Evaluating Species-Specific Replenishment Solution Effects on Plant Growth and Root Zone Nutrients with Hydroponic Arugula (Eruca sativa L.) and Basil (Ocimum basilicum L.)

Abstract

Managing nutrients in recirculating solutions can be challenging in hydroponic production, and poor management practices can rapidly result in root zone nutrient imbalances in reduced yield. Using mass balance principles to formulate hydroponic replenishment solutions is a proposed strategy to reduce nutrient imbalances and the need to periodically replace the hydroponic solution. Objectives were to (1) formulate species-specific nutrient replenishment solutions for arugula (Eruca sativa L.) and basil (Ocimum basilicum L.) using mass balance principles and (2) evaluate the effects of using these replenishment solutions on plant growth and root zone nutrients over time. In the first experiment, arugula and basil tissues were analyzed over 42 d for macronutrient concentrations which were used to custom formulate species-specific replenishment solutions for both species. In the second experiment, nutrients were resupplied for hydroponic arugula and basil over time using either the species-specific or a standard commercial hydroponic replenishment solution. Species-specific replenishment solutions resulted in decreased solution EC and concentrations of most macronutrients over time for both species. In contrast, replenishment with the standard hydroponic solution resulted in increased solution EC and concentrations of calcium, magnesium, and sulfate and decreased nitrogen, phosphorus, and potassium. Replenishment treatment had no effect on yield for arugula and basil; however, results suggested root zone imbalances may have still occurred eventually for both treatments and species. Species-specific replenishment solutions would be most effective in scenarios where solutions could be formulated for a specific crop, mixed using high-quality irrigation water, and where environmental conditions are controlled and stable, such as indoor farming systems. This study highlighted several practical challenges and considerations regarding the formulation of hydroponic solutions using mass balance.

1. Introduction

Managing nutrients in recirculating solutions can be a challenging aspect of hydroponic production [1,2], and poor nutrient management practices can rapidly result in root zone imbalances, nutritional disorders, and reduced yield [3,4,5]. Root zone nutrient imbalances are caused by the excessive accumulation or depletion of certain nutrients in solution [1], often from replenishing the hydroponic solution over time with nutrient concentrations above or below the uptake of plants, respectively [6,7,8]. To avoid nutritional problems and yield losses, a common commercial practice is to periodically discharge and replace the hydroponic solution [1,2,3,4]; however, this is wasteful and costly with negative environmental impacts. Alternatively, a mass balance approach to replenishing nutrients in recirculating hydroponic solutions is recommended as a strategy to minimize the occurrence of root zone nutrient imbalances and the need to discharge and replace the solution [1,2,4,9].

Managing nutrients using mass balance principles implies the supply and/or replenishment of nutrients in recirculating solutions are balanced with the rate of nutrient uptake by plants [1,4,9]. Since nutrients are dissolved in the hydroponic solution, the goal is to match the supplied concentration of individual nutrients in the solution with plant uptake concentrations. The result is, therefore, a scenario of relative steady-state nutrition where the supply of nutrients and water replaces the nutrients/water absorbed by the crop, minimizing the excess accumulation or depletion of nutrients from the solution [4].

In commercial hydroponic production, a common and simple nutrient management practice is to maintain a constant target electrical conductivity (EC) level in solution over time using automated fertigation injection and dosing equipment [9,10,11,12,13,14,15,16]. Although a constant EC indicates a constant concentration of total soluble salts, individual nutrients can still accumulate or become depleted [2,12]. For example, Houston et al. [2] and Miller et al. [12] found replenishing recirculating solutions to maintain a constant EC eventually resulted in nitrogen depletion and reduced growth for hydroponic lettuce (Lactuca sativa). In both studies, divalent nutrient ions such as calcium, magnesium, and sulfate-sulfur were resupplied in excess of plant demand, accumulated and became the primary contributors to the target solution EC, and caused the insufficient replenishment of nitrogen and other nutrients over time.

Formulating hydroponic solutions using mass balance principles requires prior knowledge of nutrient and water uptake for a particular crop [1,2,4,9]. Nutrient uptake requirements can be determined by analyzing plant tissues for individual nutrient concentrations [1,2,4] or by measuring the depletion of nutrients from solution over time [9]. Plant water use can be determined as the water volume lost from evapotranspiration (i.e., the combination of evaporation and transpiration) or as water-use efficiency (WUE), which is the evapotranspiration volume per unit of plant dry mass gain [1,4]. With knowledge of nutrient and water requirements, it may be possible to formulate species-specific replenishment solutions which accurately resupply nutrients and water absorbed by the crop to satisfy plant growth without leading to nutrient accumulation or toxicity/deficiency [4,9].

Formulating species-specific replenishment solutions using mass balance may be a practical strategy for managing nutrients in commercial hydroponic production. It is already common practice for growers to periodically measure nutrients in plant tissues or solutions by submitting samples to commercial horticultural testing laboratories [13,17], and water use can be measured over time using in-line flow meters. This approach would also be compatible with the strategy of maintaining a target solution EC, and excessive nutrient accumulation or depletion would be less likely since nutrient concentrations and ratios in the replenishment solution are similar to those taken up by plants [4]. Although maintaining a target EC is common in commercial practice, Langenfeld et al. [4] stated this was not necessary with species-specific replenishment and reported that relatively low EC levels could be tolerated without a reduction in yield.

Leafy greens and culinary herbs are high-value specialty crops for hydroponics [18,19,20,21,22,23,24,25,26] and also serve as good model crops for nutrient management research because of the relatively short cropping cycles and harvesting of only vegetative tissues. For this study, arugula (Eruca sativa L.) was selected because of its popularity among consumers [18,20], unique peppery and mustard-like flavor [24,25,26], and relatively high requirements for anion nutrients, particularly sulfate-sulfur [27,28]. Basil (Ocimum basilicum L.) was also selected because of consumer preference and the relatively high requirement for magnesium [18,21,22,29,30], which can accumulate in recirculating solutions.

