Resource Utilization Efficiencies in a Closed System with Artificial Lighting during Continuous Lettuce Production

New plant production systems with a low environmental impact (or a high resource utilization efficiency) are necessary for urban agriculture development. This study explores the benefits of closed plant production systems (CPPSs) with artificial lighting using a commercial CPPS at Osaka Prefecture University. Lettuce plants were produced continuously as a model for analyzing resource consumption rates, such as electricity, labor, water, and CO2, over two years. Monthly consumption rates of electricity, labor, water, and CO2 increased with the increase in the monthly production rate of the lettuce. The utilization efficiencies (=output/input) of electricity, energy, water, and CO2 were 1.0%, 1.0%, 4.0%, and 32.6%, respectively. If the commercial CPPS maintains the monthly production rate at a higher level, the energy utilization efficiency will increase. The number of air exchanges in a commercial CPPS should decrease to increase water and CO2 utilization efficiencies. Reusing water drained from the air conditioning system and employing a closed loop in the nutrient supply system also contribute to increasing the water utilization efficiency and lowering the environmental impact. Although a commercial CPPS still requires further improvements, it may be a good crop production system for urban agriculture provided resource utilization efficiencies improve.


Introduction
Two-thirds of the population are expected to live in urban areas by 2050 [1]. While urban planners understand the importance of urban agriculture [2], people living in urban areas require a new and sustainable system to be established for supplying food. Urban agriculture offers reduced environmental impact, particularly for logistics, as the distances between production and consumption sites are shorter [3]. However, crop production systems used in urban agriculture must be highly productive [4].
A potential crop production system for urban agriculture is the closed plant production system (CPPS), which uses artificial lighting [5][6][7][8][9][10]. The structure of this system allows for increased resource utilization efficiency [7]. As the inside environment can be controlled without any weather interference, crop productivity can also be increased [11].
The energy balance [12,13], water balance [14], and CO 2 balance [15,16] in CPPS systems have been previously analyzed. Moreover, researchers have also studied resource utilization efficiencies using mathematical models, analysis of energy consumption [17], transpiration, and energy balance of the crop community [4]. Although a study has compared resource use among large-and small-scale greenhouses and CPPSs [18], information about actual resource use in commercial CPPSs is still limited.
of the crop community [4]. Although a study has compared resource use among large-and smallscale greenhouses and CPPSs [18], information about actual resource use in commercial CPPSs is still limited.
This study measured electricity, labor (work hours), water, and CO2 consumption rates over two years in a commercial CPPS at Osaka Prefecture University. In this regard, lettuce plants were continuously produced. To further improve the commercial CPPS, utilization efficiencies of each resource using measurements and estimations were evaluated.

Plant Production System
For this study, a research facility at Osaka Prefecture University (approximate floor area of 1300 m 2 ) was used as a commercial CPPS model (Figure 1). The commercial CPPS consisted of two transplant production rooms and a cultivation room, workroom, pre-cooling room, and cleaning room. Approximately 12,000 LED lamps (GreenPower production module DR/W/FR, Philips Lighting Japan GK, Tokyo, Japan) were used for growing the crops. Multi-shelves (17 layers in the transplant production room and 16 or 18 layers in the cultivation room; the size of each layer: 1.5 × 27 m) were installed in the commercial CPPS for a maximum crop production rate of 6656 crops day −1 with a planting density of 34.3 plants/m 2 . Air-to-air heat-exchange-type heat pumps were installed to cool the air inside the transplant production and cultivation rooms. Hydroponic systems with the nutrient film technique (NFT) were used to apply the nutrient solution to crops in the commercial CPPS.

