Fungicide Protection as an Agrotechnical Treatment Reducing Nitrogen Gap in Winter Wheat—A Case Study

Abstract

Protection of high-yielding winter wheat (WW) with fungicides increases the productivity of nitrogen (N) present in the soil–crop system during the growing season. As a consequence of the action of fungicides, the nitrogen gap (NG) reduces. This hypothesis was verified on the basis of data from a field experiment conducted with WW during three growing seasons (2013/2014; 2014/2015, 2015/2016) in Poland. The field experiment included two crop protection systems (CP): (i) CP-0—without fungicides and CP-F—with fungicides and (ii) six N doses increased gradually by 40 kg N ha⁻¹ from 0 to 240 kg N ha⁻¹. The grain yield (GY) of WW treated with fungicides was significantly higher than that of the unprotected. The difference in yields between both CP systems was 17.3% on a plot fertilized with 200 kg N ha⁻¹ (9.13 vs. 11.2 t ha⁻¹). The fungicide yield gap increased progressively with N_f doses from 0.76 t ha⁻¹ in the N_f control plot to 2.17 t ha⁻¹ in the fertilized with 200 kg ha⁻¹. The use of fungicides increased the amount of N in grain (N_gr) from 15 kg N ha⁻¹ in the control N plot to 51 kg N ha⁻¹ in the plot with 200 kg N ha⁻¹. The main source of additional N in grain (N_gr) was inorganic N released from the soil (N_g89) during the WW growing season. The maximum N_g89 values were 64.4 and 83.0 kg N ha⁻¹. These values corresponded to N_f doses of 94.4 and 80.8 kg N ha⁻¹. The N_g89 of 70.1 kg N ha⁻¹ conditioned 100-percentage N_f recovery. As a consequence, the prediction reliability of GY and N_gr was highest when N_g89 was used as a predictor. The net increase in the absolute NG size in response to increasing N input was significantly slower and therefore smaller in fungicide-protected than in unprotected WW. It can be concluded that the use of fungicides due to the increase in inorganic N productivity in the soil–crop system reduces the potential threat of N dispersion into the environment. In the light of the results obtained, it should be concluded that the fungicidal protection of crop plants should be treated as a factor significantly reducing the nitrogen gap and, thus, the yield gap.

1. Introduction

Wheat is the most important crop, regardless of the world region, due to its use in bread production. The usefulness of wheat flour results from the specific properties of the protein, which contains glutenin and gliadin in equal proportions [1]. The quality of wheat flour depends not only on the genetic background, which determines the content of both proteins, but also from the content of total protein in the grain. High-quality wheat grain should contain at least and even more than 12.5% crude protein [2,3]. The protein content depends on the plant’s supply of nitrogen (N), more precisely on the applied dose of fertilizer N (N_f) [4]. The key primary yield-forming function of N_f is to build up the basic yield component, which is the number of grains per unit area (canopy grain density, CGD) [5]. An increase in CGD generally leads to a decrease in grain N content. This condition results from a higher CGD value or due to a larger grain weight (thousand grain weight, TGW). This phenomenon is known as “the N dilution effect”. For this reason, in agricultural practice, N_f rates are generally higher than for the real grain yield of winter wheat [3,4].

The grain yield of cereals depends on the mineral N (N_min) resources present in the soil–crop system (i) at the beginning of the growing season, (ii) applied as fertilizer (N_f) and (iii) released from the soil resources during the growing season [6,7]. The first N pool is determined at the beginning of the growing season or, as in the case of winter crops, at the beginning of the spring part of the season. The N_min method for identifying inorganic N resources has been known for over 50 years. However, its use is not common, even in leading wheat producers [8]. The advantage of the N_min method is also the ability to determine the content of other nutrients and assess their resources [9,10]. The second N pool, that is, N_f, is the only one that is directly designated by the farmer. In WW production, the key problem is not only determining the optimal N_f dose, but also its division during the growing season. This distribution must be coordinated with the plant’s N demand. In fact, the key goal of in-season N_f partitioning is to control the rate of yield component development. In WW, this process extends from the beginning of shooting to the beginning of the grain growth (BBCH 70/71) [11,12]. During this period, the CGD reaches stability and largely determines the grain yield but not the content of crude protein. The second key characteristic of wheat depends on the supply of N to plants in the period from the stage of the grain growth to its physiological maturity [13]. The third pool of N_min is the amount of inorganic N released from soil and its resources during the growing season of a given crop. This is a classic “black box” known as the in-season N gain (N_g) [7,14]. So far, attempts to determine the amount of N_min released during the growing season of crop plants have not yielded a solution that could be widely used in agricultural practice [6]. Moreover, it is unclear to what extent the in-season-released inorganic N affects the components of the yield structure. There is also a question whether it is possible to indirectly calculate this N pool in the soil. The solution in this regard is the balance method proposed by Grzebisz et al. [7].

The increase in crop plants yields over the last 70 years is largely due to advances in plant breeding. However, without a sufficiently high dose of N_f and full crop protection, the yield potential of new varieties will not be efficiently exploited [5,15]. It is well documented that plants intensively fertilized with N_f are very sensitive to pathogen attack. This applies especially to those pathogens that are equipped with a biotrophic mechanism of obtaining nutrients from the living host plant [16]. A specific characteristic of this group of fungal pathogens is the infection of physiologically active plant organs, mainly leaves and glumes of ripening cereals. These fungi redirect the flow of nutrients from the plant to the growing pathogen but at the same time keep the host plant alive. Leaf diseases significantly reduce the assimilation area of wheat, which consequently weakens the potential of the canopy to absorb solar energy and use inorganic N present in the soil. Consequently, infection of wheat by diseases in the pre-flowering period of its growth reduces the production potential of the ear or, more precisely, the number of fertile flowers [17,18]. It is generally accepted that fungicide protection is effective only at high N_f doses. Only under fertilization conditions does the applied fungicide reduce the pathogen’s pressure on photosynthetically active plant parts, for example, on wheat leaves and glumes [19].

In the light of the complexity of the crop plants’ response to the interaction of N_min resources in the soil and fungicide protection discussed above, it is justified to ask the question about the impact of both production factors on the N economy in the soil–winter-wheat system. Therefore, it should be hypothesized that insufficient development of the key yield component, i.e., CGD, leads to ineffective use of N_min in the soil during the growing season by WW, This condition, in turn, results in the formation of a N gap (NG). In agronomic practice, this phenomenon is recognized in its final form and known as the yield gap (YG) [20]. Therefore, the question arises about how to treat fungicidal protection of WW. Is this just an agronomic measure controlling the degree of yield reduction, or is it a growth factor that determines the efficiency of inorganic N resources present in the soil during the WW growing season? The impact of fungicide protection on the yield-forming physiology of crop plants, as a process dependent on the supply of inorganic N, has not been recognized yet. This is a typical “black box”. This study aims to evaluate the impact of fungicide protection on the nitrogen economy of winter wheat. Specifically, it examines whether fungicide application can reduce the N gap by improving N uptake and grain yield. We hypothesize that WW fungicide treatment will enhance N use efficiency and increase both grain yield and grain N content.

