Effects of the Combined Application of Nitrogen, Phosphorus, and Potassium Under Drip Irrigation on the Yield and Quality of Winter Wheat

Abstract

A two-year field experiment was conducted to clarify the regulatory effects of nitrogen (N), phosphorus (P), and potassium (K) combined with drip fertigation on the yield, yield components, and grain quality of winter wheat in lime concretion black soil (Calcaric Cambisols). The objective was to screen a sustainable fertilization model for coordinating high yield and quality in the Huang-Huai-Hai Plain. An L₁₆(4³) orthogonal design was adopted to investigate yield, protein content, wet gluten, test weight (TW), and grain hardness. Range analysis and ANOVA were used to evaluate factor effects and interactions. The results showed that N was the dominant factor affecting yield and quality (Rank 1), followed by K (Rank 2), while P showed the weakest effect. Compared to the control (N0P0K0), the optimized N–P–K combination increased grain yield by an average of 315.0% and enhanced grain crude protein by 55.3% over the two seasons. The optimal combination for maximum yield was N170P30K120 (kg/ha), which optimized the source–sink relationship by balancing spike density and 1000-grain weight. High N (220 kg/ha) combined with low P and high K achieved the best nutritional quality. The 3D response surface analysis confirmed significant synergistic interactions between N–K and N–P in promoting grain filling and protein synthesis. Rational NPK drip fertigation, particularly when synchronized with critical growth stages (jointing and grain filling), can simultaneously enhance grain yield and quality in this soil type. The optimized combination provides theoretical support and a robust fertilization strategy for green and efficient wheat production in the region.

1. Introduction

Winter wheat (Triticum aestivum L.) is a core staple crop globally, playing a pivotal role in safeguarding food security [1]. Wheat production is increasingly focused on improving both yield and grain quality to meet modern dietary demands. The protein components, gluten network structure, and physical properties (e.g., test weight) of grains directly determine the suitability of flour for end-use processing and its commercial value. However, at the crop physiological level, there is often a significant trade-off between high yield and high quality; specifically, a substantial increase in yield is frequently accompanied by a dilution of grain protein concentration [2]. How to break this bottleneck through scientific agronomic practices has become a frontier focus in the international fields of crop nutrition and cereal science. Specifically, managing the synergetic effects of multi-nutrients is crucial for achieving the “win-win” of high productivity and superior processing quality.

Precision nutrient management, particularly drip fertigation, serves as a key driver to mitigate the trade-off between yield and quality by ensuring nutrient supply is synchronized with the crop’s peak physiological demands [3]. Nitrogen, as the primary substrate for protein synthesis, directly defines the lower limit of wheat nutritional quality [4]; as a vital osmotic regulator and enzyme activator, potassium (K) plays a pivotal role in optimizing source–sink dynamics. By facilitating the efficient translocation of photoassimilates from leaves to grains, K enhances grain filling and synergistically modulates nitrogen metabolism through nutrient interactions [5]; phosphorus is widely involved in energy metabolism, ensuring the normal development of reproductive organs [6]. Nevertheless, in current intensive agricultural production, an empirical fertilization pattern of “overemphasizing nitrogen while neglecting potassium, and applying phosphorus blindly” is prevalent. This not only exacerbates the risk of non-point source pollution but also leads to nutrient imbalance in the late growth stage of wheat, easily causing problems such as excessive vegetative growth with delayed maturity or premature senescence under high temperature, which severely restricts the formation of high-grade processing quality.

The Huang-Huai-Hai Plain is the largest winter wheat-producing area in China, where lime concretion black soil (classified as Calcaric Cambisols according to the WRB system) is the dominant soil type. This soil type is characterized by its heavy clay texture, compact structure, poor aeration, and low nutrient availability, which together restrict root development and high-yield production [7]. Currently, most fertilization studies in this region are limited to the yield-increasing effects of single nutrients or conventional field phenotypes, and there is a lack of systematic analysis on the comprehensive impacts of the combined application of nitrogen (N), phosphorus (P), and potassium (K) on both yield and quality from a multi-factor interaction perspective. In particular, the potential for phosphorus reduction under high soil phosphorus background, and the regulation of “quality lag behind yield” by high nitrogen–potassium interaction, remain unclear.

Based on the above background, a field experiment was conducted in a typical lime concretion black soil area using an L₁₆(4³) randomized complete block orthogonal design to efficiently isolate the effects of multi-nutrient interactions with a minimized experimental scale [8]. This study systematically evaluated the effects of different N–P–K combined fertilization modes on winter wheat yield and key quality indicators. We hypothesize that appropriate reduction in phosphorus input under high soil phosphorus background, combined with precision drip topdressing of nitrogen at the critical jointing stage and sufficient potassium supply, can significantly optimize the source–sink relationship and enhance grain quality potential [9]. The ultimate aim of this study was to identify a sustainable fertilization model that achieves the dual goals of yield enhancement and quality improvement in Calcaric Cambisols areas.

2. Materials and Methods

2.1. Experimental Site and Soil Conditions

Field experiments were conducted from October 2023 to June 2025 at the Jiaozhou Modern Agricultural Demonstration Park of Qingdao Agricultural University (35.53° N, 119.58° E). The study area is characterized by a subhumid monsoon climate. The previous crop in the experimental field was summer maize. The soil at the site is classified as lime concretion black soil.

Representative soil samples from the 0–20 cm layer were collected in October 2023, prior to the sowing of the first winter wheat crop. The baseline physical and chemical properties were analyzed at the Laboratory of the College of Agronomy, Qingdao Agricultural University. The soil’s physical and chemical properties are presented in

Table 1. Basic physical and chemical properties of the 0–20 cm topsoil.

Total Nitrogen (g/kg)	Organic Matter (g/kg)	Available Phosphorus (mg/kg)	Available Potassium (mg/kg)	Alkali-Hydrolyzed Nitrogen (mg/kg)	Soil pH
0.89 ± 0.04	12.97 ± 0.64	38.20 ± 4.02	112.36 ± 9.58	84.19 ± 4.21	7.31 ± 0.10

. Detailed precipitation and daily mean temperature data for the two growing seasons are presented in Figure 1.

