Effects of Solar Radiation on Leaf Development and Yield of Tuberous Roots in Multilayered Sweet Potato Cultivation

The purpose of this study was to develop a novel method to dramatically improve the production efficiency of sweet potatoes (Ipomoea batatas (L.) Lam.) by elucidating the effect of solar radiation stress on the growth of sweet potato in a multilayer cultivation system. Twenty-five pots planted with sweet potato vine seedlings were arranged in three layers and cultivated for 160 days while supplying liquid fertilizer to the root zone. While solar radiation in the middle and lower layers decreased to 69% and 45% of that in the upper layer, respectively, the yield of tuberous roots was 0.89 kg/pot in the upper layer, 0.79 kg/pot in the middle layer, and 0.66 kg/pot in the lower layer. As a result, the productivity of tuberous roots reached 10.5 kg/m2, 4.4 times that of conventional farming. On the other hand, the amounts of leaves and stems increased in the lower layer than in the upper layer, and the biomass energy yield (photosynthetic efficiency) was 2.8% in the upper layer, 3.7% in the middle layer, and 5.1% in the lower layer. Leaves in the lower layer with less solar radiation had a lower polyphenol content and increased the amounts of low-brightness leaves. In contrast, the upper leaves were found to contain more polyphenols and have brighter, smaller leaves. These results suggest that the yield can be further increased by optimizing solar radiation stress by using the multilayer cultivation method.


Introduction
Sweet potato (Ipomoea batatas) is an important root vegetable cultivated in temperate and tropical zones, especially in Asia and Africa [1]. Storage roots of sweet potato contain relatively high amounts of carbohydrates that support the demand for food in developing countries [2,3]. In recent years, sweet potato has been also evaluated as a candidate for biofuel production [4,5].
Sweet potatoes store a lot of starch in their tuberous roots, and under the climatic conditions in Japan, calorie production per area reaches 3170 kcal/m 2 with a normal yield of 2.4 kg/m 2 , compared to 1900 kcal/m 2 for rice (yield 0.54 kg/m 2 ), 1500 kcal/m 2 for wheat (0.44 kg/m 2 ), and 1400 kcal/m 2 for sweet corn (1.5 kg/m 2 ) [6]. In addition, the leaves and stems of sweet potatoes are also consumed as edible vegetables [7]. Such edible biomass can be converted to methane in high yield by anaerobic digestion [8,9]. Therefore, mass production of sweet potatoes as a resource crop to supply biomethane for use in power generation and heat is gaining attention as an alternative to fossil fuels.
Although woody biomass is generally considered a carbon-neutral resource, annual growth in Japan's 250,500 km 2 of forest is about 68 million m 3 [10], and if we assume that 50% of solids have 22 MJ/kg of heat, the yield of photosynthesis is very low, at about 0.06% of annual solar radiation (assuming an average of 5000 MJ/m 2 y).
The current total energy quantity of 5.3 billion m 3 of forest biomass in Japan [10] is estimated to be 58 trillion MJ, equivalent to 3 years of fossil fuel consumption in Japan [11].
In other words, if all fossil fuels are replaced with domestic woody fuels, they will run out in 3 years, and even if planted there, it will take more than 50 years to regenerate.
Moreover, the thermal power generation efficiency of woody fuel is low, ranging from 25 to 30% [12], which is lower than that of coal fuel, which has a power generation efficiency of 43% [11]. As a result, by converting the fuel from coal to woody biomass, CO 2 emissions per unit of electricity generated increased to 140%-170%. In other words, woody biomass in Japan is a fuel that can neither achieve carbon neutral nor reduce greenhouse gases.
On the other hand, all of the CO 2 emitted in the process of producing biogas from sweet potatoes and consuming biomethane can be absorbed every year in the process of regenerating sweet potato fuel by the high photosynthetic yields. This allows for sweet potato biomethane to remain carbon neutral, so no amount of combustion will increase the amount of CO 2 stored in the atmosphere. Therefore, replacing natural gas, coal, and oil with this biomethane is expected to significantly reduce CO 2 emissions from fossil fuels.
In Germany, biogas power generation using dent-corn and other resource crops is widespread, and the share of biogas power generation in the total electricity has increased to 6% [13]. On the other hand, 66% of Japan's 377,727 km 2 land area is mountainous, with only 43,490 km 2 of agricultural land, or 26% of that of Germany. In addition, half of the agricultural land is devoted to rice cultivation, which is a staple food, making it difficult to mass produce resource crops.
Therefore, in order to mass produce resource crops that can replace fossil fuels, it is necessary to develop new cultivation methods that dramatically increase the biomass production per unit area. From this perspective, we studied the cultivation methods that enhance the photosynthetic efficiency of sweet potatoes [14].
Thus far, the yield of tuberous roots of sweet potatoes grown under rhizosphere irrigation has been found to reach 1.5 kg when grown in 21 cm top diameter vinyl pots filled with 4.5 L of horticultural soil [15]. The results suggest an increase in productivity far beyond the average yield of 2.4 kg/m 2 [6,16] for open-field cultivation and indicate that soil volume and land area are not essential for sweet potato production efficiency.
In addition, the photosynthetic efficiency of sweet potatoes tended to decrease under strong sunlight conditions [17][18][19][20], suggesting that solar energy cannot be fully utilized by conventional farming methods.
Based on these growth characteristics, the authors devised a multilayer cultivation method to reduce the intensity of solar radiation by increasing the spread area of leaves and decreasing the angle of the light reception of leaves, conducted cultivation tests under solar radiation, and found that the photosynthetic efficiency per field greatly improved.
Therefore, the purpose of this study was to clarify the effect of solar radiation stress on the growth of sweet potato and to develop a new method for dramatically improving production efficiency. Figure 1 shows the results of the Top (top layer), M in (inside of middle layer) and B in (inside of bottom layer) illuminance measurements in the three-layer cultivation system for a week from 22 July, which is known as the hottest week with the strongest intensity of ultraviolet rays (UV) in a typical year of the meteorological statistics.