2. Materials and Methods

2.1. Experimental Setup

This study on the impact of fungicide protection on the size of the N gap (NG) in WW was carried out in 2013/14 (acronym: first growing season, 1st), 2014/2015 (second growing season, 2nd) and 2015/2016 (third growing season, 3rd) in Smolice (52°42′ N; 17°10′ E), Poland. The field experiment was carried out on soil formed from loamy sand, classified as Albic Luvisol. The content of organic carbon (C_org) and pH were variable but at the appropriate level for WW. The content of available P, K, Fe and Zn was within the required range, at least in the medium class, while the content of Mg, Ca, Cu and Mn was insufficient for WW (low or very low classes). The amount of mineral N (N_min), measured just before the beginning of WW regrowth in spring, was variable but generally high (

Table 1. Soil agrochemical characteristics in consecutive growing seasons ^1–5.

Soil, cm	pH 1	Corg 2	P 3	K	Mg	Ca	Cu	Mn	Zn	Fe	Nmin 4
		%	mg kg−1								kg ha−1
2013/2014
0–30	6.9	1.3	234 VH 5	231 M	105 VL	988 L	0.4 L	27.2 L	3.6 M	536 H	86.4
30–60	6.7	1.1	234 VH	237 M	103 VL	876 L	0.4 L	25.7 L	3.5 M	541 H
2014/2015
0–30	7.1	2.2	185 H	185 M	165 M	2045 M	3.5 M	85.5 M	6.3 H	268 M	129
30–60	7.2	2.1	161 H	157 L	155 L	2063 M	3.5 M	93.8 M	5.6 H	26 9 M
2015/2016
0–30	6.6	1.6	202 VH	281 M	165 M	1480 L	2.8 M	61.9 M	6.1 H	347 M	110
30–60	6.6	1.4	139 H	222 M	163 L	1504 L	2.5 M	62.0 M	3.7 M	231 M

¹ 1.0 M KCl soil/solution ratio 1:2.5 m/v; ² loss-on ignition; ³ including all macro- and micronutrients Mehlich 3 [21]; ⁴ 0.01 dm⁻³ CaCl₂, soil/solution ratio 1:5 m/v; ⁵ availability classes: VL—very low; L—low; M—medium; H—high; VH—very high [22,23].

).

The climate in Western Poland is classified as transitional between Atlantic and Continental. The latter one dominates in summer (Figure 1). The early growing phase of WW in March and April 2014 was very good, but May was wet. The second part of the season was less favorable, as June was very dry, and July was dry. The 2015 growing season was favorable for WW growth despite drought throughout the entire growing season. A deficiency in rainfall covered the period from the stem elongation to early flowering. The beginning of the 2016 growing season was too wet, while May was semi-dry, and June was dry.

2.2. Experimental Design

The field experiment, arranged in a two-factor design, replicated four times, included the following:

1.

Two systems of WW protection with fungicides (CP):

(1)

without the use of fungicides—CP-0;

(2)

using the fungicides—CP-F.

2.

N control plus six doses of applied N fertilizer: 0, 40, 80, 120, 160, 200, 240 kg ha⁻¹.

N was applied in three stages of wheat growth: (i) at the beginning of plant regrowth (40 and 80 kg N ha⁻¹), (ii) the end of tillering/beginning of shoot elongation (BBCH 29/30); supplementation of respective N combinations to 120 and 160 kg N ha⁻¹, (iii) in the phase of flag leaf visible (BBCH 39; supplementation of respective N combinations to 200 and 240 kg N ha⁻¹. Phosphorus was applied at the dose of 17.2 kg P ha⁻¹ in the form of triple superphosphate (46% P₂O₅). Potassium was applied at the dose of 100 kg K ha⁻¹ as Korn-Kali (K-MgO-Na₂O-SO₃ → 40-6-3-12.5). Both fertilizers were applied two weeks before WW sowing. The foliar application of fungicides was carried out in accordance with best agronomic practice.

The WW cv. Wydma was sown annually in the fourth week of September at a dose of 300 grains m⁻². The forecrop was winter oilseed rape. The total area of a single plot was 22.5 m² (1.5 × 15 m). The next year’s crop was harvested at the end of July from an area of 19.5 m⁻².

2.3. Soil and Plant Sampling

Soil samples were taken from two layers: 0.0–0.3 m and 0.3–0.6 m. To determine the basic soil properties, they were collected after the winter oilseed rape harvest. Soil samples for mineral N (N_min) determination were collected: (i) one week before the beginning winter wheat regrowth in spring, (ii) at the physiological maturity of WW (BBCH 89). The mineral forms of inorganic N (NH₄⁺-N and NO₃⁻-N) were determined in “fresh” soil samples within 24 h after sampling. Twenty grams of soil were shaken for 1 h with 100 cm³ of 0.01 M CaCl₂ solution (soil/solution ratio 5:1; m/v). Concentrations of NH₄⁺-N and NO₃⁻-N were determined by colorimetric method using flow injection analyses (FIAstar5000, FOSS Tecator AB, Höganäs, Sweden). Total N_min content was expressed in kg ha⁻¹.

Plant material for determining dry matter and then N content was collected at WW harvest. The N content was determined in grain and straw using the standard macro-Kjeldahl procedure [24]. N content was expressed on a dry matter basis and used to calculate the amount of N in total WW biomass.

2.4. Calculated Parameters

Two sets of indicators were used to describe the N economy of WW during the growing season.

2.4.1. Nitrogen Gap Indicators

To assess the actual productivity of inorganic N present/introduced into the soil–winter-wheat system, the N gap (NG) concept was used [7,14,20]. The main components of this methodological approach for yield gap calculation are the amount of inorganic N (mineral N, N_min) and the amount of fertilizer N (N_f) applied to winter wheat during the growing season.

