ORIGINAL RESEARCH published: 25 May 2021 doi: 10.3389/fpls.2021.649304 Improving the Fertigation of Soilless Urban Vertical Agriculture Through the Combination of Struvite and Rhizobia Inoculation in Phaseolus vulgaris Verónica Arcas-Pilz 1 , Felipe Parada 1 , Gara Villalba 1,2* , Martí Rufí-Salis 1,2 , Antoni Rosell-Melé 3,4 and Xavier Gabarrell Durany 1,2 1 Sostenipra Research Group (2017 SGR 1683), Institut de Ciència i Tecnologia Ambientals (CEX2019-000940-M), Universitat Autònoma de Barcelona, Barcelona, Spain, 2 Department of Chemical, Biological and Environmental Engineering, Universitat Autònoma de Barcelona, Barcelona, Spain, 3 Institut de Ciència i Tecnologia Ambientals, Universitat Autònoma de Barcelona, Barcelona, Spain, 4 Institució Catalana de Recerca i Estudis Avançats, Barcelona, Spain Soilless crop production is a viable way to promote vertical agriculture in urban areas, Edited by: but it relies extensively on the use of mineral fertilizer. Thus, the benefits of fresher, local Jung Eek Son, food and avoiding the transportation and packaging associated with food import could Seoul National University, South Korea be counteracted by an increase in nutrient-rich wastewater, which could contribute to Reviewed by: freshwater and marine eutrophication. The present study aimed to explore the use of Pradeep Kumar, mineral fertilizer substitutes in soilless agriculture. Phaseolus vulgaris (common bean) Central Arid Zone Research Institute (ICAR), India was fertilized with a combination of slow-releasing fertilizer struvite (a source of N, P, Silvia Pampana, and Mg), which is a byproduct of wastewater treatment plants, and inoculation with University of Pisa, Italy Rhizobium (a N2 -fixing soil bacteria). The experiment included three bean-production *Correspondence: lines: (A) 2 g/plant of struvite and rhizobial inoculation; (B) 5 g/plant of struvite and Gara Villalba gara.villalba@uab.cat rhizobial inoculation, both irrigated with a Mg-, P-, and N-free nutrient solution; and (C) a control treatment that consisted of irrigation with a full nutrient solution and no Specialty section: This article was submitted to inoculation. Plant growth, development, yields, and nutrient contents were determined Crop and Product Physiology, at 35, 62, and 84 days after transplanting as well as biological N2 fixation, which was a section of the journal determined using the 15 N natural abundance method. Treatments A and B resulted in Frontiers in Plant Science lower total yields per plant than the control C treatment (e.g., 59.35 ± 26.4 g plant−1 Received: 04 January 2021 Accepted: 27 April 2021 for A, 74.2 ± 23.0 g plant−1 for B, and 147.71 ± 45.3 g plant−1 for C). For A and Published: 25 May 2021 B, the nodulation and N2 fixation capacities appeared to increase with the amount of Citation: initially available struvite, but, over time, deficient levels of Mg were reached as well Arcas-Pilz V, Parada F, Villalba G, Rufí-Salis M, Rosell-Melé A and as nearly deficient levels of P, which could explain the lower yields. Nevertheless, we Gabarrell Durany X (2021) Improving conclude that the combination of struvite and N2 -fixing bacteria covered the N needs of the Fertigation of Soilless Urban plants throughout the growth cycle. However, further studies are needed to determine Vertical Agriculture Through the Combination of Struvite the optimal struvite quantities for vertical agriculture systems that can meet the P and and Rhizobia Inoculation in Phaseolus Mg requirements throughout the lifetime of the plants. vulgaris. Front. Plant Sci. 12:649304. doi: 10.3389/fpls.2021.649304 Keywords: fertigation control, struvite, rhizobium, plant physiology, nutrient uptake, soilless agricultural system Frontiers in Plant Science | www.frontiersin.org 1 May 2021 | Volume 12 | Article 649304 Arcas-Pilz et al. Alternative Fertigation Urban Vertical Agriculture INTRODUCTION contents, which average 12.5%, 9.9%, and 5.7%, respectively (Ahmed et al., 2018) and are suitable for plant growth (Degryse From 1950 to 2018, the population living in urban areas grew et al., 2017; Ahmed et al., 2018). Due to struvite’s high nutrient more than fourfold to an estimated 4.2 billion people. This concentrations, there are many ongoing efforts to optimize unprecedented population increase has greatly increased global induced precipitation to make wastewater a valuable resource for food demand, which has exerted great pressure on natural providing a P alternative to the use of the depleting phosphate resources (United Nations, 2019). In response, new ways to rock (Massey et al., 2007; Cordell et al., 2009; Talboys et al., 2016; efficiently produce vegetables while minimizing land use are Degryse et al., 2017). being explored (Sanyé-Mengual et al., 2015, 2018). One of these A further positive aspect of struvite as an agricultural fertilizer initiatives is vertical farming with the use of soilless production substitute is its slow solubility in granular form (Talboys et al., systems with growing media or substrates (Sonneveld and Voogt, 2016) under alkaline and neutral pH soil conditions (Bhuiyan 2009), which would reduce the transportation and packaging et al., 2007). Thus, the risks of nutrient leaching and water of foodstuffs to cities (Sanyé et al., 2012). However, vertical eutrophication are rather small under these conditions when agriculture relies extensively on the use of mineral fertilizer, struvite is compared to common readily soluble fertilizers which results in nitrates and phosphate being discharged into (Ahmed et al., 2018). Furthermore, the removal of approximately wastewater, which can contribute to freshwater and marine 30–40% of N and P from wastewater to produce this substance eutrophication (Anton et al., 2005; Gopalakrishnan et al., 2015; can prevent eutrophication in urban water cycles (González Sanjuan-Delmás et al., 2018). Ponce et al., 2009; Antonini et al., 2012). The granular form This extensive use of mineral fertilizers affects not only of struvite also causes it to be easily manageable and could be the environment but can also be related to a high cost of applied in larger-scale productions by mixing it with soil or production and extraction, as is the case for nitrogen fertilizers applying it to the substrates in hydroponic production systems. due to the Haber-Bosch process (Cherkasov et al., 2015) and The use of struvite has already been tested in agriculture as for phosphorous due to phosphate rock extraction (Cordell a substitute for phosphate from other sources and has shown and White, 2013). The widespread use of these nutrients has promising results with low or even no yield losses reported caused vertical farming to rely entirely on them, which thus (González Ponce et al., 2009; Cabeza et al., 2011; Liu et al., 2011; makes this agricultural practice unsustainable in the long run. Ackerman et al., 2013; Degryse et al., 2017; Ahmed et al., 2018). The high energy cost of synthetic nitrogen production and Although struvite already contains N that is available to the ever-depleting sources of phosphate rock, when added to plants, legumes have high N demands (McKey, 1994). Therefore, the environmental cost of their disposal and emissions to the average N contents in struvite would not be sufficient for water and air (Rufí-Salís et al., 2020a,b), necessitates the search soilless crops to achieve commercial yields and would require for alternatives to further implement these technologies in a a second source of N to do so. This N could be obtained sustainable way. from Rhizobium, which is capable of forming an endosymbiotic Many strategies have been described in recent years for the interaction with leguminous plants by entering root cells and implementation of organic fertilization in vertical farming, which forming nodules. These nodules enable atmospheric N2 fixation embraces a circular economy framework to reduce new resource and ammonia (NH3 ) formation. Plants benefit from the bacteria inputs into cities. Some examples include fertilization that is that generate these compounds, while the bacteria can profit based on gray water and urine (Ikeda and Tan, 1998; Karak from photosynthesis-derived compounds (Long, 1989). This and Bhattacharyya, 2011) and the use of biofertilizers such as symbiosis, on the other hand, may entail a major requirement Rhizobium for the