The objective of this study was to formulate species-specific hydroponic replenishment solutions for arugula and basil and evaluate the effects of using this mass balance approach to resupplying nutrients on plant growth and root zone nutrients over time. We hypothesized nutrient replenishment with the species-specific solutions would result in less macronutrient accumulation and lower root zone nutrients over time compared to replenishment with the standard commercial solution.

2. Materials and Methods

2.1. Experiment #1: Evaluation of Tissue Macronutrients for Hydroponic Arugula and Basil

Experimental site. Arugula and basil were grown hydroponically for 42 d in a polycarbonate controlled-environment greenhouse located at the University of Arkansas in Fayetteville, AR (36.0764° N, 94.1608° W). The average daily temperature during the experiment was (mean ± standard deviation) 25.6 ± 1.6 °C and daily light integral (DLI) was 15.8 ± 6.9 mol·m⁻²·d⁻¹ of photosynthetically active radiation. Plants were grown in small research-scale deep water culture systems described by Dickson et al. [27], and plant tissues from independent systems for both species were destructively sampled and analyzed over time (see below).

Hydroponic setup and growing conditions. Standard arugula (Eruca sativa L.) and pelleted ‘Compact Genovese’ basil (Ocimum basilicum L.) seed (Johnny’s Selected Seeds, Waterville, ME, USA) were sown in 162-cell rockwool sheets (A/O sheets, Grodan, The Netherlands) at 1 seed or pellet per cell and germinated in the greenhouse. Rockwool sheets were sub-irrigated with a 17N-1.3P-14.1K (JR Peters; Allenstown, PA, USA) complete water-soluble fertilizer solution at 150 mg∙L⁻¹ N mixed in de-ionized water. Rockwool sheets were re-irrigated as needed with the same fertilizer. Pelleted basil seeds contained multiple seeds per pellet and were thinned to one seedling per cell after the emergence of true leaves. Seedlings of each species were then transplanted into hydroponic culture vessels.

Small research-scale hydroponic culture vessels were designed following methods described by Dickson et al. [27]. Each hydroponic culture vessel was a 4.5 L white plastic container with a snap-on plastic lid. Each seedling stem was wrapped with a neoprene collar (5 cm diameter), which fit into a black plastic hydroponic net pot (5 cm diameter). Net pots were supported in circular holes cut into the container lid, which allowed roots to be submerged in the nutrient solution. The neoprene collar reduced evaporation of the nutrient solution without constricting plant stems. A plastic air tube was inserted into the nutrient solution via a hole in the bucket lid, connected to an aquarium tank air pump, which constantly aerated the nutrient solution. Culture vessels were wrapped in aluminum foil to reduce light transmission to the nutrient solution, prevent algae growth, and help stabilize the solution temperature. Each culture vessel initially contained three plants of either arugula or compact basil. Each culture vessel held a 4.0 L nutrient solution. After transplanting into hydroponic culture vessels, all plants received nutrient solution with 100 mg∙L⁻¹ N. Vessels were placed on a greenhouse bench on 30.7 cm center spacing (30 plants per m²). See Supplemental Materials (Figure S1) for an example of the system setup.

Nutrient solution preparation and management. The hydroponic solution was formulated by mixing equal parts of commercial-grade calcium nitrate (Haifa, Matam-Haifa, Israel) and a 5N-4.8P-21.6K (JR Peters, Allentown, PA, USA) water-soluble fertilizer in de-ionized water. Nutrient concentrations from the fertilizer consisted of (in mg∙L⁻¹) 100 nitrogen (N), 25 phosphorus (P), 105 potassium (K), 93 calcium (Ca), 31 magnesium (Mg), 65 sulfate-sulfur (SO₄-S), 1.5 iron (Fe), 0.2 manganese (Mn), 0.1 copper (Cu), 0.5 molybdenum (Mo), 0.1 zinc (Zn), and 0.2 boron (B). The initial solution pH and electrical conductivity (EC) of the hydroponic solution were pH 6.0 ± 0.05 and 1.32 mS∙cm⁻¹, respectively. Solution pH was monitored every 2–3 d in each culture vessel and was maintained between pH 5.5 and 6.0 by titration with hydrochloric acid (HCl) and sodium hydroxide (NaOH) at 0.1 N. The solution level was also topped off to approximately 4 L using de-ionized water just prior to solution pH adjustment. Every 7 d, the solution in each culture vessel was replaced with 4 L of fresh solution at 100 mg∙L⁻¹ N.

Experimental design. The experiment was a 2 × 5 factorial using a randomized complete block design with four blocks (see Supplemental Materials Figure S2 for the layout). Plant species (arugula, basil) was the first factor, and harvest day after transplanting seedlings (14, 18, 21, 28, and 42 d) was the second factor. Each hydroponic culture vessel containing three seedlings was considered one experimental unit. One experimental unit per treatment was removed from each block for destructive sampling on each harvest day, after which blocks were consolidated to maintain a uniform spacing between culture vessels. The experiment started with the transfer of seedlings into the hydroponic culture vessels on 24 March 2020, and the final data collection ended on 5 May 2020.

