Nitrogen, Phosphorus, and Potassium Uptake in Rain-Fed Corn as Affected by NPK Fertilization

Abstract

Effective nutrient management requires understanding nutrient uptake at various growth stages and nutrient removal by the harvested portion. Information on nutrient accumulation was provided by some older literature, and a few researchers have focused on this issue in this modern period with modern hybrids and improved corn cultivation practices. While almost all the studies were conducted in northern states of the US, information for the Southern Great Plains is still limited. To bridge this knowledge gap, a 2-year field study was conducted in a rain-fed corn production system. The study aimed to evaluate the impact of nitrogen (N), phosphorus (P) and potassium (K) fertilization on N, P, and K contents in aboveground plants at different growth stages. Pre-plant application of N (0, 67, 133 kg N ha⁻¹), P (0 and 20 kg ha⁻¹) and K (0 and 60 kg ha⁻¹) fertilizers was done. Results from our study revealed that nutrient uptake values, pattern and dynamics depend on environmental conditions, soil type and management practices. N concentration in plants showed a linear response to N application rate while P and K concentrations were unaffected by NPK fertilization rates. Total N, P and K uptake was primarily driven by N application rate, showing a linear increase with higher N rates. Co-application of P and K with N did not significantly affect nutrient concentration and uptake.

1. Introduction

Effective nutrient management requires the knowledge of nutrient uptake at different growth stages and removal by the harvested portion. When it comes to fertilization decisions, knowing the accumulation timing and quantity of nutrients removed is helpful [1]. These values may differ from an average crop nutrient value because of different environmental conditions and agronomic practices [2]. There should be a re-evaluation of important nutrient uptakes by corn for specific locations, which can then be utilized to make more accurate fertilizer recommendations that will help current hybrids to achieve their maximum economical yield potential [3].

Information on nutrient accumulation was provided by some older literature [4,5,6,7,8], and a few researchers focused on this issue in this modern period with modern hybrids and improved corn cultivation practices [3,9,10,11,12,13]. While almost all those studies were conducted in the northern states, information for the Southern Great Plains of the US is still limited. Some research has been done on nitrogen (N) uptake in Oklahoma but is still lacking on how N, phosphorus (P) and potassium (K) fertilization practices affect these nutrients uptake and their concentrations in corn plants [14,15]. It was reported in Oklahoma that corn N uptake was 68.8 to 114 kg N ha⁻¹ [15]. According to Girma et al. [14] the maximum N accumulation in whole plant was 42 kg N ha⁻¹ in check plot to 131 kg N ha⁻¹ in a plot received 224 kg N ha⁻¹ for corn grown in Oklahoma. Nitrogen uptake and removal increases with the increases in N application rate [16,17,18]. According to Zone et al. [19], P and K fertilization marginally but consistently increased leaf P and K concentrations of corn grown in Ohio with 1.05% increase in P and 3.17% increase in K. Grain P concentration also showed a directionally positive increase in 51% (21 out of 41) of their corn trials, while grain K increased in 72% (21 out of 29) of trials. Modern hybrids take up more N and P as compared to earlier varieties [10]. They found that 178 and 213 kg N ha⁻¹ were taken up by the corn plants at R6 growth stage in 1960 and 2000 era hybrids, respectively. Phosphorus uptake in corn remains relatively constant from the V6 to R6 growth stages, while most of the K uptake occurs before R2 growth stage [3]. According to Ciampitti et al. [20], 195 kg N ha⁻¹ was accumulated by R6 for 2000 era hybrids. Bender et al. [3] conducted research in Illinois with modern corn hybrids planted at higher densities. They found that plants removed 286 kg N ha⁻¹, 114 kg P ha⁻¹, 202 kg K ha⁻¹ and 26 kg S ha⁻¹ and the grains removed 166 kg N ha⁻¹, 90 kg P ha⁻¹ and 66 kg K ha⁻¹. In a study conducted by Stammer and Mallarino [9], P concentration was 4.8 to 5.3 g kg⁻¹, and K concentration was 18.8 to 25.3 g kg⁻¹ in the whole plant at V5-V6 stage. Discrepancy in these studies suggest that nutrient concentration as well as nutrient uptake varies greatly with plant growth stage, cultivation practices, variety, soil fertility and environmental conditions. Nutrient uptake knowledge is only useful when it is specified to local growing conditions.

According to the 2010 IPNI report, macro- and micro-nutrients had been reduced in Canadian and US soils during the prior 5 years [21]. The combination of high-yielding cultivars and declining soil fertility shows that farmers did not match nutrient uptake and removal with fertilizer applications [21]. The primary goal of this study is to evaluate the effect of N, P and K fertilization on N, P and K contents in an aboveground plant at different growth stages in a rain-fed corn production system. A second objective is to evaluate the seasonal nutrient uptake pattern and accumulation as a function of time as affected by NPK fertilization. We hypothesize that increased NPK fertilization will result in significantly higher N, P and K contents in the aboveground plant tissues at each growth stage.