cultivation of legumes (Kontopoulou et al., of nutrients from the plant, such as phosphorous, to satisfy 2015; Savvas et al., 2018) for the plant nitrogen supply. Other the needs of the bacteria and successful nodulation (Olivera methods describe the use of sewage sludge (Frossard et al., et al., 2004). Possible N2 fixation depends on successful 1996), sewage sludge ash (Nanzer et al., 2014), and struvite rhizobial root colonization, which is influenced by diverse (Rech et al., 2018) as alternative P sources. While these strategies factors, such as phosphorous fertilization, salinity, drought, and may reduce the direct inputs of specific inorganic fertilizers, initial N availability (Araújo et al., 2007; Ntatsi et al., 2018; their use often results in lower crop yields and, in some cases, Savvas et al., 2018). require more infrastructure for irrigation systems. These studies Rhizobium as a second source of N was chosen due tend to focus on one particular nutrient alternative and do to the lower inputs needed to achieve nitrogen intake by not consider the combination of alternative methodologies. plants (Gopalakrishnan et al., 2015). When using the N2 -fixing Therefore, innovation to provide a solution for multiple mineral bacterium, Rhizobium, in hydroponic cultivation, Kontopoulou fertilizers while avoiding the addition of infrastructure as well as et al. (2017) described the need to apply initial N fertilization further environmental burdens due to local nutrient sourcing has until nodulation in the root medium occurs, to further not been widely studied. encourage nodulation and therefore N fixation, plant growth, and Struvite (MgNH4 PO4 .6H2 O), which is a crystalline byproduct production. Even though previous studies have reported lower of wastewater treatment plants that formes by spontaneous production capacities for N2 -fixing plants than for common or induced precipitation, usually contains high N and P beans with N fertilization (Olivera et al., 2004; Kontopoulou concentrations (Rahman et al., 2014) and is regarded as a et al., 2017), a combination of the two N sources (e.g., viable slow-releasing fertilizer due to its high P, Mg, and N struvite and N2 -fixing bacteria) was used to determine the Frontiers in Plant Science | www.frontiersin.org 2 May 2021 | Volume 12 | Article 649304 Arcas-Pilz et al. Alternative Fertigation Urban Vertical Agriculture possibility of overcoming such lower yields (Pampana et al., 2017; The inoculated plants were treated with two different struvite Savvas et al., 2018). amounts placed inside the perlite bag around the root area and To determine how effective the two alternative fertilizers are surface, varying the concentration of P and N available to the in providing N to plants, the 15 N natural abundance method plant from struvite: (A) 2 g (1.02 mmol of P and 0.46 mmol of was employed to determine the source of N throughout the N) of granulated struvite per plant and (B) 5 g (2.57 mmol of P experiment (Shearer and Kohl, 1989). While plants with N and 1.15 mmol of N) of granulated struvite per plant. The amount that is acquired from symbiotic atmospheric N2 fixation show of struvite that was best for growth was determined in a previous the lower richness of the 15 N isotope, which corresponds to experiment conducted in the same i-RTG in which 2.57 mmol the atmospheric abundance (0.3663%), plant tissues that are P was deemed sufficient for common bean fertilization to reach subjected to other N sources can exhibit greater amounts of an equivalent level of commercial production as that of mineral- the 15 N isotope, which depend on the N fertilizer applied fertilized beans. To ensure no struvite loss due to runoff, each (Robinson, 2001). plant was planted inside an additional 1 L bag containing perlite The present study aimed to add to this growing pool of and the corresponding amount of struvite, with small holes to knowledge on vertical urban agriculture by exploring the use of allow water drainage. mineral fertilizer substitutes struvite and rhizobium combined Each treatment was arranged randomly in four rows with 16 in an effort to reduce emissions of simultaneously N and P to plants each (four perlite bags with 4 plants per bag were planted the environment in urban vertical agriculture. This combination in a frame with an area of 0.125 m2 ), which resulted in a total of also aims to optimize crop yields while avoiding the installation 64 plants per treatment (e.g., A, B, and Control), with 192 plants of additional infrastructure. In this study, we analyzed the in total (Supplementary Figure 1b). Due to the irrigation and growth, development, and production of the common bean leachate recovery systems, randomization could only be achieved (Phaseolus vulgaris), which was fertilized with a combination for entire lines of four bags. of the slow-releasing fertilizer, struvite, and the soil bacteria, The plants were germinated and transplanted in duplicate and Rhizobium. A combination of these alternative fertilizers can be were thinned to one plant at 21 days after transplanting (DAT). implemented easily in terms of cost and space and promotes Greenhouse conditions were monitored during the entire nutrient recycling within cities. experiment with T107 sensing devices (Campbell Scientific) that were placed along the cropping area to measure temperature, relative humidity, and radiation (see Supplementary Table 2b). MATERIALS AND METHODS To ensure proper plant irrigation drainage volumes, the pH and electrical conductivity levels of the leachate were recorded every Experimental Site, Materials, and Growth day for each irrigation line. Conditions The phenological stages of the bean plants were determined This experiment was conducted in the Rooftop Greenhouse each week. This information was assessed to identify plant Laboratory (RTG-Lab) of the Environmental Science and growth, development, and productivity over time and provided Technology Building (ICTA-UAB), which is located in the a clear view of the plant cycle, growth, and production peaks Universitat Autònoma de Barcelona Campus (42◦ 290 2400 E, that enabled accurate comparisons of plant development between 45◦ 940 3600 N) (Sanjuan-Delmás et al., 2018). The bean variety treatments and the control. This was performed by counting used in this experiment was Phaseolus v. Pongo, which had leaves, flower buttons, and open flowers. The number of ripened previously been germinated in a commercial greenhouse 10 days bean pods was also counted and weighed for each harvest. before transplanting in the RTG-Lab. The production system These measurements were performed for each of the eight was soilless with a perlite substrate in 40 L bags and the use of plants that were in the two middle bags of each row (see fertigation through a 2 L/h drip irrigation system. Supplementary Figure 1b) and started 14 DAT. To ensure Bean seeds were treated with a commercial product (e.g., uniform counting, leaves under 5 cm length were not considered, Nadicom GmbH©) which contained a mixture of Rhizobium and only fully formed flower buttons with white coloration and phaseoli and Rhizobium giardinii strains for inoculation before fully open flowers were counted. For the bean pods, a minimum planting. The inoculation procedure was an exposure of the length of 11 cm was used for harvesting, while bean pods shorter plant seeds with the liquid commercial product before planting. than this were retained for the next harvest. The average numbers A total of 5 days after the seedling was transplanted into and bean pod weights per treatment were then calculated for the perlite substrate, 5 ml liquid commercial mix was added each week. At the same time and on a weekly basis, chlorophyll to each plant, therefore ensuring the presence of the bacteria content measurements were performed (with a SPAD CCM-200 in the substrate. Once the plants were inoculated, they were plus; Opti-Sciences, Inc.) on the same eight plants in the center of irrigated with an Mg-, P-, and N-free solution (Supplementary each row. Table 1b), and application of K2 SO4 was increased to adjust for the K requirements. The control plants, on the other Description of Plant Sampling Methods hand, were irrigated with a full nutrient solution. These To determine the changes in plant development as well as nutrient concentrations were maintained