The third component of the total N_min input to the studied soil/crop system was the amount of N_min released during the growing season, defined as N gain (N_g). The procedure for NG calculation consists of the following steps:

$$\mathrm{Partial} \mathrm{Factor} \mathrm{Productivity} \mathrm{of} {\mathrm{N}}_{\mathrm{in}} \mathrm{or} {\mathrm{N}}_{\mathrm{int}}: PF{P}_{Nf}={\displaystyle \frac{Y}{N_{in}; N_{int}}} \left({\mathrm{kg} \mathrm{kg}}^{-1} for N_{in} or N_{int}\right)$$

(1)

$$\mathrm{Yield} \mathrm{Gap} \mathrm{for} {\mathrm{N}}_{\mathrm{in}} \mathrm{and} {\mathrm{N}}_{\mathrm{int}}: YG=Y_a-Y_{attmax} \left(t {\mathrm{ha}}^{-1}\right)$$

(2)

$$\mathrm{Nitrogen} \mathrm{Gap} \mathrm{or} {\mathrm{N}}_{\mathrm{in}} \mathrm{and} {\mathrm{N}}_{\mathrm{int}}: NG={\displaystyle \frac{YG}{cPF{P}_{Nin;Nint}}}\left({\mathrm{kg} \mathrm{N} \mathrm{ha}}^{-1}\right)$$

(3)

where

N_f—is the amount of applied fertilizer N, kg N ha⁻¹;

N_min—is the amount of mineral N at the beginning of winter wheat regrowth in spring, kg N ha⁻¹;

N_in—is the sum of N_min + N_f, kg N ha⁻¹;

N_gain—is the amount of mineral N released from the soil during the growing season, kg N ha⁻¹;

N_int—is the sum of N_min + N_f + N_g, kg N ha⁻¹;

PFP-N_in—is the partial factor productivity of N_in (kg grain per kg N_in);

PFP-N_int—is the partial factor productivity of N_int (kg grain per kg N_int);

YG—is the yield gap calculated separately for N_in and N_int (t ha⁻¹);

NG—is the N gap calculated separately for N_in and N_int (kg ha⁻¹).

2.4.2. Nitrogen Gain Indicators

A set of N balance indices was calculated based on the amount of N in WW and the total content of N_min at the beginning of the crop regrowth (spring) and at crop maturity (BBCH 89) [7]:

$$\mathrm{N} \mathrm{input}: N_{in}=N_{min}+N_f, {\mathrm{kg} \mathrm{N} \mathrm{ha}}^{-1}$$

(4)

$$\mathrm{N} \mathrm{balance}: N_b=N_{in}-NA, {\mathrm{kg} \mathrm{N} \mathrm{ha}}^{-1}$$

(5)

$$\mathrm{N} \mathrm{gain}: N_g=N_{minr}-N_b, {\mathrm{kg} \mathrm{N} \mathrm{ha}}^{-1}$$

(6)

$$\mathrm{N} \mathrm{input} \mathrm{total}: N_{int}=N_{in}+N_g, {\mathrm{kg} \mathrm{N} \mathrm{ha}}^{-1}$$

(7)

where

N_min—the content of N_min at the beginning winter wheat regrowth (spring), kg N ha⁻¹;

NA—the amount of N accumulated in winter wheat biomass at harvest, kg N ha⁻¹;

N_minr—the content of N_min at winter wheat harvest, kg N ha⁻¹;

2.5. Statistical Analysis

The effects of the individual experimental factor (year, N rates and crop protection level) on N resources and indicators of N productivity were assessed by means of a three-way ANOVA, including years as a factor. Means were separated by honest significant difference (HSD) using Tukey’s method when the F-test indicated significant factorial effects at the level of p < 0.05. Trends in the responses of studied characteristics were determined using a linear and quadratic regression model. The relationships between the traits were analyzed using Pearson correlation and linear regression. STATISTICA 12 software was used for all statistical analyses (StatSoft Inc., Tulsa, OK, USA, 2013). In the second step of the diagnostic procedure, stepwise regression was applied to define an optimal set of characteristics affecting grain yield. In the computational procedure, a consecutive variable was removed from the multiple linear regressions in a step-by-step manner.

3. Results

3.1. Nitrogen Gap—First

The grain yield (GY) of winter wheat was significantly affected by the interaction of the CP × N (crop protection, CP) and N dose (N_f) (

Table 2. Primary sources and indices of nitrogen economy in winter wheat.

Variable	Level of Variable	Nin	GY	PFP-Nin	YG	NG
		kg ha−1	t ha−1	kg kg−1 N	t ha−1	kg ha−1
Year (Y)	2014	206.3 c	8.8 b	48.2 a	−3.7 a	−61.7 a
	2015	249.0 a	11.0 a	46.5 b	−4.8 b	−68.8 b
	2016	230.0 b	6.6 c	31.1 c	−7.4 c	−121.7 c
Fc, p		ne	736 ***	441 ***	611 ***	611 ***
Crop Protection	CP-0	228.4	8.1 b	38.9 b	−5.8 b	−95.0 b
(CP)	CP-F	228.4	9.4 a	45.0 a	−4.4 a	−73.1 a
Fc, p		ne	205 ***	137 ***	205 ***	205 ***
	0	108.5	6.0 e	57.0 a	−0.6 a	−9.1 a
Nitrogen rates	40	148.5	7.8 d	52.7 b	−1.3 b	−20.9 b
(N), kg N ha−1	80	188.5	8.7 c	46.4 c	−2.7 c	−44.8 c
	120	228.5	9.4 b	41.0 d	−4.5 d	−74.5 d
	160	268.4	9.5 b	35.4 e	−6.8 e	−111.9 e
	200	308.4	9.9 a b	31.9 f	−8.9 f	−146.0 f
	240	348.5	10.2 a	29.1 f	−11.0 g	−181.4 g
Fc, p		ne	137 ***	237 ***	1030 ***	1030 ***
Source of variation for the studied interactions
Y × CP		-	***	***	***	***
Y × N		-	***	***	***	***
CP × N		-	***	ns	***	***
Y × CP × N		-	ns	ns	ns	ns

Similar letters means a lack of significant differences using Tukey’s test; *** indicates significant difference at p < 0.001; ne—not estimated. Legend: N_in—mineral nitrogen input into the soil–crop system at the beginning of the growing season; PFP-N_in—partial factor productivity of N input; YG—yield gap; NG—nitrogen gap.