Insolation Rate in Three-Layers Cultivation System
In the same way, the relative values of each layer were obtained by integrating the illuminances measured from July to August during the peak growing season. Using the integrated value of Top as 100%, the solar radiation (SR) for each layer was 85% for M out (outside of middle layer), 69% for M in , 73% for B out (outside of bottom layer), and 45% for B in . The illuminance on the east side in the morning and on the west side in the afternoon was the same as the Top, so the cumulative illuminance was higher on the outside than on the inside. These relative values were used to calculate the amount of SR in each layer. As a result, the estimated SR in the upper layer was 2480 MJ/m 2 , 2108 MJ/m 2 in M out , 1711 MJ/m 2 in M in , 1810 MJ/m 2 in B out , and 1116 MJ/m 2 in B in .
Bin. The illuminance on the east side in the morning and on the west side in the afternoon was the same as the Top, so the cumulative illuminance was higher on the outside than on the inside. These relative values were used to calculate the amount of SR in each layer. As a result, the estimated SR in the upper layer was 2480 MJ/m 2 , 2108 MJ/m 2 in Mout, 1711 MJ/m 2 in Min, 1810 MJ/m 2 in Bout, and 1116 MJ/m 2 in Bin.  Figure 2 shows the estimated SR, the average fresh weight of tuberous roots (TR), and leaves and stems (LS) per pot harvested in each layer. The yield of Beniharuka (BH) was 453 g/pot for Min at 45% SR and 859 g/pot for the Top layer. On the other hand, for Suikenkintoki (SK), a peak of 903 g/pot was obtained at 83% SR of Mout, while it was almost equivalent to 893 g/pot by 100% SR of Top. The yield of SK was reduced to 662 g/pot in Min at the lowest SR.

Productivities of Sweet Potatoes in Three-Layer Cultivation Systems
In contrast, both cultivars had smaller amounts of leaves and stems in the top layer at 264 g/pot for BH and 303 g/pot for SK, and the amount increased significantly to 391 g/pot for SK in the lower layer with lower SR. Figure 3 shows the dry biomass weight per pot, the dry weight ratio of tuberous roots (TR) to leaves and stems (LS), and energy yield at each level of solar radiation. In both cultivars, the amount of dry biomass increased as the amount of solar radiation increased. The dry weight ratio of TR/LS also increased with a higher solar radiation. These results indicate that the higher the amount of solar radiation, the more likely the amount of tuberous roots is to increase.
On the other hand, the energy yield based on the amount of SR in each layer increased in the lower layer with less solar radiation. The yield of SK rose to 5.1% at Min with 45% solar radiation, while the Top layer was 2.8%. Similarly for BH, Min yield increased to 3.5%, while the Mout and Top yields were lower at 2.7% and 2.4%, respectively.
These results clearly show that the three-dimensional distributed use of sunlight increased the biomass yield of sweet potato per field area. On the other hand, even in threelayer cultivation, the top layer showed that some of the solar radiation was wasted. Bin. The illuminance on the east side in the morning and on the west side in the afternoon was the same as the Top, so the cumulative illuminance was higher on the outside than on the inside. These relative values were used to calculate the amount of SR in each layer. As a result, the estimated SR in the upper layer was 2480 MJ/m 2 , 2108 MJ/m 2 in Mout, 1711 MJ/m 2 in Min, 1810 MJ/m 2 in Bout, and 1116 MJ/m 2 in Bin.  Figure 2 shows the estimated SR, the average fresh weight of tuberous roots (TR), and leaves and stems (LS) per pot harvested in each layer. The yield of Beniharuka (BH) was 453 g/pot for Min at 45% SR and 859 g/pot for the Top layer. On the other hand, for Suikenkintoki (SK), a peak of 903 g/pot was obtained at 83% SR of Mout, while it was almost equivalent to 893 g/pot by 100% SR of Top. The yield of SK was reduced to 662 g/pot in Min at the lowest SR.