Plant tissue analysis. Plants were allowed to grow for 14 d after seedlings were transferred into the hydroponic culture vessels to obtain sufficient size for data collection, and plants were destructively sampled for data collection at 14, 18, 21, 28, and 42 d after transplant. The arugula and basil transitioned from vegetative to reproductive growth and began to produce flowers between 28 and 42 d. Harvested shoot and root tissues were washed in a dilute acid solution (0.05% HCl), immediately rinsed with de-ionized water, and oven dried at 60 °C for 72 h. Dried shoot and root tissues were combined and analyzed as whole-plant samples for macronutrient concentrations, with total N measured using persulfate digestion [6] and the remaining macronutrients measured using inductively-coupled plasma atomic emission spectrophotometry (ICP-AES) at the Fayetteville Agricultural Diagnostic Lab (University of Arkansas, Fayetteville, AR, USA).

Statistical analysis. Plant species, harvest day, and the interaction of plant species with harvest day were treated as fixed effects, and the block was treated as a random effect. Analysis of variance (ANOVA) using PROC GLIMMIX (SAS 9.4: SAS Institute, Cary, NC, USA) was used to evaluate plant species and harvest day main and interaction effects on tissue macronutrient concentrations, with means separation using Tukey’s honestly significant difference (HSD) at the α = 0.05 significance level.

2.2. Experiment #2: Hydroponic Replenishment Solution Effects with Arugula and Basil

Experimental site. The experiment was conducted in a polycarbonate controlled-environment greenhouse located at the University of Arkansas in Fayetteville, AR, USA (36.0764° N, 94.1608° W). The average daily temperature during the experiment was (mean ± standard deviation) 25.6 ± 1.6 °C and DLI was 9.38 ± 2.4 mol·m⁻²·d⁻¹ of photosynthetically active radiation. Arugula and basil seedlings were grown and transferred into hydroponic culture using the same methods and culture vessels described in Expt. #1, with the exception that two seedlings (instead of three as in Expt. #1 in Section 2.1, 20 plants per m²) were transferred to each culture vessel. Plants were allowed to establish and acclimate to hydroponic conditions for 14 d prior to the start of the experiment in the same 100 mg∙L⁻¹ N solution described in Expt. #1.

Nutrient solution preparation and management. The experiment began with the complete replacement of solution in each culture vessel with fresh fertilizer solution. All culture vessels received the same initial hydroponic fertilizer solution, which consisted of mixing equal parts of the same commercial-grade calcium nitrate (Haifa, Matam-Haifa, Israel) and 5N-4.8P-21.6K (JR Peters, Allentown, PA, USA) water-soluble fertilizer used in Expt. #1 in de-ionized water at 150 mg∙L⁻¹ N. Initial solution pH and electrical conductivity (EC) were pH 6.0 ± 0.05 and 1.98 mS∙cm⁻¹, respectively. Solution pH was monitored every 2–3 d in each culture vessel and maintained between pH 5.5 and 6.0 using HCl and NaOH at 0.1 N.

Nutrient replenishment consisted of replacing the hydroponic solution absorbed by plants by refilling the solution in each culture vessel to 4 L every 2–3 d, where the type of replenishment solution pertained to one of two treatments. The first replenishment solution treatment consisted of refilling with the initial fertilizer solution at 150 mg∙L⁻¹ N as a standard control treatment. The second replenishment solution treatment consisted of refilling with a specifies-specific solution at 150 mg∙L⁻¹ N custom formulated separately for arugula and basil using the mass balance approach described by Langenfeld et al. [4]. The Langenfeld et al. [4] method formulates replenishment solutions by calculating the optimal nutrient concentrations needed to replace both nutrients and water absorbed from the hydroponic solution by plants. The method involves dividing the expected nutrient concentrations in dried shoot tissue (in μg∙g⁻¹) by the expected WUE (in mL of water per g of total dry mass) for the crop. The nutrient ratios and concentrations in the replenishment solution are therefore determined by the ratio of nutrients in dried tissues and WUE, respectively.

The macronutrient ratios needed in the species-specific replenishment solutions were determined using the average tissue macronutrient concentrations for arugula and basil found in Expt. #1 (see

Table 1. Tissue concentrations (in % of dry mass) of nitrogen (N), phosphorus (P), potassium (K), calcium (Ca), magnesium (Mg), and sulfur (S) for hydroponically-grown arugula and basil harvested at 14, 18, 21, 28, and 42 d as well as averaged over the 42 d period.

	N		P		K		Ca		Mg		S
	Arugula	Basil	Arugula	Basil	Arugula	Basil	Arugula	Basil	Arugula	Basil	Arugula	Basil
Harvest Day	%
14	5.57 a	3.72 d	0.69 a	0.76 c	5.69 a	5.90 c	2.90 a	2.13 b	0.51 b	0.50 ab	1.02 ab	0.28 b
18	5.13 a	4.86 b	0.61 ab	1.18 a	4.62 a	5.89 c	2.81 a	2.54 a	0.54 ab	0.54 ab	0.84 c	0.37 a
21	5.11 a	5.31 a	0.65 ab	1.09 ab	4.70 a	6.62 b	2.97 a	2.32 ab	0.62 ab	0.55 a	0.94 abc	0.34 ab
28	5.83 a	5.68 a	0.75 a	1.19 a	5.75 a	7.10 a	3.06 a	2.20 b	0.65 a	0.52 ab	1.12 a	0.40 a
42	3.14 b	4.34 c	0.47 b	0.88 bc	3.28 b	6.50 b	2.11 b	1.64 c	0.57 ab	0.47 b	0.95 abc	0.37 a
Std. Error	±0.59	±0.10	±0.06	±0.10	±0.55	±0.11	±0.21	±0.08	±0.03	±0.01	±0.04	±0.02
Arugula	4.95 a		0.64 b		4.81 b		2.77 a		0.57 a		0.98 a
Basil	4.78 a		1.02 a		6.40 a		2.16 b		0.51 b		0.35 b
Std. Error	±0.12		±0.07		±0.31		±0.11		±0.01		±0.13
Species	NS		***		***		***		***		***
Harvest Day	***		***		***		***		**		**
Species*Day	***		**		***		**		*		**

