Acclimation of functional traits leads to biomass increases in leafy green species grown in aquaponics

Abstract

As human population size continues to increase and climate change effects worsen, future food security has become a primary concern for agricultural industries worldwide. Yields of traditional agricultural methods are commonly limited by water and nutrient availability and many crop yields are predicted to decline. Alternative farming practices like aquaponics, which can alleviate these negative yield pressures, may become critical to reaching food production targets. Aquaponics approaches involve the cyclic joint production of fish and hydroponic plants where the fish efflux provides nutrients to plants that then purify the water to be recycled to the fish tanks. In this study, we investigated the acclimation of physiology and functional traits of plants grown in aquaponics versus soil for three leafy green species. We compared gas exchange, stomatal anatomy, water-use efficiency, and foliar chemistry on newly formed leaves across weekly measurements. Increased photosynthetic rate, driven by higher stomatal conductance and increases in tissue nitrogen, led to higher biomass production in aquaponics for all species. Aquaponics plants adjusted stomatal behavior and to a lesser degree stomatal anatomy to become less water-use efficient than plants grown in soil. Collectively, our findings demonstrate the ability of plants to acclimate quickly to aquaponics growing systems that largely remove water and nutrient limitations to plant growth. The increased biomass production of broccoli, pak choi, and salanova by 185%, 116%, and 362% in aquaponics compared to soil-grown plants demonstrates the potential of small-scale aquaponics systems as an efficient and sustainable alternative farming practice.

As aspects of our food system are under pressure from population growth and global change, the need to investigate the efficacy of alternate agriculture practices is imminently important. Our study examined the capacity of leafy green species grown in aquaponics to acclimate to unlimited access to water and consistent supplies of nitrogen from fish efflux, compared to the same species grown in soil. We found biomass production, rates of photosynthesis, and stomatal conductance to be higher in aquaponics, while water use efficiency was lower in aquaponics for all species. Our study also found nitrogen supply to be higher in leaf tissues in plants grown in aquaponics, which correlated with the observed higher rates of photosynthesis. Our findings offer mechanistic insights into the future of sustainable farming that utilizes aquaponics to produce non-staple crops that meaningfully contribute to feed insecurity.

Introduction

Global food security

The human population is predicted to surge to over 9 billion by 2050 (Roberts 2011; Cleland 2013; Cole et al. 2018; Mote Rivas and Kalnay 2020; Siegel 2021) making sufficient food production to meet global needs an urgent problem. Within the next few decades, production of the major grain-producing staple crops including maize, wheat, and rice will need to increase by 50%–70% to feed the expanding population (Cairns et al. 2013; Ray et al. 2013; Tai Martin and Heald 2014; Long Marshall-Colon and Zhu 2015; Rockström et al. 2017; Chouchane Krol and Hoekstra 2018; Asseng et al. 2020; Gerten et al. 2020). Food security has not improved by increases in cropland efficiency as crop nutritional value has not followed the increase of production and converting natural lands to croplands has led to detrimental losses in biodiversity, soil health, and ecosystem stability (Clark and Tilman 2017; Ramankutty et al. 2018).

Additionally, biomass yields of some non-staple and legume crops are decreasing due to limited water supply and climate warming (Fahad Ali A Bajwa et al. 2017; Zhou et al. 2017; Bisbis Gruda and Blanke 2018); but impacts of climate change factors on production of non-staple crop impacts are less studied (Zhou et al. 2017; Scheelbeek et al. 2018). Further, global change factors such as pest introductions, shifts in pest phenology, and pest range expansion place additional negative pressures on crop productivity and food security (Boyer et al. 2013; Bebber Holmes and Gurr 2014; Nabout et al. 2016). These impacts will further decrease harvests and increase the urgency to enhance crop productivity to ensure food security (Boyer et al. 2013; Nabout et al. 2016; Hanes and Rhode 2050).

Globally, staple crops are impacted by nutrient and water deficiencies that restrict crop growth (Mueller et al. 2012). For example, nutrients and water can limit production of staple crops, such as wheat, soybean, and maize production by greater than 45% (Fahad Ali A. Bajwa et al. 2017). Biomass production limitations are often linked to anthropogenic disturbances like soil erosion outpacing the rate of soil replenishment in plowed fields that results in the loss of key soil nutrients such as nitrogen and phosphorus (Yuan et al. 2018; Dror Yaron and Berkowitz 2021) along with micronutrients like selenium (Jones et al. 2013). Additionally, the use of heavy agricultural machinery paired with other human field traffic destroys soil aggregates and degrades soil pore structure that can stunt root growth and decrease biomass production by as much as 50% (Shah et al. 2017). Additionally, alleviation of nutrient and water limitations in lettuce, an important non-staple crop, can improve production by more than 130% (Wang et al. 2023).