Plant Materials and Culture Conditions
Lettuce (Lacutuca sativa L. cv. Frillice and Flarebell) plants were produced in the commercial CPPS continuously. Although culture conditions and nutrient solutions were amended by facility managers based on crop growth and development, the typical growing conditions are described below.
In the transplant production room germination chamber, photosynthetic photon flux density (PPFD) and air temperature were set at 0 μmol m −2 s −1 (continuous darkness) and 24 °C, respectively. After germination, PPFD and air temperature were set at approximately 120 to 200 μmol m −2 s −1 with a 15 or 16 h photoperiod and 24/21 °C (light-/dark period). CO2 concentrations inside the transplant production rooms were maintained at 800 μmol mol −1 , except for in the germination chamber. The nutrient solution containing N, P, and K at 130, 20, and 150 mg L −1 , respectively, with a pH of 4.7, was applied during each growth stage. Detailed information on cultural conditions was provided in our previous report [19].

Plant Materials and Culture Conditions
Lettuce (Lacutuca sativa L. cv. Frillice and Flarebell) plants were produced in the commercial CPPS continuously. Although culture conditions and nutrient solutions were amended by facility managers based on crop growth and development, the typical growing conditions are described below.
In the transplant production room germination chamber, photosynthetic photon flux density (PPFD) and air temperature were set at 0 µmol m −2 s −1 (continuous darkness) and 24 • C, respectively. After germination, PPFD and air temperature were set at approximately 120 to 200 µmol m −2 s −1 with a 15 or 16 h photoperiod and 24/21 • C (light-/dark period). CO 2 concentrations inside the transplant production rooms were maintained at 800 µmol mol −1 , except for in the germination chamber. The nutrient solution containing N, P, and K at 130, 20, and 150 mg L −1 , respectively, with a pH of 4.7, was applied during each growth stage. Detailed information on cultural conditions was provided in our previous report [19].

Operations in the Research Facility
Seeds were sown on urethane foam with 300 holes (30 × 60 cm) in the transplant production room of the commercial CPPS and placed in the germination chamber for two days. After germination, trays were transferred to shelves and cultured until cotyledon expansion. Subsequently, the transplants were automatically transferred to the 153-hole panel (60 × 90 cm) and moved to another transplant production room for 12 d (cv. Frillice) or 14 d (cv. Flarebell). After establishment, transplants were manually planted in 30-hole panels (70 × 125 cm) and grown for 17 d in the cultivation room.
After cultivation, the 30-hole panels were transferred from the cultivation room to the workroom via transferring machines and a conveyor automatically. Crops were harvested manually with deteriorating and/or surplus leaves removed. After harvesting, crops were packed within cartons with plastic film and stored in the pre-cooling room until shipment to customers. During harvesting, the trays and panels used for transplant production and cultivation were transferred to the cleaning room, where they were washed for the next use.

Utilization Efficiencies
In this paper, utilization efficiencies of electricity, energy, water, and CO 2 were defined and estimated as the ratio of resource input to output. Utilization efficiencies were estimated based on monthly measured values.

Electricity Utilization Efficiency
Electricity is consumed by lights, heat pumps, and other equipment in the commercial CPPS. Electricity consumed by lighting is converted into light and heat. The light is required for crop photosynthesis; chemical energy is stored in the crops. The heat generated by lights and other equipment is removed from the commercial CPPS with heat pumps to maintain suitable air temperature for crop growth. The energy utilization efficiency in the commercial CPPS (η E ) was, therefore, estimated using the following equation: where E d : Chemical energy contained in the dry matter of marketable heads during a month (GJ), E e : Monthly electric energy consumption (GJ), k e : Energy content in dry matter, M d : Monthly cumulative value of dry weight of marketable heads in commercial CPPS (t). A more typical value for energy content in dry matter is 19 to 20 kJ g −1 [20]. Thus, we employed a value of 20 kJ g −1 .

Energy Utilization Efficiency
In cases where human labor was also considered as an energy input for the commercial CPPS, Equation (1) can be modified as follows: where E h : Energy provided by labors, k h : Energy per hour provided by human labor (0.0023 GJ h −1 ) [21], H: Monthly work hours of laborers in the commercial CPPS (h).