2.2. Experimental Materials

The tested winter wheat cultivar was ‘Denghai 206’ (National Certified No. 20190038), a semi-winter variety developed by Shandong Denghai Seeds Co., Ltd. (Laizhou, China). from a cross between Jimai 22 and Zhoumai 20. The total growth period of this cultivar is approximately 242 days. It is recognized as a high-yield and high-quality variety; in high-yield demonstration fields in Shandong, its grain yield has reached 13.45 t/ha, with a standard test weight of 825 g/L and a crude protein content of 13.4%.

2.3. Experimental Design, Fertilization Schemes, and Plot Layout

A total of 48 experimental plots (16 treatments × 3 replicates) were arranged using a completely randomized design based on an L₁₆(4³) orthogonal array design. This design, also known as the Taguchi method, was selected for its high efficiency in identifying the main effects and potential interactions among multiple fertilization factors with a reduced number of experimental trials [8]. Each experimental plot occupied an area of 40 m² (16 m × 2.5 m). To prevent the lateral movement of water and nutrients between adjacent treatments, plots were separated by 0.3 m wide buffer rows and were physically isolated by ridges. The experimental factors consisted of four levels of nitrogen (N), Phosphorus (P₂O₅), and potassium (K₂O), as specified in

Table 2. Experimental factors and their levels.

	Factor	N/ (kg/ha)	P/ (kg/ha)	K/ (kg/ha)
Level
1		0	0	0
2		120	30	40
3		170	60	80
4		220	90	120

and

Table 3. Experimental design scheme.

Treatment	Level Scheme			Implementation Scheme
	N	P	K	N/ (kg/ha)	P/ (kg/ha)	K/ (kg/ha)
1	1	1	1	0	0	0
2	1	2	2	0	30	40
3	1	3	3	0	60	80
4	1	4	4	0	90	120
5	2	1	2	120	0	40
6	2	2	1	120	30	0
7	2	3	4	120	60	120
8	2	4	3	120	90	80
9	3	1	3	170	0	80
10	3	2	4	170	30	120
11	3	3	1	170	60	0
12	3	4	2	170	90	40
13	4	1	4	220	0	120
14	4	2	3	220	30	80
15	4	3	2	220	60	40
16	4	4	1	220	90	0

.

A unified drip irrigation management system (equal volume and frequency) was strictly implemented. Supplemental irrigation was conducted twice during critical growth stages: at the jointing stage (Feekes scale 6.0) and the grain-filling stage (Feekes scale 11.0). The irrigation quota for each event was 60 mm, resulting in a total drip irrigation quota of 120 mm for the entire growth cycle. Drip tapes were placed on the soil surface between rows with a spacing of 60 cm (one tape for every three adjacent rows). The water volume for each plot was precisely measured and monitored via independently installed water meters.

2.4. Fertilization and Field Management

The fertilizers included urea (N = 46%), superphosphate (P₂O₅ = 12%), and potassium sulfate (K₂O = 40%). Basal Fertilization: 100% of P and K fertilizers, plus 50% of the N fertilizer, were uniformly incorporated into the 0–20 cm soil layer during tillage prior to sowing. Topdressing: the remaining 50% of N was applied as drip fertigation at the jointing stage (Feekes 6.0), ensuring precise nutrient delivery to the crop root zone. Wheat was sown at a rate of 185 kg/ha. Sowing dates were 24 October 2023 and 1 November 2024; harvest dates were 13 June 2024 and 17 June 2025, respectively. Integrated pest, disease, and weed management followed local high-yield cultivation standards.

2.5. Measurement of Yield and Yield Components

Before maturity, the number of spikes per unit area was determined by randomly selecting three 1-m double-row samples from each plot. These data were subsequently converted to calculate the number of spikes per hectare. The number of grains per spike was counted from 20 randomly selected plants per plot. At maturity, a 3 m² area in the center of each plot was manually harvested to determine the actual grain yield (adjusted to 13% moisture content). The 1000-grain weight was determined using three replicates per treatment.

2.6. Measurement of Grain Quality Indicators

The grain protein content was determined using the semi-micro Kjeldahl method (N × 5.7). Wet gluten content was measured using a gluten washing instrument and index tester (YUCEBAS, İzmir, Turkey) according to standard procedures. Grain test weight was recorded using a 1000 mL graduated cylinder and balance. Grain hardness was determined using a Fourier transform near-infrared spectroscopy analyzer (Antaris II, Madison, WI, USA).

2.7. Data Processing and Analysis

Initial data collation was performed in Microsoft Excel 2019. Statistical analysis was conducted using the DPS v19.05 software (Hangzhou Reifeng Technology Co., Ltd., Hangzhou, China; http://www.dpsw.cn, accessed on 2 January 2026). Figures were generated using Origin 2021 software (OriginLab Corp., Northampton, MA, USA; https://www.originlab.com, accessed on 21 February 2026). Prior to performing the analysis of variance (ANOVA), the normality of the data distribution was verified using the Shapiro–Wilk test, and the homogeneity of variance was assessed using Levene’s test. Outliers were identified using the box-plot method and managed after verifying experimental records to ensure data integrity. Post-hoc multiple comparisons were conducted using Duncan’s Multiple Range Test (p < 0.05) to identify significant differences between treatments.

3. Results

3.1. Effects of NPK Combined Application on Winter Wheat Productivity and Quality

The grain yield, yield components, and quality traits of winter wheat showed significant responses to the varying combinations of N, P, and K fertilizers over the two experimental seasons (

Table 4. Yield and Quality Indicators.