, Figure A1). The impact of this interaction on GY was independent on weather in the subsequent years of the study. Only seemingly, the GY course in response to increasing N_f doses best fit the quadratic regression model:

$$\mathrm{CP}-0: GY=-0.0009N^2+0.035N+5.0 for n=7, R^2=0.95, p \le 0.001$$

(8)

$$\mathrm{CP}-01: GY=0.035N+5.74 for n=3, R^2= 0.99, p \le 0.001$$

(9)

$$\mathrm{CP}-02: GY=0.003N+8.3 for n=4, R^2= 0.72, p \le 0.05$$

(10)

$$\mathrm{CP}-\mathrm{F}: GY=-0.000074N^2+0.036N+6.6 for n=7, R^2= 0.99, p \le 0.001$$

(11)

$$\mathrm{CP}-\mathrm{F}1: GY=0.019N+7.2 for n=3, R^2= 0.92, p \le 0.01$$

(12)

The quadratic regression model is described by two characteristics, i.e., grain yield maximum GY (GY_max) and a specific optimal N_f dose (N_fop). The first attribute of both models was 9.19 and 10.99 t ha⁻¹ for CP-0 and CP-F, respectively. These two maxima were obtained provided that N_op reached 185.6 and 245.2 kg N ha⁻¹, respectively. Therefore, the theoretical difference between both CP systems, i.e., fungicide-induced yield gap (YG) based on N_op, was 1.8 t ha⁻¹. In fact, on CP-F, the effect of N_f doses on GY was progressive, increasing linearly with subsequent N doses. Yield stagnation occurred only for the last two N_f doses. Hence, N_fop exceeded the range of N_f applied. Therefore, the linear model is only statistically correct. In the case of CP-0, the GY trend, although statistically correct, is better assessed when two linear models are used. The obtained lines intersected for N_f equal to 91.5 kg N ha⁻¹. The rate of GY increase below this critical N_f value was by 11.3-times faster than above it. An analysis of Figure A1 and the developed equations showed that the fungicide-induced yield gap (F-YG) occurred in the entire range of N_f doses. It increased progressively with N_f doses from 0.76 t ha⁻¹ in the N_f control plot to 2.17 t ha⁻¹ in the plot fertilized with 200 kg ha⁻¹.

The Partial Factor Productivity of N input (PFP-N_in) was calculated based on the N amount present in soil (N_min) at the beginning of the spring wheat regrowth and fertilizer N applied (N_f). Regardless of the growing season, the values of the index were higher by 15.7% for fungicide-protected wheat (Table 2). The PFP-N_in course was slightly more affected by the Y × N than by Y × CP interaction (Figure A2). The downward trend in the index in accordance with increasing N_f doses was fully confirmed. The highest and significant difference between years, amounting to 30 kg grain kg⁻¹, was recorded on the plot fertilized with 40 kg N ha⁻¹. At the same time, three times smaller differences were recorded in plots fertilized with 200 and 240 kg ha⁻¹. In the 1st and 3rd season, the PFP-N_in was consistent with a linear regression model, while in the 2nd season with a cubic regression model:

$$2013/2014: PFP-N_{in}=-0.18N+70.3 for n=7, R^2=0.97, p \le 0.01$$

(13)

$$2014/2015: PFP-N_{in}=0.000006N^2-0.14N+53.8 for n=7, R^2=0.98, p \le 0.01$$

(14)

$$2015/2016: PFP-N_{in}=-0.087N+41.6 for n=7, R^2=0.90, p \le 0.01$$

(15)

The equation constants for the 1st and 3rd season clearly indicate the difference between these two seasons. In the 2nd season, the downward PFP-N_in trend showed specific stability in relation to the N_f doses. This stability resulted from the maximum, amounting to 56.3 kg grain kg⁻¹ N for N_fop of 37.9 kg N ha⁻¹. The respective minimum of 35.8 kg grain kg⁻¹ N was achieved for N_fop equal to 226.4 kg N ha⁻¹. However, this value was higher than that obtained in other growing seasons.

The average PFP-N_in for the 3rd quartile of the obtained data set was then used to determine the Nitrogen Gap (NG). NG values showed significant dependence on the interaction of years (Y) and experimental factors but were considered separately. The effect of CP × N was highly significant and did not depend on weather variability in consecutive years of the study. This means that the effect of crop protection on NG, despite year-to-year variability, was consistent (Table 2). Overall, NG absolute values increased linearly with increasing N_f doses. The obtained linear regression models are as follows:

$$\mathrm{CP}-0: NG=-0.8N_f+0.75 for n=7, R^2=0.98, p \le 0.001$$

(16)

$$\mathrm{CP}-\mathrm{F}: NG=-0.7N_f+9.9 for n=7, R^2=0.98, p \le 0.001$$

(17)

In fact, lower values of NG, i.e., larger amounts of unproductive inorganic N in the soil–winter-wheat system, were recorded for the CP-0 system (Figure 2). The smallest differences between both fungicide protection systems were documented for plots fertilized with low N_f doses, i.e., from 0 to 80 kg N ha⁻¹. The advantage of CP-F over CP-0 in plots fertilized with 200 and 240 kg N ha⁻¹ reached 35 kg N ha⁻¹. It should be emphasized that both indicators of N_in productivity, i.e., PFP-N_in and NG, did not show a significant relationship with GY (

Table A1. Matrix of correlation of nitrogen soil resources, pools and indices, n = 36.

Traits	Nin	PFP-Nin	NG	Ngr	Nhrr	NA	Nminr	Nb	Ng89	Nint	PFP-Nint	NGt	NR
GY	0.57 ***	0.08	−0.10	0.90 ***	0.33 *	0.82 ***	0.49 **	−0.35*	0.52 ***	0.81 ***	0.20	−0.09	0.49 **
Nin	1.00	−0.72 ***	−0.88 ***	0.84 ***	0.83 ***	0.90 ***	0.61 ***	0.48 **	−0.24	0.89 ***	−0.58 ***	−0.79 ***	−0.35 *
PFP-Nin		1.00	0.92 ***	−0.30	−0.78 ***	−0.44 **	−0.29	−0.75 ***	0.61 ***	−0.44 **	0.90 ***	0.85 ***	0.79 ***
NG			1.00	−0.49 **	−0.82 ***	−0.60	−0.45 ***	−0.80 ***	0.60 ***	−0.61 ***	0.82 ***	0.90 ***	0.67 ***
Ngr				1.00	0.63 ***	0.98 ***	0.62 ***	−0.05	0.27	0.97 ***	−0.20	−0.49 **	0.15
Nhrr					1.00	0.77 ***	0.58 ***	0.37 ***	−0.14	0.77 ***	−0.79 ***	−0.89 ***	−0.21
NA						1.00	0.65 ***	0.05	0.19	0.99 ***	−0.36*	−0.62 ***	0.08
Nminr							1.00	0.08	0.29	0.74 ***	−0.42 **	−0.63 ***	0.18
Nb								1.00	−0.93 ***	0.06	−0.60 ***	−0.55 ***	−0.79 ***
Ng89									1.00	0.22	0.42 **	0.30	0.80 ***
Nint										1.00	−0.39 *	−0.66 ***	0.10
PFP-Nint											1.00	0.93 ***	0.55 **
NGt												1.00	0.45 **

***, **, * indicate significant differences at p < 0.001, p < 0.01 and p < 0.05, respectively. Legend: GY—grain yield; N_in—mineral nitrogen input into the soil–crop system; PFP-N_in—partial factor productivity of N input in the soil–crop system; NG—nitrogen gap; N_hrr—N in straw and harvest residues; NA—nitrogen accumulation in wheat; N_minr—residual mineral N at harvest; N_b—nitrogen balance; N_g89—nitrogen gain in the soil–crop system; N_int—total nitrogen in the soil–crop system; PFP-N_int—productivity of total N input in the soil–crop during the whole growing season; NR—nitrogen recovery.