Data are least-square means of four and 20 replicates per harvest day and averaged over 42 d for each species, respectively. Means separated using Tukey’s honestly significant difference (HSD) at α = 0.05. Means followed by the same letter were determined not to be statistically different. NS, *, **, *** refer to non-significance or significance at p < 0.05, p < 0.01, and p < 0.0001, respectively.

arugula and basil main effects on tissue nutrients). At the time of this study, there were no published reports of WUE for hydroponic arugula and basil, and WUE was not measured in Expt. #1. Nitrogen concentration in the replenishment solutions was therefore set to 150 mg∙L⁻¹ N following recommendations for hydroponic leafy greens and herbs by Mattson and Peters [31]. Species-specific solutions were formulated using commercial-grade potassium nitrate, calcium nitrate, mono-potassium phosphate, potassium sulfate, magnesium sulfate, magnesium nitrate, and nitric acid mixed in de-ionized water.

It was not possible to achieve the exact target concentrations needed for all macronutrients (see Results and Discussion,

Table 2. Initial electrical conductivity (EC) and target concentrations (mg·L⁻¹) of total nitrogen (N), nitrate nitrogen (NO₃-N), ammonium nitrogen (NH₄-N), phosphorus (P), potassium (K), calcium (Ca), magnesium (Mg), and sulfate-sulfur (SO₄-S) for the standardized and species-specific replenishment solution treatments for arugula and basil. Micronutrient concentrations for standard and species-specific solutions were (mg·L⁻¹) 2.28 iron, 0.34 manganese, 0.12 zinc, 0.34 boron, 0.12 copper, and 0.08 molybdenum. Macronutrient ratios in the replenishment solutions in Experiment #2 and plant tissues in Experiment #1.

Replenishment Solution	EC (mS∙cm−1)	Total N	NO3-N	NH4-N	P	K	Ca	Mg	SO4-S
Standard x	1.54	150	141	9	38	157	140	46	54
Arugula y	0.86	150	146	4	19	141	67	16	35
Basil	1.00	150	146	4	31	194	67	17	35
Nutrient ratios in replenishment solution					N:P	N:K	N:Ca	N:Mg	N:S
Standard					3.9	1.0	1.1	3.3	2.8
Arugula					7.8	1.1	2.2	9.4	4.3
Basil					4.8	0.8	2.2	8.7	4.3
Nutrient ratios in plant tissues z					N:P	N:K	N:Ca	N:Mg	N:S
Arugula					7.8	1.0	1.8	8.6	5.1
Basil					4.7	0.7	2.2	9.4	13.6

^x Formulated using equal parts of a commercial 5N-4.8P-21.6K water-soluble fertilizer and calcium nitrate (15.5% N, 19% Ca). ^y Arugula and basil solutions formulated using individual commercial-grade fertilizer salts and nitric acid (see Materials and Methods). Micronutrient concentrations were the same as those for the standard solution formulated with iron-EDTA, manganese-EDTA, copper-EDTA, zinc-EDTA, boric acid, and sodium molybdate. ^z Average macronutrient ratios in shoot tissues harvested over a 42 d period during Experiment #1.

), and therefore species-specific solutions were formulated to minimize the differences between target and actual macronutrient concentrations. Micronutrient concentrations in species-specific refill solutions were the same as those in the standard refill solution (see Table 2) and were formulated using Fe-EDTA, Mn-EDTA, Cu-EDTA, Zn-EDTA, boric acid, and sodium molybdate (JR Peters, Inc., Allentown, PA, USA).

Experimental design. The experiment was a 2 × 2 factorial arranged using a randomized complete block design with three blocks (see Supplemental Materials Figure S3 for the layout). Plant species (arugula, basil) was the first factor, and replenishment solution (standard or species-specific replenishment) was the second factor. Each culture vessel was considered one experimental unit and treatment replicate, and there were three replicate culture vessels for each species-replenishment solution treatment combination. The experiment started with the initiation of the replenishment strategy treatments on 20 October 2020 and ended with final data collection on 10 November 2020.

Data collection and analysis. Initial data collected at the start of the experiment consisted of destructively sampling three extra culture vessels per species not used in experimentation. Final data were collected for each treatment replicate 21 d after the start of the experiment (35 d after seedling transfer into culture vessels). At final data collection, flower buds were emerging and visible in arugula and basil. Initial and final data included shoot and root fresh and dry mass, macronutrient concentrations in plant tissues, macronutrient concentrations in solution including nitrogen forms, solution pH and EC, and total volume of solution replenished during the experiment.

After weighing fresh root and shoot tissues harvested per culture vessel for fresh mass determination, tissues were then washed in 0.05% HCl, rinsed with de-ionized water, and oven dried at 60 °C for 72 h for dry mass determination. Dried tissues of whole-plant samples were ground to a fine powder and analyzed for N concentration using persulfate digestion [32], and the remaining macronutrients using inductively-coupled plasma atomic emission spectrophotometry (ICP-AES) at the Fayetteville Agricultural Diagnostic Lab (University of Arkansas, Fayetteville, AR, USA). Solution samples collected from each culture vessel were measured for solution pH and EC using a benchtop pH/EC meter (Orion Versa Star Pro, Thermo Fisher Scientific, Waltham, MA, USA), NH₄-N and NO₃-N using semi-automated and automated colorimetry, and remaining macronutrients using inductively-coupled plasma atomic emission spectrophotometry (ICP-AES). Change in solution EC and nutrient levels over time was calculated by subtracting final solution EC and macronutrient concentrations at 21 d from the initial values supplied in the starting hydroponic solution (standard hydroponic solution) at 0 d for each treatment replicate. One-sample t-tests (α = 0.05) were then used to determine if the final solution EC and macronutrient concentrations differed from the starting values.