2. Materials and Methods

2.1. Field Trial, Experimental Design and Treatment Structure

Initially, this field study was conducted at four locations over a span of two years, encompassing 2021 and 2022. The experiment was established at EFAW Research Station in Stillwater, Oklahoma (EFAW21, 36°08′14.1″ N 97°06′22.4″ W) on Ashport silty clay loam soil, and Lake Carl Blackwell Research Farm (LCB), Stillwater, OK (LCB21, no results due to raccoon damage, 36°09′04.8″ N 97°17′21.5″ W) on Pulaski fine-sandy loam in 2021. While in 2022, the trail was conducted at LCB (LCB22, 36°09′1.64″ N 97°17′23.30″ W) and Perkins research station (PRK22, no results due to damages by wild hogs, 35°59′37.32″ N 97°2′31.41″ W) on Teller sandy loam soil [22]. In this article, we will discuss the results from only two site years, EFAW site of 2021 (referred to as EFAW21) and LCB site of 2022 (abbreviated as LCB22).

A randomized complete block design was used for all locations, which included 12 treatments and three replications. Each treatment plot was measured 3 × 6 m and consisted of four rows of corn plants, with an alley of 3 m between each replication. The study encompassed a total of 12 distinct treatments, systematically exploring the factorial combinations of nitrogen (N), phosphorus (P) and potassium (K) fertilizer rates applied preplant for corn (

Table 1. Nitrogen (N), phosphorus (P) and potassium (K) application rates employed at both EFAW and LCB locations.

Treatment Code	Nutrient Rates (kg ha−1)
	N	P	K
N0P0K0	0	0	0
N0P0K1	0	0	60
N0P1K0	0	20	0
N0P1K1	0	20	60
N1P0K0	67	0	0
N1P0K1	67	0	60
N1P1K0	67	20	0
N1P1K1	67	20	60
N2P0K0	133	0	0
N2P0K1	133	0	60
N2P1K0	133	20	0
N2P1K1	133	20	60

). Nitrogen rates evaluated included 0, 67 and 133 kg N ha⁻¹. The rates were 0 and 20 kg P ha⁻¹ and 0 and 60 kg K ha⁻¹. This rigorous factorial arrangement enabled a robust evaluation of the interactive effects of these essential nutrients on the corn crop under study. All fertilizers were applied as pre-plant using a barber metered feed fertilizer spreader using appropriate settings to achieve the desired fertilizer rates. After applying a particular nutrient, the hopper of the spreader was cleaned with pressurized air. Fertilizer sources for N, P and K included urea (46-0-0), triple super phosphate (0-46-0) and Muriate of potash (0-0-60).

Approximately four days after pre-plant fertilizer applications were made, when the soil moisture level and soil temperature were both adequate, corn hybrid “DKC66-29” (DeKalb Genetics Corporation, DeKalb, IL, USA) was planted at both sites using a John Deere Max Emergence 2 7300 four row planter. Row spacing was 76 cm with a population of 49,400 seeds ha⁻¹ at EFAW21 and 69,160 seeds ha⁻¹ at LCB22. Integrated Pest Management was performed according to Oklahoma State University recommendations.

2.2. Soil Sampling and Analysis

Composite pre-plant soil samples were taken from both locations at 0–15 cm depth from each replication. Soil samples were dried at 65 °C for 12 h and passed through a 2-mm sieve in preparation for the determination of pH, ammonium-N (NH₄-N), nitrate-N (NO₃-N), plant available P and K. Analysis of soil nitrate and ammonium nitrogen was performed using 5 g of soil sample extracted with 25 mL of 1 M KCl solution [23]. These samples were filtered after shaking for 30 min and analyzed using a Lachat flow injection autoanalyzer. Mehlich 3 (M3) extractant (Mehlich, 1984) was used for P and K extraction and quantification by an Inductively Coupled Plasma-Atomic Emission Spectroscopy. Results of the pre-plant routine soil tests are shown in

Table 2. Basic soil properties at EFAW in 2021 and Lake Carl Blackwell (LCB) in 2022.

Year	Location	pH	Organic Carbon %	NH4-N	NO3-N mg kg−1	P ǂ	K ǂ
2021	EFAW	5.6	0.83	42.5	23.3	24	201
2022	LCB	6.1	0.63	3.3	<0.1	11.5	69.5

^ǂ P& K are plant available phosphorus (P) and potassium (K) using Mehlich 3, respectively.

.

2.3. Plant Samples Collection and Analysis

Plant samples were collected from the side two rows of each plot at V6, VT, R2 and R6 growth stage for nutrient analysis [24]. A 1-meter length of row was randomly selected; then, plants were cut at 5 cm above ground level. Fresh weight was taken after collecting plant samples and then weighed again after oven-drying to determine subsample aliquot dry weight and dry biomass accumulation. Dry plant samples were then ground and passed through a 1-mm sieve.