Year	Treatment	Spike Number (×104)	Grains Number	1000-Grain Weight (g)	Grain Yield (kg/ha)	Grain Crude Protein (%)	Wet Gluten (%)	Test Weight (g/L)	Grain Hardness Index
2023–2024	N1P1K1	309.3 e	22 e	43.1 e	2960 f	8.9 e	19.8 e	808.5 a	49.1 a
	N1P2K2	328.5 e	23.5 e	44.3 d	3353.3 f	9.1 e	20.3 e	778.6 c	42 cd
	N1P3K3	313.5 e	24.2 e	45.2 d	3440 f	9.2 e	20.6 e	775.2 c	41.5 d
	N1P4K4	336 e	23.8 e	46.5 cd	3693.3 f	9.2 e	20.5 e	773.6 c	40.1 d
	N2P1K2	563.6 d	34.7 d	46.1 cd	8713.3 e	10.5 d	24.1 fg	771.5 cd	39.5 de
	N2P2K1	552.8 d	35.2 cd	42.2 e	8126.7 e	10.3 e	23.8 d	768.5 d	34.5 e
	N2P3K4	545.1 d	36.7 b	49.5 b	10,226.7 cd	10.8 d	24.9 d	768.1 d	34.2 e
	N2P4K3	559.4 d	37.3 b	48.7 bc	10,073.3 cd	10.6 d	24.5 d	765.4 d	33.8 e
	N3P1K3	628.7 bc	36.3 bc	49.2 b	11,033.3 c	11.9 c	27.6 cd	762.3 d	32.5 e
	N3P2K4	640.3 b	38.5 a	51.7 a	12,653.3 a	12.5 bc	29.1 bc	801.4 ab	48.5 ab
	N3P3K1	617.8 c	37.2 b	43.3 e	10,060 cd	12 c	27.9 cd	797.2 ab	48 b
	N3P4K2	639.5 b	36.8 b	47.8 bc	11,100 bc	12.1 c	28.3 c	795.3 ab	47.5 bc
	N4P1K4	671.9 a	34.7 d	48.1 bc	11,486.7 b	13.3 ab	30.8 ab	792.1 ab	47.2 bc
	N4P2K3	656.2 ab	35.3 cd	47.3 c	11,166.7 bc	13.6 a	31.5 a	789.2 b	46.8 bc
	N4P3K2	667.8 a	33.5 d	45.4 d	10,313.3 c	13.2 b	30.5 b	788.5 b	46.5 c
	N4P4K1	671.9 a	34.2 d	42.6 e	9693.3 d	12.9 b	29.9 bc	785.6 b	46.2 c
2024–2025	N1P1K1	302.5 e	22.4 e	42.2 f	3114.9 g	9 e	20.5 d	768.5 d	33.9 e
	N1P2K2	317.5 e	23.8 e	43.8 e	3574.8 g	9.2 e	20.8 d	772.1 d	35.1 e
	N1P3K3	310.2 e	24.5 e	44.5 e	3414.5 g	9.4 e	21.2 d	776.4 cd	36.5 e
	N1P4K4	325.8 e	24 e	45.2 d	3659.8 g	9.3 e	21 d	779.8 c	37 e
	N2P1K2	553.8 d	34.1 d	45.4 d	8873.6 e	10.9 d	25.3 c	785.4 c	41.9 d
	N2P2K1	546.7 d	35.5 cd	41.4 f	8084.5 f	10.9 d	25.1 c	771.2 d	41.5 d
	N2P3K4	565.5 d	36.5 bc	48.6 b	9723.8 cd	11.2 d	26.2 c	798.5 b	43.5 d
	N2P4K3	555.8 d	37.2 b	47.9 bc	10,087.4 c	11.2 d	25.8 c	792.3 b	43.1 d
	N3P1K3	611.3 c	36.8 bc	48.3 b	11,067.7 b	12.4 c	29.7 bc	801.4 ab	47.9 bc
	N3P2K4	633.8 bc	38.6 a	50.9 a	12,533.6 a	13.1 b	30.9 b	814.8 a	48.8 bc
	N3P3K1	620.8 c	37.5 ab	42.2 f	9642.2 cd	12.6 c	29.8 bc	780.2 c	47.2 c
	N3P4K2	638.8 b	36.5 c	46.2 cd	10,811.1 b	12.8 bc	30.2 bc	796.5 b	48.2 bc
	N4P1K4	676.3 a	35.2 cd	47.4 bc	11,013.9 b	13.9 a	32.1 a	808.1 ab	49.5 ab
	N4P2K3	656.7 ab	35.8 cd	46.6 c	10,897.7 b	14.2 a	32.9 a	803.7 ab	50.2 a
	N4P3K2	660 a	34.5 d	44.4 e	9902.5 cd	13.8 ab	31.8 ab	791.4 bc	49.1 ab
	N4P4K1	662.5 a	34.8 d	41.2 f	9402.9 d	13.5 ab	31.2 ab	775.1 cd	48.2 bc

Different letters indicate significant differences at p < 0.05.

).

3.1.1. Yield and Yield Components: Optimal vs. Control

The data analysis revealed a substantial productivity gap between the optimized fertilization treatments and the control (N1P1K1).

Grain Yield: The N3P2K4 treatment (N 170 kg/ha, P₂O₅ 30 kg/ha, K₂O 120 kg/ha) was identified as the optimal combination for maximizing productivity. In 2023–2024, the yield of N3P2K4 (12,653.3 kg/ha) was 327.5% higher than that of the N1P1K1 control (2960.0 kg/ha). A similar trend was observed in 2024–2025, with N3P2K4 achieving a yield of 12,533.6 kg/ha, representing a 302.4% increase over the control (3114.9 kg/ha, p < 0.05).

Sink Capacity (Spike and Grains): This yield surge was primarily driven by the expansion of sink capacity. Compared to N1P1K1, the spike number of the optimal N3P2K4 treatment increased by 107.0–109.5%, while the grains per spike increased by 72.3–75.0%. Notably, while N4 level treatments (e.g., N4P1K4) produced the highest spike density, N3P2K4 achieved a superior balance between spike number and grain number per spike.

Grain Filling (1000-Grain Weight): The 1000-grain weight was significantly enhanced by potassium application. The N3P2K4 treatment (K 120 kg/ha) reached 51.7 g (2023–2024) and 50.9 g (2024–2025), which was 19.9–20.6% higher than the control.

3.1.2. Grain Quality Traits: Optimal vs. Control

Nutritional and processing qualities reached their maximum levels under higher nitrogen inputs, typically at the N4 level (220 kg/ha).