).

3.2. Nitrogen Pools at Harvest

N resources in the soil–WW system at harvest include three pools: (i) grain N, i.e., N accumulated in grain (N_gr), (ii) the amount of N in harvest residues (N_hrr), (iii) the amount of residual mineral N in the soil (N_minr). All of these N pools were significantly and positively associated with GY, but the strongest correlation was noted for N_gr (r = 0.90 ***). There was also a positive relationship between this N trait and the other N pools (Table A1). The N_gr course was strongly affected by the CP × N interaction (Figure 3). Significantly higher amounts of N_gr for the CP-F system were recorded for each N plot, except for the one fertilized with 80 kg N ha⁻¹. Grain N increased by 14.7% in the N control plot and by 24.2% in the plot fertilized with 240 kg N ha⁻¹. In general, the rate of N accumulation in grain was significantly stronger for WW protected with fungicides, as shown for the obtained regression models:

$$1.\mathrm{CP}-0:N_{gr}=0.5N_f+114.6forn=7,R^2=0.96,\le 0.001$$

(18)

$$2.\mathrm{CP}-\mathrm{F}:N_{gr}=0.66N_f+124.6forn=7,R^2=0.99,\le 0.001$$

(19)

The value of the slope coefficient of CP-F system compared to the CP-0 system was higher by 32% and the equation constant by 8.7%.

The second N pool concerns the amount of N in the harvest residues (N_hrr). This N pool was highly dependent on the years and experimental factors examined, but no interaction between them was found (

Table 3. Nitrogen pools, resources and indices at winter wheat harvest.

Variable	Level of Variable	Ngr	Nhrr	NA	Nminr	Nb	Ng89	Nint	PFP-Nint	NGt	NR
		kg N ha−1							kg kg−1 N	kg N ha−1	%
Year (Y)	2014	174.9 b	45.8 b	220.7 c	25.7 c	−14.4 b	40.0 b	246.3 c	36.6 a	−22.2 a	87.5 b
	2015	229.9 a	66.0 a	295.9 a	48.4 a	−46.9 c	95.5 a	344.4 a	32.4 b	−64.8 b	132.2 a
	2016	162.3 c	68.0 a	230.2 b	37.3 b	−0.2 a	37.5 b	267.5 b	25.3 c	−99.3 c	69.3 c
Fc, p		264 ***	45.5 ***	214 ***	48.9 ***	73.2 ***	91.4 ***	228 ***	306 ***	179 ***	61 ***
Crop Protection	CP-0	184.4 b	62.6 a	237.0 b	35.8	−8.6 a	44.3 b	272.8 b	30.5 b	−65.8 b	94.3
(CP)	CP-F	203.6 a	57.2 b	260.9 a	38.5	−32.4 b	70.9 a	299.4 a	32.4 a	−58.5 a	98.4
Fc, p		131 ***	6.7 *	54.4 ***	2.1 ns	54.5 ***	45.3 ***	45.3 ***	25.4 ***	4.8 *	0.7 ns
	0	109.3 f	34.8 e	144.1 g	28.7 d	−35.6 d	64.3 ab	172.7 f	35.4 a	−18.6 a	-
Nitrogen rates	40	145.3 e	49.3 d	194.5 f	30.8 cd	−46.1 d	76.8 a	225.3 e	34.4 a	−27.5 a	134.1 a
(N), kg N ha−1	80	171.6 d	55.1 cd	226.7 e	32.7 cd	−38,2 d	70.9 a	259.3 d	33.8 ab	−36.5 ab	103.2 b
	120	195.6 c	61.7 bc	257.3 d	36.0 b–d	−28.8 cd	64.8 ab	293.3 c	32.0 b	−54.5 b	94.3 bc
	160	214.9 b	71.1 ab	286.0 c	39.1 a–c	−17,6 c	56.7 ab	325.1 b	29.4 c	−82.4 cd	88.7 c
	200	236.6 a	70.9 ab	307.5 b	45.0 ab	0.9 b	44.1 bc	352.5 a	27.9 cd	−100.5 cd	81.7 cd
	240	249.9 a	76.6 a	326.6 a	47.7 a	21.0 a	25.8 c	374.2 a	27.2 d	−114.9 d	76.0 d
Fc, p		222 ***	28.0 ***	231 ***	8.3 ***	32.2 ***	11.2 ***	189 ***	44.4 ***	72.6 ***	12.6 ***
Source of variation for the studied interactions
Y × CP		**	ns	*	*	*	*	*	**	ns	***
Y × N		***	ns	***	***	***	***	***	**	ns	*
CP × N		***	ns	ns	ns	ns	*	*	ns	ns	ns
Y × CP × N		ns	ns	ns	ns	ns	ns	ns	ns	ns	ns

Similar letters means a lack of significant differences using Tukey’s test; ***, **, * indicate significant differences at p < 0.001, p < 0.01 and p < 0.05, respectively. Legend: N_gr—N amount in grain; N_hrr—N amount in harvest residues; NA—N accumulated in wheat at harvest; N_minr—residual mineral N at harvest; N_b—nitrogen balance; N_g89—nitrogen gain in the soil–winter-wheat system; N_int—total nitrogen input in the soil–winter-wheat system; PFP-N_int partial factor productivity of the total N input into the soil–winter-wheat system; NG_t—actual nitrogen gap; NR—nitrogen recovery.

). The lowest N_hrr was recorded in the 1st season. In other seasons, it was higher by 50%. A significantly higher N_hrr value was recorded for not-fungicide-protected WW (+9.4%). The effect of N_f doses, regardless of other factors examined, was progressive, showing a linear trend.

Both of these two wheat N pools constitute the total stock of N in WW at harvest (NA). The variability of this N characteristic depended on the Y × N and Y × CP interaction, but the former was more significant (Table 3, Figure S1). In each year, the amount of N in WW increased in line with the increase in N_f doses. Significantly greater values, starting from the plot fertilized with 40 kg N ha⁻¹, were recorded in the 2nd season. NA was significantly associated with primary N pools at harvest, being the strongest for N_gr (r = 0.98 ***) and slightly weakly for N_hrr (r = 0.77 ***). At the same time, it was significantly affected by N_in (r = 0.90 ***) (Table A1).