Productivities of Sweet Potatoes in Three-Layer Cultivation Systems
In contrast, both cultivars had smaller amounts of leaves and stems in the top layer at 264 g/pot for BH and 303 g/pot for SK, and the amount increased significantly to 391 g/pot for SK in the lower layer with lower SR. Figure 3 shows the dry biomass weight per pot, the dry weight ratio of tuberous roots (TR) to leaves and stems (LS), and energy yield at each level of solar radiation. In both cultivars, the amount of dry biomass increased as the amount of solar radiation increased. The dry weight ratio of TR/LS also increased with a higher solar radiation. These results indicate that the higher the amount of solar radiation, the more likely the amount of tuberous roots is to increase.
On the other hand, the energy yield based on the amount of SR in each layer increased in the lower layer with less solar radiation. The yield of SK rose to 5.1% at Min with 45% solar radiation, while the Top layer was 2.8%. Similarly for BH, Min yield increased to 3.5%, while the Mout and Top yields were lower at 2.7% and 2.4%, respectively.
These results clearly show that the three-dimensional distributed use of sunlight increased the biomass yield of sweet potato per field area. On the other hand, even in threelayer cultivation, the top layer showed that some of the solar radiation was wasted.
: B in . Figure 2 shows the estimated SR, the average fresh weight of tuberous roots (TR), and leaves and stems (LS) per pot harvested in each layer. The yield of Beniharuka (BH) was 453 g/pot for M in at 45% SR and 859 g/pot for the Top layer. On the other hand, for Suikenkintoki (SK), a peak of 903 g/pot was obtained at 83% SR of M out , while it was almost equivalent to 893 g/pot by 100% SR of Top. The yield of SK was reduced to 662 g/pot in M in at the lowest SR.

Productivities of Sweet Potatoes in Three-Layer Cultivation Systems
x FOR PEER REVIEW 4 of 14  In contrast, both cultivars had smaller amounts of leaves and stems in the top layer at 264 g/pot for BH and 303 g/pot for SK, and the amount increased significantly to 391 g/pot for SK in the lower layer with lower SR. Figure 3 shows the dry biomass weight per pot, the dry weight ratio of tuberous roots (TR) to leaves and stems (LS), and energy yield at each level of solar radiation. In both cultivars, the amount of dry biomass increased as the amount of solar radiation increased. The dry weight ratio of TR/LS also increased with a higher solar radiation. These results indicate that the higher the amount of solar radiation, the more likely the amount of tuberous roots is to increase. Figure 2. The amount of solar radiation in each layer, and the fresh weight of tuberous roots, leaves, and stems per pot harvested in each layer. Symbols: □; Suikenkintoki, ■; Beniharuka. Data are means and error bars represent standard error of the mean. Different letters indicate statistically significant differences between cultivation layers (p < 0.05).  Figure 4 shows the chlorophyll content (SPAD), total phenolic content index, and absorbance at 540 nm of leaf extract at each level of solar radiation. In both cultivars, it was revealed that the chlorophyll content reached a maximum in Bout and the top layer decreased. In contrast, the phenol content increased with higher solar radiation.

Photosynthetic Efficiency in Three-Layers Cultivation
Polyphenols are known to absorb and attenuate ultraviolet rays (UV) to prevent the degradation of chlorophyll by UV [21]. For this reason, plants need to produce more polyphenols at higher levels of solar radiation, and UV is known to enhance polyphenol production [22,23]. The results in the figure also show that the polyphenol content increased with UV, depending on the amount of solar radiation in each layer.
On the other hand, it was supposed that the Top layer was insufficiently protected from UV rays by polyphenols, which increased the degradation of chlorophyll. These results suggest that a high solar radiation rate decreases photosynthetic efficiency because chlorophyll is degraded by UV.
The figure also shows the absorbance at 540 nm. The absorption at 540 nm is thought to depend on the amounts of carotenoids [24] and flavonoids including anthocyanins [25,26]. In plants, yellowish carotenoids are found in chloroplast membranes [27] and function to absorb and reflect excess light and UV radiation, and to inhibit the production of reactive oxygen species. Both are thought to increase production to reduce the degradation of chlorophyll by UV, but the details of their effects on sweet potato leaves are On the other hand, the energy yield based on the amount of SR in each layer increased in the lower layer with less solar radiation. The yield of SK rose to 5.1% at M in with 45% solar radiation, while the Top layer was 2.8%. Similarly for BH, M in yield increased to 3.5%, while the M out and Top yields were lower at 2.7% and 2.4%, respectively.
These results clearly show that the three-dimensional distributed use of sunlight increased the biomass yield of sweet potato per field area. On the other hand, even in three-layer cultivation, the top layer showed that some of the solar radiation was wasted. Figure 4 shows the chlorophyll content (SPAD), total phenolic content index, and absorbance at 540 nm of leaf extract at each level of solar radiation. In both cultivars, it was revealed that the chlorophyll content reached a maximum in B out and the top layer decreased. In contrast, the phenol content increased with higher solar radiation. unclear. The figure shows a significant increase in absorption at 540 nm in the Top layer with high polyphenol content and low SPAD value, suggesting a higher content of polyphenols and carotenoids compared to chlorophyll. Degradation of chlorophyll may have made the carotenoid pigments easier to extract, but in any case, degradation of chlorophyll by the strong UV in the upper layer may have reduced the photosynthetic efficiency. Figure 5 shows the average values of the leaf surface brightness (L*), yellowness (b*), and individual leaf weights for each insolation. In the Top layer, L* and b* were higher and the leaf size was smaller, but when the light intensity decreased to 73% in Bout, L* and b* decreased and the leaf growth increased. These results indicate that strong sunlight conditions increase leaf L* and b* and decrease leaf size.