Protein and Gluten: The N4P2K3 treatment demonstrated the strongest impact on protein accumulation. Compared to the N1P1K1 control, the grain crude protein in N4P2K3 increased from 8.9–9.0% to 13.6–14.2%, representing a relative gain of 52.8–57.8%. Similarly, wet gluten content increased by 59.1–60.5% over the control.

Test Weight and Hardness: The optimized treatment (N3P2K4) not only maximized yield but also maintained high physical quality, with a test weight consistently exceeding 800 g/L (a significant improvement over the low-NPK treatments). This suggests that the N170K120 interaction facilitates dense starch packing and grain vitreousness.

3.2. Orthogonal Experimental Analysis of Fertilization Effects

The grain yield, yield components, and quality traits of winter wheat under different NPK combinations are presented in Table 4. To further clarify the contribution of each nutrient, range analysis and factor effect evaluations were performed (

Table 5. Main effect table of mean values from orthogonal experimental analysis for yield.

(a) Yield (kg/ha)
Year	Levell	N	P	K
2023–2024	1	3440.97	8517.53	7561.1
	2	9192.31	8772.62	8290.5
	3	11,013.68	8170.74	8866.8
	4	10,304.24	8490.31	9232.79
	Delta	7572.71	601.88	1671.69
	Rank	1	3	2
2024–2025	1	3361.67	8548.33	7710
	2	9285	8825	8370
	3	11,211.67	8510	8928.33
	4	10,665	8640	9515
	Delta	7850	315	1805
	Rank	1	3	2
(b) Spike number (×104/ha)
Year	Levell	N	P	K
2023–2024	1	321.83	543.39	537.94
	2	555.23	544.43	549.86
	3	631.58	536.05	539.44
	4	666.95	551.71	548.35
	Delta	345.12	15.66	11.92
	Rank	1	2	3
2024–2025	1	314	535.94	533.12
	2	555.1	538.64	542.5
	3	626.15	539.01	533.28
	4	663.85	545.51	550.2
	Delta	349.85	9.57	17.08
	Rank	1	3	2
(c) Grains per spike
Year	Levell	N	P	K
2023–2024	1	23.38	31.93	32.15
	2	35.98	33.13	32.13
	3	37.2	32.9	33.28
	4	34.43	33.03	33.43
	Delta	13.82	1.2	1.3
	Rank	1	3	2
2024–2025	1	23.68	32.13	32.55
	2	35.78	33.43	32.23
	3	37.35	33.25	33.53
	4	35.08	33.08	33.58
	Delta	13.67	1.3	1.35
	Rank	1	3	2
(d) 1000-grain weight (g)
Year	Levell	N	P	K
2023–2024	1	43.91	45.8	41.73
	2	45.83	45.66	44.97
	3	46.89	44.92	46.83
	4	44.9	45.15	48.01
	Delta	2.98	0.88	6.28
	Rank	2	3	1
2024–2025	1	44.75	46.63	42.78
	2	46.61	46.34	45.89
	3	47.98	45.83	47.59
	4	45.85	46.4	48.94
	Delta	3.23	0.8	6.16
	Rank	2	3	1

and

Table 6. Main Effect Table of Mean Values from Orthogonal Experimental Analysis for Quality.

(a) Grain crude protein (%)
Year	Levell	N	P	K
2023–2024	1	9.06	11.1	11
	2	10.53	11.36	11.19
	3	12.1	11.26	11.31
	4	13.23	11.19	11.41
	Delta	4.17	0.26	0.41
	Rank	1	3	2
2024–2025	1	9.22	11.56	11.49
	2	11.03	11.84	11.67
	3	12.71	11.72	11.79
	4	13.83	11.68	11.85
	Delta	4.61	0.28	0.36
	Rank	1	3	2
(b) Wet gluten (%)
Year	Levell	N	P	K
2023–2024	1	20.29	25.58	25.35
	2	24.31	26.16	25.79
	3	28.23	25.96	26.05
	4	30.68	25.8	26.31
	Delta	10.39	0.58	0.96
	Rank	1	3	2
2024–2025	1	20.87	26.91	26.66
	2	25.59	27.4	27.02
	3	30.13	27.24	27.39
	4	31.99	27.03	27.52
	Delta	11.12	0.49	0.86
	Rank	1	3	2
(c) Test weight (g/L)
Year	Levell	N	P	K
2023–2024	1	769.28	784.4	767.45
	2	781.15	784.8	780.38
	3	791.65	780.7	788.13
	4	788.18	780.35	794.3
	Delta	22.37	4.45	26.85
	Rank	2	3	1
2024–2025	1	774.2	790.85	773.75
	2	786.85	790.45	786.35
	3	798.23	786.63	793.45
	4	794.58	785.93	800.3
	Delta	24.03	4.92	26.55
	Rank	2	3	1
(d) Grain hardness index
Year	Levell	N	P	K
2023–2024	1	33.75	41.825	41.425
	2	40.775	42.475	42.175
	3	46.75	42.75	42.825
	4	48.2	42.425	43.05
	Delta	14.45	0.925	1.625
	Rank	1	3	2
2024–2025	1	35.625	43.2975	42.685
	2	42.4975	43.9	43.57
	3	48.025	44.075	44.4275
	4	49.2375	44.1125	44.7025
	Delta	13.6125	0.815	2.0175
	Rank	1	3	2

; Figure 2 and Figure 3).

3.2.1. Effects of NPK Combinations on Grain Yield and Its Components

The range analysis for yield indicators (Table 5) revealed that nitrogen (N) was the primary factor regulating grain yield and its core components (spike number and grains per spike).

Grain Yield: The magnitude of factor influence followed the order of N > K > P consistently over both years. Peak yield was achieved under the N3P2K4 combination (N 170 kg/ha, P₂O₅ 30 kg/ha, K₂O 120 kg/ha).

Spike Number and Grains per Spike: N exerted the greatest impact (Rank 1) on both indicators. Maximum values were observed at the N3 level (170 kg/ha). While the ranking of K and P for spike number showed interannual variability, their influence on grains per spike remained stable in the order of N > K > P.