Plant growth was measured by subtracting the average initial shoot and root mass from the final shoot and root mass per plant for each treatment replicate. The total volume of replenished solution was measured per replicate by summing the volumes of solution needed to top off each culture vessel during the experiment, including just before final data collection at 21 d, and was evaluated on a per-plant basis. Nutrient concentrations in the replenished solutions were determined using the methods previously described. Nutrient concentrations in the replenishment solutions were multiplied by the volume of replenished solution per vessel and were evaluated as nutrients replenished per plant per treatment replicate.

Statistical analysis. Plant species and replenishment solution were treated as fixed effects, and the block was treated as a random effect. Analyses were conducted by plant species since arugula and basil each had an independently formulated species-specific replenishment solution. Analysis of variance (ANOVA) with PROC GLIMMIX (SAS 9.4: SAS Institute, Carry, NC, USA) was used to evaluate treatment effects on fresh and dry mass gain per plant, final solution pH and EC, the final solution and tissue macronutrients, the mass of individual macronutrients replenished per plant, and volume of replenished solution per plant. At the end of the experiment, certain macronutrients in solution for the species-specific treatments were measured at low concentrations near the minimum detection limit. These data were studentized in the statistical model by standardizing the residual error by the standard deviation for each species to achieve homogenous treatment variances for ANOVA. Means separation used Tukey’s HSD at the α = 0.05 significance level.

3. Results and Discussion

3.1. Experiment #1: Evaluation of Tissue Macronutrients for Hydroponic Arugula and Basil

Tissue macronutrient percentages differed between harvest dates for arugula and basil in Expt. #1 (Section 2.1, Table 1). For both species, macronutrient percentages tended to increase from 14 to 28 d, followed by a decrease from 28 to 42 d (Table 1). The decrease in macronutrients after 28 d may have resulted from increased growth rates for both species later in the experiment, which can dilute nutrient concentrations in tissues. In general, tissue macronutrients in Table 1 remained within or slightly above the sufficiency ranges reported by Bryson et al. [17] for arugula (2.86–3.97% N, 0.61–0.72% P, 4.80–5.60% K, 2.40–2.46% Ca, 0.28–0.29% Mg, 0.52–0.55% S) and basil (4.0–6.0% N, 0.62–1.00% P, 1.55–2.05% K, 1.25–2.00% Ca, 0.60–1.0% Mg, 0.2–0.6% S). Overall, species differed in percentages of all macronutrients except N, for which both arugula and basil had similar tissue N (Table 1).

3.2. Experiment #2: Hydroponic Replenishment Solution Effects with Arugula and Basil

The standard fertilizer and species-specific replenishment solutions used in Expt. #2 (Section 2.2) differed in resupplied macronutrient concentrations and EC (Table 2). All solutions were formulated with 6% or less of total N as NH₄-N, with the remainder as NO₃-N, which is common for hydroponic solutions used in commercial production [3,4,9,31]. Species-specific replenishment solutions had a lower EC and resupplied lower concentrations of P, Ca, Mg, and SO₄-S compared to the standard fertilizer replenishment solution. Species-specific solutions for arugula and basil resupplied lower and greater concentrations of K, respectively, when compared to the standard fertilizer replenishment solution (Table 2).

Limitations in achieving target nutrient concentrations. Species-specific replenishment solutions in Expt. #2 were formulated separately for arugula and basil using the tissue results from Expt. #1 (Table 1), with the intent that macronutrient ratios in the solution would match the ratios in plant tissues for each crop, following the recommendations of Langenfeld et al. [4]. However, the fertilizer salts used to formulate these solutions limited the ability to achieve target solution ratios for certain macronutrients, as shown in Table 2, with ratios standardized to N.

For example, the N:Ca ratio in arugula tissue from Expt. #1 was 1.8 (Table 2), which resulted in a target Ca concentration of 83 mg∙L⁻¹. However, the N:Ca ratio in the arugula replenishment solution was greater (N:Ca ratio of 2.2, Table 2) because calcium nitrate was used to formulate the solution and solutions were standardized to 150 mg∙L⁻¹ N, therefore limiting Ca to only 67 mg∙L⁻¹. Similarly, the N:S ratios in arugula and basil tissues were 5.1 and 13.6, respectively (Table 2), resulting in target concentrations of 29 and 11 mg∙L⁻¹ SO₄-S for the species-specific solutions. However, it was not possible to achieve <35 mg∙L⁻¹ SO₄-S using magnesium sulfate without compromising the Mg concentration. Macronutrient ratios in the standard replenishment solution tended to be lower compared to shoot tissues (Table 2).

Species-specific solutions were formulated using de-ionized water and fertilizer salts common in commercial hydroponics [9]. The use of chloride-containing fertilizer salts, such as calcium chloride and magnesium chloride, may have made it possible to achieve the target macronutrient ratios in Table 2. However, these salts are less desirable and less frequently used in commercial hydroponic production because they contribute an abundance of chloride [13,31], which can rapidly accumulate to toxic levels. Irrigation water high in dissolved salts, such as Ca and Mg, may also make it impossible to achieve lower nutrient concentrations without water treatment or switching to a higher-quality water source [9,33]. In addition, irrigation water high in alkalinity may require a mineral acid injection to neutralize bicarbonates and lower pH, which also contributes nutrients such as NO₃-N, P, or SO₄-S, depending on the acid type [33]. These issues highlight practical challenges in formulating hydroponic solutions and achieving target nutrient concentrations and ratios.