Photosynthetic Efficiency in Three-Layers Cultivation
On the other hand, the leaf L* and b* and leaf size of the inner Min (69%) were similar to those of Top, while that of Bin (45%) decreased the leaf brightness. These results suggest that there were other factors that limited the growth of sweet potatoes on the inner side of the multi-layered shelves besides the rate of sunlight.   Polyphenols are known to absorb and attenuate ultraviolet rays (UV) to prevent the degradation of chlorophyll by UV [21]. For this reason, plants need to produce more polyphenols at higher levels of solar radiation, and UV is known to enhance polyphenol production [22,23]. The results in the figure also show that the polyphenol content increased with UV, depending on the amount of solar radiation in each layer.
On the other hand, it was supposed that the Top layer was insufficiently protected from UV rays by polyphenols, which increased the degradation of chlorophyll. These results suggest that a high solar radiation rate decreases photosynthetic efficiency because chlorophyll is degraded by UV.
The figure also shows the absorbance at 540 nm. The absorption at 540 nm is thought to depend on the amounts of carotenoids [24] and flavonoids including anthocyanins [25,26]. In plants, yellowish carotenoids are found in chloroplast membranes [27] and function to absorb and reflect excess light and UV radiation, and to inhibit the production of reactive oxygen species. Both are thought to increase production to reduce the degradation of chlorophyll by UV, but the details of their effects on sweet potato leaves are unclear. The figure shows a significant increase in absorption at 540 nm in the Top layer with high polyphenol content and low SPAD value, suggesting a higher content of polyphenols and carotenoids compared to chlorophyll.
Degradation of chlorophyll may have made the carotenoid pigments easier to extract, but in any case, degradation of chlorophyll by the strong UV in the upper layer may have reduced the photosynthetic efficiency. Figure 5 shows the average values of the leaf surface brightness (L*), yellowness (b*), and individual leaf weights for each insolation. In the Top layer, L* and b* were higher and the leaf size was smaller, but when the light intensity decreased to 73% in B out , L* and b* decreased and the leaf growth increased. These results indicate that strong sunlight conditions increase leaf L* and b* and decrease leaf size.
reduced the photosynthetic efficiency. Figure 5 shows the average values of the leaf surface brightness (L*), yellowness (b*), and individual leaf weights for each insolation. In the Top layer, L* and b* were higher and the leaf size was smaller, but when the light intensity decreased to 73% in Bout, L* and b* decreased and the leaf growth increased. These results indicate that strong sunlight conditions increase leaf L* and b* and decrease leaf size.
On the other hand, the leaf L* and b* and leaf size of the inner Min (69%) were similar to those of Top, while that of Bin (45%) decreased the leaf brightness. These results suggest that there were other factors that limited the growth of sweet potatoes on the inner side of the multi-layered shelves besides the rate of sunlight.  Comparing the results in Figures 4 and 5, it can be seen that SPAD and L* and b* including the inside of the multilayer shelves showed the opposite trend, with leaves with high chlorophyll content having a lower amount of reflected light due to increased light absorption. In contrast, leaf size and SPAD showed the same trend, indicating that leaves On the other hand, the leaf L* and b* and leaf size of the inner M in (69%) were similar to those of Top, while that of B in (45%) decreased the leaf brightness. These results suggest that there were other factors that limited the growth of sweet potatoes on the inner side of the multi-layered shelves besides the rate of sunlight.
Comparing the results in Figures 4 and 5, it can be seen that SPAD and L* and b* including the inside of the multilayer shelves showed the opposite trend, with leaves with high chlorophyll content having a lower amount of reflected light due to increased light absorption. In contrast, leaf size and SPAD showed the same trend, indicating that leaves with high chlorophyll content increased the size to absorb light efficiently under low sunlight conditions. Figure 6 shows the CO 2 concentration, temperature, and humidity of each layer measured in September. Data on clear days from 24-29 September were averaged to show the time changes. CO 2 concentration at B out was maintained within 15 ppm of Top, while the inside M in was 30 to 40 ppm lower than Top, and B in was a further 20 ppm lower. with high chlorophyll content increased the size to absorb light efficiently under low sunlight conditions. Figure 6 shows the CO2 concentration, temperature, and humidity of each layer measured in September. Data on clear days from 24-29 September were averaged to show the time changes. CO2 concentration at Bout was maintained within 15 ppm of Top, while the inside Min was 30 to 40 ppm lower than Top, and Bin was a further 20 ppm lower.
