100	20	-
Spring barley+undersown red clover	200	36.5	91.5	20	50	18.3	45.8	10	-
Red clover	-	36.5	91.5	-	-	18.3	45.8	-	-
Winter wheat	100	36.5	91.5	20	50	18.3	45.8	10	-

and

Table 4. Fertilization in crop rotation treatment without legumes (B).

Plants	Fertilization
	NPK			FM	½ NPK + ½ FM			FM	Control
	N	P	K		N	P	K
	kg ha−1			t ha−1	kg ha−1			t ha−1
Sugar beet	200	56.0	200	40	100	28.0	100	20	-
Spring barley	100	36.5	91.5	20	50	18.3	45.8	10	-
Winter rapeseed	100	36.5	91.5	20	50	18.3	45.8	10	-
Winter wheat	100	36.5	91.5	20	50	18.3	45.8	10	-

Mineral fertilization (NPK); farmyard manure (FM); mixed mineral and organic fertilization (½ NPK + ½ FM); and control without any fertilization (Control).

. Over the course of the study, the crop rotation sequence was varied from year to year. Since 1990, the following crop rotations have been used: (A) Norfolk rotation: sugar beet, spring barley with undersown red clover, red clover, and winter wheat; (B) crops without legumes: sugar beet, spring barley, winter rapeseed, and winter wheat.

Soil samples were regularly taken from the experimental plots. The basic properties of the soil are examined—N, P, K, Corg, and pH in KCl. Organic carbon content (Corg) in soil was determined after dry combustion, while the content of soil total nitrogen (N) was determined using the Kjeldahl method. Plant available forms of phosphorus and potassium were determined using the Egner−Riehm method. The measurement of soil pH was conducted potentiometrically in 1M solution of KCl.

The long-term effect of the crop rotation and fertilization treatments on major soil properties is shown in

Table 5. Effect of crop rotation and fertilization on soil properties.

Experimental Objects	pH KCl	C org g kg−1	N tot g kg−1	P mg kg−1	K mg kg−1
NONleg_NPK	6.10	8.43	0.82	90.58	107.80
NONleg_FM	6.55	11.11	0.93	103.90	183.93
NONleg_½NPK + ½FM	6.58	9.95	0.87	102.60	154.48
NONleg_0	6.55	7.25	0.73	60.93	62.33
Means for crop rotation	6.44	9.19	0.84	89.50	127.13
NOR_NPK	5.78	11.66	1.11	80.25	79.10
NOR_FM	6.35	13.88	1.27	79.23	118.88
NOR_½NPK + ½FM	6.03	12.41	1.16	77.63	107.88
NOR_0	5.95	10.76	0.97	41.33	49.63
Means for crop rotation	6.03	12.17	1.13	69.61	88.87

Nonleg—crop rotation without legumes; NOR—Norfolk rotation; mineral fertilization (NPK); farmyard manure (FM); mixed mineral and organic fertilization (½ NPK + ½ FM); and control without any fertilization (0).

. The organic carbon content and nitrogen content were the greatest in Norfolk rotation treatment. In both crop rotation treatments, experimental objects in manure-fertilized sites were characterized by increased organic carbon content and nitrogen content. These were lowest in both control groups (NONleg_0 and NOR_0). The Norfolk rotation treatment had a lower phosphorus and potassium content compared to the rotation treatment without the legume, independent of fertilization. The objects to which farmyard manure was applied were characterized by the highest content of phosphorus and potassium, except NOR_FM. In Norfolk rotation treatment, the phosphorus content was slightly higher in NOR_NPK. The soil pH was lower in Norfolk rotation treatment (6.03) in comparison to crop rotation treatment without legumes (6.44).

2.2. Methodology for Determining Yield and Yield Components

The grain yield (GY) was recorded by hand-harvesting the grain from one-square meter sample of each experimental plot and adjusting this to a 15% moisture level. Thousand grain weight (TGW) was calculated as the sum of the weight of two samples of 500 seeds/plot, according to Polish norm PN68/R-74017 (1968). The number of spikes (NS) was determined by counting and cutting off the spikes from samples taken from one square meter. The average number of grains per spike (NGS) was calculated by dividing the grain yield from one square meter by the number of spikes and the average grain weight.