With altered soil nutrient cycling, the agriculture industry heavily relies on fertilizers to supplement these growth-limiting nutrients (Grzebisz et al. 2013; Paulot and Jacob 2014; Houlton et al. 2019) but it is difficult to balance the environmental harm of unstainable fertilizer use on enhanced agricultural production (X. Zhang et al. 2015). Fifty percent of the added nitrogen fertilizer in conventional soil agriculture ends up in the harvested biomass with the remaining 50% susceptible to nitrate leaching into groundwater or runoff, or as ammonia or nitrous oxide emissions to the atmosphere (Philip Robertson et al. 2013). These permanent removals of elemental resources via harvesting biomass and erosion drive a continuous, accelerating feedback loop towards poor soil quality, increased fertilizer use, and further environmental damage (Davidson et al. 2015). Moreover, other soil constraints such as low moisture content and high salinity contribute to biomass production shortfalls by constraining plant responses to vapor pressure deficits, disrupting ion homeostasis, reducing photosynthesis, and diverting energy for growth towards soil water uptake (Cemek et al. 2011; Zörb Geilfus and Dietz 2019; Rigden et al. 2020).

Groundwater depletion makes any increase in irrigated croplands unsustainable (Dalin et al. 2017; Perrone and Jasechko 2017) and projected increases in drought across major agriculture zones coupled with scarce groundwater supplies reduce the potential of irrigating new lands (Peña-Gallardo et al. 2018; Leng and Hall 2019). The complexity of tracking consumption, restricting extractions, and negotiating with irrigators makes increasing current irrigation efficiency an ineffective, slow solution (Grafton et al. 2018). To overcome water concerns, farming industries are turning to altering irrigation practices like plant-centric irrigation (Zhang et al. 2021) and water-saving irrigation (Xu et al. 2015b). However, reliance on irrigation to solve the climate water crisis is impossible with current renewable water resources falling short of the irrigation demand by as much as 47% (Wang et al. 2021).

Photosynthetic constraints of crops

Photosynthesis is fundamental to the production of food as it generates the raw materials for all plant products (Richards 2000). Photosynthesis in plants is tied to species-specific leaf traits such as leaf nitrogen content that regulates photosynthetic capacity and sensitivity to water stress through transpiration rate and stomatal conductance that regulates stomatal behavior. Photosynthesis requires many proteins, especially Rubisco, that account for the majority of nitrogen content in leaves (Evans and Clarke 2019). Photosynthesis is well-known to be limited by nitrogen availability, especially because photosynthetic enzymes require high nitrogen investments to build and maintain (Evans 1989; Evans and Seemann 1989). In addition, under water restriction, plants must optimize stomatal behavior to minimize transpiration losses while maximizing the CO₂ drawdown needed to drive photosynthetic reactions. Current and projected climate instability, including higher temperatures paired with decreasing precipitation, drives increased stomatal closure via changes in soil hydraulic conductivity and higher vapor pressure deficits (Carminati and Javaux 2020; Grossiord et al. 2020; Abdalla et al. 2022). Overall, fluctuating levels of water stress and nitrogen availability drive changes in photosynthetic traits that vary widely between species (Ahanger et al. 2016; Wang Wang and Shangguan 2016; Richards 2018).

Bioengineering and genetic manipulation research focused on drought tolerance and nutrient deficiency traits has increased photosynthetic rates and crop yields (Bouzid et al. 2019; Conti 2019; Condon 2020; Martignago et al. 2020; Dietz Zörb and Geilfus 2021; Ozeki Miyazawa and Sugiura 2022). However, these approaches are often limited by unexpected side effects of inserting transgenes and biological constraints of limited availability of genetic resources (Ladics et al. 2015; Kumar et al. 2020). Improved traditional agriculture techniques like cover cropping (Rosa-Schleich et al. 2019), permaculture (Suh 2014), and biodynamic farming (Santoni et al. 2022) practices will likely still fall short of our rising food targets. Shortfalls in these improvements to traditional agriculture have driven the refinement of alternative soilless farming methods like aeroponics, hydroponics, and aquaponics (AlShrouf 2017). Aquaponics, a highly efficient system with low pollution and water consumption, combines hydroponics and aquaculture (controlled production of aquatic organisms) and is one of the highest-yield animal production systems (Palm et al. 2018; Wei et al. 2019). In aquaponics systems, fish waste efflux supplies a hydroponic source of nitrogen-based nutrients to plants, which eliminates water stress and potentially decreases nitrogen limitation for leaf photosynthetic functional traits. Aquaponics plants subsequently filter the fish efflux, which is then recycled to the fish tanks, vastly reducing water requirements (Janni and Jadhav 2022). Although soilless growth systems are shown to increase the yield and quality of some crops compared to soil (Verdoliva et al. 2021), the limited spatial capacity of those systems to contribute to crop yield demands is largely untested. Consequently, soilless agricultural practices, like aquaponics systems with cyclic water and nutrient renewal from fish efflux, need to become a research priority.

