Water and fertilizer efficiency in a polyculture cropping system under three production systems

Approximately 40% of the water used in intensive agriculture is discarded as ‘drainage,’ which contains high amounts of ions that pollute the environment. This work aimed at investigating polyculture (tomatoes, cucumbers, and lettuce) under three production systems, namely, open (OPS), soil (SPS), and closed (CPS), in which the drainage water from the first crop was used to feed the second crop, and the water from the second crop was used to feed the third crop. The water and fertilizer efficiencies and some physical–chemical properties of the plants and soil were measured. The results showed no significant difference in the yield for polycultures between the CPS and OPS systems. The most efficient system for water use was the CPS, with 54.85 kg m , with a water savings of 55.69% compared to the OPS. The efficiency of fertilizers, such as N, P, K, Ca, and Mg, was statistically higher in the CPS, providing more kilograms of fruit per kilogram of nutrients. The reuse of drainage water in a polyculture not only increased the efficiency of the water and fertilizers but also increased the yields produced per cubic meter of water used, thereby minimizing environmental


INTRODUCTION
In 2009, the Food and Agriculture Organization of the United Nations (FAO) estimated that the global population in 2050 will be 9.3 billion; therefore, greater demands for water and food are expected. A 90% increase in food production due to increased yields from intensive production units is expected, whereas an increase of only 10% is expected due to the expansion of productive land (FAO ). However, the outlook for water use is not very encouraging, especially considering that the agricultural sector is responsible for 70% of the global freshwater withdrawal and more than 90% of the global consumption (FAO ). Therefore, the efficient use of this vital liquid is important. In intensive agriculture, to oxygenate the roots and wash excess salts from hydroponic substrates, the application of up to 40% extra nutrient solution to the plant is common (Schwarz et al. ). These solutions are commonly called 'drainage' and are mostly discharged to the ground, and they contain high amounts of ions, such as NO 3 À , PO 4 À , K þ , Ca 2þ , and Mg 2þ , which cause substantial contamination of the soil and aquifers.
To increase the efficiency of agricultural production and avoid contamination, it is necessary to not only use drainage containing high amounts of nutrients but also measure production in terms of the product produced per surface area In contrast, in the present work, the polyculture of these crops is conducted wherein the drainage water of the first crop (tomato) is used to feed the second crop (cucumber), and the drainage water of the second is used to feed the third crop (lettuce). In this approach, the use of the initial water and the reused water in the second and third crops and the nutrients they contain are maximized. The objective of this study was to determine the WUE and NUE and several physiological variables in a polyculture (tomatoes, cucumbers, and lettuce) under three greenhouse production systems: a CPS, an OPS, and an SPS.

Field characteristics and agronomic conditions
This research was performed in a greenhouse using the indeterminate tomato (Solanum lycopersicum L.) cultivar 'El and for the lettuce, two harvests were made (two production cycles).

Treatments and experimental design
Three types of production systems were used (treatments). The

Soil chemical and physical analyses
The following physical properties were determined: saturation (SAT; %), FC (%), wilting point (WP; %), HC (cm hr À1 ), BD (g cm À3 ), cation exchange capacity (CEC, meq 100 g À1 ), and ECs (dS m À1 ). The chemical properties determined included organic matter (OM; %), pH, and the nutrients (mg kg À1 ) inorganic nitrogen (NO 3 -N) and total P, K, Ca, Mg, and Na. The K, Ca, Mg, and Na contents were analyzed by atomic absorption after wet digestion of the material. P was determined after digestion and calcination.

Plant and fruit analyses
For each crop, the measured physiological variables were stem diameter (SD); plant height; leaf chlorophyll level, which was reported in SPAD units (every 15 days, as measured on the fourth developed young leaf) using a chlorophyll meter (SPAD 502, USA); and total yield (t ha À1 ).
Using a sample of 10 fruits per treatment, the following parameters were measured: equatorial diameter (cm); fruit length (cm); fresh fruit weight (g); content of soluble solids

Water and fertilizer efficiency
Water use efficiency and total net water The WUE (kg m À3 ) was calculated as where Y is the yield (kg ha À1 ), and I is the amount of water used in the crop (m 3 ).
The total net water (NWT; m 3 ha À1 ) was calculated as where W I is the net water applied to the plant (m 3 ), and W d is the reused drainage water (m 3 ).

Fertilizer use efficiency
The use efficiency of fertilizer (NUE, kilogram of fruit per kilogram of nutrient) was calculated as where Y is the total yield (kg), and F is the nutrient used (N, P, K, Ca, and Mg) in kg.

Statistical analysis
Analysis of variance (ANOVA) was performed using OriginPro 8.0 software (USA). Where differences were found among treatments, pairwise comparisons were performed using Tukey's method at the 0.05 probability level.