2.3. Statistical Methods

The analysis of the study traits was performed using a single-stage approach for a linear mixed model (LMM):

y_ijklhn = µ + g_k + l_j + a_i + ga_ki + gl_kj + la_ji + gla_kji + b_jihn + e_ijklhn

(1)

where:

• y_ijklhn is the traits;
• µ is the overall mean;
• g_k is the fixed effect of kth fertilization treatment;
• l_j is the fixed effect of jth crop rotation;
• a_i is the random effect of ith growing season;
• ga_ki is the random interaction effect of kth fertilization treatment and ith growing season;
• gl_kj is the fixed interaction effect of kth fertilization treatment and jth crop rotation;
• la_ji is the fixed interaction effect of ith growing season and jth crop rotation;
• gla_kji is the random interaction effect of kth fertilization treatment, jth crop rotation, and ith growing season;
• b_jihn is the random effect of nth block nested in hth replication at jth crop rotation and ith growing season;
• e_ijklhn is the random error associated with the trait observation y_ijklhn.

The restricted maximum likelihood (REML) method estimated the linear mixed model. Wald’s test was used to test the significance of fixed effects; the significance of random effects was evaluated by variance components with their standard errors. Based on linear mixed model, we adjusted means of yield and study grain quality traits for main effects and appropriate combinations of factors (e.g., fertilization treatment × crop rotation × growing season). These averages allowed us to determine the correlation between the soil’s total precipitation and chemical properties. The Shukla variance stability parameter was also determined. This parameter evaluates the stability of both the yield and its components; the lower its value, the greater the stability. We ranked Shukla parameter values, in which one means the highest degree of trait stability.

Calculations were performed using R 4.4.2, and the stability measure was determined using the metan package.

3. Results

3.1. Yield and Yield Component Stability of Winter Wheat

The highest yield stability of winter wheat grown without legumes was recorded for NONleg_FM and NONleg_½NPK+½FM, while in the Norfolk rotation treatment, the highest yield stability was recorded for NOR_½NPK + ½FM and NOR_NPK. The variant of manure fertilization in the Norfolk rotation treatment (NOR_FM) adversely affected the wheat yield stability. Considering the yield components in the crop rotation treatment without legumes, the NGS and TGW were the most stable in the NONleg_½NPK + ½FM combination, while the NS showed the greatest stability in NONleg_NPK. In the Norfolk rotation treatment, all the studied elements of winter wheat yield structure (the NS, NGS, and TGW) had the highest stability in the object fertilized with NPK only (NOR_NPK). In the objects fertilized only with manure in the Norfolk rotation treatment (NOR_FM), the NGS and NS showed less stability. The sum of the stability rankings indicates that the most stable system for the yield and all components were the facilities fertilized with NPK only and with combined fertilization (mineral and organic). The NOR_FM experimental object was distinguished by a higher level of instability compared to NONleg_FM. The control objects (NONleg_0 and NORleg_0) were the most unstable in both rotation treatments (

Table 6. Yield and yield component stability of winter wheat and spring barley depending on the fertilization and crop rotation.

Experimental Objects	Ranking and Its Sum of Shukla Stability Variance for Yield and Yield Componets
	Winter Wheat					Spring Barley
	GY	NGS	NS	TGW	Sum	GY	NGS	NS	TGW	Sum
NONleg_NPK	3	2	1	2	8	3	3	2	1	9
NONleg_FM	1	3	2	3	9	1	1	1	3	6
NONleg_½NPK + ½FM	2	1	3	1	7	2	2	3	2	9
NONleg_0	4	4	4	4	16	4	4	4	3	15
NOR_NPK	2	1	1	1	5	1	2	2	2	7
NOR_FM	4	4	3	2	13	3	3	4	1	11
NOR_½NPK + ½FM	1	3	2	3	9	2	1	1	4	8
NOR_0	3	2	4	4	13	4	4	3	3	14

Nonleg—crop rotation without legumes, NOR—Norfolk rotation; mineral fertilization (NPK); farmyard manure (FM); mixed mineral and organic fertilization (½ NPK + ½ FM); and control without any fertilization (0).