The objective of this study is to compare leaf-level functional traits in plants grown in soil to plants grown in aquaponics to investigate the capacity of aquaponics plants to acclimate to the resource environment created by fish efflux and hydroponics. We used three leafy green crop species, common in aquaponics systems, to test for acclimation of leaf gas exchange, stomatal anatomy, water-use efficiency, and foliar chemistry on newly formed leaves. We hypothesized that plants in an aquaponics system will acclimate stomatal behavior and anatomy to upregulate photosynthesis despite the increased water cost of carbon gain via transpiration.

Methods

Study site

The Shepherd University Aquaponics Laboratory is located at Shepherd Farm, a 160-acre agricultural innovation center for teaching and demonstrating small-scale agricultural techniques, in Shepherdstown WV, USA (39.44549°N, 77.83048°W). The indoor aquaponics facility uses two 4542-liter fish tanks that provide fish efflux to separate hydroponic plant growth systems (Supplementary Figure S1). This study utilized the nutrient water from the Tilapia tank that contains 60 Oreochromis aureus fed with Purina Aquamax Pondfish 4000 (Nestlé Purina PetCare, St. Louis, Missouri, USA). The large particulate fish waste from the tank is removed in the settling tank with the excreted ammonia converted to nitrate in the biofilter. Fish tank temperature ranged from 24°C to 25°C with a pH of 7.5–8.0. Water from the Deep-Water Culture beds (DWC) flows to a sump tank and is pumped back into the fish tanks. The water flows at a rate of ~3785 liters per hour through the growing systems. Aquaponic plants were grown in the DWC beds that were 1.2 m wide × 5.0 m long × 0.15 m deep. Each bed fits eight floating Styrofoam raft boards (0.6 m × 1.2 m each) that contain 18 2.5 cm. square cutouts. Nitrate levels were measured twice over the study, varied negligibly, and averaged 11.2 mg/l.

Study design

We tested three commonly used species in aquaponics, Brassica oleracea var. italica (broccoli), Brassica rapa subsp. chinensis (pak choi), and Lactuca sativa (salanova). Plants in the aquaponics treatment were seeded in a 60:40 Coco Coir:Vermiculite media into 2.54 cm. net pots. Each net pot fit into the square cutouts in each floating raft, allowing roots to be exposed to fish efflux from the tilapia tank. Plants in the soil treatment were sowed directly into square 10 cm diameter × 8 cm deep pots with organic topsoil. Pots were placed on top of adjacent floating rafts at the same planting density and spacing (max 18 plants per raft). This ensured that the two experimental groups experienced similar airflow, microclimates, and light conditions. Three 650-watt Scynce Raging Kush II LED lights (ScynceLED, Mesa, Arizona, USA) were mounted directly above the floating rafts, with a 12:12 hour daylight and night cycle light regime with the Cool and Red channels set at 100%. Incident light Photosynthetic Photon Flux density, measured with a Licor 250-A with a LI-193 sensor (LI-COR Biosciences Inc. Lincoln, Nebraska, USA), averaged 410 μmol/m²/s at 15 cm above the floating raft. This irradiance level is within ranges shown to increase biomass accumulation, while remaining near the light saturation point of photosynthesis for leafy green species (Wong et al. 2020). For each species, we conducted two experimental trials that ran for 3–4 weeks with 10 individuals for each treatment. Trials with broccoli were run simultaneously, while pak choi and salanova trails were run consecutively during the spring and fall semesters, respectively. Broccoli trials occurred from April 19 to May 3, 2022, pak choi from February 8to March 29, 2022, and salanova from October 19 to November 23, 2021. Measurements were initiated when plants had fully formed leaves, approximately 3 weeks for each species. Growth trials were kept relatively short to prevent plants grown in soil from becoming root bound or nutrient limited in small pots. Plants in soil were watered every Monday, Wednesday, and Friday to field capacity and measurements were taken one day after watering. At the end of the trials, all plants were harvested to measure biomass growth. For the soil treatment, plant roots were carefully removed from the soil and washed to remove all soil particles. The dry biomass of shoot and root components of each plant were weighed after oven drying at 60°C to a constant mass.