Water Utilization Efficiency
In the commercial CPPS, water is used for crop irrigation. If the number of air exchanges is 0 h −1 and all water drained from the heat pumps is collected and reused for irrigation, 100% water recycling can be achieved with no impact on the outside environment caused by erosion of fertilizers. In a commercial CPPS, however, water is also used to maintain sanitary conditions (washing hands, cleaning trays, panels, and other tools). To increase water utilization efficiency, minimizing water used for sanitation is essential. In this paper, water utilization efficiency in the commercial CPPS (η W ) was evaluated for overall water inputs (irrigation and sanitary) and outputs using the following equation: Agronomy 2020, 10, 723 4 of 11 where W f : Water contained in the marketable heads in a month (t), W c : The monthly water consumption in the closed system (t), and M F : The monthly cumulative value of the fresh weight of marketable heads in the commercial CPPS (t).

CO 2 Utilization Efficiency
CO 2 is generally supplied in a commercial CPPS to maintain CO 2 concentration at an appropriate level for crop photosynthesis (c.a., 1000 µmol mol −1 ). If the number of air exchanges in the commercial CPPS is 0 h −1 , all CO 2 supplied is absorbed by crops, indicating 100% CO 2 utilization efficiency. However, if the number of air exchanges is higher than 0 h −1 , CO 2 supplied to the commercial CPPS has escaped from the closed system as the CO 2 concentration inside the commercial CPPS is higher than the concentration outside. In this paper, the CO 2 utilization efficiency of the closed system (η C ) was estimated using the following equation: where C d : The monthly cumulative value of CO 2 absorbed by crops (t), C c : The monthly CO 2 consumption in the closed system (t), and k c : The ratio of molecular weights of sugar (CH 2 O) and CO 2 (0.68) [22].

Measurements and Estimations
The monthly consumption rates of electricity, labor (work hours), water, and CO 2 were determined in the commercial CPPS. Monthly consumption rates of electricity, labor, water, and CO 2 per fresh weight of marketable head were estimated based on the crop growth and energy, water, and CO 2 contained in each marketable head. These values were applied to Equations (1)-(4) to estimate the utilization efficiencies of electricity, energy, water, and CO 2 .
The monthly consumption rate of electric energy in the commercial CPPS was measured at a transformer substation at Nakamozu Campus, Osaka Prefecture University, and laborer work hours were collected and summarized monthly by a time recorder system in the commercial CPPS. The data on work hours collected from 0 to 181 days after the beginning of the operation have already been reported in our previous paper [19]. The monthly water consumption was determined using a water meter installed in the commercial CPPS, while the CO 2 consumption rate of the plant production system was estimated monthly based on the number of gas cylinders ordered during the research period.
The monthly production rate of lettuce in the commercial CPPS was estimated by multiplying the monthly number of plants harvested by the fresh weight of the marketable head. The number of lettuce plants harvested monthly was recorded manually throughout the two-year research period. The fresh and dry weights were measured at 37 (cv. Frillice) or 39 days (cv. Flarebell) after sowing using an electric balance (PL602, Mettler Toledo, Greifensee, Switzerland). The fresh weight of the marketable head was determined after trimming deteriorated and surplus leaves. To determine fresh and dry weights and the fresh weight of the marketable head, lettuce plants were irregularly sampled 28 times for Frillice (n = 104) and 31 times in Flarebell (n = 116) during the research period. At each sampling period, 3-5 plants were harvested.

Statistical Analyses
Regression analyses (using R package 'lm') were conducted to clarify the effect of the monthly production rate on the resource consumption rates. The effects of the monthly production rate on resource consumption per fresh weight of marketable head and resource utilization efficiencies were also subjected to regression analyses. The regression analyses were conducted using R statistical software 3.6.0 [23].