Because a rain-fed crop often loses biomass per unit area over time, data for nutrient uptake curves and seasonal nutrient uptake patterns were determined per plant basis. We counted the number of plants in the 1-m rows each time we took a biomass sample. To determine the amount of biomass per plant, the total biomass from these meter rows was determined and then divided by the number of plants.

Upon crop maturity, the middle two rows of each plot were harvested using a Kincaid 8XP plot combine equipped with a Harvest Master Yield monitor in which the final grain yield reported was adjusted to 15.5% moisture.

For grain N and crude protein (CP) analysis, a LECO CN 828 instrument was used (LECO Corp. St. Joseph, MI, USA). Total P and K in plant samples were determined by nitric acid digestion and inductively coupled plasma spectrometry quantification. Nutrient concentrations (Nc) were provided in percentage and the values of total aboveground macronutrient uptake were calculated by Equation (1). While Equations (2) and (3) were used for the estimation of nutrient removal by grains and whole plants, respectively.

$${\mathrm{Plant}\mathrm{Macronutrient}\mathrm{Uptake}(\mathrm{kg}\mathrm{ha}}^{-1})={\mathrm{Dry}\mathrm{Biomass}(\mathrm{kg}\mathrm{ha}}^{-1})\times \mathrm{Nc}$$

(1)

$${\mathrm{Grain}\mathrm{Macronutrient}\mathrm{Uptake}(\mathrm{kg}\mathrm{ha}}^{-1})={\mathrm{Yield}(\mathrm{kg}\mathrm{ha}}^{-1})\times \mathrm{Nc}$$

(2)

$${\mathrm{Plant}\mathrm{Macronutrient}\mathrm{Uptake}(\mathrm{kg}\mathrm{ha}}^{-1})={\mathrm{Dry}\mathrm{Biomass}(\mathrm{g}\mathrm{pl}}^{-1})\times \mathrm{Nc}$$

(3)

Nutrient harvest index is the partitioning efficiency of nutrient to grain [3]. Following grain nutrient harvest, indices are calculated based on grain nutrient uptake and total aboveground plant nutrient uptake.

The Grain Nitrogen Harvest Index (GNHI) was calculated to estimate the amount of nitrogen portioned to the grain by plant

$$\mathrm{G}\mathrm{N}\mathrm{H}\mathrm{I}=\frac{\mathrm{N}\mathrm{i}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{g}\mathrm{e}\mathrm{n}\mathrm{u}\mathrm{p}\mathrm{t}\mathrm{a}\mathrm{k}\mathrm{e}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{r}\mathrm{a}\mathrm{i}\mathrm{n}}{\mathrm{N}\mathrm{i}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{g}\mathrm{e}\mathrm{n}\mathrm{u}\mathrm{p}\mathrm{t}\mathrm{a}\mathrm{k}\mathrm{e}\mathrm{b}\mathrm{y}\mathrm{w}\mathrm{h}\mathrm{o}\mathrm{l}\mathrm{e}\mathrm{p}\mathrm{l}\mathrm{a}\mathrm{n}\mathrm{t}}\times 100$$

(4)

The Grain Phosphorus Harvest Index (GPHI) was calculated to determine how much K was portioned to the grain by plant.

$$\mathrm{G}\mathrm{P}\mathrm{H}\mathrm{I}=\frac{\mathrm{P}\mathrm{h}\mathrm{o}\mathrm{s}\mathrm{p}\mathrm{h}\mathrm{o}\mathrm{r}\mathrm{u}\mathrm{s}\mathrm{u}\mathrm{p}\mathrm{t}\mathrm{a}\mathrm{k}\mathrm{e}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{r}\mathrm{a}\mathrm{i}\mathrm{n}}{\mathrm{P}\mathrm{h}\mathrm{o}\mathrm{s}\mathrm{p}\mathrm{h}\mathrm{o}\mathrm{r}\mathrm{u}\mathrm{s}\mathrm{u}\mathrm{p}\mathrm{t}\mathrm{a}\mathrm{k}\mathrm{e}\mathrm{b}\mathrm{y}\mathrm{w}\mathrm{h}\mathrm{o}\mathrm{l}\mathrm{e}\mathrm{p}\mathrm{l}\mathrm{a}\mathrm{n}\mathrm{t}}\times 100$$

(5)

The Grain Potassium Harvest Index (GKHI) was estimated to quantify the amount of K portioned to the grain by plant.