).

3.2. Yield and Yield Component Stability of Spring Barley

The results for spring barley in the crop rotation treatment without legumes showed that the yield stability was the highest for NONleg_FM and NONleg_½NPK + ½FM. With manure fertilization in this rotation treatment (NONleg_FM), the yield components NGS and NS also had the highest stability. The TGW was the most stable in NONleg_NPK. The most stable spring barley yields were obtained in NOR_NPK and NOR_½NPK + ½FM in the Norfolk rotation treatment. With combined fertilization, the yield components NGS and NS also showed the greatest stability. The TGW was the most stable in this rotation treatment at NOR_FM. All yield components achieved a relatively high level of stability in NOR_NPK. The sum of the stability rankings shows that the NONleg_FM combination had the highest stability for yield and all components in the non-bean crop rotation treatment, while the NOR_NPK combination had the highest stability in the Norfolk rotation treatment. Similarly, as in the case of winter wheat, less stability was observed in NOR_FM for the yield and components of spring barley (Table 6).

3.3. Winter Wheat Yield and Yield Component Response to Crop Rotation and Fertilization

The winter wheat in the Norfolk rotation treatment showed the highest mean yields (5.39 t ha⁻¹) compared to the yields in the non-legume rotation treatment (4.34 t ha⁻¹). The highest mean yield of wheat was obtained in the objects with only NPK fertilization in both crop rotation treatments. In the crop rotation treatment without legumes, there was no significant difference in the winter wheat yield between NONleg_FM and NONleg_½NPK + ½FM. The yields in plots without fertilization in the non-legume rotation treatment (NONleg_0) were significantly lower than those in the control in the Norfolk rotation treatment (NOR_0). Considering the yield components, there were no statistically significant differences between the different fertilizer combinations in the non-beavertail rotation treatment for the NGS and TGW. Only the control object (NONleg_0) for these components differed significantly. All fertilizer combinations differed in terms of the NS, and significantly higher values for the NS were obtained in NONleg_NPK. In the Norfolk rotation treatment, the control (NOR_0) also obtained significantly lower values for all yield components tested. The highest NGS values were recorded in NOR_NPK, and no significant differences existed between NOR_FM and NOR_½NPK + ½FM. Also, for the NS, the highest values were recorded in NOR_NPK, and no differences were found between NOR_NPK and NOR_½NPK + ½FM. The TGW had the highest values on NOR_NPK and NOR_FM; significantly lower values were recorded on NOR_½NPK + ½FM (

Table 7. Mean comparison of the winter wheat and spring barley grain yield and yield components depending on the crop rotation and N fertilization treatment.