throughout the entire nutritional states and 15 N concentrations, samples were taken experiment. The crops were irrigated four times a day for 3 min, during the three different crop stages. The first sampling took which provided a total amount of 400 ml per day per plant. place 35 DAT, immediately before bean pod production started; Frontiers in Plant Science | www.frontiersin.org 3 May 2021 | Volume 12 | Article 649304 Arcas-Pilz et al. Alternative Fertigation Urban Vertical Agriculture the second sampling took place 62 DAT, during the productive δ15 N values provide an indication of the N sources in plant phase of the plants; and the last sampling took place 84 DAT, tissues. Values close to 0 indicate that the plant N sources are at the end of the productive stage, which marked the last day mainly due to atmospheric N2 fixation, while higher values can of the experiment. be interpreted as indicating mixed sources or those dominated The samples consisted of eight randomly chosen plants per by the N obtained from struvite. The δ15 N value obtained for treatment (excluding the eight central plants of each row that the struvite used in the experiment was 7.1h. To determine were kept for phenological analysis). Each plant was washed the relative contributions from the two sources considered, we with deionized water, excess water was dried off and each used Eq. 2, which yields an estimate of the percentage of N that plant was separated into the four main organs: leaves, shoots, was derived from N2 - fixation (%Ndfa) (Shearer and Kohl, 1993; roots, and nodules. These were then weighed separately to Unkovich et al., 2002; Arndt et al., 2004) determine their fresh weights (FW). All organs were placed separately in envelopes and left to dry in an oven at 65◦ C until δ15 N Source 2 − δ15 N Sink %Ndfa = × 100 stabilized dry weights (DW) were obtained, which occurred after δ15 N Source 2 −0 B0 value approximately 7–8 days. The means of the obtained values were Equation 2: %Ndfa (nitrogen derived from N2 fixation from the calculated for each treatment, each organ, and time. The numbers atmosphere), δ15N Source 2 (h) corresponds to the δ15N value of of nodules were counted prior to drying to determine the mean struvite, δ15N Sink (h) corresponds to the δ15N value from the nodulation of each plant. In addition, fruit samples from each sample, and the “B” value corresponds to the δ15N of N2 fixation treatment were taken at three different times (49, 62, and 77 taking into account possible fractionation. DAT), which closely matched the three plant harvests. Moreover, 25% of the total sampled leaves for each plant were The “B” value is the isotopic fractionation observed in N2 - separated to determine their areas before the drying process. fixing P. vulgaris was set to –1.16h, which corresponded to the To do so, these fresh leaves were scanned with a reference lowest δ15 N value obtained (Shearer and Kohl, 1989; Peoples pixel to obtain leaf areas using ImageJ software (Rueden et al., et al., 2002; Kermah et al., 2018) and was similar to the values 2017). These leaf areas were further extrapolated to 100% of the determined by Kontopoulou et al. (2017) in common bean that leaf biomass of the plant. The leaf area index (LAI) was then was fertilized without N and inoculated with Rhizobium. calculated by dividing the total leaf area by the area of the planting The biologically fixed nitrogen (BNF) levels were further frame of our crop (0.125m2 ). calculated with the obtained %Ndfa values as well as the obtained values for the nitrogen contents in the plants. To extrapolate to Nitrogen Isotopic (δ15 N) Analysis kg/ha, a theoretical plant density of eight plants/m2 was used. The goal of inoculating treatments A and B with Rhizobium Finally, the nitrogen use efficiency (NUE) for all treatments was for the plants to indirectly fix N2 from the air and meet was estimated. The methodology that was followed to perform their N needs in this way. To determine how much of the N these calculations was given by Weih (2014), who provided a assimilated by the plants came from the atmosphere, we used tool to successfully calculate the NUE. To accomplish this, the the natural abundance method (Shearer and Kohl, 1989) to information provided was as follows: identify the origin of the N that was obtained by the plants, – N content at the initial stage of the plant in g/m2 (previous which in our case, should be either struvite or atmospheric to the main production stage at 35 DAT), N. While treatments A and B were actively inoculated with – N content at the main productive stage in g/m2 (chosen at Rhizobium strains and fertilized with struvite containing N, the 84 DAT), control treatment was fertilized through standard N fertilization – N content in the harvested yield in g/m2 , that was administered through irrigation. Additional nitrogen – Biomass yield g/m2 , sources were not considered due to the laboratory conditions and – Added N to the soil in g/m2 (in this case, perlite). production of inert perlite. Analysis was performed with an elemental analyzer isotopic ratio mass spectrometer (EA-IRMS; Thermo Fisher Scientific). Plant Nutritional Analysis The devices used were a Flash EA 1112 analyzer and Delta Dried and ground plant organs were weighed (up to 0.25 g) and V Advantage spectrometer that was coupled with a Conflo III digested using a single reaction chamber microwave (Milestone interface. The plant and struvite samples were weighed in tin Ultrawave) with concentrated HNO3 . The digested samples were capsules and were introduced into the EA-IRMS system to obtain then diluted with 1% HNO3 (v/v) and were analyzed by optical the δ15 N values, as calculated with the following equation (Eq. 1) spectrometry (ICP-OES) (Perkin-Elmer, Optima 4300DV). All (Robinson, 2001): samples were weighed, digested, and analyzed in duplicate. Sample atom %15 N − 0.3663 Statistical Analysis δ15 N = × 1000 0.3663 All statistical analyses in this experiment were performed with Equation 1: δ15 N is the natural tracer for our N sources, the R studio software. Data normality in our values was tested sample atom %15N is the previously obtained value of our plant with Shapiro-Wilk test p > 0.05, and to ensure homogeneity sample, and the value 0.3663 is a standard value that represents the of variance the Levene test was performed p > 0.05. When percentage of 15 N in the atmosphere. both criteria were validated Duncan’s multiple range test was Frontiers in Plant Science | www.frontiersin.org 4 May 2021 | Volume 12 | Article 649304 Arcas-Pilz et al. Alternative Fertigation Urban Vertical Agriculture used to assess the statistical significance of treatments. The treated with struvite and rhizobium produced approximately half Kruskal-Wallis test was used for no parametric data. The the yield of the mineral-fertilized plants, it is important to note significance between the treatments was tested for each harvest that they were healthy throughout the experiment. The average time separately. yields provided per plant were 59.35 ± 26.4 ga plant−1 for A, 74.24 ± 23.0 ga plant−1 for B, and 147.71 ± 45.3 gb plant−1 for the control treatment C. These production differences can RESULTS also be seen in Figure 1 where the obtained yields are shown as a function of time and show greater production peaks and a Phenology, Biomass, and Yield more rapid ability to develop flower buttons and open flowers Weekly recordings of the phenological growth of the bean after each harvest. plants exhibited differences among all treatments (Figure 1). In this figure, we can see the evolution throughout the crop δ15 N, %Ndfa and Biologically Fixed N development of biomass production as well as flower production The results obtained for the δ15 N values of plant tissues and finally bean pod production. The control plants (Treatment and bean pods as seen in Figure 2 (and Supplementary C) showed greater biomass growth and faster development in Figure 3b) show great variability in the enrichment of all their transitions from flower buttons to open flowers and bean organs except for the nodules. While treatment C showed pod production. Although the treatment performances were clear enrichment over time, the pattern for treatments A and similar in the earlier growth stages, once the production stage B was the opposite. For the nodules, all three