$$\mathrm{G}\mathrm{K}\mathrm{H}\mathrm{I}=\frac{\mathrm{P}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{s}\mathrm{s}\mathrm{i}\mathrm{u}\mathrm{m}\mathrm{u}\mathrm{p}\mathrm{t}\mathrm{a}\mathrm{k}\mathrm{e}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{r}\mathrm{a}\mathrm{i}\mathrm{n}}{\mathrm{P}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{s}\mathrm{s}\mathrm{i}\mathrm{u}\mathrm{m}\mathrm{u}\mathrm{p}\mathrm{t}\mathrm{a}\mathrm{k}\mathrm{e}\mathrm{b}\mathrm{y}\mathrm{w}\mathrm{h}\mathrm{o}\mathrm{l}\mathrm{e}\mathrm{p}\mathrm{l}\mathrm{a}\mathrm{n}\mathrm{t}}\times 100$$

(6)

2.4. Statistical Analysis

Data were analyzed using SAS version 9.4 (SAS Institute, Cary, NC, USA). ANOVA was applied using PROC-GLIMMIX procedure, and the mean separation procedure was done by Tukey. Each site and year were analyzed separately. Treatments were used as fixed effect and replications as random. All nutrient uptake values were calculated on dry weight basis.

3. Results and Discussions

3.1. Nutrient Concentration in Whole Plant at Early Growth Stage (V6/V7) and Plant Components at Physiological Maturity (R6)

3.1.1. Nitrogen Concentration

There was a main effect of N rate on N concentration at V6/V7 growth stage at EFAW21 and LCB22 (Table S1). Nitrogen concentrations in plants increased with the increase in N rates (Figure 1). The addition of P, K or both did not have any significant effect of decreasing or increasing N concentration. Terman et al. [25] also documented that P had no effect of increasing N concentration in corn plants.

At maturity, there was a main effect of N rate on N concentration in stover, ear and grain at both locations. Nitrogen concentrations in these plant components were significantly greater when a higher rate of N was applied (Figure 1). This could be due to sufficient availability of N in soil with higher N input, which further led to better N uptake and utilization in cell metabolic processes. These results were similar to those of Bruns and Ebelhar [26], who found that N concentrations in stover and grain increased as the N fertility level increased. Kurtz et al. [27] also found that increasing N fertility increased protein content in corn grain, which is the major form of N found in corn grain. This positive effect of increased N concentration by N rates has been often reported. The maximum grain N concentration in our study was 14.0 g kg⁻¹ which is consistent with the findings of Heckman et al. [2], Setiyono et al. [28] and Bender et al. [3].

3.1.2. Phosphorus and Potassium Concentrations

There was no significant effect observed at V6 growth stage for P and K concentrations at EFAW21. The main effect of N (p ≤ 0.001) and main effect of K (p = 0.013) was observed at LCB22 for P concentration. The similar effect of N (p = 0.005) and K (p = 0.003) was observed for K concentration at this site. The phosphorus concentration in whole plant ranged from 2.49 g kg⁻¹ to 2.82 g kg⁻¹ at EFAW21 and 3.89 g kg⁻¹ to 4.97 g kg⁻¹ at LCB22. While K concentration ranged from 39.9 g kg⁻¹ to 52.4 g kg⁻¹ at EFAW21 and 33.6 g kg⁻¹ to 50.5 g kg⁻¹ at LCB22 (Table S1). The disparity in P concentrations between sites can be attributed to the dilution effect caused by more biomass at the EFAW21 site. However, this effect was not observed for K concentration, which might suggest the luxury uptake of K.

At maturity, stover P concentration was significantly affected by the N rate at LCB22 site only. Stover P concentration decreased linearly with N rate at this site. Stover P concentration was reduced by half where 133 kg N ha⁻¹ was applied. At EFAW21 site, P concentration in stover tended to decrease as N rate increased but was not significantly different (Figure 1). Stover P concentration ranged from 0.83 g kg⁻¹ to 2.5 g kg⁻¹ at LCB22 and 0.37 g kg⁻¹ to 0.65 g kg⁻¹ at EFAW21. Stover P concentration at EFAW21 falls within the range reported by Setiyono et al. [28] while ear P concentration was similar among all treatments at both locations, which varied from 1.9 g kg⁻¹ to 2.5 g kg⁻¹ at EFAW21 and 2.6 g kg⁻¹ to 3.3 g kg⁻¹ at LCB22 (Table S1).