Experimental Objects	Winter Wheat				Spring Barley
	GY	NGS	NS	TGW	GY	NGS	NS	TGW
NONleg_NPK	5.48 ± 0.35 c	30.4 ± 1.23 b	549 ± 13 d	40.9 ± 3.87 b	5.67 ± 0.34 c	20.2 ± 1.21 b	626 ± 14 c	45.8 ± 4.32 b
NONleg_FM	4.32 ± 0.31 b	29.9 ± 1.63 b	488 ± 11 b	41.5 ± 3.91 b	4.54 ± 0.29 b	20.3 ± 1.42 b	604 ± 12 b	47.3 ± 4.43 c
NONleg_½NPK+½FM	5.13 ± 0.33 c	29.7 ± 1.32 b	509 ± 13 c	41.8 ± 3.98 b	4.98 ± 0.38 bc	20.2 ± 1.34 b	662 ± 11 d	45.8 ± 4.28 b
NONleg_0	2.44 ± 0.29 a	23.6 ± 1.09 a	372 a ± 10	37.7 ± 378 a	1.90 ± 0.23 a	16.2 ± a1.41	382 ± 9.2 a	42.7 ± 3.82 a
Mean for crop rotation without legume	4.34 ± 0.33	28.4 ± 1.12	480 ± 12	40.4 ± 3.82	4.27 ± 0.31	19.2 ± 1.43	569 ± 11	45.4 ± 3.28
NOR_NPK	6.12 ± 0.36 c	33.3 ± 1.28 c	527 ± 14 c	41.2 ± 4.11 c	4.83 ± 0.35 bc	19.9 ± 1.28 b	569 ± 14 b	45.7 ± 3.72 b
NOR_FM	5.51 ± 0.32 b	31.2 ± 1.34 b	499 ± 13 b	41.1 ± 4.08 c	4.51 ± 0.43 b	20.5 ± 1.39 b	559 ± 11 b	46.4 ± 2.45 b
NOR_½NPK+½FM	5.86 ± 0.31 bc	31.6 ± 1.35 b	525 ± c15	39.4 ± 3.93 b	5.22 ± 0.31 c	19.7 ± 1.37 b	622 ± 12 c	44.5 ± 2.46 a
NOR_0	4.05 a ± 0.30	26.8 ± 1.33 a	417 ± 9.4 a	36.0 ± 3.87 a	2.54 ± 0.27 a	17.0 ± 2.01 a	378 a ± 9.6	43.9 ± 3.82 a
Mean for Norfolk crop rotation	5.39 ± 0.32	30.7 ± 1.27	492 ± 12	39.4 ± 4.01	4.28 ± 0.34	19.3 ± 1.41	532 ± 10	45.1 ± 2.74

Data followed by the same letter within a column or group of treatments do not differ significantly at the p < 0.05 level; Nonleg—crop rotation without legumes; NOR—Norfolk rotation; mineral fertilization (NPK); farmyard manure (FM); mixed mineral and organic fertilization (½ NPK + ½ FM); and control without any fertilization (0).

).

3.4. Spring Barley Yield and Yield Component Response to Crop Rotation and Fertilization

There were no statistically significant differences in barley yields between the crop rotation treatments. The highest yield of spring barley in the NONleg crop rotation treatment was when fertilizing only with NPK (NONleg_NPK), but in the Norfolk rotation treatment, the highest yield was obtained by applying mineral and organic fertilization together (NOR_½NPK + ½FM). As with wheat, significantly lower yields were obtained in the control in the NONleg crop rotation treatment (NONleg_0) in comparison to NOR_0. In the crop rotation treatment without legumes, there were no significant differences between fertilizer combinations for the NGS; only the NGS values in the control object (NONleg_0) were significantly lower. The highest NS values were obtained in NONleg_NPK, while the highest TGW values were obtained in NONleg_FM and differed significantly from the other experimental plots. In the Norfolk rotation treatment, the NGS values were not significantly different from each other, except for the control (NOR_0), in which they were much lower. Significantly a higher NS value was observed for the NOR_½NPK + ½FM combination, and NOR_NPK and NOR_FM did not differ from each other for this yield component. There were no significant differences between NOR_NPK and NOR_FM for the TGW, and the highest values for this trait were recorded on these objects. No significant differences were found for the TGW between the control (NOR_0) and NOR_½NPK + ½FM (Table 7).

3.5. Relationship Between Selected Soil Properties and Yield and Yield Component Stability of Winter Wheat and Spring Barley

The organic carbon content and total nitrogen content were strongly correlated with the wheat yield stability. When considering the components of the winter wheat yield, there is a correlation between Corg and the NGS stability and a strong correlation between Corg and N_tot with the NS stability. For the TGW stability, a correlation with the P content was noted. For spring barley, a correlation of the P and K content with the yield stability was observed, as well as the NGS. For the NS and TGW stability, there were no strong correlations with soil properties (

Table 8. Correlation coefficient between selected soil properties and stability of yield and yield components of winter wheat and spring barley.