treatments started, greater differences were observed. exhibited clear enrichment over time. Treatment B exhibited At 40 DAT and 50 DAT, treatments A and B began to perform intermediate δ15 N values that were between those of A and worse for leaf production as well as for the formation and C, with decreasing δ15 N values that were not as abrupt when opening of flower buttons than the control plants (C). Between compared to the tissues that were exposed to treatment A. 60 DAT and 70 DAT, a second production peak can be seen It was also interesting to observe that the major decrease in for the control treatment as well as rapid generation of flower δ15 N values for treatment A occurred between days 35 and buttons, while treatments A and B showed a declining pattern for 62 after transplanting and remained rather constant at 84 bean pod production. DAT. For the plants in treatment B, the value at 62 DAT Table 1 shows the changes in the plant measurement results did not fall as drastically and experienced a more significant that were conducted on the sampled plants at three different change at 84 DAT. developmental stages. While the first period of plant sampling, When calculating the percentage of fixed atmospheric N 35 DAT, showed very few significant differences among the during our three sampling periods, we obtained the values shown treatments (only in the case of dry weight), the later samplings at in Figure 3. This figure shows the approximate percentages of N 62 and 84 DAT showed greater differences between treatments. that were derived from atmospheric fixation relative to the total At this point, the leaf and shoot dry weights were greater for N obtained by the plants. the control treatment, as was the measured leaf area index. The As shown in the figure, the percentages of fixed N2 in all only parameter without significant differences among treatments three tissues were higher for the plants in treatment A, with throughout the entire experiment was the root dry weight at values of 65-80% at 35 DAT, which reached 90% by the end 62 DAT. The dry weights of the nodules exhibited persistent, of the experiment (84 DAT). On the other hand, treatment B significant differences for the three samplings among treatments exhibited lower values throughout the experiment, with initial A and B and control treatment C and reached maximum values of values close to 50% to 60% (35 DAT), which reached final values 0.16 g, 0.12 g, and 0.05 g for treatments A, B, and C, respectively. of 80% at 84 DAT. On the other hand, treatment B (with higher struvite quantities) While the plants with less struvite in the root medium also exhibited significantly greater numbers of nodules as well (treatment A) increased their percentages of fixed N2 more as higher weights than the other two treatments during the rapidly (from 70% (35 DAT) to 90% (62 DAT) in the leaves), the third sample period. plants in treatment B took longer to reach this value (from 60% When examining the SPAD measurements (Supplementary (35 DAT) to 71% (62 DAT) in leaves). This corresponds to the Figure 2b), some differences in chlorophyll content were results for the δ15 N values shown in Figure 2. observed throughout the experiment. Initially, we can see a Table 2 shows the results of the estimations of biological significant difference between the A and B treatments and the fixed nitrogen (BNF) contents expressed in kg/ha. These control marking a greater chlorophyll content in the latter results show the extrapolations of total N found in the plants that is sustained until 35 DAT. From 42 DAT to 63 DAT, for each treatment to kg/ha values. The N percentages that the chlorophyll content in treatments A and B increases while were of atmospheric origin (obtained previously) were further treatment C remains stable. While differences toward the end used to attain the kg/ha of biologically fixed nitrogen for of the experiment remain small, we can appreciate a greater each treatment as well as the N from struvite that was chlorophyll content in the struvite fertilized treatments. used by the plants. The final production amounts that were obtained for all three Here, we can see that as the percentages of atmospheric- treatments were 1899.2 g, 2375.6 g, and 4726.7 g of green bean derived N and total N that were found in the plants increased, as pods for treatments A, B, and C, respectively. Although the plants well as the kg/ha values of biologically fixed N. While the plants in Frontiers in Plant Science | www.frontiersin.org 5 May 2021 | Volume 12 | Article 649304 Arcas-Pilz et al. Alternative Fertigation Urban Vertical Agriculture FIGURE 1 | Graphic representation of the mean numerical count per plant for each organ (Leaves, flower buttons, open flowers) and yield in g/plant on a weekly basis, DAT representing the days after transplanting inside the iRTG. The colors represent the three treatments: A = N.free solution with Rhizobium inoculation and 2g of struvite per plant. B = N.free solution with Rhizobium inoculation and 5g of struvite per plant. C = Complete nutrient solution without struvite and no inoculation treatment. TABLE 1 | Results for the mean values (n = 8) per plant of fresh weight (FW) and dry weight (DW) of the different organs as well as the Leaf Area Index (LAI) m2 plant−1 of the three treatments (A = 2g Struvite + Rhizobium; B = 5g Struvite + Rhizobium; C = Control) in three different time periods: 35 DAT (1), 62 DAT (2), and 84 DAT (3). Leaf DW (g) per plant Shoot DW (g) per plant Roots DW (g) per plant Nodules n per plant Nodules DW per plant (g) LAI (1) A 1.12a ± 0.22 0.46a ± 0.08 0.44a ± 0.10 132.50a ± 80.35 0.16b ± 0.07 0.57a ± 0.12 B 1.31a ± 0.46 0.56a ± 0.19 0.51a ± 0.15 156.75a ± 60.82 0.12b ± 0.06 0.62a ± 0.23 C 1.33a ± 0.57 0.58a ± 0.18 0.53a ± 0.14 148.75a ± 48.23 0.05a ± 0.02 0.65a ± 0.27 (2) A 3.97a ± 1.25 2.02a ± 0.72 0.80a ± 0.28 127.88a ± 63.85 0.14b ± 0.09 1.28a ± 0.51 B 3.69a ± 1.53 2.24a ± 1.01 0.87a ± 0.34 172.25a ± 132.66 0.15b ± 0.14 1.29a ± 0.64 C 6.44b ± 3.09 3.85b ± 1.95 0.95a ± 0.44 82.25a ± 62.47 0.01a ± 0.01 2.64b ± 1.33 (3) A 5.86a ± 2.96 3.09a ± 1.45 1.77a ± 0.79 136.88b ± 106.31 0.15b ± 0.13 1.74a ± 0.92 B 7.40a ± 2.17 4.53a ± 1.48 2.49a ± 0.57 186.25c ± 48.79 0.24b ± 0.11 1.80a ± 0.67 C 11.11b ± 1.51 6.91b ± 1.42 3.35b ± 0.88 39.13a ± 24.76 0.02a ± 0.02 3.72b ± 0.87 Significant differences (p < 0.05) between treatments marked with different letter (a,b,c). treatment A had higher values of biologically fixed N during the Nutrient Content first two sampling periods at 84 DAT, the increase in the fixation The nutrient contents in the aboveground plant organs are percentage and total N in the plants in treatment B increased presented in Figure 4 (Supplementary Figure 5b for differences their amounts of biologically fixed N. On the other hand, the use between harvests). The observed concentrations of nutrients in of N from struvite increased only for treatment B and remained leaves for the three treatments were at sufficient levels except constant for treatment A. for the less than optimal Mg2+ concentrations at 62 DAT for Frontiers in Plant Science | www.frontiersin.org 6 May 2021 | Volume 12 | Article 649304 Arcas-Pilz et al. Alternative Fertigation Urban Vertical Agriculture FIGURE 2 | Boxplot representing the obtained δ15 N values (n = 4) for treatments: A = 2g of struvite + Rhizobium inoculation + P, Mg, N-free nutrient solution, B = 5g of struvite + Rhizobium inoculation + P, Mg, N-free nutrient solution and C = standard nutrient solution - Rhizobium inoculation. These observed values are given by plant organs in three different time periods: 35 days after transplanting, 62 days after transplanting, and 84 days after transplanting. Significant differences (p < 0.05) between dates marked with different letter (a,b,c). treatments A and at 84 DAT for treatments A and B and were DISCUSSION close to deficient levels P in both treatments A and B at 62 and 84 DAT according to Hochmuth et al. (2018). In the case of N, Plant Growth and Development in both leaf and shoot tissues, no deficient levels were found for The results indicate that once the first production peak was any of the treatments, and no significant differences were found reached, the control plants were more capable of continuing among treatments. On the other hand, a clear decline in P and to produce flower