Leaf gas exchange

Weekly sampling was conducted on two newly formed leaves; one leaf for gas exchange and the second leaf for stomatal anatomy and stoichiometry. The species used in this experiment produced new, fully formed leaves between weekly measurements. Gas exchange was measured with a LI-6800 Portable Photosynthesis System (LI-COR Biosciences Inc. Lincoln, Nebraska, USA), fitted with a 1 × 3 cm cuvette. Gas exchange measurements were conducted with chamber conditions of T_air at 20°C, 420 µmol mol⁻¹ reference [CO₂], 60% humidity, photosynthetically active radiation of 1500 mmol m⁻² s⁻¹, and internal lights set to a 9:1 red to blue light ratio. Once water vapor and [CO₂] values stabilized within the sample chamber, gas exchange parameters including light-saturated rates of photosynthesis (A_n), stomatal conductance (g_s), and transpiration (E) were logged once for each leaf. Intrinsic water use efficiency (WUE_g) was calculated as A_n divided by g_s.

Foliar stoichiometry and anatomy

For elemental and isotope analysis, representative subsamples of dried leaf and root samples from the final harvest were ground to a fine powder using a Bead Ruptor 96 (OMNI International, Kennesaw, Georgia, USA). Elemental tissue contents (carbon and nitrogen) and shoot δ¹³C were measured with a Carlo Erba NA elemental analyzer coupled with a Thermo Delta C IRMS (Thermo Fischer Scientific, Waltham, Massachusetts, USA). Percentages of carbon and nitrogen in samples calculated by comparison with certified standards and tissue nitrogen (N) is reported on a mass basis (g g⁻¹). Isotopic signatures of dry matter are reported relative to standard Vienna Pee Dee Belemnite.

Stomatal peels were taken weekly for stomata anatomical traits on the same age cohort of leaves used for gas exchange measurements. For each peel, a liquid bandage (New Skin, Advantice Health New Jersey, USA) was applied on the abaxial side of the leaf and allowed to dry. Clear tape was used to remove and transfer each peel onto a microscope slide. Two peels (~8 cm²) were taken from random locations on the fully formed leaf not chosen for gas exchange. Stomatal density (SD, # mm^-2) was measured by counting stomata under 400× magnification (field of view diameter 0.50 µm) for salanova and 1000× magnification (field of view diameter 0.2 µm) for broccoli and pak choi. For each leaf, SD was counted three times for randomly selected non-overlapping fields of view across the two available peels.

Ten stomata were imaged in either one or two peels per leaf for each plant during weekly measurement trials. All stomatal pictures were taken at 400× magnification. Stomatal size was measured using ImageJ (NIH, Bethesda, Maryland, USA). The width of both guard cells (mm) was measured from the edge of the stomatal aperture to the outside edge of each guard cell. Stomatal length (mm) was measured from the top to the bottom edge of each stomata. The combined stomatal width was multiplied by stomatal length to calculate stomatal size (mm²) as in (Franks and Beerling 2009).

Statistical analyses

A Welch’s t-test was first used to compare differences in treatments for each weekly measurement campaign. Across the weekly campaigns, data for all variables were consistently different between treatments. Thus, we report statistical comparisons (t-tests) of treatments as pooled data for each species. To investigate acclimation, a two-way repeated measures ANOVA (Type II) was performed to evaluate the effect of treatment (aquaponics or soil) and time (weekly measurements on fully developed leaves) on functional traits and gas exchange parameters for each species separately. The “car” (Fox and Weisberg 2018) package was used to extract model coefficients. Post-hoc pairwise comparisons were computed with the “emmeans” package (Lenth et al. 2022) to determine when significant differences developed within a treatment and across treatments. For pairwise comparisons across treatments, we primarily focused on identifying when treatment comparisons diverged (e.g. week 2) for a given variable to evaluate trait acclimation.

