Two experimental units (EU), EU-1 and EU-2, were used. They were built with plastic tanks of 0.60 m x 0.4 m x 0.4 m (length x width x height). Eight aeroponic baskets were installed for each EU on top of EU-1 and EU-2. The aeroponic baskets are made of plastic and grooved with a total volume of 232.26 cm³, filled with Arlite (inert material) for the development of the root. Lilium ‘Tresor’ was selected as the cut flower because it has a commercial market and also, it has shown tolerance to irrigation with water with an electrical conductivity of up to 2.3 dS/m [25]. The experimental units were installed in the experimental station of the Center for Research and Development in Water Resources (CIDERH) in the city of Iquique, in the north of Chile, in the coastal Desert of Atacama, and separated from the outside only by an anti-aphid mesh and without temperature regulation. Figure 1 shows the configuration of the proposed aeroponic unit.

Figure 1. Figure 1. View of the experimental unit used. Measures in meters. Not to scale Source: Own elaboration

Wastewater treated by means of SS-CWs was used as irrigation water. The CWs were fed with filtered wastewater (10 mm) from a nearby treatment plant. EU-1 was irrigated with water composed of an equivalent mixture (50 %) of effluents to two SS-CWs planted with Schoenoplectus americanus, operated in parallel, with two hydraulic retention times (HRT): 3.5 d and 7.5 d. On the other hand, EU-2 was also irrigated by an equivalent mixture (50 %) of effluents from two SS-CWs, but in this case, planted with Cyperus papyrus, operated in parallel, with two HRTs: 3.5 d and 7.5 d. The effluents used for irrigation were manually put in each EU. The effluents used correspond to months 4 to 7 of operation of the SS-CWs. Further details on the CWs experimental system can be seen in Vera et al. [26].

The Lilium ‘Tresor’ flowers were planted in August 2014 and the systems were in operation for 9 weeks. Bulbs with a 12 cm to 14 cm circumference purchased from national producers were used for this purpose. The bulbs were disinfected to prevent fungus formation prior to planting. During plant growth, the height and diameter of the stem were manually measured every week for the 9 weeks of the experiment. The height was measured from the clay surface to the apex with a measuring tape. Diameter was measured at the midpoint of the plant height using a manual caliper. The harvest was made when 50 % of the flower buds showed pigmentation. In the EU-1 and EU-2, irrigation water was manually changed every 7 d, using 13 L for each change. The volume for the water change was defined according to the immersion condition required by the pump used for irrigation (figure 1). Prior to its use in irrigation, the effluent water from the SS-CWs was measured for Chemical Oxygen Demand (COD), Total Suspended Solids (TSS), Total Nitrogen (TN), Total Phosphorus (TP), and Fecal Coliforms (FC). Irrigation was carried out continuously during 24 h. During irrigation, pH and Electrical Conductivity (EC) was measured in the stored water on days 0, 3, and 7 in each EU. After 7 d, the remaining water was assessed in terms of volume and discarded. The Temperature (T) and Relative Humidity (RH) were gathered from the “Diego Aracena” Weather Station at the Iquique airport (Northern Chile) [27].

Samples were filtered (0.45 µm) for the COD, TSS, TN, and TP readings. COD, TN and TP were measured photometrically (Hanna photometer model HI-83214) using the kits: a) COD, HI 93754B-25; b) TN, HI 93767A-50; and c) TP, HI 93763B-50 (analytical methods of adaptations of APHA-AWWA-WEF [28]). The TSS were determined according to APHA-AWWA-WEF [17]. Fecal Coliforms (FC) were determined by multiple tube fermentation [29]. The pH and Electrical Conductivity (EC) were measured with a Hanna portable multiparameter probe model HI-9829. Water volumes were measured with a test tube. Height was measured with a measuring tape and diameter was measured with a manual caliper.

Statistical tests were used to evaluate the differences between: a) development of the plant, comparison between the height and diameter of the stem taking into account the values at the time of harvest, b) variations in pH and EC in the water used for EU-1 and EU-2 on day 0, and c) variations in pH and EC during irrigation, comparing between days 0, 3, and 7. The Wilcoxon test was used for the comparisons of numerals a and b. The Kruskal Wallis test was used for the comparison of numeral c. All statistical analyzes were carried out using the Infostat software with a significance level α = 0.05 [30].