Replenishment treatment effects on plant growth, solution nutrients, and EC. In Expt. #2, the total plant fresh and dry mass at the end of the experiment was not affected by replenishment treatment for either arugula or basil. Total fresh mass gain per plant was 304.4 and 271.7 g for arugula and basil, respectively. Total dry mass gain per plant was 29.0 and 30.0 g for arugula and basil, respectively. Similarly, replenishment treatment had no effect on the volume of solution replenished per plant for arugula or basil (

Table 3. Volume of solution replenished per plant and mass of total nitrogen (N), nitrate nitrogen (NO₃-N), ammonium nitrogen (NH₄-N), phosphorus (P), potassium (K), calcium (Ca), magnesium (Mg), and sulfate-sulfur (SO₄-S) replenished per plant (in mg) from the replenishment solution at 21 d for arugula and basil.

Replenishment Solution	Replenished Solution (L)	Macronutrients Supplied in Replenishment Solution Per Plant (in mg)
		Total N	NO3-N	NH4-N	P	K	Ca	Mg	SO4-S
Arugula
Standard	7.0	913 a	859 a	54 a	232 a	956 a	851 a	279 a	326 a
Species-specific	7.7	1004 a	977 a	28 b	129 b	944 a	451 b	107 b	237 b
	NS	NS	NS	**	***	NS	***	***	***
Std. Error	±0.2	±34	±36	±4	±18	±12	±28	±17	±11
Basil
Standard	5.5	717 a	676 a	42 a	183 a	751 a	669 a	220 a	256 a
Species-specific	4.5	587 a	570 a	17 b	122 b	761 a	264 b	68 b	138 b
	NS	NS	NS	**	*	NS	***	***	***
Std. Error	±0.2	±30	±28	±4	±11	±8	±26	±16	±15

Data are least-square means of three replicates, with means separated using Tukey’s honestly significant difference (HSD) at α = 0.05. NS, *, **, *** refer to non-significance or significance at p < 0.05, p < 0.01, and p < 0.0001, respectively. Means followed by the same letter were determined not to be statistically different.

), which averaged 7.4 L for arugula and 5.0 L for basil.

Replenishment treatment influenced the mass of certain macronutrients resupplied per plant during replenishment (Table 3). For arugula, the species-specific replenishment solution resupplied lower amounts of P (p = 0.0025), Ca (p ≤ 0.0001), Mg (p ≤ 0.0001), and SO₄-S (p ≤ 0.0001) per plant compared to the standard replenishment solution (Table 3). Similar trends were observed for basil (Table 3), where lower amounts of P (p = 0.0398), Ca (p ≤ 0.0001), Mg (p ≤ 0.0001), and SO₄-S (p ≤ 0.0001) were resupplied with the species-specific solution. The resupply of K was not affected by replenishment treatment for either species (Table 3). Replenishment solution treatment had no effect on supplied N for arugula or basil (Table 3), likely because treatments had little influence on plant water use, and replenishment solutions were standardized to 150 mg∙L⁻¹ N.

The species-specific replenishment solution for arugula resulted in a lower final solution EC (p ≤ 0.0001) and lower concentrations of Ca (p ≤ 0.0001), Mg (p ≤ 0.0001), and SO₄-S (p ≤ 0.0001) compared to the standard replenishment solution at 21 d (

Table 4. Electrical conductivity (EC) and concentration (mg∙L⁻¹) of nitrate-nitrogen (NO₃-N), phosphorus (P), potassium (K), calcium (Ca), magnesium (Mg), and sulfate (S) measured in solution for the standardized and species-specific replenishment solution treatments at 21 d.

Replenishment Solution	EC (mS∙cm−1)	NO3-N	P	K	Ca	Mg	SO4-S
Arugula
Standard	2.75 a	1.9	41.2 a	10.4 a	258.3 a	219.3 a	98.7 a
Species-specific	0.69 b	N/A	41.9 a	52.2 a	31.0 b	38.3 b	4.5 b
	***		NS	NS	***	***	***
Std. Error	±0.28		±4.1	±39.3	±6.2	±5.4	±0.8
Basil
Standard	2.71 a	105.5 a	27.7 a	1.4 a	426.7 a	144.3 a	202.3 a
Species-specific	1.19 b	58.5 b	16.7 a	49.9 a	97.8 b	19.7 b	42.1 b
	***	*	NS	NS	***	***	***
Std. Error	±0.18	±6.6	±2.8	±41.9	±17.0	±3.2	±4.9

Data are least-square means of three replicates, with means separated using Tukey’s honestly significant difference (HSD) at α = 0.05. NS, *, *** refer to non-significance or significance at p < 0.05 and p < 0.0001, respectively. Means followed by the same letter were determined not to be statistically different.

). For arugula, NO₃-N concentrations were near zero with both replenishment treatments (Table 4) and below the detectable limit with the species-specific solution, indicating depletion of N from the solution. Final solution EC (p ≤ 0.0001) and concentrations of NO₃-N (p = 0.0134), Ca (p ≤ 0.0001), Mg (p ≤ 0.0001), and SO₄-S (p ≤ 0.0001) were also lowest for basil with species-specific replenishment (Table 4). Replenishment treatment had no effect on final P concentrations for arugula or basil (p = 0.9152 and p = 0.7961 for arugula and basil, respectively, Table 4). Concentrations of NH₄-N were below detectable limits, and therefore, these data were not shown in Table 4.

Final solution K concentrations appeared to differ substantially between standard and species-specific replenishment solutions for both species (Table 4). However, there were no statistical differences (p = 0.3618 and p = 0.4957 for arugula and basil, respectively), resulting from variability in K concentrations between treatment replicates, some of which were near zero. Variability in final solution K concentrations may have resulted from differences in plant uptake of K between culture vessels or from variability in measuring solution K using ICP-AES, as was reported by Bugbee [1].