Soil Properties	Correlation Coefficient
	Winter Wheat				Spring Barley
	GY	NGS	NS	TGW	GY	NGS	NS	TGW
pH KCl	0.4990 **	0.3912 *	0.6322 *	−0.4455 *	0.2236	0.0912	0.3578 *	−0.4686 **
C org g/kg	−0.8038 **	−0.5054 **	−0.7959 **	0.1988	−0.4450 *	−0.5369 **	0.2208	0.3079 *
N tot g/kg	−0.7513 **	−0.4195 *	−0.8218 **	0.2109	−0.3422 *	−0.4897 **	0.2647	0.3389 *
P mg/kg	0.0425	−0.4567 **	−0.2435	−0.8669	−0.6699 **	−0.8054 **	−0.0242	−0.4946 **
K mg/kg	−0.1072	−0.4211 *	−0.1812	−0.6298 **	−0.6145 **	−0.7007 **	0.0244	−0.3546 *

*—significant at the p < 0.05 level.; **—significant at the p < 0.001 level; Nonleg—crop rotation without legumes; NOR—Norfolk rotation; mineral fertilization (NPK); farmyard manure (FM); mixed mineral and organic fertilization (½ NPK + ½ FM); and control without any fertilization (0).

).

3.6. Relationship Between the Total Rainfall in the Spring Months and the Stability of Yields and Yield Components of Winter Wheat and Spring Barley

In the fertilizer plots to which no manure was applied in both crop rotation treatments, a strong correlation was found between the yield stability and total rainfall in May. In the plots to which manure was applied (NONleg_FM, NONleg_½NPK + ½FM, NOR_FM, and NOR_½NPK + ½FM), the correlations between the yield stability and total rainfall in May were much weaker. For barley, strong negative correlations between the yield stability and total precipitation in April were observed for all fertilizer combinations except NONleg_NPK (

Table 9. Correlations between winter wheat yield and total precipitation in spring for four fertilizer objects and two crop rotations.

Fertilizer Objects		Winter Wheat		Spring Barley
	April	May	June	April	May	June
NONleg_NPK	0.3599 *	0.8788 **	0.0651	−0.4719 *	0.6722 **	−0.3646 *
NONleg_FM	0.0910	0.3136 *	−0.445 *	−0.7513 **	0.5059 **	−0.5369 **
NONleg_½NPK + ½FM	−0.1065	0.3243 *	−0.0305	−0.6920 **	0.3084 *	−0.5951 **
NONleg_0	−0.1779	0.8157 **	0.3404 *	−0.7784 **	0.5637 **	−0.3958 *
NOR_NPK	0.2779	0.7233 **	−0.1506	−0.9410 **	0.4227 *	−0.7297 **
NOR_FM	0.0560	0.4520 *	−0.325 *	−0.7608 **	0.6020 **	−0.4367 *
NOR_½NPK + ½FM	0.3871 *	0.2622	−0.6317 **	−0.9177 **	0.6163 **	−0.7443 **
NOR_0	0.3441 *	0.7528 **	0.1873	−0.8344 **	0.7480 **	−0.5111 **

*—significant at the p < 0.05 level.; **—significant at the p < 0.001 level; Nonleg—crop rotation without legumes; NOR—Norfolk rotation; mineral fertilization (NPK); farmyard manure (FM); mixed mineral and organic fertilization (½ NPK + ½ FM); and control without any fertilization (0).

).

3.7. Effects of Crop Rotation, Fertilization, Growing Season, and Their Interaction on the Grain Yield Variability of Winter Wheat and Spring Barley

For the winter wheat and spring barley, fertilization explains 57% and 56% of yield variability, respectively, in the non-legume crop rotation treatment. In contrast, in the Norfolk crop rotation treatment, the growing season determines most of the variability: 48.7% for winter wheat and 52.3% for spring barley (

Table 10. Effects of fertilization, growing season, and their interaction on the grain yield variability of winter wheat and spring barley depending on the crop rotation.