Results
For this study, the monthly electricity consumption rate ranged from 428 to 1080 GJ and increased with the monthly production rate in the commercial CPPS (y = 47.9x + 234, R 2 = 0.762, Figure 2a and Table S1). As the monthly production rate increased, work hours increased from 1360 to 2610 h (y = 87.7x + 1090, R 2 = 0.774, Figure 2b and Table S1). Monthly consumption rates of water and CO 2 also increased with the monthly production rate (y = 23.3x + 10.4, R 2 = 0.922, Figure 2c and Table S1; y = 0.204x − 0.403, R 2 = 0.793, Figure 2d and Table S1).
Agronomy 2020, 10, x FOR PEER REVIEW 5 of 11 also subjected to regression analyses. The regression analyses were conducted using R statistical software 3.6.0 [23].
The fresh and dry weights of the two lettuce cultivars are shown in Table 1. Based on these measurements, the electricity utilization efficiency was estimated to be 0.79% to 1.25% and tended to increase with the monthly production rate (y = 0.0335x + 0.628, R 2 = 0.531, Figure 4a). A similar trend was observed for energy utilization efficiency ( Figure 4b); however, the water and CO2 utilization  Table S1.
The monthly electricity consumption rate per fresh weight of marketable head tended to decrease as the monthly production rate increased (y = −59.0x + 2360, R 2 = 0.570, Figure 3a and Table S2). Similarly, work hours per fresh weight of marketable head decreased as the monthly production rate increased (y = −0.721x + 19.9, R 2 = 0.840, Figure 3b and Table S2). The monthly consumption of water and CO 2 , however, was not affected by the monthly production rate (y = −0.163x + 26.3, R 2 = 0.0652, Figure 3c and Table S2; y = 0.000344x + 0.126, R 2 = 0.129, Figure 3 and Table S2).
The fresh and dry weights of the two lettuce cultivars are shown in Table 1. Based on these measurements, the electricity utilization efficiency was estimated to be 0.79% to 1.25% and tended to increase with the monthly production rate (y = 0.0335x + 0.628, R 2 = 0.531, Figure 4a and Table  S3). A similar trend was observed for energy utilization efficiency (Figure 4b and Table S3); however, the water and CO 2 utilization efficiencies were not affected by the monthly production rate (y = 0.0230x + 3.72, R 2 = 0.0548, Figure 4c; y = −0.755x + 41.0, R 2 = 0.144, Figure 4d; Table S3).

Discussion
To adopt the commercial CPPS for urban agriculture, resource utilization efficiency must be increased to minimize the environmental impact of crop production on urban areas. To improve the commercial CPPS for adaptation to urban agriculture, we determined the resource consumption rates and methods for increasing utilization efficiencies within the commercial CPPS.