It also showed lower temperatures than Top after 10:30 a.m. in Min and Bin. These results suggest that the three-dimensional dense planting condition of sweet potato reduces the CO2 concentration in the middle and lower layers of the growth shelf, and that low temperature also affects the suppression of leaf growth and chlorophyll accumulation, resulting in low SPAD and high L* and b*. The relationship between the chlorophyll content (SPAD) of each layer and the leaf lightness (L*) is shown in Figure 7. In both cultivars, leaves with lower L* tended to have higher SPAD. This indicates that leaves with higher SPADs have more chlorophyll, which absorbs light, and lower brightness due to less reflected light.
The figure also shows the correlation coefficient between SPAD and L* and the approximate line obtained by the least-squares method. R45-100 denotes the correlation coefficient for all data, and R73-100 denotes the correlation coefficient for data excluding Min and Bin. Both varieties showed a high negative correlation with R73-100 of −0.67, indicating that under conditions where CO2 concentration and temperature are well maintained, the upper part with higher solar radiation has higher L* and lower SPAD.
From these results, it can be concluded that leaves in the low layer with low solar radiation increased the absorption rate of sunlight by decreasing L* and increasing the chlorophyll content, thus increasing the yield of photosynthesis. It also showed lower temperatures than Top after 10:30 a.m. in M in and B in . These results suggest that the three-dimensional dense planting condition of sweet potato reduces the CO 2 concentration in the middle and lower layers of the growth shelf, and that low temperature also affects the suppression of leaf growth and chlorophyll accumulation, resulting in low SPAD and high L* and b*.
The relationship between the chlorophyll content (SPAD) of each layer and the leaf lightness (L*) is shown in Figure 7. In both cultivars, leaves with lower L* tended to have higher SPAD. This indicates that leaves with higher SPADs have more chlorophyll, which absorbs light, and lower brightness due to less reflected light.
The figure also shows the correlation coefficient between SPAD and L* and the approximate line obtained by the least-squares method. R 45-100 denotes the correlation coefficient for all data, and R 73-100 denotes the correlation coefficient for data excluding M in and B in . Both varieties showed a high negative correlation with R 73-100 of −0.67, indicating that under conditions where CO 2 concentration and temperature are well maintained, the upper part with higher solar radiation has higher L* and lower SPAD.
From these results, it can be concluded that leaves in the low layer with low solar radiation increased the absorption rate of sunlight by decreasing L* and increasing the chlorophyll content, thus increasing the yield of photosynthesis.  Figure 8 shows the relationship between the weight of individual leaves and their phenolic content. In both cultivars, the larger the leaf size, the lower the phenol content. It can also be seen that the upper layer with high solar radiation has many small leaves with high phenol content, while the lower layer has many large leaves with low phenol content.  Figure 7. The approximate lines were determined by the least-squares method, R45-100 denotes the correlation coefficient for all data, and R73-100 denotes the correlation coefficient for data excluding Min and Bin.
These results suggest that the upper layer with high insolation carries out photosynthesis while increasing light reflectivity through polyphenols and carotenoid dyes and reducing light stress by reducing leaf size so as to not absorb excess light.
In contrast, it was considered that leaves with low solar radiation increased the amount of light absorbed by expanding the leaf area and decreasing the reflectance, thereby making efficient use of less light. Figure 9 shows that the total dry biomass production of each layer increased against the photosynthetic reaction rate index multiplied by three factors: chlorophyll content, leaf and stem volume (kg/pot), and solar radiation (%). In the three-layer cultivation Bin. The illuminance on the east side in the morning and on the west side in the afternoon was the same as the Top, so the cumulative illuminance was higher on the outside than on the inside. These relative values were used to calculate the amount of SR in each layer. As a result, the estimated SR in the upper layer was 2480 MJ/m 2 , 2108 MJ/m 2 in Mout, 1711 MJ/m 2 in Min, 1810 MJ/m 2 in Bout, and 1116 MJ/m 2 in Bin.  Figure 2 shows the estimated SR, the average fresh weight of tuberous roots (TR), and leaves and stems (LS) per pot harvested in each layer. The yield of Beniharuka (BH) was 453 g/pot for Min at 45% SR and 859 g/pot for the Top layer. On the other hand, for Suikenkintoki (SK), a peak of 903 g/pot was obtained at 83% SR of Mout, while it was almost equivalent to 893 g/pot by 100% SR of Top. The yield of SK was reduced to 662 g/pot in Min at the lowest SR.