Table 1 presents the quality values of the effluent water from the CWs used as irrigation water, along with the limits of the standards included in the reuse guides. In this regard, COD (> 170 mg/L) is the water quality parameter that presented the greatest difference when compared to the standards in table 1. The TSS, the TN and the TP showed values within the standards in table 1, but they would have restrictions regarding crop and soil type [31]. Therefore, the CWs proposed in Vera et al. [26] would not be sufficient by themselves to achieve the reuse standards in table 1, therefore, more stages of treatment or recirculation must be integrated as it has already been done by other authors in CWs systems to treat wastewater under arid conditions [10], [32]. In addition, it is necessary to integrate some subsequent disinfection process into the proposed CWs systems for FC to reach the standards in table 1. Finally, regarding the nutritional requirements of Lilium regarding the macronutrients nitrogen and phosphorus, this is a plant that does not require large amounts of phosphorus, so the concentration reported in table 1 would be sufficient for its development [33]. Regarding nitrogen, Vera-Puerto et al. [5] indicated that the relationship between the forms of TN present in the treated wastewater (ammonium and nitrate) would not have adverse effects on the development of Lilium.

Table 1.

n = 5 for DQO, SST, NT and PT; n = 3 for CF ^a(Min.-Max.).

During planting, the temperature showed average values between 14 °C and 16 °C, with minimums above 12 °C and maximums below 20 °C. In this regard, the ideal temperature for growing Lilium should have a minimum value of 8 °C to 10 ºC and a maximum value of 23 ºC to 25 ºC [33]. Therefore, the temperature conditions during the planting time were adequate for the development of this crop. Regarding relative humidity, it varied between 55 % and 85 %. The optimal relative humidity for Lilium should be between 80 % and 85 % [35]. However, other cut flowers such as roses have not shown issues with a relative humidity as low as 70 % [36]. This would indicate that the relative humidity during planting would not affect the quality of the flower. On the other hand, Olave et al. [28] showed that the light intensity (luminosity variable) must be controlled to improve the height of the cut flowers cultivated in the Atacama Desert. The reason behind this is that under the arid conditions of this desert, light intensity can exceed 120 klux [23]. However, in this study, luminosity was not controlled, therefore, it is considered as one of the factors that affected the final height reached by the Lilium ‘Tresor’ flowers grown in this work.

Figure 2 presents the average growth per week for EU-1 and EU-2. The average height at harvest (week 9) varied by less than 0.1 m between the two experimental units, with a higher development in EU-1, but this difference was not significant (p > 0.05) when compared with EU-2. Height is defined as one of the most important indicators of commercial quality for this cut flower [37]. Values higher than 0.65 m are considered for export, but lower values are considered for national markets. Thus, the heights obtained in this work, between 0.5 m and 0.6 m, would indicate that the flowers produced can be sold only in the national market [37]. Furthermore, these values for height are similar to those achieved by Schiappacasse et al. [38], who noted heights of 0.59 m for Lilium ‘Dreamland’ and ‘Alhambra’ in controls without shading. Regarding stem diameter, EU-1 and EU-2 were similar, with average values between 0.6 cm and 0.7 cm, similar to those reported by Schiappacasse et al. [38].

Figure 2. Figure 2. Average growth of the plants per week (a) EU-1 (b) EU-2 Source: Vera-Puerto et al. [5].

Between 1.71 L (EU-1) and 2.35 L (EU-2) of water are required to produce a cut flower of Lilium ‘Tresor’ under the study conditions described in this work. These values are more than 50 % higher than those reported by Safi et al. [21], who, when reusing effluents, reported water consumption values between 0.70 L and 1.11 L for a soilless cultivation system managed under arid conditions. The difference between this work and the one developed by Safi et al. [21] could be explained by the salinity of the water and the irrigation mode. Safi et al. [21] reported an EC of 2002 µs/cm, while the EC was above 2300 µs/cm for the water in this study. Furthermore, Safi et al. [21] reported that programmed irrigation was used in their work, while continuous irrigation was carried out in this work. Despite this, other authors such as the International Flower Bulb Centre (IFBC) [35] indicate that for dry conditions (similar to arid conditions), water consumption can be as high as 9 L/m2-d, which means that up to 0.163 L/unit are required on a daily basis. Therefore, taking into consideration a cultivation time of 65 d, the water required to produce a Lilium flower would be 10.6 L. This would indicate that the water used to produce a Lilium flower in this work corresponds to only a 20 % of the value indicated by the IFBC [35], thereby showing the potential advantage of the aeroponic cultivation system in terms of water usage. This would be consistent with the result of Treftz and Omaye [39], who found a reduction of up to 90 % in the use of water for aeroponic lettuce crops when compared to conventional cultivation techniques. Despite this, the results of this work will require further scaling and optimization of irrigation to be more conclusive on the subject.