Table 5. Change in solution electrical conductivity (EC) and concentrations (in mg∙L⁻¹) of total nitrogen (N), nitrate nitrogen (NO₃-N), ammonium nitrogen (NH₄-N), phosphorus (P), potassium (K), calcium (Ca), magnesium (Mg), and sulfate-sulfur (SO₄-S) from initial concentrations in the standard hydroponic solution at 21 d for arugula and basil replenished with standard and species-specific replenishment solutions. Positive and negative values represent an increase or decrease, respectively, in EC or concentration.

Replenishment Solution	EC (mS∙cm−1)	Total N	NO3-N	NH4-N	P	K	Ca	Mg	SO4-S
Arugula
Standard	1.21	−148.1	−139.1	−9.0	3.2	−146.6	118.3	173.3	44.7
	***	***	***	***	NS	*	***	***	***
Std. Error	±0.14	±1.0	±4.5	±0.01	±2.4	±35.5	±6.9	±18.2	±4.2
Species-specific	−0.85	−150.0	−141.0	−9.0	3.9	−104.8	−109.0	−7.7	−49.5
	**	***	***	***	NS	*	***	NS	***
Std. Error	±0.12	±1.2	±2.9	±0.01	±2.2	±42.6	±12.1	±6.8	±5.2
Basil
Standard	1.17	−44.5	−35.5	−9.0	−10.3	−155.6	286.7	98.3	148.3
	***	*	*	***	*	**	***	***	***
Std. Error	±0.14	±17.0	±8.9	±0.01	±2.2	±34.1	±25.2	±58.9	±13.0
Species-specific	−0.35	−91.5	−82.5	−9.0	−21.3	−107.1	−42.2	−26.3	−11.9
	NS	***	***	***	**	*	*	**	NS
Std. Error	±0.15	±10.1	±8.6	±0.02	±4.1	±31.2	±15.3	±7.5	±5.5

Initial EC was 1.54 mS/cm, and initial macronutrient concentrations (in mg∙L⁻¹) were 150 N, 141 NO₃-N, 9 NH₄-N, 38 P, 157 K, 140 Ca, 46 Mg, and 54 SO₄-S in the standard hydroponic solution. Data are the least-square means of three replicates. Means were compared to the initial EC and macronutrient concentrations in the standard hydroponic solution (see Materials and Methods) supplied at the beginning of the experiment using one-sample t-tests with α = 0.05. NS, *, **, *** refer to non-significance or significance at p < 0.05, p < 0.01, and p < 0.0001, respectively. Means followed by the same letter were determined not to be statistically different.

shows solution EC increased over time with the standard replenishment solution and decreased with species-specific solutions for arugula and basil. Both replenishment treatments resulted in decreased solution N and K for both species, whereas solution P decreased only for basil (Table 5). The standard replenishment treatment resulted in increased solution Ca, Mg, and SO₄-S over time for both species, indicating an oversupply of these nutrients [12]. In contrast, the species-specific treatment resulted in decreased solution Ca and SO₄-S for arugula and decreased solution Ca and Mg for basil (Table 5). Overall, the decreased solution EC and macronutrient concentrations with species-specific replenishment indicated nutrients were taken up by plants more rapidly than they were resupplied.

Replenishment treatments differed significantly in resupplied and measured nutrients and EC over time (Table 3, Table 4 and Table 5). However, the lack of statistical differences in plant growth and shoot tissue nutrients between treatments indicated these crops absorbed nutrients efficiently over a wide concentration range and were tolerant of varying root zone nutrient conditions, a finding consistent with past hydroponic fertilizer research [9,14,20,34,35]. It also suggests nutrients supplied in the species-specific replenishment solutions were closer to the concentrations taken up by arugula and basil compared to the standard replenishment solution.

Similar to findings in Table 5, Langenfeld et al. [4] also reported a decrease in solution EC to <0.1 mS∙cm⁻¹ and near-zero concentrations of macronutrients (particularly N, P, and K) when using species-specific replenishment solutions. Like in this study, Langenfeld et al. [4] found no reduction in yield for lettuce (Lactuca sativa L.) and tomato (Solanum lycopersicum L.) from low EC, presumably because nutrients were resupplied at the same rate as plant uptake.

However, in this study and Langenfeld et al. [4], low EC conditions occurred when plants were more mature and had relatively large root systems, a development stage at which plants would have been more efficient at taking nutrients up at low concentrations. In contrast, seedlings with small root systems are less efficient at absorbing nutrients from the root zone, and low EC and nutrients can result in reduced seedling growth [1]. In production scenarios where seedlings and mature crops are grown in the same hydroponic system and recirculating solution, it is likely necessary to control the replenishment solution strength and maintain a constant target EC level to ensure adequate nutrient uptake by young plants [1,9].

Replenishment treatment effects on tissue nutrients. Percent macronutrient concentrations in dried tissues at the end of Expt. #2 were not affected by replenishment solutions for either arugula or basil. Tissue macronutrient concentrations consisted of 5.03% N, 0.51% P, 4.32% K, 2.98% Ca, 0.77% Mg, and 1.17% S for arugula when averaged across replenishment strategies. For basil, tissue macronutrients averaged 4.36% N, 1.01% P, 5.13% K, 2.45% Ca, 0.57% Mg, and 0.37% S across replenishment strategies. Overall, these tissue macronutrient concentrations were slightly different from the values reported in Table 1.