At maturity, only the main effect of K was observed in stover for increasing stover K concentration at EFAW21. The K concentration in stover ranged from 14 to 17.5 g kg⁻¹ while K concentration in ear ranged from 4.4 to 5.4 g kg⁻¹ at this site (Table S1). The presence of K in the NPK combination increased K concentration in stover by 12.5%. The positive effect of K may be due to the fact that K enhances tissue turgor pressure, which regulates the opening and closing of stomata [29]. However, this effect of K application was not seen at LCB22 site. This might be due to dry weather condition during 2022, which could have impeded K uptake in plants due to reduced soil moisture and limited water mobility, ultimately leading to lower K concentrations in plant components [30]. Restricted root activity and competition for water resources during drought further contribute to the difficulty of K absorption [31]. Additionally, the dilution effect may occur, as plants prioritize water uptake over nutrient uptake, resulting in decreased K concentrations in plant tissues. Stover K concentration and ear K concentration varied from 8.6 to 12.5 g kg⁻¹ and 5.5 to 8.5 g kg⁻¹, respectively, at this site (Table S1). Stover K concentration in our study is close to that reported by Ciampitti et al. [20], which was 15 g kg⁻¹. However, stover K concentration was lower than the 21.8 g kg⁻¹ reported by Setiyono et al. [28]. Phosphorus and K concentrations of stover and ear were similar among different treatments, even if they were accumulated differently among biomass samples.

The treatment effect was non-significant for P and K concentrations in grains. This may be due to increased grain yield by sufficient fertilization, which resulted in P and K dilution. Adequate nitrogen fertilization enhances plant growth and grain yields. As crops grow and biomass increases, the demand for nutrients rises. The nutrient uptake rate may not match the increased biomass production, causing nutrient dilution in plant tissues, including grains. Consequently, the concentration of certain nutrients like P and K may decrease relative to biomass accumulation, resulting in no significant increase in their concentrations in grains despite sufficient fertilization. This dilution effect could cause similar P and K concentration, even when compared with controls. Bélanger and Richards [32] also observed this effect. Phosphorus concentration decreased with increasing N rates due to dilution, which was consistent with the results of Schlegel and Havlin [33]. Mallarino and Higashi [13] also supported that K concentration in corn grain was not affected by K application. Phosphorus concentrations in grain ranged from 2.20 g kg⁻¹ to 2.93 g kg⁻¹ at EFAW21 and 2.86 g kg⁻¹ to 3.89 g kg⁻¹ at LCB22. While K concentration in grain ranged from 3.77 g kg⁻¹ to 4.31 g kg⁻¹ at EFAW21 and 5.21 g kg⁻¹ to 6.39 g kg⁻¹ at LCB22 (Table S1). The difference between these two sites could be due to the difference of the test weight of grains. The mean test weight for LCB22 site was higher as compared to EFAW21, which could have caused dilution of nutrients in grain. This suggests that nutrients’ dilution occurs in corn grain in high yielding environment. This was also observed by Lollato et al. [34] in wheat. Grain P concentration in our study was consistent with that reported by Setiyono et al. [28], which was 2.6 g kg⁻¹. Other studies reported higher grain P concentration. Ferreira et al. [35] conducted experiments with different cultivars in Brazil and reported P concentration ranged from 2.9 g kg⁻¹ to 5.1 g kg⁻¹ and K concentration ranged from 3.7 g kg⁻¹ to 10.3 g kg⁻¹. Bender et al. [3] reported 3.3 g kg⁻¹ P and 4.4 g kg⁻¹ K in grain. According to Heckman et al. [2], the average grain P concentration was 3.34 g kg⁻¹, and the average grain K concentration was 4.8 g kg⁻¹. These values were calculated from corn cultivars grown in northeastern US states. Ciampitti et al. [20] reported grain P concentration from 3.4 to 4.0 g kg⁻¹ and grain K concentration between 4.9 and 5.1 g kg⁻¹ while Mallarino and Higashi [13] reported grain K concentration of 3.5 g kg⁻¹ in their study. Phosphorus and K concentration from the LCB22 site falls within these ranges reported by different researchers. Even the highest P concentration at the EFAW21 site was below these values while the K concentration was within the published ranges. The reason for this could be discrepancy between yield level and study environment. Almost all those comparable studies were conducted in high-yielding environments, especially in the Midwest of US. Favorable environment conditions, such as availability of irrigation and high soil organic matter, can provide better conditions for diffusion of these nutrients from soil to roots [2].

3.2. Total Plant Biomass Accumulation and NPK Uptake at Early Growth Stage and Maturity

3.2.1. Plant Biomass Accumulation

Analysis of Variance shows that there was no interaction effect of N × P × K on biomass at V6 growth stage at both locations (p ≤ 0.05) (Table S2). Application of K had significant effect at EFAW21 (p ≤ 0.05) as the biomass was reduced by 15.48% with the application of K fertilizer. Only the EFAW21 site did not respond to N rate for biomass accumulation at V6 (p = 0.08). Biomass at EFAW21 ranged from 1138 kg ha⁻¹ to 1665 kg ha⁻¹. At LCB22, biomass ranged from 307.6 kg ha⁻¹ to 1014 kg ha⁻¹. Application of K with N increased biomass seldomly at LCB22. This may be due to the positive response of K to provide plant drought resistance, which is beneficial to rain-fed crops. A study by Martineau et al. [36] reported that K fertilizer enhanced water use efficiency under water stress conditions by enhancing stomatal sensitivity to drought. Hsiao and Lauchli [37] found K played a vital role in making plants more resistant to water deficit.