Discussion
To adopt the commercial CPPS for urban agriculture, resource utilization efficiency must be increased to minimize the environmental impact of crop production on urban areas. To improve the commercial CPPS for adaptation to urban agriculture, we determined the resource consumption rates and methods for increasing utilization efficiencies within the commercial CPPS.  The electricity consumption rate increased with the monthly production rate (Figure 2a). Generally, the electricity used by lighting increased linearly with the monthly production rate, and  Table S2. The electricity consumption rate increased with the monthly production rate (Figure 2a and Table S1). Generally, the electricity used by lighting increased linearly with the monthly production rate, and the electricity used by other equipment, particularly air conditioning, mainly caused the residual between the linear regression line and observations. The electricity used by air conditioning was affected by the air temperature outside the plant production system [24,25], the cooling load for air conditioning [24], and the water vapor pressure deficit inside the plant production system [25]. However, our results showed a linear relationship between the electricity consumption rate and the monthly production rate, probably because the electricity used for lighting was three times or more of those used by the other equipment [12,26,27].
Similar to the trend of monthly electricity consumption (Figure 2a and Table S1), the monthly consumption rates of labor, water, and CO 2 increased linearly with the increase in the monthly production rate in the commercial CPPS (Figure 2b-d and Table S1). Monthly electricity and labor consumption rates were lower when the monthly production rate was higher because the electricity and energy utilization efficiencies increased with the monthly production rate (Figure 4a,b and Table S3). This indicates that a synergistic effect on the monthly electricity and labor consumption rates can be obtained at the higher monthly production rate. Moreover, no similar effect was observed for water and CO 2 in the commercial CPPS because utilization efficiencies of water and CO 2 were not affected by the monthly production rate (Figure 4c,d and Table S3) Theoretically, synergistic effects could also be obtained in the monthly consumption rates of water and CO 2 . Leakage of water and CO 2 from the commercial CPPS may be the reason that these effects were not observed in the current situation.
Agronomy 2020, 10, x FOR PEER REVIEW 7 of 11 the electricity used by other equipment, particularly air conditioning, mainly caused the residual between the linear regression line and observations. The electricity used by air conditioning was affected by the air temperature outside the plant production system [24,25], the cooling load for air conditioning [24], and the water vapor pressure deficit inside the plant production system [25]. However, our results showed a linear relationship between the electricity consumption rate and the monthly production rate, probably because the electricity used for lighting was three times or more of those used by the other equipment [12,26,27]. Similar to the trend of monthly electricity consumption (Figure 2a), the monthly consumption rates of labor, water, and CO2 increased linearly with the increase in the monthly production rate in the commercial CPPS (Figure 2b-d). Monthly electricity and labor consumption rates were lower when the monthly production rate was higher because the electricity and energy utilization efficiencies increased with the monthly production rate (Figure 4a,b). This indicates that a synergistic effect on the monthly electricity and labor consumption rates can be obtained at the higher monthly production rate. Moreover, no similar effect was observed for water and CO2 in the commercial CPPS because utilization efficiencies of water and CO2 were not affected by the monthly production rate (Figure 4c,d). Theoretically, synergistic effects could also be obtained in the monthly consumption rates of water and CO2. Leakage of water and CO2 from the commercial CPPS may be the reason that these effects were not observed in the current situation. The average monthly electricity consumption rate per fresh weight of marketable head was 70.8 GJ/kg (= 19.7 kWh/kg) (Figure 3a). The monthly labor, water, and CO2 consumption rates per fresh  Table S3. The average monthly electricity consumption rate per fresh weight of marketable head was 70.8 GJ/kg (=9.7 kWh/kg) (Figure 3a and Table S2). The monthly labor, water, and CO 2 consumption rates per fresh weight of marketable head were 11.6 min/kg, 24.4 L/kg, and 0.165 kg/kg, respectively (Figure 3b-d and Table S2). Little information on these data are currently available in the literature. Thus, we could not make direct comparisons to other studies. Accumulating these data will be essential for adapting commercial CPPSs to urban agriculture, as well as for minimizing the environmental impact of crop production.
In this study, the electricity utilization efficiency was 0.74% to 1.25% (Figure 4a and Table S3). These values were almost comparable to or higher than the values reported in our previous study (0.6%) [13]. Based on crop growth data shown in Table 1, the estimated electricity consumption per dry weight of marketable head ranged from 1330 to 2240 MJ kg (dry weight) −1 . This was almost the same or slightly higher than the reported value obtained using the mathematical model (1410 MJ kg (dry weight) −1 ) [4]. When the production rate was higher, the electricity consumption per dry weight was lower (data not shown) as the ratio of electricity consumption for the other equipment (such as air conditioners, pumps, and fans) was higher at a lower monthly production rate. Hence, maintaining the monthly production rate at a higher level is essential for obtaining lower electricity consumption per dry weight, as well as higher electricity utilization efficiency.
As the contribution of energy from labor to energy utilization efficiency was relatively small compared to electricity, the value of electricity utilization efficiency was almost equal to that of energy utilization efficiency (Figure 4a,b and Table S3). However, labor cost is one of the major costs for producing crops in the CPPS. Thus, from an economic point of view, work hours should be reduced.
Theoretically, the CPPS could achieve 100% water utilization efficiency [7]. Water utilization efficiency increased when the number of air exchanges was lower [28]. Hence, a high water utilization efficiency (90%) was obtained in the laboratory-level CPPS [14]. However, in this study, the water utilization efficiency averaged only 4% (Figure 4c). The primary reason for this low water utilization efficiency was that water was used not only for crop cultivation but also for sanitation of the commercial CPPS. In this research, the water utilization efficiency was not affected by the monthly production rate, indicating that a significant amount of water used for sanitation escaped from the commercial CPPS. According to our preliminary examination, approximately only 30% of water was used for cultivation in the commercial CPPS. Thus, minimizing water use for sanitation is essential for increasing water utilization efficiency in the commercial CPPS.
In this research, all water drained from the air conditioners was discarded, which also lowered the water utilization efficiency in the commercial CPPS. If the water drained from the air conditioners was reused for irrigation, the water utilization efficiency could be increased [14]. Irregular replacement of the nutrient solution also decreased the water utilization efficiency in the commercial CPPS. As employing a closed-loop for the nutrient solution contributes to decreasing the environmental impact [29][30][31][32][33], replacing the nutrient solution should be minimized to increase the water utilization efficiency, as well as to reduce the environmental load. These adjustments will be beneficial for adapting the commercial CPPS to urban agriculture.
Similar to water utilization efficiency, CO 2 utilization efficiency in the commercial CPPS is strongly affected by the number of air exchanges [4]. It has been reported that a high level of CO 2 utilization efficiency (90%) can be achieved in a closed plant production system because of its airtight structure [16]. However, our result was lower than the reported value (32.4% on average, Figure 4d), because the production rate was lower, and the number of air exchanges was higher, than in an ideal situation [15].
In this study, the number of air exchanges was significantly higher than those in the previous two works (roughly estimated value was 0.2 h −1 ). Hence, the CO 2 utilization efficiency did not increase with the monthly production rate, because the amount of escaped CO 2 was significantly higher than that in the previous literature. Although a completely airtight structure is ideal for increasing CO 2 utilization efficiency, its establishment was technically infeasible [34]. In future studies, an acceptable Agronomy 2020, 10, 723 9 of 11 number of air exchanges should be determined, both for increasing the CO 2 utilization efficiency in the commercial CPPS, as well as for decreasing the cost for CO 2 .