Productivities of Sweet Potatoes in Three-Layer Cultivation Systems
In contrast, both cultivars had smaller amounts of leaves and stems in the top layer at 264 g/pot for BH and 303 g/pot for SK, and the amount increased significantly to 391 g/pot for SK in the lower layer with lower SR. Figure 3 shows the dry biomass weight per pot, the dry weight ratio of tuberous roots (TR) to leaves and stems (LS), and energy yield at each level of solar radiation. In both cultivars, the amount of dry biomass increased as the amount of solar radiation increased. The dry weight ratio of TR/LS also increased with a higher solar radiation. These results indicate that the higher the amount of solar radiation, the more likely the amount of tuberous roots is to increase.
On the other hand, the energy yield based on the amount of SR in each layer increased in the lower layer with less solar radiation. The yield of SK rose to 5.1% at Min with 45% solar radiation, while the Top layer was 2.8%. Similarly for BH, Min yield increased to 3.5%, while the Mout and Top yields were lower at 2.7% and 2.4%, respectively.
These results clearly show that the three-dimensional distributed use of sunlight increased the biomass yield of sweet potato per field area. On the other hand, even in threelayer cultivation, the top layer showed that some of the solar radiation was wasted.
: Top (SR = 100%), : M out (85%), : B out (73%), : M in (69%), and : B in (45%). The approximate lines were determined by the least-squares method, and R 45-100 denotes the correlation coefficient for all data, and R 73-100 denotes the correlation coefficient for data excluding M in and B in . Figure 8 shows the relationship between the weight of individual leaves and their phenolic content. In both cultivars, the larger the leaf size, the lower the phenol content. It can also be seen that the upper layer with high solar radiation has many small leaves with high phenol content, while the lower layer has many large leaves with low phenol content.  Figure 8 shows the relationship between the weight of individual leaves and their phenolic content. In both cultivars, the larger the leaf size, the lower the phenol content. It can also be seen that the upper layer with high solar radiation has many small leaves with high phenol content, while the lower layer has many large leaves with low phenol content.  Figure 7. The approximate lines were determined by the least-squares method, R45-100 denotes the correlation coefficient for all data, and R73-100 denotes the correlation coefficient for data excluding Min and Bin.
These results suggest that the upper layer with high insolation carries out photosynthesis while increasing light reflectivity through polyphenols and carotenoid dyes and reducing light stress by reducing leaf size so as to not absorb excess light.
In contrast, it was considered that leaves with low solar radiation increased the amount of light absorbed by expanding the leaf area and decreasing the reflectance, thereby making efficient use of less light. Figure 9 shows that the total dry biomass production of each layer increased against the photosynthetic reaction rate index multiplied by three factors: chlorophyll content, leaf and stem volume (kg/pot), and solar radiation (%). In the three-layer cultivation  Figure 7. The approximate lines were determined by the least-squares method, R 45-100 denotes the correlation coefficient for all data, and R 73-100 denotes the correlation coefficient for data excluding M in and B in .
These results suggest that the upper layer with high insolation carries out photosynthesis while increasing light reflectivity through polyphenols and carotenoid dyes and reducing light stress by reducing leaf size so as to not absorb excess light.
In contrast, it was considered that leaves with low solar radiation increased the amount of light absorbed by expanding the leaf area and decreasing the reflectance, thereby making efficient use of less light. Figure 9 shows that the total dry biomass production of each layer increased against the photosynthetic reaction rate index multiplied by three factors: chlorophyll content, leaf and stem volume (kg/pot), and solar radiation (%). In the three-layer cultivation method, it was concluded that the photosynthetic yield increased compared to the upper layer because the amounts of leaves and chlorophyll increased as the solar radiation including strong ultraviolet rays decreased at the lower layer. method, it was concluded that the photosynthetic yield increased compared to the upper layer because the amounts of leaves and chlorophyll increased as the solar radiation including strong ultraviolet rays decreased at the lower layer. Based on the above results, the effects of the three-layer cultivation method in increasing the production of sweet potato biomass were summarized in Table 1. The upper row of the table shows the fresh weight of tuberous roots and stems and leaves, and the lower row shows the dry weight.
In a mass production system combining a number of three-layer cultivation systems, the productivity was estimated to be 10.5 kg/m 2 for SK and 8.6 kg/m 2 for BH from the results of this study. As a result, SK is expected to be 4.4 times more productive and BH 3.7 times more productive than conventional farming at 2.4 kg/m 2 .
On the other hand, the energy yield of the dried biomass of stems, leaves, and tuberous roots was calculated to be 2.5% for SK and 2.0% for BH relative to the insolation energy. Conventional sweet potato production yields 2-3 kg/m 2 of stems and leaves for every 2.4 kg/m 2 of tuberous roots, with an estimated total dry weight of 1.0 kg/m 2 (17 MJ/m 2 ) and a photosynthetic yield of 0.7% (= 17 MJ/2480 MJ). Therefore, it was shown that the three-layer cultivation method could increase the photosynthetic efficiency of sweet potatoes to three times that of the conventional method.
However, compared with the highest yield at 5.1% of effective solar radiation in Bin, as shown in Figure 3, the actual photosynthetic yield was only about half, ranging from 2.0% to 2.5%. Further productivity gains are therefore expected by improving methods to increase the efficiency of the distributed use of solar radiation. Based on the above results, the effects of the three-layer cultivation method in increasing the production of sweet potato biomass were summarized in Table 1. The upper row of the table shows the fresh weight of tuberous roots and stems and leaves, and the lower row shows the dry weight. Total yield * = mean yield of Top × 5 pots + mean yield of M in × 10 pots + mean yield of B in × 10 pots.
In a mass production system combining a number of three-layer cultivation systems, the productivity was estimated to be 10.5 kg/m 2 for SK and 8.6 kg/m 2 for BH from the results of this study. As a result, SK is expected to be 4.4 times more productive and BH 3.7 times more productive than conventional farming at 2.4 kg/m 2 .
On the other hand, the energy yield of the dried biomass of stems, leaves, and tuberous roots was calculated to be 2.5% for SK and 2.0% for BH relative to the insolation energy. Conventional sweet potato production yields 2-3 kg/m 2 of stems and leaves for every 2.4 kg/m 2 of tuberous roots, with an estimated total dry weight of 1.0 kg/m 2 (17 MJ/m 2 ) and a photosynthetic yield of 0.7% (=17 MJ/2480 MJ). Therefore, it was shown that the threelayer cultivation method could increase the photosynthetic efficiency of sweet potatoes to three times that of the conventional method.
However, compared with the highest yield at 5.1% of effective solar radiation in B in , as shown in Figure 3, the actual photosynthetic yield was only about half, ranging from 2.0% to 2.5%. Further productivity gains are therefore expected by improving methods to increase the efficiency of the distributed use of solar radiation.