buttons, while the inoculated and struvite- Mg2+ over time can be seen for treatments A and B in the leaves fertilized plants took longer. The relationship between their as well as for P in the shoots. The control treatment (C), on the development and the amount of struvite given to the plants other hand, remained stable. seems to be directly correlated. Generally, the biomass and Supplementary Figure 4b also indicates the total nutrient contents that are bound to the total biomass of the sampled bean pod production was higher in the control plants, while plants. Here, it is apparent that treatment B, with more struvite, treatment B had a greater amount of struvite (5 g). Treatment provided results that were between those of the other treatments. A, with the lowest amount of struvite (2 g), was determined to In the case of Mg at 35 DAT in leaves, treatment B showed be the treatment with the lowest growth and production rates. levels as high as those for the control treatment, but while the These findings agree with those presented in previous literature latter remained constant over time, both A and B decreased. (Nanjareddy et al., 2014), for which lower KNO3 availability was The same trend can be seen for P in both leaves and shoots. directly linked to a reduction in leaf and flower formation. This In the case of N, we can see an increase for all treatments reduction also seemed to be related to the P and Mg availability that was faster for control C, while A and B increased in a over time due to struvite depletion, considering that the initial similar fashion. performance was similar in all three treatments. Finally, the NUEs obtained for all three treatments were By observing the SPAD measurements, the chlorophyll 1.32 g/g, 0.55 g/g, and 0.29 g/g for A, B, and C, respectively. The contents in all three treatments indicated that the N contents calculation methodology considered that N was in the soil, while in the leaves were not strongly affected by the treatments but the fixed nitrogen was not considered; therefore, the use efficiency rather the LAI. Lower P availability resulted in a reduction in can be very different for all three treatments. LAI as well as in overall plant growth, which was observed in Frontiers in Plant Science | www.frontiersin.org 7 May 2021 | Volume 12 | Article 649304 Arcas-Pilz et al. Alternative Fertigation Urban Vertical Agriculture FIGURE 3 | Percentage of Nitrogen derived from atmospheric N2 fixation (%Ndfa) represented for treatments: A = 2g of struvite + Rhizobium inoculation + P, Mg, N-free nutrient solution, and B = 5g of struvite + Rhizobium inoculation + P, Mg, N-free nutrient solution. These observed values are given for two plant organs (leaf; shoot) as well as the bean pods in three different time periods: 35 days after transplanting, 62 days after transplanting, and 84 days after transplanting. treatments A and B. These differences were not as great as those to be insufficient to reach production levels as high as those for root weights (compared to the other plant organs), which have of conventionally fertilized beans. However, struvite fertilization been reported in the previous literature to be less affected by P seemed to increase the production of inoculated plants by up to reductions (Chaudhary and Fujita, 1998; Rao et al., 2008). 25% when treatments with 2 g and 5 g per plant were compared The lower nodule dry weights in the control treatment, compared to treatments A and B, have previously been reported in other studies, in which the nodule fresh and dry TABLE 2 | Results for percentage of Nitrogen derived from atmospheric N2 weights were found to be considerably reduced when inorganic fixation (%Ndfa) in plant, Total amount of N in plant expressed in kg/ha NO3 − fertilization was not restricted (Nanjareddy et al., 2014; (Leaves+Shoot+Root+Beans) and Biologically fixed N expressed in kg/ha. Kontopoulou et al., 2017). On the other hand, other authors report that the nodule number was not affected when exposing Date Treatment % Ndfa Total N in Kg/ha Kg/ha N plant−1 plant kg/ha biologically from the crop to mineral and organic N sources but rather affected in fixed N struvite size and weight (Pampana et al., 2017). The increasing nodule numbers and weights throughout 35 DAT A 68% 7.5 ± 1.0a 5.4 ± 1.0a 2.2 the experiment for the B treatment (with greater struvite), B 60% 8.6 ± 2.2a 5.3 ± 1.4a 3.3 when compared to treatment A, confirm Kontopoulou et al. 62 DAT A 89% 24.7 ± 5.0a 22.9 ± 4.0b 1.8 (2017). findings that low initial N fertilization can restrict B 73% 24.6 ± 6.2a 18.7 ± 5.1a 5.9 successful colonization. These differences, however, could also 84 DAT A 90% 27.3 ± 12.8a 25.4 ± 13.0a 1.9 be due to the lower P amounts in treatment A compared to B 82% 35.0 ± 9.2a 29.2 ± 7.8a 5.8 treatment B since P is a limiting factor for successful nodulation Results given for three treatments (n = 8 each) (A) 2g of struvite + Rhizobium (Olivera et al., 2004). inoculation + P, Mg, N-free nutrient solution (B) 5g of struvite + Rhizobium The lower bean productivities were similar to those in the inoculation + P, Mg, N-free nutrient solution at three different time periods. 35 days after transplanting, 62 days after transplanting and 84 days after study reported by Olivera et al. (2004), where bean production transplanting. Significant differences (p < 0.05) between treatments marked with with lower P fertilization and Rhizobia inoculation turned out different letter (a,b,c). Frontiers in Plant Science | www.frontiersin.org 8 May 2021 | Volume 12 | Article 649304 Arcas-Pilz et al. Alternative Fertigation Urban Vertical Agriculture FIGURE 4 | Nutrient concentration in Phaseolus vulgaris leaves and shoots, expressed in mg/g. Boxplot (n = 4) results given for three treatments: A = 2g of struvite + Rhizobium inoculation + P, Mg, N-free nutrient solution, B = 5g of struvite + Rhizobium inoculation + P, Mg, N-free nutrient solution and, C = standard nutrient solution - Rhizobium inoculation at three different time periods: 35 days after transplanting, 62 days after transplanting, and 84 days after transplanting. Significant differences (p < 0.05) between treatments marked with different letter (a,b,c). (59.35 ± 26.4 g plant−1 in treatment A and 74.23 ± 23.0 g plant−1 P for treatment B than for treatment A (Olivera et al., 2004; in treatment B). Kontopoulou et al., 2017). The effect of the struvite treatment on the increasing nodule The δ15 N reductions in treatments A and B over time number and dry weight indicates successful nodulation and a corresponded to the availability of N provided by the struvite, greater fixation capacity with the given N. The slow release of assuming that it decreased over time. These reductions can be N has presented itself as sufficient to increase the nodulation seen when the NO3 − concentrations in the drained water were capacity as well as production capacity, without inhibiting N2 examined (see Supplementary Table 3b). While initially greater fixation by the bacteria. amounts of N were detected in the leached water, by the end of the experiment, very low concentrations were seen. Therefore, while the δ15 N values for the control treatment C remained The Effect on Atmospheric N Fixation constant over time (except in the nodules), the δ15 N values for Capacity treatments A and B decreased progressively over time, which The aboveground organs showed a clear pattern throughout the corresponded to the available N that was provided by struvite in three measurements in terms of N assimilation. 