To examine bivariate trait relationships between treatments, responses of dependent variables were analyzed with either linear mixed-effect models or generalized additive models (GAM), with treatment as categorical fixed effects and species as a random effect. Akaike Information Criterion scores were used to evaluate and select the best fit model. For linear relationships, differences in slopes of significant relationships between bivariate traits by treatment or species were tested by calculating estimated marginal means and computing pairwise comparisons with the “emmeans” package (Lenth et al. 2022). Explained variance (R²) of mixed models were computed as in Nakagawa and Schielzeth (2013), in which the marginal R² represents variance explained by fixed factors and the conditional R² represents variance explained by both fixed and random factors. For non-linear trait relationships, confidence intervals were estimated by fitting a generalized additive model to the data with the “mgcv” package (Wood 2017). For significant GAM fits, we report the percent deviance explained and significance of smooth terms. All tests of statistical significance were conducted at an α level of 0.05. All analyses were performed with R 4.2.2 (R Development Core Team 2017).

Results

Plant growth

Harvested total biomass was significantly greater in aquaponics compared to soil for each species. Total biomass for broccoli, pak choi, and salanova increased by 170%, 114%, and 279%, respectively, in aquaponics (Fig. 1A, all P values 0.001). In broccoli, pak choi, and salanova, aboveground shoot biomass increased by 185%, 116%, and 362%, respectively when grown in aquaponics compared to soil (Supplementary Table S1 all P values < 0.001). Similarly for roots, broccoli, pak choi, and salanova biomass increased in aquaponics by 96%, 102%, and 86% (Supplementary Table S1, all P values < 0.001). The ratio of roots to shoots in broccoli and salanova decreased by 33% and 58%, respectively, in aquaponics treatments (both P < 0.001), while biomass partitioning did not change in pak choi (Fig. 1B).

Figure 1.

Boxplots of total biomass (A) and root:shoot ratios (B) for three leafy green species in aquaponics and soil treatments. Asterisks represent significant differences between growth treatments for a given species. Each box represents the interquartile range, the line is the median, and the whiskers extend to the lower and upper limits of the data.

Photosynthetic parameters

Photosynthetic parameters broadly changed for all species between aquaponics and soil treatments (Supplementary Table S2). Overall, A_n for broccoli, pak choi, and salanova increased by 39%, 71%, and 153%, respectively, in aquaponics compared to soil treatments (Fig. 2A, P < 0.001). For broccoli and salanova, A_n declined in the third week for both treatments (A_n * week, P < 0.001). In pak choi, photosynthesis decreased after the first week in soil treatments only (A_n * week, P = 0.005).

Figure 2.

Box plots of (A) light saturated photosynthesis rates (A_n) and (B) stomatal conductance (g_s) for three leafy green species in aquaponics and soil treatments. Asterisks represent significant differences between growth treatments for a given species. Each box represents the interquartile range, the line is the median, and the whiskers extend to the lower and upper limits of the data.

Overall, g_s increased in aquaponics by 138%, 219%, and 357% in broccoli, pak choi, and salanova, respectively (Fig. 2B, P < 0.001). Stomatal conductance in broccoli was variable between weeks in soil treatments only (g_s * week, P < 0.001). For pak choi, stomatal conductance declined after week one for both treatments (g_s * week, P = 0.004). In salanova, stomatal conductance declined in week three in aquaponics treatments only (g_s * week, P = 0.037).

Intrinsic water use efficiency (WUE_g) decreased significantly in aquaponics by 52%, 54%, and 44% in broccoli, pak choi, and salanova, respectively, compared to soil treatments (Fig. 3A, P < 0.001). Intrinsic water-use efficiency in salanova, decreased by week three in aquaponics treatments only (WUE_g * week, P = 0.037). For broccoli, WUE_g was variable from week to week in soil treatments only (WUE_g * week, P < 0.001).

Figure 3.

Box plots of (A) intrinsic water use efficiency (WUE_g) from gas exchange, (B) bulk shoot δ¹³C from harvested aboveground biomass and (C) stomatal density for three leafy green species grown in aquaponics and soil treatments. Asterisks represent significant differences between growth treatments for a given species. Each box represents the interquartile range, the line is the median, and the whiskers extend to the lower and upper limits of the data.

Leaf stoichiometry and anatomy

Tissue nitrogen content of both shoots and roots was higher in aquaponics across all three species. Shoot N content increased by 367%, 491%, and 287% in broccoli, pak choi, and salanova respectively (Fig. 4A, all P < 0.001). Consequently, the C:N ratio of shoots broadly decreased across all species in aquaponics (Fig. 4B, all P < 0.001). Root N content also increased for each species in aquaponics by 167%, 100%, and 444%, respectively for broccoli, pak choi, and salanova (Supplementary Table S1, all P < 0.001). Similarly, the C:N ratio of roots decreased across all species in aquaponics (Supplementary Table S1, all P < 0.001). Bulk shoot δ¹³C was significantly lower in broccoli and pak choi plants grown in aquaponics compared to soil (P = 0.004 and <0.001, respectively), while values were similar for salanova between treatments (Fig. 3B).