At crop maturity, the interaction effect of N × P and the main effect of N (p < 0.0001) and K (p = 0.02) were observed for biomass accumulation for EFAW21 while only the main effect of N was observed at the LCB22 site (

Table 3. ANOVA results for total biomass, N-Uptake, P-uptake and K-uptake by whole aboveground plant at R6 growth stages (physiological maturity).

			Biomass	Whole Plant			Grain
Site-Year	Source of Variation	Df		N	P	K	N	P	K
EFAW21	N	2	<0.0001	<0.0001	<0.0001	<0.0001	<0.0001	<0.0001	<0.0001
	P	1	0.31	0.28	0.47	0.54	0.28	0.02	0.06
	N × P	2	0.04	0.01	0.002	0.23	0.01	0.0008	0.01
	K	1	0.02	0.26	0.57	0.003	0.26	0.18	0.11
	N × K	2	0.13	0.69	0.14	0.07	0.69	0.94	0.78
	P × K	1	0.17	0.96	0.59	0.07	0.96	0.14	0.30
	N × P × K	2	0.33	0.08	0.04	0.11	0.08	0.45	0.40
LCB22	N	2	<0.0001	<0.0001	<0.0001	<0.0001	<0.0001	<0.0001	<0.0001
	P	1	0.17	0.02	0.74	0.20	0.54	0.78	0.88
	N × P	2	0.32	0.16	0.40	0.47	0.15	0.22	0.16
	K	1	0.93	0.81	0.60	0.36	0.76	0.59	0.58
	N × K	2	0.64	0.49	0.55	0.65	0.83	0.86	0.86
	P × K	1	0.75	0.24	0.36	0.72	0.93	0.85	0.50
	N × P × K	2	0.20	0.65	0.10	0.50	0.24	0.19	0.21

). Potassium application increased dry biomass by 6.25% at EFAW21. The highest aboveground biomass at EFAW21 was observed where 20 kg ha⁻¹ P was applied with 133 kg ha⁻¹ N, which is similar to the findings of Karlen et al. [38]. The maximum biomass at maturity was found at LCB22 when K was applied with 133 kg ha⁻¹ N. Nitrogen rate increased biomass by 36% and 50% at EFAW21 and 80% and 123% at LCB22 when N was applied at 67 kg N ha⁻¹ and 133 kg N ha⁻¹, respectively, as compared to the control. The role of K in increasing dry biomass can also be explained by the fact that K provides strength to stalk [39] and enhances resistance to plant diseases and insects [40]. These could be the main reasons why the corn plant produced higher biomass when K was applied. However, K did not increase grain yield, which suggests that K was able to enhance stover biomass only. This can also be confirmed through low GKHI (Table S3), which proved that luxury uptake of K occurred.

3.2.2. Total Nitrogen Uptake

There was a significant difference for N uptakes during the early growth stage at both locations attributed to N rates. Applied P or K did not have any significant effect at this stage at either location (Table S2). Nitrogen uptake ranged from 22.2 kg ha⁻¹ to 43.3 kg ha⁻¹ at EFAW21. It almost doubled when 133 kg N ha⁻¹ was applied as compared to the control. At LCB22, N uptake increased more than three folds when the higher rate of N was applied and it ranged from 8.1 kg ha⁻¹ to 31.4 kg ha⁻¹. When P, K or both were applied with N in 2021, N uptake decreased. In contrast, N uptake increased in 2022 when P and K were present with N. Both the reduction and increment were insignificant and inconsistent.

At maturity stage (R6), there was the interaction effect of N × P × K and main effect of N rate at EFAW21 (Table 3). The addition of P, K or both with 133 kg ha⁻¹ N increased the nitrogen uptake by 3 to 13% as compared to 133 kg ha⁻¹ N only. The application of 67 kg N ha⁻¹ resulted in 34.5% less N uptake as compared to 133 kg ha⁻¹ N. The main effect of N and main effect of P was observed at LCB22. As compared to 0-N, the addition of 67 kg N ha⁻¹ increased N uptake by two folds and 133 kg N ha⁻¹ increased N uptake by three folds at LCB22 (Figure 2). The application of P decreased N uptake by 7 kg ha⁻¹, which was not expected. But it is clear from this study that total N uptake at maturity depends greatly on N input. This can be attributed to increased biomass due to higher N rates. Nitrogen uptake at maturity in the fertilized plots was 60% higher at EFAW21 and 43% lower at LCB22 than the findings of Freeman et al. [15]. They computed that total N uptake for irrigated corn in Oklahoma was 108.2, 108.5 and 114.4 kg N ha⁻¹ when N was applied at the rate of 118, 236 and 354 kg ha⁻¹, respectively. Our current mean value was 162 kg N ha⁻¹ at EFAW21 and 68.15 kg N ha⁻¹ when 133 kg ha⁻¹ N was applied alone or in NPK combination (Figure 2). The highest N accumulated across two sites when P was applied with 133 kg ha⁻¹ N, but this value was far less than the findings of Woli et al. [10], Bender et al. [3] and Karlen et al. [38]. This uptake value was even less than that reported by Hanway [4] in 1962 (mean = 201 kg N ha⁻¹).