Proportion of Variation (%) Depending on Crop Rotation System
	Winter Wheat		Spring Barley
Factor	NONleg	NOR	NONleg	NOR
Growing Season	24.0 *	48.7 **	32.9 **	52.3 **
Fertilization	57.1 **	39.7 **	56.1 **	39.2 **
Growing season × Fertilization	18.9 *	11.6 *	11.0 *	8.5 *

*—significant at the p < 0.05 level.; **—significant at the p < 0.001 level; Nonleg—crop rotation without legumes; NOR—Norfolk rotation.

).

In the NPK fertilized plots for the winter wheat, the growing season explained 77.6% of the yield variation; for the plots fertilized with manure (FM) and with mixed fertilization (½NPK + ½FM), the growing season explained 55.3% and 53.4% of the variation, respectively. The cropping system only explained 59.8% of the variability in the control. For the spring barley, the growing season explained 82.2% of the variability for the NPK-fertilized objects and 91.6% for the FM-fertilized objects (

Table 11. Effects of cropping system, growing season, and their interaction on the grain yield variability of winter wheat and spring barley depending on the fertilization.

Proportion of Variation (%) Depending on Crop Rotation System
	Winter Wheat				Spring Barley
Factor	NPK	FM	½NPK + ½FM	Control	NPK	FM	½NPK + ½FM	Control
Growing season	77.6 **	55.3 **	53.4 **	28.9 **	82.2 **	91.6 **	88.5 **	88.4 **
Crop rotation	8.4 *	30.6 **	36.7 **	59.8 **	6.2	1.0	2.6	1.2
Growing season × Crop rotation	13.9 **	14.1 *	10.0 *	11.3 *	11.6 *	7.4 *	8.9 *	10.4 *

*—significant at the p < 0.05 level.; **—significant at the p < 0.001 level; mineral fertilization (NPK); farmyard manure (FM); mixed mineral and organic fertilization (½ NPK + ½ FM); and control without any fertilization (0).

).

4. Discussion

In this study, we found that for both species (winter wheat and spring barley), the combination of mineral fertilization and farmyard manure (NONleg_½NPK + ½FM) provided the best yield stability out of the crop rotation treatments without legumes. Additionally, the application of manure (NONleg_FM) has a stabilizing effect on the wheat and barley yields in this crop rotation treatment. Although NPK’s application may have a positive effect on the yield and yield components, the regular application of manure is effective in ensuring the yield stability. The highest yield stability of the winter wheat cultivated without legumes was noted for NONleg_FM and NONleg_ ½ NPK + ½ FM, while for the spring barley, the yield stability was highest for NONleg_FM and NONleg _ ½ NPK + ½ FM. In the case of the winter wheat, the number of grains per spike (NGS) and thousand grain weight (TGW) were most stable in the combination of NONleg_ ½ NPK + ½ FM, while the NS showed the greatest stability in NONleg_NPK. For the spring barley, the NGS and number of spikes (NS) showed the highest stability in NONleg_FM. The TGW was the most stable in the NONleg_NPK rotation treatment. Macholdt et al. [8] also indicated that the yield stability was improved when applying a mineral supply of N + P + K with additional manure for winter wheat. The same research team determined the impact ranking of mineral fertilizers on the production risk and yield stability, indicating N as the most important mineral, followed by K and P. Mixed manure with fertilizer use can improve crop yield stability and sustainability [21]. The beneficial effect of balanced fertilization with manure and NPK on the yield stability of soybeans and wheat was also recorded by Wankhede et al. [17], who, when studying the effect of a crop rotation treatment after 43 years of adding manure in combination with NPK in India, noticed less yield variability compared to a control with NPK and fallow plots.