Figure 4.

Boxplots of (A) shoot nitrogen content and (B) shoot C:N ratio for three leafy green species grown in aquaponics and soil treatments. Asterisks represent significant differences between growth treatments for a given species. Each box represents the interquartile range, the line is the median, and the whiskers extend to the lower and upper limits of the data.

Acclimatory changes in stomatal traits between aquaponics and soil treatments varied by species (Supplementary Table S1). The stomatal density of aquaponics-grown broccoli and salanova was lower by 31% and 28%, respectively, compared to soil-grown plants (Fig. 4C, P < 0.001). Stomatal density in leaves of pak choi did not significantly change between treatments. Stomatal density (SD) for broccoli and salanova did not change through time in aquaponics, however, SD in both species did increase when grown in soil between weeks 1 and 2 (SD * week, P < 0.001 and P = 0.014, respectively). Stomatal density in pak choi in aquaponics increased from week 1 to 4 (SD * week, P = 0.024) while pak choi in soil increased from week 1 to 2 (SD * week, P < 0.001). No changes were detected in stomatal size for pak choi and salanova, while stomatal size for broccoli increased by 16% in aquaponics compared to soil treatments (P < 0.001).

Bivariate relationships between functional traits

A positive relationship between A_n and shoot nitrogen (N) existed across all plants (P < 0.001, R²  _marginal = 0.37, R²  _conditional = 0.90). However, the A_n—N relationship was not apparent within treatments (aquaponics nor soil), due to a large amount of variation across species. Across treatments (Fig. 5A), increases in shoot nitrogen were positively correlated with increases in A_n for broccoli (P < 0.001, R² = 0.79), pak choi (P < 0.001, R² = 0.78), and salanova (P < 0.001, R² = 0.80). Light-saturated photosynthesis increased with higher g_s across species for both treatments until A_n plateaued at larger rates of g_s (P < 0.001, deviance explained = 84.1%). Importantly, three-fold increases in g_s in some aquaponics plants led to much higher rates of A_n (P < 0.001, Fig. 5B). Light-saturated photosynthesis rates were also positively related to stomatal density despite large amounts of variation between species (P = 0.001, R²  _marginal = 0.25, R²  _conditional = 0.81). The slopes of the A_n—SD relationship also differed across treatments, as A_n aquaponics plants responded more strongly to changes in SD (P = 0.001, Fig. 5C). Intrinsic water use efficiency was not related to SD for either treatment.

Figure 5.

Relationships between light saturated photosynthesis (A_n) and (A) shoot nitrogen content, (B) stomatal conductance, and (C) stomatal density. For each treatment, dashed lines represent significant linear or generalized additive model fits and gray shaded areas are 95 % confidence intervals for the mean.

Discussion

In a novel aquaponics experiment, we tested the capacity of three leafy green species to physiologically acclimate to unrestricted water availability and consistent nitrogen supply from fish efflux compared to growth in limited soil volume. We found biomass production, rates of photosynthesis, and stomatal conductance to be higher in aquaponics, while intrinsic water use efficiency was lower in aquaponics for all species. Although the relationship between photosynthetic rates (A_n) and either foliar nitrogen (N) or stomatal conductance (g_s) is well understood, our results show the potential of plants to acclimate key photosynthetic traits to optimize physiology in a unique growth system. Our findings offer mechanistic insight into the future of sustainable farming that utilizes aquaponics to produce non-staple crops that meaningfully contribute to feed insecurity.

Changes in growth and biomass partitioning in aquaponics

Full biomass harvests showed that total biomass production of leafy green crops vastly increased in aquaponics compared to unfertilized soil. Utilizing aquaponics systems, as a substitute or complement to soil-based agriculture, contributes to stable food production while also increasing efficiency of water usage and reduction in reliance on fertilizer. For example, compared to conventional agricultural methods paired with pond-based or lake-based aquaculture, aquaponics was found to conserve water usage by greater than 75% (Cohen et al. 2018). Surprisingly, comparisons of biomass growth for key crop species between aquaponics systems and soil are largely unrepresented in the aquaponics literature. Aquaponic crop production is more commonly compared to hydroponic systems, with aquaponics systems producing higher yields (Johnson et al. 2017; Lennard and Ward 2019). Many species grow well in aquaponics systems due to their nutritional needs being supplied by the fish efflux (Krastanova et al. 2022), however, plant growth in aquaponics is linked to tradeoffs associated with maintaining fish and plant needs simultaneously, including pH, temperature, and nutrient compositions (Delaide et al. 2019). Although our study does indicate increased biomass production is possible in aquaponics, the magnitude of the increase is very likely overestimated compared to soil-based agriculture with fertilizer regimes. Nonetheless, Ranawade Tidke and Kate (2017) found the biomass production of spinach to be the highest when grown in an aquaponics system compared to both a hydroponics system and a traditional soil method with several fertilizers.