15 N enrichment the root medium. levels in the A and B treatments were lower than that in This information indicates that a change in the source of the C treatment, which means that treatments A and B N for the plants took place during the time span of 35 to obtained most of their N from the atmosphere. This difference 62 DAT. We can therefore assume that the availability of became even greater as time progressed and reflected a greater struvite and therefore N in the root medium was depleted dependence on N2 fixation in the A and B treatments. The mainly during that time, which forced the plants to rely on differences between these two treatments (A and B) themselves atmospheric N2 fixation. The results obtained for %Ndfa also can be due to the greater availability of struvite in the root confirm that the levels of N2 fixation increased over time in medium and therefore a greater availability of initial N and both treatments. Frontiers in Plant Science | www.frontiersin.org 9 May 2021 | Volume 12 | Article 649304 Arcas-Pilz et al. Alternative Fertigation Urban Vertical Agriculture The indicated timespan of 35 to 63 DAT corresponds to the These findings lead to the concept that a lack of N is not the initial pod production of the plant, maximizing its nutritional limiting factor that is entirely responsible for the lower yields needs. Therefore, an N and P source capable to uphold these of the A and B treatments, but the limiting factor is instead the needs during this stage must be contemplated. As seen in progressive loss of P and Mg in the root medium as well as the Supplementary Table 3b a major reduction of NO3 − in the reduced cation uptake. When examining the NUEs that were leachate water is found between day 35 and 49 for treatments A obtained for all treatments, it is evident that plants with lower and B, indicating that the administered struvite was insufficient N inputs have greater use efficiency. This difference is very clear to further support a greater production. in treatment A with a three-times higher efficiency compared The nodules appeared to be highly enriched with 15 N during to treatment B. These differences can also be influenced by all three harvests, especially for treatments A and B. These results atmospheric N fixation, which was not provided as “Soil” N agree with previous literature that attributes this enrichment to in the calculation tool (Weih, 2014). A higher fixation capacity the export of 15 N-depleted ureides and import of 15 N-enriched can therefore generate a higher NUE, which corresponds to amino acids. Nevertheless, these values do not have a great effect our BNF results. on the total plant enrichment if the nodule biomass is considered For production on larger-scale vertical farms, fertilization (Shearer and Kohl, 1986; Unkovich, 2013; Craine et al., 2015). with struvite and Rhizobium seems possible, especially with The quantity of fixed nitrogen did not reach 40-50 kg/ha, greater struvite quantities, as in treatment B, which shows great which corresponds to low ranges, as reported in previous research compatibility with soil bacteria and produces larger yields than (Farid and Navabi, 2015). While treatment A, with less struvite, those crops fertilized with only 2 g of struvite. The initial had higher BNF values at the first two sampling times, treatment fixation capacity of the control treatment and appearance of B’s BNF value had increased by the end of the experiment. nodules during the first sampling stage indicate that nodulation These findings are in agreement with those mentioned in the could occur even with naturally occurring Rhizobium, which literature, where BNF was found to be restricted in the presence could simplify the fertilization process in soil-based agriculture. of plant-available NO3 − , and the BNF values increased during A limitation for larger-scale production could be providing the mature stages of the plant with sufficient NO3 − fertilization precise applications of struvite in the root areas. As seen in during early plant growth (Müller et al., 1993; Hungria et al., this study, there were large production differences between the 2006; Kontopoulou et al., 2017). applications of 2 g and 5 g of struvite, and large-scale production in a vertical farm would mean precise weighting of the struvite Plant Health and Nitrogen Assimilation amounts per plant and direct applications to each rhizosphere We conclude that all treatments had sufficient N since there of each plant. As stated by Degryse et al. (2017), the location of were minimal differences in N concentrations in the shoots and this slow-releasing fertilizer can have a great impact on successful leaves during plant growth and at the end of the experiment, as nutrient delivery to plants. These could thus be highly time- and was also found by Kontopoulou et al. (2015). We consider that resource-consuming applications. the lower yields were caused by the reduced uptake of K+ and Mg2+ cations, which was cased by the electrochemical imbalance generated by the reduced presence of NO3 − in the root medium. CONCLUSION This idea is reinforced by the results shown in Supplementary Figure 4b, where N gradually increased in all three treatments This work aimed to study the feasibility of using struvite and throughout the experiment, which indicated that fixation was inoculation with Rhizobium bacteria as alternative Mg, N, and taking place for treatments A and B. The values increased from P fertilization methods for vertical agriculture systems. For less than 0.1 g N at 35 DAT up to 0.2 g at 84 DAT for both the A this purpose, we quantified the nitrogen sources, production, and B treatments. and evolution of the phenological stages of Phaseolus vulgaris The slight increase in K by the end of the experiment in the with Rhizobium inoculation and different quantities of struvite plants with less struvite (treatment A) was most likely due to and compared the results to a control treatment. Three main the lower availability of the Mg2+ cation, which facilitated cation conclusions can be drawn from this study. uptake (Marschner, 2002). First, both alternative fertilizer treatments supplied the The declining N concentrations in the leachates led us to necessary nutrients to fulfill the plant cycle in soilless growing believe that the decreases in P and Mg concentrations in the media. The lower yields compared to the control suggest the aboveground organs could also be related to the depletion necessity for evaluating higher struvite quantities to fulfill of struvite in the medium. This depletion occurred faster in plant requirements to achieve higher yields. Since previous treatment A than in treatment B, which was related to the initial experiments conducted with struvite suggested successful amounts of struvite provided in each treatment (2 g and 5 g, performance with 5 g/plant, its combination with the soil respectively). It was seen that for the inoculated plants, greater bacteria, Rhizobium, causes this quantity to be insufficient due amounts of P were needed to support symbiosis and nodulation, to the additional nutritional requirements of the bacteria. This as has also been observed by other researchers (Olivera et al., can be seen by the great reduction in yields of treatments A and 2004; Ntatsi et al., 2018; Savvas et al., 2018). Whether the B in comparison to the control. additional required P can be assimilated by adding more struvite Second, while nodulation seemed to not be hindered by to the substrate is worth pursuing in future studies. nitrogen input through struvite in the root medium, it did not Frontiers in Plant Science | www.frontiersin.org 10 May 2021 | Volume 12 | Article 649304 Arcas-Pilz et al. Alternative Fertigation Urban Vertical Agriculture significantly improve it either, although BNF appeared to increase conceptualization, performed the methodology, wrote, reviewed, in the later stages for plants grown under the treatment with a and edited the manuscript, supervised the data, and carried greater initial quantity of struvite. out the funding acquisition. AR-M performed the methodology Third, the limiting factor for struvite-fertilized and rhizobia- and reviewed and edited the manuscript. XG contributed to the inoculated treatments did not seem to be nitrogen, which was conceptualization, performed the methodology, wrote, reviewed, maintained at sufficient concentrations in the plants throughout and edited the manuscript, supervised the data, and carried out the experiment, but rather was potassium, due to the lower the funding acquisition. All authors contributed to the article and uptake capacity that was caused by an electrochemical imbalance approved the submitted version. that was generated by the reduced presence of NO3 − in the root medium as well as by magnesium and phosphorus, given that struvite depletion was reflected by the reduced plant nutrient FUNDING concentrations over time. An increase in the amount of applied struvite might be The authors are grateful to the Spanish Ministry of Economy, a solution for a more sustained phosphorus and magnesium Industry, and Competitiveness (Spain) for the grant that supply for vertical agriculture but could also interfere with the was awarded to VA-P (FPI-MINECO 2018), the National nodulation capacity of the plants. Furthermore, we encourage Commission for Scientific and Technological Research (Chile) the addition of nutrients in the form of anions to ensure the for the grant that was awarded to FP (PFCHA-CONICYT electrochemical balance in the root area in case NO3 − is removed. 