1000-Grain Weight: Unlike yield, 1000-grain weight was predominantly influenced by potassium (K), followed by N and P (rank: K > N > P). The highest 1000-grain weight was recorded at the K4 level (120 kg/ha) and N3 level (170 kg/ha), while P showed a negative effect, with the maximum achieved at the P1 level (0 kg/ha).

3.2.2. Effects of NPK Combinations on Grain Quality Traits

The influence of NPK application on nutritional and processing quality is summarized in Table 6.

Protein and Gluten Properties: Nitrogen (N) was the most critical factor (Rank 1) affecting crude protein (CP), wet gluten content (WGC), and grain hardness. These indicators showed a linear upward trend with N application, reaching their maximum at the N4 level (220 kg/ha). Potassium (K) and phosphorus (P) followed N in influence (rank: N > K > P), with the highest quality typically achieved at K4 (120 kg/ha) and P2 (30 kg/ha).

Test Weight: Similar to 1000-grain weight, test weight was primarily regulated by K (Rank 1), followed by N and P. Optimal test weight was consistently obtained under the K4 (120 kg/ha) and N3 (170 kg/ha) treatments across both years.

Grain Hardness: The dominance of N was evident, with the magnitude of influence following N > K > P. Maximum grain hardness index was observed at N4 (220 kg/ha), K4 (120 kg/ha), and P3 (60 kg/ha).

3.3. Factor Effects and Analysis of Variance (ANOVA)

The ANOVA results (

Table 7. Analysis of variance (ANOVA) for yield indicators.

(a) Grain yield (kg/ha)
	Source	df	SS	MS	F	P
2023–2024	N	3	142,585,181.5	47,528,393.85	185.83	0.0001
	P	3	730,171.13	243,390.38	0.95	0.4733
	K	3	6,385,409.07	2,128,469.69	8.32	0.0147
	Error	6	1,534,601.75	255,766.96
	Total		151,235,363.5
2024–2025	N	3	155,962,272.8	51,987,424.25	205.72	0.0001
	P	3	236,763.33	78,921.11	0.31	0.8163
	K	3	7,144,880.33	2,381,626.78	9.42	0.0109
	Error	6	1,516,249.49	252,708.25
	Total		164,860,165.9
(b) Spike number (×104)
	Source	df	SS	MS	F	P
2023–2024	N	3	289,093.1	96,364.37	2439.71	0.0001
	P	3	492.49	164.16	4.16	0.0652
	K	3	442.71	147.57	3.74	0.0796
	Error	6	236.99	39.5
	Total		290,265.29
2024–2025	N	3	296,257.14	98,752.38	3955.27	0.0001
	P	3	198.07	66.02	2.64	0.1435
	K	3	810.32	270.11	10.82	0.0078
	Error	6	149.8	24.97
	Total		297,415.33
(c) Grains per spike
	Source	df	SS	MS	F	P
2023–2024	N	3	483.6	161.2	465.28	0.0001
	P	3	3.68	1.23	3.54	0.0879
	K	3	5.93	1.98	5.7	0.0344
	Error	6	2.08	0.35
	Total		495.28
2024–2025	N	3	471.52	157.17	767.48	0.0001
	P	3	4.04	1.35	6.58	0.0252
	K	3	5.62	1.87	9.15	0.0117
	Error	6	1.23	0.2
	Total		482.41
(d) 1000-grain weight (g)
	Source	df	SS	MS	F	P
2023–2024	N	3	19.38	6.46	4.68	0.0516
	P	3	2.07	0.69	0.5	0.6952
	K	3	90.12	30.04	21.77	0.0013
	Error	6	8.28	1.38
	Total		119.85
2024–2025	N	3	22.03	7.34	5.41	0.0384
	P	3	1.37	0.46	0.34	0.8002
	K	3	84.84	28.28	20.82	0.0014
	Error	6	8.15	1.36
	Total		116.39

and

Table 8. Analysis of variance (ANOVA) for quality indicators.

(a) Grain crude protein (%)
	Source	df	SS	MS	F	P
2023–2024	N	3	39.73	13.24	517.84	0.0001
	P	3	0.15	0.05	1.95	0.2237
	K	3	0.38	0.13	4.94	0.0463
	Error	6	0.15	0.03
	Total		40.41
2024–2025	N	3	48.49	16.16	496.48	0.0001
	P	3	0.16	0.05	1.68	0.2701
	K	3	0.3	0.1	3.08	0.1119
	Error	6	0.2	0.03
	Total		49.15
(b) Wet gluten (%)
	Source	df	SS	MS	F	P
2023–2024	N	3	248.9	82.97	589.98	0.0001
	P	3	0.74	0.25	1.76	0.2538
	K	3	2.02	0.67	4.79	0.0493
	Error	6	0.84	0.14
	Total		252.51
2024–2025	N	3	296.62	98.87	767.26	0.0001
	P	3	0.59	0.2	1.52	0.3028
	K	3	1.81	0.6	4.67	0.0519
	Error	6	0.77	0.13
	Total		299.79
(c) Test weight (g/L)
	Source	df	SS	MS	F	P
2023–2024	N	3	1170.54	390.18	19	0.0018
	P	3	66.99	22.33	1.09	0.4234
	K	3	1607.53	535.84	26.1	0.0008
	Error	6	123.2	20.53
	Total		2968.26
2024–2025	N	3	1354.75	451.58	20.49	0.0015
	P	3	77.86	25.95	1.18	0.3938
	K	3	1543.69	514.56	23.35	0.001
	Error	6	132.24	22.04
	Total		3108.54
(d) Grain hardness index
	Source	df	SS	MS	F	P
2023–2024	N	3	520.0869	173.3623	959.7912	0.0001
	P	3	1.8219	0.6073	3.3622	0.0962
	K	3	6.4019	2.134	11.8143	0.0063
	Error	6	1.0838	0.1806
	Total		529.3944
2024–2025	N	3	463.7424	154.5808	1789.4749	0.0001
	P	3	1.7089	0.5696	6.5943	0.025
	K	3	9.9833	3.3278	38.5233	0.0003
	Error	6	0.5183	0.0864
	Total		475.953

) provide quantitative evidence for the regulatory roles of nitrogen (N), phosphorus (P), and potassium (K) on winter wheat productivity and grain quality.