RESULTS AND DISCUSSION
Yield and fruit variables The highest yields of the production systems were obtained for tomatoes, at 143.41 t ha À1 , in the SPS and cucumbers. at 130.81 t ha À1 , in the CPS. For lettuce, production did not significantly differ between the OPS and CPS, with yields of 68.40 and 62.79 t ha À1 , respectively ( Figure 1). However, the total yield of the polyculture (the average sum of tomato, cucumber, and lettuce yield) significantly differed between CPS and OPS ( Figure 1). These results showed that the drainage water from some crops used to feed other crops following readjustment for nutrients was of good quality and improved yields relative to those under the OPS, in which the drainage was discarded. Tomato fruit weight was higher in the SPS system than in the other systems, possibly because the fruit required a longer time to mature in this system than in the other systems; this longer time was due to the low temperature in the root zone (data not shown), which delayed ripening in this system relative to that in the other systems of production. The number of fruits did not significantly differ between the SPS and CPS systems, but the number was significantly higher in both of these systems than in the OPS (Table 1). The data represent the mean ± standard error (n ¼ 12). Different letters within a column indicate significant differences (p < 0.05).   Table 3 shows that no significant differences in Brix and firmness were found between the conventional production system (OPS) and the CPS.

Water and fertilizer efficiency
For cucumbers and lettuce, the highest WUE values among the systems were obtained in the CPS. In addition, regarding the average total WUE of the three crops (polyculture), the CPS yielded an average of 55.69% water savings relative to the OPS and a 34.70% savings relative to the SPS (Table 4). Furthermore, the NWT applied to the polyculture was greater under the OPS than in the other systems in this investigation, which was expected since the OPS does not benefit from the drainage and the SPS underuses the water because much of the solution (up to 50%) infiltrates the subsoil. Thus, the best results were obtained with the CPS, with a net water application that was 23.46% and 21.01% lower than that in the OPS and SPS, respectively (Table 4).
These results are very important because water scarcity The data represent the mean ± standard error (n ¼ 12). Different letters within a column indicate significant differences (p < 0.05). The data represent the mean ± standard error (n ¼ 10). Different letters within a column indicate significant differences (p < 0.05). The data represent the mean ± standard error (n ¼ 8). Different letters within a column indicate significant differences (p < 0.05). The polyculture is the mean ± standard error of the tomato, cucumber, and lettuce variables within a production system.
has become a major concern, posing serious threats to both food security and the economy in many parts of the world (Alam ). However, for producers to adopt water-saving technologies, understandable technical information is needed (Negatu & Parikh ). The nutrient measurement equipment for drainage used in the present study is economically accessible to producers; thus polyculture can easily be adapted for use in any production area.
However, an NUE of 141.34 kg of tomatoes was obtained for each kilogram of N used in the CPS; this value was higher than that in the OPS but not that in the SPS. In contrast, the highest NUE in cucumbers and lettuce was obtained in the CPS. The NUE for nitrogen in the cucumber crop was 75.73% higher in the CPS than in the OPS and 27.72% higher in the CPS than in the SPS. In lettuce, in the CPS, the NUE of N was 17.85% higher than that in the OPS and 51.31% higher than that in the SPS. Furthermore, for polyculture, the CPS yielded the highest NUE of N, with an average 40.89% higher than that in the OPS and 26% higher than that in the SPS (Table 5). The primary reason for the lower NUE in the SPS is that 50-70% of the total N applied to the soil is lost via runoff, denitrification, volatilization, and leaching (Montemurro & Diacono ).
Various techniques to improve the efficiency of N have been developed, such as for grafts in tomato plants (Djidonou et al. ). However, it is problematic to use drainage in an OPS because drainage is known to contain several macroelements that are potentially harmful, and this practice risks the agricultural sustainability of the system and the health of the environment (Yasuor et al. ). At the polyculture level, the NUE of P in the CPS was 9% higher than that in the SPS and 21.87% higher than that in the OPS. For K, the greatest NUE was obtained with the CPS, being 22.92% higher in this system than in the SPS and 37.22% higher in this system than in the OPS. Calcium was the element with the highest use efficiency, with an NUE higher in the CPS than in the SPS (216%) and the OPS (249%). Similar results were obtained for magnesium, with the CPS yielded the highest NUE among the systems (Table 5).
Physical and chemical properties of the soil Table 6 shows that the improper use of fertilizers can affect the physical-chemical structure of the soil, leaving large crops. Therefore, CPS is a good production method that not only can enhance the efficiencies of water and fertilizer consumption but also shows evidence of suitability regarding other indicators, such as oxygen concentration. On average, in the three production systems, dissolved oxygen levels higher than 8 ppm were observed in the nutrient solution ( Figure 2). These levels are higher than  Figure 3 | Concentration of Na (mean ± standard error, n ¼ 50) in the input nutrient solution by crop and by production system. Bars with different letters within a production system are significantly different (p < 0.05).