Two of the species, broccoli and salanova, also exhibited decreased root:shoot ratios when grown in aquaponics. Optimal partitioning theory describes changes in biomass allocation to environmental factors (e.g. water, light, and CO₂) by correlating belowground deficiencies with higher root growth and above-ground deficiencies with an increase in aboveground biomass (Reynolds and Thornley 1982; Johnson and Thornley 1987; Poorter et al. 2012; Ledo et al. 2018). For example, shifts in biomass allocation are often linked to water stress. Increases in belowground biomass as a response to drought conditions can result in increases in root:shoot ratios (Xu et al. 2015a). In this experiment, without belowground stressors for water or nutrients driving an increased carbon sink in roots for resource uptake, aquaponic broccoli, and salanova partitioned growth more towards aboveground biomass. Simply, aquaponics plants of these species allocated more resources into aboveground biomass without concurrent increased investment in root tissue.

Acclimation of gas exchange in aquaponics

Photosynthetic rate and stomatal conductance significantly increased in all three leafy green species when grown in aquaponics. Our results support the stomatal optimality theory that stomata regulate functions to balance carbon uptake with the penalties of open stomata (Buckley and Schymanski 2014), albeit with a phenomenon that is less often observed. With a continuous supply of water, aquaponics plants acclimated stomatal behavior by maintaining high rates of g_s because the consequences of transpiration were negligible compared to plants grown in soil. This acclimation of stomatal behavior in aquaponics plants contributed to concurrent higher rates of A_n. As instantaneous photosynthetic rates are a function of stomatal opening and the biochemical parameters (V_cmax and J_max) that regulate photosynthetic capacity (Wang et al. 2020), further research on the capacity of N-rich fish efflux to enhance rates of Rubisco carboxylation and/or electron transport for RuBP regeneration in aquaponics plants are still needed.

Importantly, stomatal responses can respond slower than photosynthetic responses to environmental factors, often limiting the capacity of A_n (Lawson and Blatt 2014; Lawson and Vialet-Chabrand 2019). Here, the lag between stomatal responses to external factors was inconsequential for plants grown in the controlled aquaponics environment, allowing for optimization of carbon gain. Thus, the larger rates of g_s exhibited across species in aquaponics did result in some apparent degree of saturation of A_n at a given light and temperature environment. Future studies in aquaponics should prioritize optimizing light regimes and growth temperatures to test if crop plants can further optimize the acclimation of gas exchange.

Our study found N supply to be significantly higher in aboveground and belowground tissues in plants grown in aquaponics. Crop plants, like the ones used in this study, are sensitive to changes in N supply and N limitation can drive changes in leaf traits and N investment to maintain photosynthetic capacity (Vos Van Der Putten and Birch 2005). For example, leaves rely on sufficient N intake to produce essential photosynthetic proteins like Rubisco, creating an extensive N sink to support photosynthesis and subsequent growth (Gastal et al. 2015; Evans and Clarke 2019). Here, higher rates of A_n for all three species in aquaponics were correlated with increased leaf nitrogen content. Aquaponics plants simply had access to more N to allocate to photosynthetic machinery, without the need to invest in higher root production. The access to a larger pool of N likely allowed the upregulated A_n rates to meet the sink demands of aboveground organs (Zhang et al. 2015; Tegeder and Masclaux-Daubresse 2018). The access to a continuous pool of N also resulted in a consistently lower C:N ratio of both above- and belowground tissues in aquaponics plants. With lower C:N ratios, plants can prioritize growth in nitrogen-sufficient environments compared to the need to prioritize nitrogen use efficiency (Zhang et al. 2019). Further work should determine the degree to which typical fertilization regimes in soil-based agriculture close the observed gap in plant nitrogen economy with aquaponics.