Discussion
Due to the low light saturation of sweet potatoes, the search for varieties that enhance photosynthetic efficiency and research to improve the yield are actively underway. In reality, however, the average yield of sweet potatoes on agricultural land has not increased in half a century.
On the other hand, the calorie productivity of sweet potatoes was nearly twice as high as that of wheat, corn, and rice, which have higher light saturation points than sweet potatoes, and the same is true overseas. Therefore, while sweet potatoes may be wasting some solar energy in the summer, when the sun's intensity increases, they are very capable of producing food and fuel. Therefore, we investigated the effect of solar radiation intensity on the photosynthetic reaction of sweet potato, and hypothesized that if the solar radiation intensity is optimized, the photosynthetic yield could be further enhanced, and we conducted this study to test this hypothesis.
As a result, even when the insolation rate was 45%, the dry biomass production of both stems and potatoes reached 81% for Suikenkintoki and 60% for Beniharuka compared to the top layer. The effects of solar radiation stress on sweet potato, which is used as a staple food source worldwide, became partially apparent, resulting in a tuberous root production of 10.5 kg/m 2 and 13,900 kcal/m 2 in the three-layer cultivation. This was 4.4 and 7.3 times higher than the production of sweet potato and rice under the conventional farming described in the introduction, respectively.
As the world's population continues to grow beyond 8 billion, there is strong concern that deforestation for agricultural land expansion will continue to increase, further accelerating global warming. However, the results of this study are the first in the world to demonstrate that food production can be significantly increased without increasing the area of agricultural land.
The photosynthetic yield of sweet potato biomass reached 2.0% to 2.5%. This is 33-41 times more efficient than the current 0.06% photosynthetic yield of Japanese forests. In other words, it became clear that the same area could absorb 33 to 41 times more CO 2 if the forest was converted to a multilayer cultivation field for sweet potatoes.
On the other hand, when the amount of solar radiation was low, the amounts of leaves and stems increased relative to the tuberous roots. To increase biomass productivity, it is desirable to grow tuberous roots efficiently with fewer leaves, so the growing conditions of sweet potatoes need to be further investigated.
Lin et al. investigated the color and photosynthetic efficiency of sweet potato leaves and reported that green leaves accumulated more biomass, while yellow leaves had less biomass production and increased heat release [28]. These results are consistent with the above results, indicating that when the solar radiation intensity is high, photosynthetic efficiency is reduced by producing more yellow pigments such as flavonoids and polyphenols than chlorophyll.
The effect of solar radiation stress such as UV irradiation on tuberous root enlargement in sweet potato is largely unknown and appears to vary among varieties [29,30]. However, the results presented here suggest that solar stress may have promoted the enlargement of tuberous roots, while leaf light use decreased with increasing solar radiation intensity.
Sweet potatoes barely bloom and produce no seeds in the warm, humid climate of spring and summer in Japan. For this reason, the accumulation of tuberous roots is considered to be one of the important means of the survival of sweet potatoes during the cold and dry season from autumn to winter. It is also known that in ordinary farming methods, high amounts of nitrogen fertilizer overgrow stems and leaves, reducing crop yields without enlarging tuberous roots.
The common cultivation know-how of sweet potato farmers suggests that the upper layer promotes tuberous root enlargement to avoid the risk of survival due to strong UV stress, while the middle and lower layers prioritize the growth of stems and leaves with the UV stress of sunlight relieved. However, it is difficult to elucidate the effect of solar UV stress on the enlargement of tuberous roots by the present experimental results alone. Further research on solar radiation and UV stress is expected to significantly improve the production efficiency of biofuel resources as well as food crops.