2018 – Folio 72180248), and Universitat Autònoma de Barcelona In this sense, further studies should aim to determine the for awarding a research scholarship to MR-S (PIF-UAB optimal struvite quantities for hydroponic bean production in 2017). This work was supported by the Spanish Ministry combination with Rhizobium inoculation. of Economy, Industry, and Competitiveness (AEU/FEDER)- [CTM2016-75772-C3-1-R] and the “María de Maeztu” program for Units of Excellence in R&D [MDM-2015-0552] [CEX2019- DATA AVAILABILITY STATEMENT 000940-M]. This research received funding from the European Union’s Horizon 2020 research and innovation program under The raw data supporting the conclusions of this article will be grant agreement No 862663 (FoodE) and ERC grant agreement made available by the authors, without undue reservation. n◦ 818002 URBAG, which was awarded to GV. The publication reflects the authors’ views. The Research Executive Agency (REA) is not liable for any use that may be made of the information AUTHOR CONTRIBUTIONS contained therein. VA-P contributed to the conceptualization, performed the methodology, carried out the formal analysis and data curation, SUPPLEMENTARY MATERIAL and wrote the original draft. FP performed the methodology, carried out the data curation, and wrote, reviewed, and edited The Supplementary Material for this article can be found the manuscript. MR-S performed the methodology, wrote, online at: https://www.frontiersin.org/articles/10.3389/fpls.2021. reviewed, and edited the manuscript. GV contributed to the 649304/full#supplementary-material REFERENCES Bhuiyan, M. I. H., Mavinic, D. S., and Beckie, R. D. (2007). A solubility and thermodynamic study of struvite. Environ. Technol. 28, 1015–1026. doi: 10. Ackerman, J. N., Zvomuya, F., Cicek, N., and Flaten, D. (2013). Evaluation of 1080/09593332808618857 manure-derived struvite as a phosphorus source for canola. Can. J. Plant Sci. Cabeza, R., Steingrobe, B., Römer, W., and Claassen, N. (2011). Effectiveness of 93, 419–424. doi: 10.4141/cjps2012-207 recycled P products as P fertilizers, as evaluated in pot experiments. Nutr. Cycl. Ahmed, N., Shim, S., Won, S., and Ra, C. (2018). Struvite recovered from various Agroecosyst. 91:173. doi: 10.1007/s10705-011-9454-0 types of wastewaters: characteristics, soil leaching behaviour, and plant growth. Chaudhary, M. I., and Fujita, K. (1998). Comparison of phosphorus deficiency Land Degrad. Dev. 29, 2864–2879. doi: 10.1002/ldr.3010 effects on the growth parameters of mashbean, mungbean, and soybean. Soil Anton, A., Montero, J. I., Muñoz, P., and Castells, F. (2005). LCA and tomato Sci. Plant Nutr. 44, 19–30. doi: 10.1080/00380768.1998.10414423 production in Mediterranean greenhouses. Int. J. Agric. Resour. Gov. Ecol. 4, Cherkasov, N., Ibhadon, A. O., and Fitzpatrick, P. (2015). A review of the existing 102–112. doi: 10.1504/IJARGE.2005.007192 and alternative methods for greener nitrogen fixation. Chem. Eng. Process. 90, Antonini, S., Arias, M. A., Eichert, T., and Clemens, J. (2012). Greenhouse 24–33. doi: 10.1016/j.cep.2015.02.004 evaluation and environmental impact assessment of different urine-derived Cordell, D., and White, S. (2013). Sustainable phosphorus measures: strategies struvite fertilizers as phosphorus sources for plants. Chemosphere 89, 1202– and technologies for achieving phosphorus security. Agronomy 3, 86–116. doi: 1210. doi: 10.1016/j.chemosphere.2012.07.026 10.3390/agronomy3010086 Araújo, A. S. F., Monteiro, R. T. R., and Carvalho, E. M. S. (2007). Effect Cordell, D., Drangert, J. O., and White, S. (2009). The story of phosphorus: global of composted textile sludge on growth, nodulation and nitrogen fixation of food security and food for thought. Glob. Environ. Change 19, 292–305. doi: soybean and cowpea. Bioresour. Technol. 98, 1028–1032. doi: 10.1016/j.biortech. 10.1016/j.gloenvcha.2008.10.009 2006.04.028 Craine, J. M., Brookshire, E. N. J., Cramer, M. D., Hasselquist, N. J., Koba, K., Arndt, S. K., Kahmen, A., Arampatsis, C., Popp, M., and Adams, M. (2004). Marin-Spiotta, E., et al. (2015). Ecological interpretations of nitrogen isotope Nitrogen fixation and metabolism by groundwater-dependent perennial plants ratios of terrestrial plants and soils. Plant Soil 396, 1–26. doi: 10.1007/s11104- in a hyperarid desert. Oecologia 141, 385–394. doi: 10.1007/s00442-004-1655-7 015-2542-1 Frontiers in Plant Science | www.frontiersin.org 11 May 2021 | Volume 12 | Article 649304 Arcas-Pilz et al. Alternative Fertigation Urban Vertical Agriculture Degryse, F., Baird, R., da Silva, R. C., and McLaughlin, M. J. (2017). Dissolution rate sludge ashes as studied by 33P labeling techniques. Plant Soil 377, 439–456. and agronomic effectiveness of struvite fertilizers – effect of soil pH, granulation doi: 10.1007/s11104-013-1968-6 and base excess. Plant Soil 410, 139–152. doi: 10.1007/s11104-016-2990-2 Ntatsi, G., Karkanis, A., Yfantopoulos, D., Olle, M., Travlos, I., Thanopoulos, R., Farid, M., and Navabi, A. (2015). N 2 fixation ability of different dry bean et al. (2018). Impact of variety and farming practices on growth, yield, weed genotypes. Can. J. Plant Sci. 95, 1243–1257. doi: 10.4141/CJPS-2015-084 flora and symbiotic nitrogen fixation in faba bean cultivated for fresh seed Frossard, E., Sinaj, S., and Dufour, P. (1996). Phosphorus in urban sewage sludges production. Acta Agric. Scand. B 68, 619–630. doi: 10.1080/09064710.2018. as assessed by isotopic exchange. Soil Sci. Soc. Am. J. 60, 179–182. doi: 10.2136/ 1452286 sssaj1996.03615995006000010029x Olivera, M., Tejera, N., Iribarne, C., Ocaña, A., and Lluch, C. (2004). Growth, González Ponce, R., López-de-Sá, E. G., and Plaza, C. (2009). Lettuce response to nitrogen fixation and ammonium assimilation in common bean (Phaseolus phosphorus fertilization with struvite recovered from municipal wastewater. vulgaris): effect of phosphorus. Physiol. Plant. 121, 498–505. doi: 10.1111/j. HortScience 44, 426–430. 0031-9317.2004.00355.x Gopalakrishnan, S., Sathya, A., Vijayabharathi, R., Varshney, R. K., Gowda, C. L., Pampana, S., Scartazza, A., Cardelli, R., Saviozzi, A., Guglielminetti, L., Vannacci, and Krishnamurthy, L. (2015). Plant growth promoting rhizobia: challenges and G., et al. (2017). Biosolids differently affect seed yield, nodule growth, nodule- opportunities. 3 Biotech 5, 355–377. doi: 10.1007/s13205-014-0241-x specific activity, and symbiotic nitrogen fixation of field bean. Crop Pasture Sci. Hochmuth, G., Maynard, D., Vavrina, C., Hanlon, E., and Simonne, E. (2018). Plant 68, 735–745. doi: 10.1071/CP17166 tissue analysis and interpretation for vegetable crops in Florida. Hortic. Sci. Dep. Peoples, M. B., Boddey, R. M., and Herridge, D. F. (2002). “Quantification of 2012:55. nitrogen fixation,” in Nitrogen Fixation at the Millennium, ed. G. Jeffery Leigh Hungria, M., Franchini, J. C., Campo, R. J., Crispino, C. C., Moraes, J. Z., Sibaldelli, (Amsterdam: Elsevier B.V.), 357–389. doi: 10.1016/b978-044450965-9/50013-6 R. N. R., et al. (2006). Nitrogen nutrition of soybean in Brazil: contributions Rahman, M. M., Salleh, M. A. M., Rashid, U., Ahsan, A., Hossain, M. M., and of biological N 2 fixation and N fertilizer to grain yield. Can. J. Plant Sci. 86, Ra, C. S. (2014). Production of slow release crystal fertilizer from wastewaters 927–939. doi: 10.4141/p05-098 through struvite crystallization – a review. Arab. J. Chem. 7, 139–155. doi: Ikeda, H., and Tan, X. (1998). Urea as an organic nitrogen source for 10.1016/j.arabjc.2013.10.007 hydroponically grown tomatoes in comparison with inorganic nitrogen Rao, I. M., Fredeen, A. L., and Terry, N. (2008). Leaf phosphate status, sources. Soil Sci. Plant Nutr. 44, 609–615. doi: 10.1080/00380768.1998.1041 photosynthesis, and carbon partitioning in sugar beet: III. Diurnal changes in 4484 carbon partitioning and carbon export. Plant Physiol. 