The third basic component of N resources in WW after its harvest is the amount of N_min in the soil, the so-called residual N_min (N_minr). This N pool was significantly affected by the Y × N interaction (Table 3). Therefore, weather was the dominant factor influencing the residual N_min content (Figure S2). Moreover, the amount N_min in the soil varied in a relatively narrow range from 45 kg N ha⁻¹ in a plot fertilized with 40 kg N ha⁻¹ to 58 kg N ha⁻¹ in the plot fertilized with 200 kg N ha⁻¹. The largest N_min amount, which dominated in this range, was found in 2015. The lowest N_min values recorded in 2014 ranged from 22 to 33 kg N ha⁻¹. A completely different N_min trend was observed in 2016, when a linear N_min increase was demonstrated in response to the applied N_f doses. In this particular year, the N_minr increased from 17 kg N ha⁻¹ in the N control plot to 66 kg N ha⁻¹ in the plot fertilized with 240 kg N ha⁻¹.

3.3. Nitrogen Gain

The analyses of N sources for WW presented above are incomplete. One of the inorganic N pools was omitted, namely, the amount of N_min released from soil resources in the spring part of the WW growing season. Three indices have been developed to describe this specific N pool. The first is the N balance (N_b), based on the amount of inorganic N (N_min + N_f) introduced and removed (NA = N_gr + N_hrr) from the soil–wheat system. The variability of this N characteristic depended on the Y × N and Y × CP interactions, with the former being statistically stronger (Table 3). The decisive factor, affecting N_b variability, was the weather in consecutive growing seasons (Figure A3). The most negative N balance, regardless of N doses, was recorded in 2015. In this particular year, its values first decreased in the N_f range from 0 to 80 kg N ha⁻¹, then recovered. However, over the whole range of the tested N_f doses, they were always negative. A completely different course of this N characteristic was recorded in other years. In 2014, the N_b basically followed the linear regression model, but positive values were first noted in the plot with 200 kg N ha⁻¹. In 2016, this trend was almost identical, but positive N_b values were achieved first in the plot fertilized with 160 kg N ha⁻¹.

The pool of N_min released from the soil N stock during the growing season was called N gain 89 (N_g89). This N pool showed a significant response to experimental factors and the years. The highest N_g89 values were recorded in 2015. In the remaining years, they were significantly lower (approximately by 50%) (Table 3). However, the amount of N_g89 actually depended the CP × N interaction (Figure 4). The application of fungicides to WW, averaged over N_f doses, increased this N pool by 60%. The use of fungicides in the control N plot increased the N_g89 pool by 30.3% (from 56 to 73 kg N ha⁻¹). The N_g89 pool on the CP-0 object was the highest on plots fertilized with 40 and 80 kg N ha⁻¹ and then strongly decreased. Compared to the N control plot, it increased by 23.2% (from 56 to 69 kg N ha⁻¹). On the CP-F object, the highest N_g89 was recorded in the plot fertilized with 40 kg N ha⁻¹. It was higher by 15% (from 73 to 84 kg N ha⁻¹) compared to the corresponding N control plot. At the same time, it was higher by 21.7% to the corresponding plot on the CP-0. In addition, this particular N trait showed high stability for the N_f doses ranging from 0 to 160 kg N ha⁻¹. The course of N_g89, as shown in Figure 4, can be presented in the form of a regression model. The obtained data fit the quadrate regression model well:

$$\mathrm{CP}-0: N_{g89}=-0.0018N^2+0.17N+60.4 for n=7, R^2=0.97, p \le 0.01$$

(20)

$$\mathrm{CP}-\mathrm{F}: N_{g89}=-0.0013N^2+0.21N+74.5 for n=6, R^2=0.95, p \le 0.05$$

(21)

The maximum N_g89 value for the CP-0 system of 64.4 kg N ha⁻¹ was recorded for N_f of 94.4 kg N ha⁻¹. At the same time, for the CP-F system, these values were 83.0 kg N ha⁻¹ and 80.8 kg N ha⁻¹, respectively. This means that they were larger at lower N_f doses.

The relationship between N_g89 and N_b clearly explains the effect of fungicide use on N economy in WW (Figure A4). Regardless of the level of this crop protection with fungicides, the obtained relationships showed a downward trend. The slope coefficient was more negative for CP-F than CP-0, but started from a significantly higher level, as indicated by a constant value (75.9 vs. 45.0 for CP-F and CP-0, respectively). Negative N_b values indicate a net increase in the content of N_min. N_g89, assuming no change in N_min resources during the spring vegetation of WW, i.e., for N_b = 0.0 kg N ha⁻¹, reached +37.2 and +49.6 kg N ha⁻¹ for the non-protected and fungicide-protected crop.

The total amount of N_in and N_g89 pools constitute the total inorganic N stock (N_int) in the spring part of WW vegetation. This aggregate index was largely driven by years and experimental factors. The impact of years was quite clear, and the highest N_int in the soil–WW system was recorded in 2015. In 2014, it was lower by 28.5% and by 22.3% in 2016. Fungicide protection of wheat increased N_int significantly by 9.8% (Table 3). However, a decisive impact on the variability of N_int was by the CP × N interaction (Figure 5). Its values increased gradually, regardless of the level of crop protection, in accordance with the increased N_f doses. Significant differences between both treatments began first with the plot fertilized with 120 kg N ha⁻¹. The resulting linear models are as follows:

$$\mathrm{CP}-0: N_{int}=0.75N_{int}+183.1 for n=7, R^2=0.97, p \le 0.001$$

(22)

$$\mathrm{CP}-\mathrm{F}: N_{int}=0.90N_{int}+190 for n=7, R^2=0.99, p \le 0.001$$

(23)

The key differences between both fungicide treatments concern the initial N_int and the slope coefficient values. Both characteristics were much lower for the CP-0 system by 4.4% and 16.7% compared to the CP-F system.

3.4. Nitrogen Gap—Second

N_int was then used to calculate the actual productivity of inorganic N in the soil–WW system (Partial Factor Productivity of N_int, PFP-N_int) during the spring part of the growing season. The variability of PFP-N_int depended on the interaction of both experimental factors with years, but no interaction between them was found (Table 3). In 2014 and 2015, relative stability of the index was observed in the N_f range of 0–80 kg N ha⁻¹). In general, regardless of growing seasons, the course of the PFP-N_int showed a decreasing trend in response to increasing N_f doses. In the assessment of this N trait, the most important were attributes of the obtained linear models, reflecting the effect of years (Figure S3). The constants of the developed equations decreased in the order: 2014 (42.7 kg N ha⁻¹) > 2015 (36.1 kg N ha⁻¹) > 2016 (29.1 kg N ha⁻¹). At the same time, the slope coefficient was much larger in 2014 compared to other years, indicating a stronger PFP-N_int decline in response to the increasing N_int. PFP-N_int was weakly related to both N_b and N_g89, but strongly to N_in (Table A1).