The stability was different in the Norfolk rotation treatment, in which the combination of NPK and manure (NOR_½NPK + ½ FM), as well as mineral fertilization alone (NOR_NPK), stabilized wheat and barley yields the most. For the winter wheat, the highest NGS and NS values were recorded for NOR_NPK, and there were no significant differences between NOR_FM and NOR_ ½ NPK + ½ FM. In the Norfolk crop rotation treatment, the NGS values in the treatments with fertilizer application did not differ significantly from each other. A significantly higher NS was observed in the NOR_ ½ NPK + ½ FM combination treatment, and NOR_NPK and NOR_FM did not differ from each other in this yield component. There were no significant differences between NOR_NPK and NOR_FM in terms of TGW. However, in the Norfolk rotation treatment, fertilization with farmyard manure (NOR_FM) had a negative impact on the stability of both the winter wheat and spring barley yields. Similar results were obtained by Hlisnikovský et al. [22] and Macholdt et al. [18]. In contrast, Berzsenyi et al. [2] indicated that in the Norfolk rotation treatment, the stability and yield of farmyard manure were the highest, while the stability of the NPK mineral fertilizer was lower. In the overall ranking of stability in the non-legume rotation treatment, the NONleg_FM combination has the most favorable effect, followed by the Norfolk rotation treatment (NOR_NPK).

Organic and mineral fertilization can significantly influence the yield of winter and spring cereals. Our research shows that the addition of manure over a period of many years can improve not only the height but also the stability of yields over time and effectively build up soil carbon and nitrogen reserves. However, it has also been shown that relying solely on manure fertilization without mineral NPK supplementation is risky and can decrease the stability of yield components and the overall yield. In a long-term fertilization experiment in Germany by Macholdt et al. [23], it was confirmed that the supply of NPK minerals to soil with additional manure ensured the best stability of winter wheat yields. Analyses of the effects of different fertilization regimes on plant yields and their stability in other climatic conditions also confirm that the combined use of mineral fertilizers and manure can improve yield stability in years with variable environmental conditions. Shi et al. [24] demonstrated in a moderate temperate semi-arid climate that the combined use of manure and mineral fertilizers improves soil carbon stocks and can reduce the impact of climatic conditions on crop yields over the years. Similarly, the study by Wankhede et al. [17] in a subtropical region confirms that NPK + manure fertilization can lead to yield stability while improving the soil carbon content and other soil properties.

On the other hand, the study carried out by Studnicki et al. [25] does not fully confirm this pattern in the case of winter rye, for which high and stable yields were ensured by NPK mineral fertilization alone. The authors indicate that manure has a positive effect on soil carbon reserves and may be more important in shaping yield stability under conditions of limited nitrogen supply, while under optimal NPK fertilization conditions, the impact of manure on the grain yield stability decreases. In our experience, this may explain the observed lower soil nitrogen concentration and the improvement in yield stability under conditions of a regular supply of manure only in the non-leguminous crop rotation treatment (NOR_FM and NONleg_NPK) (Table 5). In the crop rotation treatment with an additional nitrogen source from a legume plant, the stability of cereal yields was greatest in the variant with only NPK mineral fertilization (NOR_FM and NONleg_NPK).

In our experience, the yields of the winter wheat and the yield components in all fertilizer variants, including the control facility, were better in the rotation treatment with a legume. In general, the pre-crop value of legumes for cereals, including winter wheat, is very high, which has been confirmed by numerous studies. At the same time, the authors point out that the beneficial effect of legumes on the winter wheat’s yield stability depends on the fertilization regime used and may be limited by high levels of nitrogen in the fertilizer. Furthermore, the positive effect of pre-crops from different legume species on winter wheat yields and their stability may be more pronounced with longer crop rotation treatment cycles, especially with low N fertilization [26,27]. In particular, the beneficial effects of red clover on crops in the crop rotation treatment and on the soil are well known. They are mainly due to its ability to fix N and transfer it to subsequent crops and to improve the quality of arable soils, including soil organic matter [3,28]. The spring barley, on the other hand, had an average yield similar to that in the crop rotation treatment with and without a legume. Still, under the treatments with NPK fertilization only and manure only, lower yields of barley undersown with red clover (NOR_NPK and NOR_FM) were recorded. Research by Kunelius et al. [29] indicates that sowing clover in barley can reduce its yield by 10% compared to sowing it alone. However, studies on the combined sowing of barley and clover under conditions of limited nitrogen supply from synthetic fertilizers, despite some downward trends, do not indicate a significant limiting effect of the presence of live clover on the yield of spring barley [30]. In the study by Känkänen et al. [31], no significant effect of undersown red clover on spring barley yields was found. At the same time, a beneficial effect of the residual nitrogen from the clover on plants in crop rotation treatments in subsequent crop cycles was indicated [32]. Improved soil fertility and the supply of residual N in the legume-based crop rotation treatment may have been the reason for better barley yields in NOR_0, to which residual nitrogen was of key importance.