92, 29–36. doi: 10.1104/ Karak, T., and Bhattacharyya, P. (2011). Human urine as a source of alternative pp.92.1.29 natural fertilizer in agriculture: a flight of fancy or an achievable reality. Resour. Rech, I., Withers, R., Jones, D. L., and Pavinato, P. (2018). Solubility, diffusion and Conserv. Recycl. 55, 400–408. doi: 10.1016/j.resconrec.2010.12.008 crop uptake of phosphorus in three different struvites. Sustainability 11:134. Kermah, M., Franke, A. C., djei-Nsiah, S. A., Ahiabor, B. D. K., Abaidoo, R. C., and doi: 10.3390/su11010134 Giller, K. E. (2018). N2-fixation and N contribution by grain legumes under Robinson, D. (2001). δ15N as an integrator of the nitrogen cycle. Trends Ecol. Evol. different soil fertility status and cropping systems in the Guinea savanna of 16, 153–162. northern Ghana. Agric. Ecosyst. Environ. 261, 201–210. doi: 10.1016/j.agee.2017. Rueden, C. T., Schindelin, J., Hiner, M. C., DeZonia, B. E., Walter, A. E., Arena, 08.028 E. T., et al. (2017). ImageJ2: imageJ for the next generation of scientific image Kontopoulou, C. K., Giagkou, S., Stathi, E., Savvas, D., and Iannetta, P. P. M. data. BMC Bioinformatics 18:529. doi: 10.1186/s12859-017-1934-z (2015). Responses of hydroponically grown common bean fed with nitrogen- Rufí-Salís, M., Parada, F., Arcas-Pilz, V., Petit-Boix, A., Villalba, G., and Gabarrell, free nutrient solution to root inoculation with N2-fixing bacteria. HortScience X. (2020a). Closed-loop crop cascade to optimize nutrient flows and grow low- 50, 597–602. doi: 10.21273/HORTSCI.50.4.597 impact vegetables in cities. Front. Plant Sci. 11:596550. doi: 10.3389/fpls.2020. Kontopoulou, C. K., Liasis, E., Iannetta, P. P. M., Tampakaki, A., and Savvas, 596550 D. (2017). Impact of rhizobial inoculation and reduced N supply on biomass Rufí-Salís, M., Petit-Boix, A., Villalba, G., Sanjuan-Delmás, D., Parada, F., Ercilla- production and biological N2 fixation in common bean grown hydroponically. Montserrat, M., et al. (2020b). Recirculating water and nutrients in urban J. Sci. Food Agric. 97, 4353–4361. doi: 10.1002/jsfa.8202 agriculture: an opportunity towards environmental sustainability and water use Liu, Y. H., Rahman, M., Kwag, J. H., Kim, J. H., and Ra, C. S. (2011). Eco- efficiency? J. Clean. Prod. 261:121213. doi: 10.1016/j.jclepro.2020.121213 friendly production of maize using struvite recovered from swine wastewater Sanjuan-Delmás, D., Llorach-Massana, P., Nadal, A., Ercilla-Montserrat, M., as a sustainable fertilizer source. Asian-Australas. J. Anim. Sci. 24, 1699–1705. Muñoz, P., Montero, J. I., et al. (2018). Environmental assessment of an doi: 10.5713/ajas.2011.11107 integrated rooftop greenhouse for food production in cities. J. Clean. Prod. 177, Long, S. R. (1989). Rhizobium-legume nodulation: life together in the 326–337. doi: 10.1016/j.jclepro.2017.12.147 underground. Cell 56, 203–214. doi: 10.1016/0092-8674(89)90893-3 Sanyé, E., Oliver-Solà, J., Gasol, C. M., Farreny, R., Rieradevall, J., and Gabarrell, Marschner, H. (2002). “Ion uptake mechanisms of individual cells and roots: X. (2012). Life cycle assessment of energy flow and packaging use in food short-distance transport,” in Marschner’s Mineral Nutrition of Higher Plants, purchasing. J. Clean. Prod. 25, 51–59. doi: 10.1016/j.jclepro.2011.11.067 (Cambridge, MA: Academic Press), 6–78. doi: 10.1016/b978-0-08-057187-4. Sanyé-Mengual, E., Martinez-Blanco, J., Finkbeiner, M., Cerdà, M., Camargo, M., 50008-4 Ometto, A. R., et al. (2018). Urban horticulture in retail parks: environmental Massey, M. S., Davis, J. G., Sheffield, R. E., and Ippolito, J. A. (2007). “Struvite assessment of the potential implementation of rooftop greenhouses in production from dairy wastewater and its potential as a fertilizer for organic European and South American cities. J. Clean. Prod. 172, 3081–3091. doi: production in calcareous soils,” in Proceedings of the International Symposium 10.1016/J.JCLEPRO.2017.11.103 on Air Quality and Waste Management for Agliculture, 16-19 September 2007, Sanyé-Mengual, E., Oliver-Solà, J., Montero, J. I., and Rieradevall, J. (2015). (Broomfield, CO). An environmental and economic life cycle assessment of rooftop greenhouse McKey, D. (1994). Legumes and nitrogen: the evolutionary ecology of a nitrogen- (RTG) implementation in Barcelona, Spain. Assessing new forms of urban demanding lifestyle. Adv. Legume Syst. 5, 211–228. doi: 10.1007/s13398-014- agriculture from the greenhouse structure to the final product level. Int. J. Life 0173-7.2 Cycle Assess. 20, 350–366. doi: 10.1007/s11367-014-0836-9 Müller, S., Pereira, P. A. A., and Martin, P. (1993). Effect of different levels of Savvas, D., Ntatsi, G., Vlachou, M., Vrontani, C., Rizopoulou, E., Fotiadis, C., mineral nitrogen on nodulation and N2 fixation of two cultivars of common et al. (2018). Impact of different rhizobial strains and reduced nitrogen supply bean (Phaseolus vulgaris L.). Plant Soil 152, 139–143. doi: 10.1007/BF00016343 on growth, yield and nutrient uptake in cowpea grown hydroponically. Acta Nanjareddy, K., Blanco, L., Arthikala, M. K., Affantrange, X. A., Sánchez, F., Hortic. 1227, 417–424. doi: 10.17660/ActaHortic.2018.1227.52 and Lara, M. (2014). Nitrate regulates rhizobial and mycorrhizal symbiosis Shearer, G., and Kohl, D. H. (1986). N2 -fixation in field settings: estimations based in common bean (Phaseolus vulgaris L.). J. Integr. Plant Biol. 56, 281–298. on natural 15 N abundance. Aust. J. Plant Physiol. 13, 699–756. doi: 10.1016/ doi: 10.1111/jipb.12156 0003-2697(92)90302-N Nanzer, S., Oberson, A., Berger, L., Berset, E., Hermann, L., and Frossard, E. (2014). Shearer, G., and Kohl, D. H. (1989). “Natural 15N abundance as a method The plant availability of phosphorus from thermo-chemically treated sewage of estimating the contribution of biologically fixed nitrogen to N2-fixing Frontiers in Plant Science | www.frontiersin.org 12 May 2021 | Volume 12 | Article 649304 Arcas-Pilz et al. Alternative Fertigation Urban Vertical Agriculture systems: potential for non-legumes,” in Proceedings of the Fourth International Unkovich, M. J., Pate, J. S., Lefroy, E. C., and Arthur, D. J. (2002). Nitrogen Symposium on ‘Nitrogen Fixation with Non-Legumes’, Rio de Janeiro, 23-28 isotope fractionation in the fodder tree tagasaste (Chamaecytisus proliferus) and August 1987, eds F. A. Skinner, R. M. Boddey, and I. Fendrik (Dordrecht: assessment of N2 fixation inputs in deep sandy soils of Western Australia. Funct. Springer Netherlands), 289–299. doi: 10.1007/978-94-009-0889-5_33 Plant Biol. 27, 921–929. doi: 10.1071/pp99201 Shearer, G., and Kohl, D. H. (1993). “Natural abundance of 15N: fractional Weih, M. (2014). A calculation tool for analyzing nitrogen use efficiency in annual contribution of two sources to a common sink and use of isotope and perennial crops. Agronomy 4, 470–477. doi: 10.3390/agronomy4040470 discrimination,” in Nitrogen Isotope Techniques, eds R. Knowles and T. H. Blackburn (Cambridge, CA: Academic Press), 89–125. doi: 10.1016/B978-0-08- Conflict of Interest: The authors declare that the research was conducted in the 092407-6.50009-2 absence of any commercial or financial relationships that could be construed as a Sonneveld, C., and Voogt, W. (2009). Plant Nutrition of Greenhouse Crops. potential conflict of interest. Dordrecht: Springer Netherlands. doi: 10.1017/CBO9781107415324.004 Talboys, P. J., Heppell, J., Roose, T., Healey, J. R., Jones, D. L., and Withers, Copyright © 2021 Arcas-Pilz, Parada, Villalba, Rufí-Salis, Rosell-Melé and Gabarrell P. J. (2016). Struvite: a slow-release fertiliser for sustainable phosphorus Durany. This is an open-access article distributed under the terms of the Creative management? Plant Soil 401, 109–123. doi: 10.1007/s11104-015-2747-3 Commons Attribution License (CC BY). The use, distribution or reproduction in United Nations (2019). World urbanization prospects the 2018 revision. Demogr. other forums is permitted, provided the original author(s) and the copyright owner(s) Res. 12, 197–236. doi: 10.4054/demres.2005.12.9 are credited and that the original publication in this journal is cited, in accordance Unkovich, M. J. (2013). Isotope discrimination provides new insight into biological with accepted academic practice. No use, distribution or reproduction is permitted nitrogen fixatio. New Phytol. 198, 643–646. which does not comply with these terms. Frontiers in Plant Science | www.frontiersin.org 13 May 2021 | Volume 12 | Article 649304