Re-estimation of WUE and target N concentrations for replenishment solutions. Although not part of the original objectives, at the end of Expt. #2, WUE was estimated (in mL∙g⁻¹) for each treatment by dividing the total evapotranspiration of solution (summarized by treatment in Table 3) by the total dry mass gain of roots and shoots per plant for each replicate, with species and replenishment solution effects on WUE analyzed using ANOVA with PROC GLIMMIX at α = 0.05. There was a plant species main effect (p ≤ 0.0001) but no replenishment strategy main effect (p = 0.8714) on WUE. WUE for arugula and basil in this experiment was 255 and 180 mL∙g⁻¹, respectively, which was near the lower limit of the WUE range reported by Bugbee [1] for most crops grown hydroponically (200 to 400 mL∙g⁻¹ of dry mass gain).

Using the Langenfeld et al. [4] mass balance approach, the target nutrient concentrations to supply in the replenishment solution depend on and are largely affected by changes in WUE. Species tissue N levels from Expt. #1 (Table 1) and WUE values calculated at the end of Expt. #2 (255 mL∙g⁻¹ for arugula, 180 mL∙g⁻¹ for basil) were used to re-estimate the target N concentrations needed in replenishment solutions for Expt. #2. Using the same approach described in the Materials and Methods, the re-estimated N concentration was 194 and 266 mg∙L⁻¹ N for arugula and basil, respectively. Houston [8] estimated WUE values of 313 and 263 mL∙g⁻¹ for arugula and basil, respectively, from a hydroponic growth chamber experiment with different environmental conditions, which would have resulted in N concentrations of 158 mg∙L⁻¹ N for arugula and 182 mg∙L⁻¹ N for basil in this study.

These findings suggest the standardized N concentration used in replenishment solutions for Expt. #2 (150 mg∙L⁻¹ N) may have been sub-optimal for these crops, although whether deficiency symptoms would have occurred long-term remains unclear. They also highlight the large effect estimated WUE can have on hydroponic solutions formulated using this approach and the potential variability in resupplied nutrients from fluctuations in WUE over time and between crops.

Potential of using mass balance to formulate replenishment solutions and manage nutrients. The Langenfeld et al. [4] approach to formulating hydroponic solutions relies on tissue nutrient levels and WUE estimates to predict the nutrient/water requirements for a particular crop. The mass balance approach described by Sonneveld and Voogt [9] relies on changes in solution volume (i.e., evapotranspiration) and nutrient concentrations but otherwise results in the same nutrient/water requirement predictions. Both are useful as a starting point to determine initial nutrient concentrations to supply in a standard or replenishment hydroponic solution. However, the uptake of nutrients and water are not physiologically related processes within plants and may fluctuate independently with changing climate and cultural conditions [5,8,9,17], also shown in Table 1. Therefore, the actual nutrient concentrations taken up by plants over time will likely differ from the concentrations supplied in hydroponic solutions formulated using simple mass balance principles.

Results of this study suggest hydroponic replenishment solutions formulated using simple mass balance principles based on estimated tissue nutrient levels and WUE can reduce nutrient accumulation risks and the need to periodically discharge and replace solution. This approach would be most effective in scenarios where replenishment solutions could be formulated to a specific crop and variety, mixed using very high-quality irrigation water, and where environmental conditions were stable over time, such as indoor or vertical farming systems where the climate is strictly controlled [36]. The nutrient concentrations in these replenishment solutions would likely require regular adjustment since plant nutrient/water uptake and growth rates change significantly over time, particularly in greenhouses and outdoor hydroponics, where environmental conditions fluctuate throughout growing seasons [13].

The species-specific replenishment treatments had macronutrient ratios based primarily on expected plant tissue ratios and allowed certain nutrient concentrations and EC to decrease over time, sometimes to near-zero levels. In some scenarios, it is desirable to maintain certain solution nutrient ratios, such as maintaining a relatively high Ca:K ratio to prevent physiological disorders related to Ca uptake and translocation within the plant, including “tip-burn” in lettuce and “blossom-end-rot” in tomato [9,37]. Maintaining certain NH₄-N:NO₃-N ratios helps stabilize solution pH, counteract high water alkalinity, and increase N uptake and growth of leafy greens and herbs [27,38]. In addition, controlling and maintaining a minimum target EC level is a common and effective strategy to control crop quality, such as “toning” plants under low-light conditions and increasing soluble solids and flavor components in hydroponic tomato fruit [9,11].

Optimizing nutrient replenishment in recirculating solutions is challenging because root zone nutrients are influenced by multiple interacting factors, including plant species, environmental conditions (light, temperature, humidity, CO₂), irrigation water quality, injection of mineral acids/bases for pH control, fertilizer N forms, soilless substrate components, and water treatment technologies [8,9,13]. Formulating hydroponic solutions using a combination of mass balance and grower experience, periodic monitoring of root zone and tissue nutrient levels, and adjustment of the replenishment solution formulation as needed may be the most practical and effective strategy for managing nutrients in recirculating solutions.

4. Conclusions

This study showed hydroponic replenishment solutions could be formulated for arugula and basil using simple mass balance principles and potentially reduce nutrient accumulation and the need to periodically discharge and replace the hydroponic solution in closed systems. This study also highlighted several practical challenges with formulating efficient hydroponic solutions intended to balance nutrient and water supply with plant demand. Predicting the optimal nutrient concentrations to supply by using mass balance principles and from past measurements of plant tissue nutrients and water-use efficiency may still lead to root zone nutrient imbalances over time because plant nutrient and water uptake fluctuate with changing environmental conditions and between crops. Overall, managing nutrients in recirculating solutions is complex since multiple factors interact to affect solution nutrient concentrations, including plant species, environmental conditions, irrigation water quality, injection of mineral acids/bases, fertilizer N forms, substrate components, and water treatment technologies.