Water-use “”inefficiency” in aquaponics plants

As hypothesized, one of the strongest acclimatory responses in aquaponics plants was a functional shift towards water-use inefficiency in leaf gas exchange. Stomatal behavior defines a plant’s water use efficiency, as the pore acts as the resistant force towards atmosphere flux and internal stimuli interact with CO₂ flux and transpiration (Lawson and Blatt 2014; de Santana et al. 2015; Zhao et al. 2020). We detected aspects of these shifts in both adjustments in physiological behavior and anatomical traits. Instantaneous transpiration efficiency was lower in all three leafy green species when grown in aquaponics. For broccoli and pak choi, the removal of water limitation in aquaponics plants also drove decreases in water-use efficiency across the leaf life span (lower foliar δ¹³C content), revealing likely sustained higher uptake of CO₂ and decreased water-use efficiency across the experiment.

Contrary to our hypothesis, stomata density (SD) decreased in aquaponics. Water vapor losses from leaves are functions of the size, density, and distributions of stomata (Franks and Beerling 2009; Dow Berry and Bergmann 2014; Fanourakis et al. 2015; Devi and Reddy 2018). Our results relate to evidence highlighted by Lawson and Blatt (2014) that decreased SD can be compensated for with an increase in stomatal aperture. Here, aquaponic plants increased g_s by leaving stomata pores more open, while also investing less into creating more stomata on newly created leaves. Our study uniquely focused on an environment without water stress and consistent light and CO₂ levels that is seldom found in literature as investigations of stomatal traits often involve responses to drought, light regimes, or altered CO₂ levels (Bertolino Caine and Gray 2019). For example, changes to stomatal aperture size, without acclimation of SD, was detected in Arabidopsis plants in response to drought (Doheny-Adams et al. 2012). Additionally, Xu and Zhou (2008) found a positive correlation between SD and water-use efficiency, g_s, and A_n in false wheatgrass experiencing drought conditions. In contrast to these drought studies, the ability of aquaponics plants to maintain consistent levels of water-use “inefficiency“ via high g_s, allowed concurrent upregulation of A_n and down regulation of stomatal production. Experiments in both aquaponics and hydroponics systems should continue to prioritize understanding how acclimation of plant water-use strategies can be harnessed to optimize crop production.

Acclimation potential of aquaponics plants

Plants are known to acclimate to a variety of changing environmental conditions such as cold temperatures (Hassan et al. 2021; Juurakko et al. 2021a, b), diffused light (Li et al. 2014; Liu and Su 2016), high salinity (Hossain et al. 2017; Pandolfi et al. 2017), and drought (Pandey and Shukla 2015; Ahmad et al. 2016; Menezes-Silva et al. 2017). Currently, evidence of plant acclimation in hydroponics growth environments compared to soil-based growth is mostly absent in the literature. Overall, these leafy green species were capable of adjustments to the aquaponic growth environment for key processes that regulate photosynthesis and growth by the time the initial cohort of leaves was fully formed. Here, gas exchange variables of all species had acclimated to aquaponics when the first cohort of leaves was measured, and stomatal anatomy had acclimated by the second cohort of leaves. As future food security concerns become a driving force in agricultural research, the acclimation potential exhibited by the hydroponically grown crops in this study should be more broadly investigated to improve precision farming.

Summary

Connections between decreasing human health from malnourishment resulting from climate change continue to rise as concerns grow for future food and freshwater availability, soil health, and biodiversity (Mcmichael 2013). Our experimental findings that functional traits of different leafy green species all acclimated to enhance photosynthesis in aquaponics provide evidence that aquaponics has the capacity to improve production on large and small scales and that aquaponics systems could become a key contributor to global food security as the system is unaffected by urbanization, environmental degradation, and climate change (Ghandar et al. 2021). Although not part of the study, dietary deficiencies can be combatted through increased consumption of fish which are high in protein and contain vitamin A, B, and D along with essential micronutrients like calcium, iron, and zinc (Béné et al. 2015). Thus, the capacity of aquaponics to support sustainable vegetarian and pescetarian style diets could combat health concerns, climate change complications like drought, and food security risks by relying on the improved vegetable production capacities of aquaponics systems (Springmann et al. 2016). Higher investments in aquaponics growing systems could meaningfully contribute to the supply of crucial greens, herbs, vegetables, and fish needed to support the growing human population in an agriculture world plagued by global change (Shreejana et al. 2022).

Conflict of interest: None declared.

Funding

This material is based upon work supported by Shepherd University professional development mini-grants and the NASA Undergraduate Affiliate Fellowship program.

Data Availability

The data that support the findings of this study are openly available at https://figshare.com/projects/Aquaponics_Physiology_dataset_Nicholes_et_al_2025_AoB_Plants/233471.