Conclusions
To investigate the feasibility of using a CPPS for urban agriculture with minimal environmental impact, this paper determined the consumption rates of resources, such as electricity, labor, water, and CO 2 , in a commercial CPPS at Osaka Prefecture University. From these data, resource utilization efficiencies were estimated, and methods for increasing resource utilization efficiencies were discussed.
The electricity utilization efficiency in the commercial CPPS was almost comparable to or slightly higher than the reported value [13]. The electricity and energy (electricity + labor) utilization efficiencies increased with the monthly production rate. Electricity consumption per dry weight was also comparable to the value obtained using a mathematical model [4]. The electricity and energy utilization efficiencies can be maintained at a high level if a high monthly production rate is also maintained.
On the other hand, the water and CO 2 utilization efficiencies in the commercial CPPS were lower than the reported values. Approximately 70% of water in the commercial CPPS was used for sanitation (washing hands, cleaning equipment, and trays) and was wasted. The water utilization efficiency would increase if wastewater was reduced. Reusing water drained from the air conditioning system and employing a closed loop in the nutrient supply system would also contribute to increasing the water utilization efficiency. The high number of air exchanges of the CPPS also lowered the water and CO 2 utilization efficiencies in the commercial CPPS. To increase the water and CO 2 utilization efficiencies, the number of air exchanges should be minimized.
To adopt the commercial CPPS for use in urban agriculture, further improvements should be achieved. However, the commercial CPPS is a potential candidate for the urban agriculture crop production system because, theoretically, the environmental impact can be minimized during crop production. To completely exploit the potential of CPPS, it is necessary to consider the basic design more thoroughly.
Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4395/10/5/723/s1, Table S1: Monthly production rate of lettuce and monthly consumption rates of electricity, labor (work hours), water, and CO 2 , Table S2: Monthly consumption rates of electricity, labor (work hours), water, and CO 2 per fresh weight of marketable lettuce head, Table S3: Utilization efficiencies of electricity, energy (electricity + labor), water, and CO 2 .