3.3.1. Dominant Role of Nitrogen in Yield Formation and Nutritional Quality

Nitrogen was identified as the primary driver (Rank 1) for yield formation and protein accumulation.

Productivity and Sink Size: In both experimental seasons, N fertilization exerted an overwhelming influence on grain yield (F = 185.83–205.72, p < 0.001) and spike number (F = 2439.71–3955.27, p < 0.0001). This confirms that in the Calcaric Cambisols of the Huang-Huai-Hai Plain, N supply is the fundamental factor for establishing a high-capacity “sink” (Table 7a,b).

Protein and Gluten Synthesis: N was also the most critical factor for grain crude protein (F = 496.48–517.84) and wet gluten content (F = 589.98–767.26), with all values reaching the significance level. These results indicate that the drip fertigation system, by optimizing N delivery, effectively enhances nitrogen metabolism and the accumulation of storage proteins, thus overcoming the “quality lag” in high-yielding wheat.

3.3.2. Potassium-Mediated Regulation of Grain Filling and Physical Traits

While N builds the framework, potassium (K) is essential for grain filling and density.

1000-Grain Weight and Test Weight: ANOVA revealed that K was the Rank 1 factor determining 1000-grain weight (F = 20.82–21.77, p < 0.0014) and test weight (Table 8c, F = 23.35–26.11, p < 0.001).

Physiological Mechanism: The significance of K highlights its critical role in optimizing the source–sink relationship. Potassium facilitates the “flow”—the translocation of assimilates from leaves (source) to grains (sink). The synergy between N (establishing sink size) and K (ensuring sink filling) explains the superior grain weight and test weight (exceeding) observed in the optimized N3P2K4 treatment.

3.3.3. Phosphorus Response and Nutritional Balance

Phosphorus Sensitivity: Phosphorus (P) showed a generally non-significant impact on grain yield (F = 0.31–0.95, p > 0.47) and quality (p > 0.22). This low sensitivity is consistent with the high background available P (32.8 mg/kg) in the experimental soil. The drip irrigation management likely maintained sufficient P mobility, minimizing the marginal benefits of additional P applications.

Interannual Consistency: Despite minor fluctuations in F-values between years, the rank of each factor remained remarkably stable. This high interannual consistency provides strong evidence for the robustness of the proposed optimized fertilization model.

3.4. Interaction Analysis via 3D Response Surfaces

The 3D response surface plots (Figure 4, Figure 5 and Figure 6) were generated by fixing the factor with the lowest priority at its optimal level, allowing for a visual exploration of the complex interactions among nitrogen (N), phosphorus (P), and potassium (K) on wheat productivity and quality.

3.4.1. Synergy of N–K and N–P Interactions on Grain Yield

As illustrated in Figure 4, the response surfaces for grain yield exhibit a high degree of interannual consistency across the two experimental seasons. Nitrogen application served as the foundational driver, with yield increasing significantly as N levels rose. On this basis, potassium (K) fertilization further amplified productivity, demonstrating a significant synergistic effect. The peak yield was consistently achieved within the interval of medium nitrogen (N170) and high potassium (K120). In contrast, the response to phosphorus (P) exhibited a typical “diminishing returns” pattern; yield initially increased with P levels but plateaued or slightly declined beyond the P30 level. The optimal fertilization model for yield maximization was identified as the combination of N170 and P30.

3.4.2. Regulation of Physical Quality via N–K Interaction

Regarding physical grain quality (Figure 5), a significant positive interaction was observed between nitrogen and potassium on test weight (TW) and 1000-grain weight (TKW). The 3D surfaces reveal that while increasing a single factor (either N or K) had a limited impact, their combined application triggered a steep, non-linear upward trend. This indicates that the N–K synergy is not a simple additive effect but a physiological optimization of the source–sink relationship; potassium facilitates the efficient translocation of photoassimilates to grains under sufficient nitrogen supply. Consistent with yield trends, the optimal regime for maximizing TW and TKW remained N170 combined with K120 over both years.

3.4.3. Synergy of N–P Interaction on Nutritional Quality

For nutritional quality (Figure 6), the interaction between nitrogen and phosphorus significantly influenced crude protein (CP) and wet gluten content (WGC). The response surfaces highlight a joint regulation mechanism where N and P synergistically enhanced grain protein synthesis. While the independent effect of increasing phosphorus was relatively modest, its combination with high nitrogen (N220) facilitated a marked increase in both CP and WGC. To achieve superior nutritional quality, a fertilization regime centered on N220 and P30 proved most effective. These results demonstrate that maintaining a high nutritional balance through N–P interaction is essential for breaking the yield-quality trade-off in the Calcaric Cambisols of this region.

4. Discussion

4.1. Core Regulatory Role of Nitrogen and Optimization of Source–Sink Relationship

Regarding nitrogen’s influence, the yield ranges across the two-year study reached 7572.71 kg/ha and 7850.0 kg/ha, respectively, confirming N as the most dominant factor. The grain yield followed a typical parabolic response, peaking at N3 (170 kg/ha), where spike numbers and grains per spike were simultaneously maximized. This result validates the regulatory mechanism where N strengthens the “source” (photosynthetic capacity) to provide a material foundation for spike development, while optimizing the “sink” structure to promote tillering and floret setting. These findings align with the established consensus in the North China Plain by Qu et al. [1], and reinforce the “promoting early growth and controlling late growth” principle of N demand in winter wheat [10]. Furthermore, the success of the “50% base + 50% drip topdressing” method mirrors the results of Wang et al. [11] and Yan et al. [12,13], demonstrating that precision topdressing can significantly improve nitrogen use efficiency (NUE) while mitigating environmental loss. This conclusion is also supported by Yang et al. [14] regarding the effects of drip fertigation on crop yield and quality.