3.2.3. Total Phosphorus Uptake

The statistical data presented in Table S2 revealed that, at early growth stage, P uptake was not significantly affected by any interaction effect and main effect at EFAW21. There was the main effect of N (p < 0.0001) and the main effect of K (p = 0.03) on total P uptake at LCB22. Phosphorus uptake increased with the increase of N rate. Application of 133 kg N ha⁻¹ resulted in P uptake increase by almost two folds when compared to treatments with 0-N at LCB22. The co-application of K also increased P uptake at early growth stage at LCB22.

At maturity, there was interaction effect of N × P × K, N × P and the main effect of N for P uptake in above ground whole plant at EFAW21 (Table 3). Grain P uptake at this site was affected significantly (p ≤ 0.05) by the interaction of N × P, the main effect of N and the main effect of P. Only the main effect of N was observed for whole plant P uptake as well as grain P uptake at LCB22 (Table 3). Grain P uptake increased by 11.7% with P application at EFAW21, but this effect was not observed at LCB22. The maximum whole plant P uptake was 32.28 kg ha⁻¹, and the maximum grain P uptake was 31.54 kg ha⁻¹ at EFAW21 (Figure 2) while, at LCB22, the highest whole plant P uptake was 16.31 kg ha⁻¹ and the grain P uptake was 17.68 kg ha⁻¹. Increases in total aboveground biomass resulted in more P uptake by whole plant as well as grain. Positive interaction of N × P for P absorption was also noted by Fageria [41]. Phosphorus uptake in stover and grain increased with the increase in N supply, which was also supported by Setiyono et al. [28], Ma et al. [42] and Ciampitti et al. [20]. According to Wilkinson et al. [43], P uptake was increased by N rate because of increase in the root length and the ability of roots to explore and absorb more P. NH₄⁺ ions from N fertilizer compete with other cations, which then increase soil P solubility by releasing P fixed on oxide surfaces of clay minerals. Grain N, P and K uptake increased with N rates linearly, which were similar to the observations of Ciampitti et al. [20] and Hanway [4].

3.2.4. Total Potassium Uptake

During the early growth stage, neither interaction effect nor main effect was observed for K uptake at EFAW21 while, at LCB22, there was the main effect of N and the main effect of K for K uptake. Nitrogen application increased K uptake at early growth stage at LCB22. Application of K also increased K uptake by 20.7% at LCB22 (Table S2).

At maturity, the whole plant K uptake was affected significantly by the main effect of N and main effect of K while grain K uptake was affected by interaction of N × P and main effect of N at EFAW21. At LCB22, both whole plant K uptake and grain K uptake were solely affected by N rate (Table 3). Total K uptake in whole plant and grain was significantly (p ≤ 0.05) greater at 133 kg N ha⁻¹ rate than lower levels of N (Table 3) (Figure 2). However, K concentration in the whole plant and grain was unaffected by varying N application rates. (Table S1). This suggests that the uptake was affected due to increase in dry biomass. In our study, we found that N, P and K contents were controlled by N rates. This finding is consistent with previous studies [27,44,45].

3.3. Nutrient Harvest Index

Nitrogen Harvest Index (NHI), Phosphorus Harvest Index (PHI) and Potassium Harvest Index (KHI) were computed and presented in Table S3. Results showed that NHI increased linearly with N rates while PHI and KHI were similar among treatments at EFAW21 site. At LCB22, all these nutrient harvest indices increased with increased N application rates. In the study of Bender et al. [46], NHI ranged from 0.51 to 0.62; PHI ranged from 0.70–0.82; and KHI ranged from 0.27 to 0.37. Ciampitti et al. [20] documented the mean NHI value of 0.55 to 0.70, the mean PHI value of 0.70 to 0.85 and the KHI as 0.28. While, Setiyono et al. [28] reported the NHI 0.64, the PHI 0.84 and the KHI 0.17. Similar values were reported by Bender et al. [46]. NHI, PHI and KHI, in our study, were higher than what have been reported by published studies at both locations when 133 kg N ha⁻¹ was applied.

3.4. Seasonal Nutrient Uptake Pattern

Nitrogen uptake (mg per plant) was significantly affected by N rate across all growth stages at EFAW21 (Table S4) and LCB22 (Table S5), except for R6 growth stage at EFAW21, where interaction effect of N × P × K was also observed along with the main effect of N. Applied P or K did not have significant effect on seasonal nutrient uptake at any growth stages. The Nitrogen rate significantly increased N uptake at all growth stages at both locations (Figure 3).