The total N gap (NG_t) was calculated on the basis of PFP_int. This N pool was highly dependent on the years and experimental factors examined, but no interaction between them was found (Table 3). With respect to years, the highest NG_t of almost 100 kg N ha⁻¹ was recorded in 2016. In 2014, it was almost 5-times lower. In 2015, NG_t was much lower compared to 2016, but at the same time, it was almost three times larger than in 2014. The effect of fungicide use was quite obvious. A significantly higher absolute NG_t value was found for the CP-0 system. Increasing N_f doses significantly increased absolute NG_t values, consistent with a linear regression model. It should be emphasized that NG_t significantly correlated with NG (r = 0.90 ***). At the same time, it correlated negatively with the main N wheat traits, especially with N_hrr (r = −0.89 ***) and also with N_minr (r = −0.63 ***) (Table A1).

The N recovery rate (NR) showed a response to the interaction of years with experimental factors, but no interaction between them was found (Table 3). Of the two significant interactions, Y × CP was the stronger (Figure A5). The effect of CP varied in subsequent years of study. In 2014, significantly higher NR was recorded for the CP-F system. An opposite situation was recorded in 2015, while in 2016, no difference between crop protection systems was found. The NR trend in subsequent years of study was highly specific. In each season, the highest NR values were recorded for the plot fertilized with 40 kg N ha⁻¹ and the lowest for the plot fertilized with 240 kg N ha⁻¹. However, in 2014 and 2015, the NR trend followed a quadratic regression model (Figure S4). Additionally, in 2015, NR values in the entire range of tested N_f doses were above the threshold value of 100%. In 2016, such a level was found only for a plot fertilized with 40 kg N ha⁻¹. In 2016, the NR trend followed a linear regression model, decreasing with increasing N_f doses.

4. Discussion

4.1. Nitrogen Pools

The total amount of inorganic N in the soil–winter-crop system in the spring part of the growing season consists of three pools [7]. The main question is as follows: to what extent do these resources determine the grain yield? The effect of both initial N pools (N_min + N_f), defined as N input (N_in), on the GY of WW was significant but not dominant (R² = 0.32 ***). A much stronger relationship was found for N_in with the grain N (N_gr89, R² = 0.70 ***). Regardless of weather conditions in subsequent growing seasons, the key factor affecting both wheat traits was plant protection using fungicides. The net increase in N_gr (ΔN_gr) due to the use of fungicides was consistent with the amount of N_in in the soil–winter-wheat system:

$$\mathsf{\Delta }Ngr=0.16N_{in}+7.31 for n=7, R^2=0.77, p \le 0.001$$

(24)

The slope coefficient of the above equation clearly suggests a higher N supply to WW treated with fungicides. The difference between both fungicide protection systems in the plots with the highest N_in reached approximately 50 kg N ha⁻¹. This value is equal with the GY increase of 2.2 t ha⁻¹. This increase was confirmed in the CP-F system. In addition, this difference clearly indicates a good supply of N to the fungicide-treated WW during the grain filling period (GFP), which was confirmed by Szczepaniak [25]. As is known, only when there is an excess of N accumulated in the grain in relation to its mass, the grain N concentration increases [26,27].

To explain the phenomenon obtained, it is first necessary to recognize the association between fungicide protection of WW and the amount of N_g89. This is the third main pool of inorganic N for WW, which appears in the spring part of the growing season [7,14]. This N pool, considered separately, was important in GY formation but was not a key factor (R² = 0.27 ***). At the same time, no significant effect of N_g89 on N_gr was found. It can therefore be concluded that the effect of N_g89 on WW must have occurred before flowering. These relationships indirectly suggest that the yield-forming effect of N_g89 resulted from its effect on CGD of WW, while being neutral on the protein content in the grain [28,29].

The ΔN_g89 trend, confirming the significant impact of fungicide protection on WW, showed a quite specific response in relation to N_f doses. The entire course of N_g89, considered in relation to N_f doses, can be divided into two parts. The obtained relationships are illustrated by the following set of equations:

$$\mathrm{Low} \mathrm{N} \mathrm{doses}: \mathsf{\Delta }{N}_{g89}=-0.16N_f+18.4 for n=3, R^2=0.87, p \le 0.05$$

(25)

$$\mathrm{High} \mathrm{N} \mathrm{doses}: \mathsf{\Delta }{N}_{g89}=-0.0016N_f^2+0.75N_f+42.4 for n=4, R^2=0.92, p \le 0.05$$

(26)

Generally, as shown in Figure 4, the net amount of N_g89, regardless of WW protection with fungicides, increased in the N_f range from 0 to 80 kg N ha⁻¹. The recorded decrease in ΔN_g89, as shown in Equation (25), indicates the occurrence of a phenomenon called “the N priming effect” [30]. This phenomenon was revealed in the reaction of WW plants to insufficient N_f supply during the critical stages of yield formation. The plants suffering from N deficiency stimulated the mineralization processes of organic N in the soil. This hypothesis is justified by the fact that the highest demand for N in WW occurs in the phase of intensive stem and ear growth (BBCH 40–49), which determines CGD [31]. In the second, much larger range of N_f doses examined, the ΔN_g89 trend was consistent with the quadratic regression model. For N_fop of 234 kg N ha⁻¹, a ΔN_g89 amounted to 45.5 kg N ha⁻¹. This amount of inorganic N released from soil N resources on the fungicide-treated object is almost the same as the amount of N accumulated in the grain (Equation (24)). These two relationships, obtained independently of each other, indicate that the available N resources during the grain filling period of GY development are the basic source of grain N and, thus, protein in the grain. However, their use depends on the wheat canopy index, i.e., CGD, which is formulated before flowering [28,29,32].

However, only the inclusion of all three inorganic N resources in the aggregated index, defined as the total N input (N_int), allowed us to explain the variability of WW grain yields in subsequent growing seasons (R² = 0.66 ***) and, even more strongly, the variability of N_gr (R² = 0.94 ***). The strength of these two relationships confirms the importance of N supply for the grain N content [13].

An important resource of inorganic N in the soil that requires consideration, mainly from the point of view of environmental protection, is residual N (N_minr) [33]. In the studied case, the N_minr content depended mainly on the course of weather during the consecutive growing seasons and on the N_f doses applied. In general, the obtained values were low (<50 kg N ha⁻¹) according to EU standards, ranging from 30 to 70 kg N ha⁻¹ [34]. Even more, in 2014, they were below the threshold range, which proves the high yield-forming efficiency of inorganic N in the soil–WW system. Moreover, the N_minr content showed a significant and relatively high correlation with main wheat traits, including the most important N_gr (R² = 0.42 ***) and much lower with GY (R² = 0.24 **). This is further evidence of the importance of available N resources for the N content in grain during GFP.