At the same time, the grain crude protein and wet gluten contents increased continuously with the increase in nitrogen application rate, reaching the highest values at N4 (220 kg/ha). The crude protein contents in the two years were 13.2% and 13.8%, and the wet gluten contents were 30.68% and 31.99%, respectively. This continuous increase parallels the physiological mechanism where N is directly allocated to grain protein synthesis [11], and mutually corroborates with the research conclusions of Duncan et al. [15] and Janczak-Pieniazek et al. [4]. In this study, the N3 (170 kg/ha) level achieved a synergistic balance [16], which not only ensured high yield but also maintained the crude protein and wet gluten contents at high-quality levels (crude protein ≥ 12%, wet gluten ≥ 28%), avoiding the environmental risks of excessive nitrogen application and conforming to the requirements of green and sustainable agricultural development. This balance supports the premise that reasonable N application can harmonize productivity and quality [17,18,19], an observation also corroborated by Yang et al. [20,21] regarding optimized NPK management.

Potassium (K) exerted the primary influence on physical grain quality, specifically governing 1000-grain weight and test weight. The ranges of 1000-grain weight in the two years were 6.28 g and 6.16 g, and the ranges of test weight were 26.85 g/L and 26.55 g/L, respectively. The highest 1000-grain weight and test weight were achieved at K4 (120 kg/ha), confirming that potassium is a key factor ensuring yield and commercial quality by promoting the transport of photosynthates to grains and improving grain plumpness [22]. The peak values achieved at K4 (120 kg/ha) confirm that K acts as a vital “pump” for photosynthate translocation, effectively optimizing the source–sink dynamics to ensure grain plumpness. This role echoes the results reported by Yin et al. [14] in other cereal crops. The N3K4 combination in this study reinforces the significance of the nitrogen–potassium interaction, echoing the source–sink optimization trends documented by Xu et al. [23] and Gaj et al. [24].

Potassium (K) also played a secondary but critical role in enhancing crude protein and wet gluten contents. The optimal quality indicators observed at K4 (120 kg/ha) align with the established physiological role of K in promoting protein synthesis and grain filling [17,24], while echoing the conclusions of Bouacha et al. [25] regarding N and K effects on durum wheat quality. Furthermore, the influence of K on yield ranked second in this study, with two-year yield ranges of 1671.6 kg/ha and 1805.0 kg/ha, confirming that K is a vital supplement to the nitrogen-induced yield-increasing effect [17]. Moreover, sufficient K availability effectively prevented the “quality lag” often triggered by high N application, a benefit particularly significant in heavy-textured Calcaric Cambisols where nutrient mobility is limited. This observation aligns with the long-term findings of Firmano et al. [26] and parallels the evidence from Shahid et al. [27] on K’s ability to alleviate heat stress and enhance wheat quality.

Regarding phosphorus (P), yield ranges across the two years were 601.88 kg/ha and 315.0 kg/ha, respectively, with its influence ranking third among the factors. The optimal yield was achieved at P2 (30 kg/ha), as excessive P provided no significant gain. Physiologically, P promotes root development in Calcaric Cambisols, yet excessive application risks inducing trace element deficiencies such as zinc [6], corroborating the findings of Dhaliwal et al. [6] In this study, the effectiveness of 30 kg/ha P aligns with the life cycle assessment by Chen et al. [28] in the North China Plain, illustrating that appropriate P application ensures productivity while mitigating environmental emissions. This strategy for sustainable agricultural development mirrors the conclusions of Ceclan et al. [29,30] regarding nitrogen and phosphorus fertilization effects.

The effect of P application on 1000-grain weight was weak, aligning with its primary role as a foundational guarantee for productivity rather than a core driver of grain filling. This observation parallels the existing research on nitrogen–phosphorus combined application [31,32], and echoes the results of Feng et al. [31] concerning the synergistic regulation of wheat yield and protein content.

4.2. Synergistic Regulatory Effects of Nitrogen–Potassium and Nitrogen–Phosphorus Interactions

A primary advantage of the orthogonal design is its ability to reveal the comprehensive impacts of nutrient interactions, which in this study proved more decisive for yield and quality than single-factor effects. The synergistic regulation of N–K and N–P interactions emerged as the critical mechanism for optimizing winter wheat performance in Calcaric Cambisols.

Regarding N–K synergy, N dominates source-end photosynthetic production while K facilitates sink-end transport, forming a critical regulatory nexus [33]. In this study, when the nitrogen application rate was at N3 (170 kg/ha), increasing the potassium application rate from K1 (0 kg/ha) to K4 (120 kg/ha) increased the yield by 18.2% and 19.1%, and the 1000-grain weight by 13.7% and 13.2% in the two years, respectively, confirming that sufficient K availability is the prerequisite for N to reach its full potential in Calcaric Cambisols [34]. On the contrary, when the potassium application rate was at K4 (120 kg/ha), increasing the nitrogen application rate from N1 (0 kg/ha) to N3 (170 kg/ha) increased the yield by 124.3% and 127.6%, and the number of spikes per hectare by 98.7% and 101.2% in the two years, respectively, indicating that adequate N provides the necessary sink capacity for K to regulate grain filling. This interaction aligns with the findings of Yin et al. [14], which highlight K’s role in promoting the transport of photoassimilates to grains. The optimal combination of N3 and K4 in this study reflects the maximization of N–K interaction effects: N strengthens source-end supply while K improves the “flow” and sink filling, paralleling the established view that N–K interaction is a core mechanism for productivity [23,29], and mirroring the results of Chen et al. [30] regarding quality enhancement.

At the same time, the regulatory effect of nitrogen–potassium interaction on grain quality was significant: at N3 level, the grain test weight of K4 treatment was 2.8% and 2.7% higher than that of K1, and the hardness was 11.3% and 10.8% higher; at K4 level, the crude protein content of N3 treatment was 34.6% and 35.2% higher than that of N1, and the wet gluten was 38.9% and 39.5% higher, underscoring that N–K combined application improves both nutritional and commercial quality. This synergy echoes the consensus that K can enhance N’s regulatory effect on grain protein [17], an observation further corroborated by the evidence from Bouacha et al. [25] regarding N and K effects on durum wheat and in agreement with Rawal et al. [2] concerning NPK-mediated nutrient absorption.