Our research has shown that in the fields where NONleg_FM, NONleg_½NPK + ½FM and NOR_FM, and NOR_½NPK + ½FM manure was used, the yield stability of both species correlated with the total rainfall in the spring months. This may indicate that sites with a high organic matter content are less sensitive to water shortages in May, when the water demand is high, and can better cope with drought and rainfall shortages. In the case of winter wheat, strong correlations were found with rainfall in May in the control plots and with exclusive NPK fertilization in both crop rotation treatments. The stability of barley yields showed a strong relationship with the total rainfall in April, especially in the crop rotation treatment with clover sowing.

Based on the research of many authors, it can be concluded that sustainable mineral fertilization in combination with organic fertilization has a positive effect on the SOC. Seremesic et al. [33] found that soil organic carbon preservation, coupled with proper management, such as crop rotation and fertilization, is important for preserving soil productivity, and when soil organic carbon increases, this could increase winter wheat’s yield. Wankhede et al. [17] indicated that the total SOC stock increased in objects with mineral and manure fertilization in comparison to plots with only NPK and fallow plots. However, this study also shows that organic carbon and nitrogen content strongly correlated with winter wheat yield stability. Such a strong correlation was not observed for spring barley. Our research conducted on the basis of a static field experiment with a varied crop rotation and fertilization method indicates that from a long-term perspective, the crop rotation treatment based on legumes and manure has significant potential in reducing the dependency of crops on mineral fertilizers in the stabilization of winter and spring cereal yields. This is due to the fact that repeated incidents of the presence of legumes in crop rotation treatments not only have a beneficial effect on the cereal plants occurring immediately after them, but also support the supply of nitrogen to the soil and crops in subsequent rotations. Legume-based crop rotation treatments together with manure are a recommended long-term sustainable tool for building the health of arable soils and resistant soil-crop systems in rainfed conditions, which plays a critical role in strengthening crop resistance to progressive climate change. In the future, research should be extended to analyze the relationship between soil and plant health indicators under varied rotation and fertilization treatments in variable environmental conditions.

5. Conclusions

In this long-term experiment, it was shown that the combined application of mineral fertilizers and manure can be a tool for improving the stability of cereal yields in years with variable environmental conditions (Q1). Relying solely on manure fertilization without mineral NPK supplementation is risky and can generally result in a decrease in the stability of yield components and yields. On the other hand, it has been shown that organic matter and nitrogen content are strongly correlated with the stability of winter wheat yields. The addition of red clover undersown in spring barley stabilized its yield, and the combinations that were additionally suitably fertilized showed the greatest yield stability, number of grains per spike (NGS), and number of spikes (NS) (Q2). Organic-matter-rich sites, i.e., those with the addition of manure, were less sensitive to water stress in May, when the water demand of cereals is high, and this allowed them to better tolerate the lack of rainfall (Q3). In addition, our research confirms that long-term studies are needed to synthesize the benefits of diversified crop rotation and fertilization treatments under different land management practices and climatic conditions for the development of climate-resilient and stable cropping systems. Long-term experiments are important as a source of data for decision makers and could be a foundation for climate change adaptation strategies.