Phosphorus uptake was not affected by N rate until maturity at EFAW21 (Table S4). At R6 growth stage, N rate increased P uptake (Figure 4). The application of 67 kg N ha⁻¹ increased P uptake by 14.7% while 133 kg N ha⁻¹ increased P uptake by 54.6% as compared to the control at this site. While at LCB22, P uptake was affected by N rate at V6 and interaction of N × K at VT growth stage. Bennett et al. [47] reported that, when the N uptake of the plant increased, it became physiologically active, which further caused higher uptake of P. A large amount of N compounds are formed in the plant due to a high uptake of N, and some of these compounds contains P while some of the other plant compounds require P, even for their formation. These physiological changes in the plant cause the plant to uptake higher P if available. This explanation was also supported by Cole et al. [48].

Potassium uptake was not significantly affected (p ≤ 0.05) at V6 and R2 growth stage at EFAW21 (Table S4). At VT growth stage, only the main effect of K was observed while, at R6, the main effect of N (p = 0.000) as well as the main effect of K (p = 0.003) was observed at this site. Potassium application increased the K uptake by plants. Potassium concentration in grain was not significantly affected by the K rate, but K concentration in stover was affected by the K rate (Table S1). The application of K increased stover K concentration, which further caused higher K accumulation. At LCB22, K uptake was affected at V6 growth stage only, where K uptake was increased by N rate (Figure 5).

The maximum uptake values of N, P and K per plant at early growth stage were higher at both locations than found by Rosa et al. [49] and Kaiser et al. [50]. To our knowledge all studies have reported uptake pattern on the basis of total nutrient uptake per hectare [3,4,10,38,51]. There was only N rate, which was influencing N, P and K uptake at almost all growth stages at both sites; therefore, the only effect of N rate was shown in seasonal nutrients uptake curves (Figure 3, Figure 4 and Figure 5).

3.5. Nutrient Accumulation Timing

The timing of nutrient accumulation is presented on nutrient contribution during particular growth period (Figure 6, Figure 7 and Figure 8). There was no significant difference of applied P or K with timing of nutrients uptake at any growth stage. There was a difference of N rate at some growth stages at EFAW21. This difference was due to the difference in total uptake of that particular nutrient. When 133 kg N ha⁻¹ was supplied, 39% of total N uptake was done by VT stage whereas 43.2% and 47.4% of total N uptake was observed when N input decreased to 67 kg N ha⁻¹ and 0 kg N ha⁻¹, respectively, at EFAW21. About half of N was taken in well fertilized plots between VT and R2 stage, which was R1 growth stage. About 14% to 25% of P uptake was done till VT growth stage and 60 to 80% of total K uptake was done at VT stage at EFAW21. Nitrogen and P uptake follows a different pattern while K follows the same pattern reported by Karlen et al. [38], Woli et al. [51], Ciampitti et al. [20] and Bender et al. [3].

Karlen et al. [38] reported that 65% N, 46% P and 88% uptake were done by R1 growth stage. According to Hanway et al. [4], 65% N, 50% and 90% of K uptake occurred by R1 growth stage. Ciampitti et al. [20] reported that 70% N, 49% P and 122% K uptake was done by R1. According to Bender et al. [3], 67% N, 46% P and 66% K uptake was completed by R1 growth stage. The variability among treatments for uptake percentage depends on total nutrient uptake at maturity and amount of available nutrients. Nutrient uptake pattern differences from other studies could be explained by environmental difference, which includes differences in weather, soil-type, hybrids, irrigation status, agronomic management and other factors. Further field research is necessary considering these varied results.

4. Conclusions

It is important to adapt sound nutrient management practices, which balance inputs and outputs of nutrients. To achieve this goal, nutrients removed by the crop should be replaced. Local nutrient removed values are an important part of making an effective nutrient management plan and sustainable agriculture. In this study we observed that nutrient uptake and concentration in plants depends greatly on environment and management. Nitrogen concentration in the whole plant at early growth stage was linearly affected by N application. A similar trend in N concentration was observed for all plant components at maturity. Phosphorus and K concentration was not affected by NPK rate. Total N, P and K uptake was primarily driven by N rate at all growth stages. Almost all K uptake occurred in the vegetative stage of the corn plant. The uptake pattern in our rain-fed corn study was very different from published studies because of the difference in environment and management practices. Almost all those studies were conducted in high-yielding environments with more rain or better irrigation facilities. Removal values in our study question the usefulness of average values from those empirical studies in rainfed environment. More studies are required on the variety of soils in different environments to compute the nutrient uptake and removal values for rain-fed corn. Additional site years would provide useful nutrient uptake and removal data that will be valuable for making sound nutrient management plans.