4.2. Indicators of Inorganic N Economy in the Soil–Winter-Wheat System

The yield-forming effect of inorganic N on N resources in the soil–WW system was assessed using a diverse set of indicators. The basic, simple indicator is gross N productivity, relating GY to the mass of N introduced or present in the soil–crop system. So far, only the N_f dose is used in the calculation of N use efficiency (NUE) [35]. Unfortunately, this calculation procedure is subject to errors because it does not take into account soil N resources, at least the initial ones [7].

In the presented case, the initial amount of inorganic N in the N control plot (N_f control) was 108.5 kg N ha⁻¹, allowed the production of 6 t grain ha⁻¹. This level of GY is very high and confirms the high value of winter oilseed rape as a forecrop for WW [36,37]. The calculated PFP-N_in was 57 kg grain per kg N_in. This value is high, but in essence, it is not true. The total amount of inorganic N in the soil during the growing season of WW on the N_f control, including its amount released from soil resources, was 173 kg N ha⁻¹. Therefore, the actual N productivity in the examined system was much lower and amounted to 34 kg grain kg⁻¹ N. It is obvious that increasing N_f doses resulted in a progressive decline of PFP-N index. It is worth emphasizing the stability of the discussed index in the N_f ranging from 0 to 80 kg N ha⁻¹. The reason for this phenomenon was the increase in the amount of inorganic N released from the soil, observed in this set of N_f plots (N_g89; Figure 4).

Regardless of the N resources, included in the PFP-N index calculations, its value was always higher for the CP-F system. Moreover, both PFP-N indicators were related to each other, but at the same time, they were negatively related to N_in. This relationship indicates an oversupply of inorganic N into the soil–WW system. In such a case, the only practical solution is to reduce the N_f dose applied [38]. However, this recommendation does not apply to wheat grown for flour, but it fully applies for the so-called fodder wheat [3].

The developed PFP-N indices were then used to calculate the N gap (NG), which is an indicator of N_f inefficiency in the soil–crop plant system [7,20,39]. However, it can also be used to assess the inefficiency of inorganic N present in the soil–crop system during the growing season. Regardless of the calculation basis, both examined NG indicators clearly showed a significant and negative relationship with the initial amount of inorganic N in the soil–winter-wheat system. The developed models are as follows:

$$\mathrm{CP}-0: NG=-0.8N_{in}+87.4 for n=7, R^2=0.98, p \le 0.001$$

(27)

$$\mathrm{CP}-\mathrm{F}: NG=-0.7N_{in}+85.0 for n=7, R^2=0.98, p \le 0.001$$

(28)

In general, the amount of saved N (ΔNG) due to the protection of WW with fungicides increased in accordance with the N_f dose increase. The dependence obtained is described by a linear regression model:

$$\mathsf{\Delta }NG=0.11N_{in}-2.3 for n=7, R^2=0.82, p \le 0.001$$

(29)

The obtained models clearly showed that NG increased in line with increasing N_in. The observed trend was both much slower and, at the same time, much lower for WW treated with fungicide. For example, NG on the plot fertilized with 200 kg N ha⁻¹ decreased by 36 kg N ha⁻¹. The practical significance of NGt resulted from a clearer differentiation of the variability of this characteristics in subsequent years of the study.

The analysis of N management in the soil–winter-wheat system was summarized with the classic N indicator, i.e., N recovery (NR). This study revealed four important messages. Firstly, NR, despite variability between years, was very high, even in 2016 with a low yield. Secondly, its trend in the first two seasons followed a quadratic regression model, which indirectly indicates an exponential decline in NR in the low ranges of N_f doses. This type of tendency was related to the “N priming effect”, observed for N doses of 40 and 80 kg N ha⁻¹. Third, this observation was confirmed by a significant linear relationship between N_g89 and NR (R² = 0.64). In fact, a better fit was obtained with the quadratic regression model (Figure A6). The most important information resulting from this relationship is N_g89 of 70.1 kg N ha⁻¹, as the required amount of inorganic N released from soil resources during the winter wheat growing season to meet the NR threshold of 100%. This condition was met in 2015 for the entire range of N_f doses and in 2014 for a plot fertilized with 40 kg N ha⁻¹ but protected with fungicide. Fourth, NR showed no significant relationships with basic N pools during winter wheat harvest (N_gr, N_hrr, N_minr). This means that regardless of weather conditions in a given growing season, there was a balance between the supply of N from the applied N_f and its use by WW. These relationships indeed indicate that the amount of inorganic N released from soil resources during the winter wheat growing season was the main cause of seasonal variability in NR. Therefore, the variability of N_g89 should be treated as a factor explaining the seasonal variability of NR, observed in fertilizer experiments [40,41]. Full NG coverage is an extremely challenging task for the farmer due to the many factors, some out of control, responsible for N use efficiency [42]. In the conducted study, this effect was achieved for NR at the level of 199%. This result was recorded for a plot with N_f dose of 40 kg N ha⁻¹ (Figure S5).

5. Conclusions

Fungicidal protection of WW is an agrotechnical solution that allows the effective use of inorganic N present in the soil in the spring part of the growing season. The amount of N “saved” in the wheat grain increases with the N_f doses applied. The key factor influencing the positive effect of fungicide use in WW was the greater amount of N released from N soil stock during the growing season. These resources acted in two ways. Firstly, they increased the density of grain in the wheat canopy (larger GY), and secondly, they covered the N demand of the growing grains during the grain filling period. The observed “N priming effect” showed a longer and more stable duration for the fungicide-treated crop.

Single, inorganic N pools can be used as predictors of grain yield, but their reliability is relatively low (significant but low R² values). Only data taking into account the full pool of inorganic N in the soil–winter-wheat system allowed for a high and reliable prediction of GY. An even better prediction can be obtained for the amount of N stored in the grain.

The primary indicator of N economy in the soil–crop system is the Partial Factor Productivity of N applied or present in the soil. This indicator confirmed both the positive impact of the use of fungicides in WW on the productivity of inorganic N and, consequently, on the decrease in the N gap. The values of N gap were closely but negatively related to the amount of inorganic N released from the soil during the growing season. As a consequence, the fungicide protection of winter wheat increased the recovery of inorganic N introduced and present in the soil during the growing season. It can be concluded that the seasonal variability of nitrogen recovery observed in fertilizer experiments is in fact due to the variability of the amount of inorganic N released during the growing season for a given crop. Therefore, the essential challenge for scientists is to develop a reliable model for predicting the amount of inorganic N released from soil resources during the growing season of a crop. This is a necessary task to effectively control the use of N from applied fertilizer.