Phosphorus (P) provides a foundational basis for the optimization of nitrogen (N) effects by promoting root architecture development and enhancing nutrient uptake capacity, a synergy that serves as a vital support for stable winter wheat yields. In this study, at the N3 level (170 kg/ha), increasing P from P1 (0 kg/ha) to P2 (30 kg/ha) improved yield by 3.1% and 2.8%, and spike number by 2.5% and 2.3%. This indicates that appropriate P application synergistically improves the tillering and spike formation effect of N. Conversely, at P2 (30 kg/ha), the yield increase achieved by moving from N1 to N3 was 1.2 to 1.5 percentage points higher than at P1, reinforcing the premise that P amplifies N-induced yield effects by strengthening root absorption [35]. This result aligns with established findings suggesting that N–P combined application effectively enhances agronomic performance and wheat yield [31,32]. The P2N3 combination represents an optimal balance point for N–P interaction, ensuring robust root vigor while avoiding the trace element deficiencies associated with excessive P application in Calcaric Cambisols [6]. This regulatory trend reinforces the conclusions of Ceclan et al. [29,30] and parallels the evidence from Assefa et al. [36] regarding N–P fertilization in stable production systems [36].

N–P Synergy: P provides a foundational support for N uptake by promoting root development, especially vital in the compact and poorly aerated Calcaric Cambisols. Our results illustrate that appropriate P (30 kg/ha) can amplify the N-induced yield-increasing effect by strengthening root nutrient absorption, a trend previously identified by Feng et al. [31]. Furthermore, N–P interactions appeared to regulate floret retention and seed setting rate, thereby optimizing the final grain number per spike.

The regulatory effect of the nitrogen–phosphorus (N–P) interaction on the number of grains per spike proved significant, representing a critical synergy for optimizing yield components in Calcaric Cambisols. At the N3 level, the grain number per spike in the P2 treatment was 4.2% and 3.9% higher than that of the P1 treatment across the two experimental years. This confirms that P synergistically optimizes spike structure with N by reducing floret abortion and enhancing the seed setting rate, thereby effectively stabilizing the sink capacity [32]. Furthermore, N–P interactions modulate starch synthesis and the kinetics of grain filling, a regulatory mechanism that echoes the findings of Jiang et al. [37] regarding nutrient-mediated grain weight and quality improvement. In the compact and nutrient-limited Calcaric Cambisols, this coordination is essential for ensuring that the enhanced root vigor provided by P maximizes the nitrogen use efficiency (NUE) required for reproductive organ development.

4.3. Optimal Fertilization Scheme for the Synergy of High-Yield and High-Quality, and Sustainability Analysis

Synthesizing the results from this two-year field study, the optimal NPK fertigation regime for synergistic yield and quality enhancement was identified as N 170 kg/ha (50% base + 50% drip topdressing), P₂O₅ 30 kg/ha (base), and K₂O 120 kg/ha (base). The strategic scheduling of drip fertigation during the jointing (Feekes 6.0) and grain-filling stages ensured that nutrient availability was synchronized with the peak physiological demands (ETc) of the crop. Under this scheme, N3 serves as the optimal yield threshold, while K4 secures both 1000-grain weight and test weight (TW), ensuring both physical and nutritional grain quality. This synergistic interaction effect, maximizing the N–K and N–P nexus, aligns with the established consensus that nutrient interactions are the fundamental drivers of high crop productivity [23,30]. While ensuring productivity, this scheme effectively reduces the excessive application of N and P, thereby enhancing nutrient use efficiency and mitigating environmental footprints such as nitrogen leaching and greenhouse gas emissions [38]. This approach echoes the global research objective of “achieving efficiency through reduction” [28,39,40] and parallels the optimized fertilization strategies proposed by Jiang et al. [37]. Furthermore, the incorporation of micronutrients (e.g., Zn, Fe) and testing across diverse wheat cultivars in subsequent studies will be essential to refine this model and ensure its broader applicability. Such refinements would further support the findings of Jayara et al. [41] regarding the integration of mineral nutrients to bolster agricultural sustainability.

The stability of the results of this study was verified by the two-year experiment. The priority of factor influence and the optimal level of each indicator were completely consistent in the two years, confirming that this fertilization scheme has good yield stability and can cope with interannual climate fluctuations. This is consistent with the research conclusion of Addy et al. [39] that interannual weather changes will affect the response of crops to nitrogen application, but reasonable fertilization schemes can ensure stable yield. It also mutually corroborates the research conclusions of Iwanska et al. [42] and Sarker et al. [32] on the interannual variability of wheat yield, and is consistent with the research conclusion of Grzezbisz et al. [43] on the prediction of wheat yield and gluten content.

5. Conclusions

Under the experimental conditions of this study, conducted in Calcaric Cambisols, nitrogen (N) was the primary factor regulating grain yield, spike components, and core quality indicators, followed by potassium (K), while phosphorus (P) exerted the weakest influence. Notably, potassium (K) was the dominant regulatory factor for 1000-grain weight and grain test weight (TW).

The fertilization combinations of N170K120P30 (N3K4P2) and N220K120P30 (N4K4P2) were identified as the optimal models for maximizing grain yield and nutritional quality, respectively. By prioritizing nutritional balance over single-nutrient maximization, these optimized regimes ensure that winter wheat achieves superior quality standards without excessive fertilizer input. The precision of drip fertigation, particularly when synchronized with critical growth stages such as jointing and grain filling, is essential for optimizing the source–sink relationship and enhancing fertilizer use efficiency.

Significant positive interactions were observed between N–K and N–P, which synergistically regulated both yield and quality. This study confirms that balanced fertilization via drip irrigation is the key to achieving the synergy of high yield and high quality in winter wheat. To further enhance the applicability of these findings, future research should explore the integration of micronutrients and validate the stability of this fertilization model across diverse wheat cultivars.