Three-Layer Cultivation System
Sweet potato (Ipomoea batatas) cultivars Suikenkintoki (SK) and Beniharuka (BH) were used. Regarding the sweet potato seedlings, commercial vine seedlings were purchased in the same way as general farmers, planted in each pot in early June, grown for 160 days and harvested in early November. Figure 10 shows the three-layer cultivation system used in this study. Five plastic pots (capacity; 8 L) were placed on an irrigation vinyl chloride (PVC) pipe (inner diameter; 10 cm) and each pot was filled with 7 L of gardening soil to plant the sweet potato seedlings. This set of five pots was placed in a row in the upper layer and two rows in the middle and lower layers for three-layer cultivation. The occupied area of the three-layer-25 pots cultivation system was 1.8 m 2 .  Figure 11 shows the sweet potato cultivation in the three-layer cultivation system and the position of each layer. The top layer is Top, the outside of the middle layer is Mout, the inside of the middle layer is Min, the outside of the lower layer is Bout, and the inside of the lower layer is Bin. A monolayer culture system was also used for comparison.

Solar Radiation Rates in a Three-Layer Cultivation System
The illuminance of each layer was measured using an illuminance UV recorder (TR Type-74 Ui, T&D, Tokyo, Japan) during July-August, when sweet potato grows actively, and the cumulative values were compared to the Top layer. The solar radiation ratio (SR) of each layer was given as a relative value compared to the Top layer. The CO2 concentration, temperature, and humidity of each layer were measured using a CO2 datalogger (TR Type-76 Ui, T&D, Tokyo, Japan).
The cumulative value of solar radiation at the top during the entire 160 d cultivation period was set at 2480 MJ/m 2 by referring to measurements taken at the nearest meteorological observatory (Shizuoka City). Each SR was then multiplied by 2480 MJ/m 2 to calculate the cumulative solar radiation inside and outside the middle and lower layers. Using the method introduced in a previous paper [14,15], a water-absorbing sheet was inserted into the PVC pipe from the bottom of the pot and guided so that the thin roots of the sweet potatoes extended into the pipe containing water and liquid fertilizer.
The liquid fertilizer for the gardening of flowers and vegetables was 500 times diluted and supplied once a day through the PVC pipe. The approximate concentrations of the diluted fertilizer were: N: 120 mg/L, H 3 PO 4 : 200 mg/L, K: 100 mg/L, Mg: 1 mg/L, Mn: 0.02 mg/L, B: 0.01 mg/L. Figure 11 shows the sweet potato cultivation in the three-layer cultivation system and the position of each layer. The top layer is Top, the outside of the middle layer is M out , the inside of the middle layer is M in , the outside of the lower layer is B out , and the inside of the lower layer is B in . A monolayer culture system was also used for comparison.

Solar Radiation Rates in a Three-Layer Cultivation System
of each layer was given as a relative value compared to the Top layer. The CO2 concentration, temperature, and humidity of each layer were measured using a CO2 datalogger (TR Type-76 Ui, T&D, Tokyo, Japan).
The cumulative value of solar radiation at the top during the entire 160 d cultivation period was set at 2480 MJ/m 2 by referring to measurements taken at the nearest meteorological observatory (Shizuoka City). Each SR was then multiplied by 2480 MJ/m 2 to calculate the cumulative solar radiation inside and outside the middle and lower layers.

Photosynthetic Efficiency in Three-Layer Cultivation
Harvested tuberous roots, leaves, and stems were dried at 90 °C for one day and the dry matter content measured. The unit heat of dry matter was set at 17 MJ/kg referring to previous reports [31]. To analyze the effect of SR on leaf photosynthetic efficiency, the rate of energy conversion to biomass was calculated by dividing the thermal energy stored in the tuberous roots and stems by the cumulative amount of solar radiation in each layer. The heat yield of each layer was converted to productivity at 1 m 2 by the equation of The illuminance of each layer was measured using an illuminance UV recorder (TR Type-74 Ui, T&D, Tokyo, Japan) during July-August, when sweet potato grows actively, and the cumulative values were compared to the Top layer. The solar radiation ratio (SR) of each layer was given as a relative value compared to the Top layer. The CO 2 concentration, temperature, and humidity of each layer were measured using a CO 2 datalogger (TR Type-76 Ui, T&D, Tokyo, Japan).
The cumulative value of solar radiation at the top during the entire 160 d cultivation period was set at 2480 MJ/m 2 by referring to measurements taken at the nearest meteorological observatory (Shizuoka City). Each SR was then multiplied by 2480 MJ/m 2 to calculate the cumulative solar radiation inside and outside the middle and lower layers.

Photosynthetic Efficiency in Three-Layer Cultivation
Harvested tuberous roots, leaves, and stems were dried at 90 • C for one day and the dry matter content measured. The unit heat of dry matter was set at 17 MJ/kg referring to previous reports [31]. To analyze the effect of SR on leaf photosynthetic efficiency, the rate of energy conversion to biomass was calculated by dividing the thermal energy stored in the tuberous roots and stems by the cumulative amount of solar radiation in each layer. The heat yield of each layer was converted to productivity at 1 m 2 by the equation of "harvested quantity kg/pot × 25 pots/1.8 m 2 × 17 MJ/kg", and the energy yield for the estimated solar radiation (SR × 2480 MJ/m 2 ) of each layer was obtained.
To analyze the photosynthetic efficiency during the active growing season, three leaves of each size (S = 4-10 cm, M = 10-13 cm, L = 13 cm or more) were collected from each pot 90 days after planting, and each leaf was weighed. The lightness (L*), redness (a*), and yellowness (b*) of each leaf were measured using a color reader, CR-10 (Konica Minolta Co. Ltd., Minato, Tokyo) [28]. In addition, the relative chlorophyll content by SPAD-502 (Konica Minolta, Inc.) was determined [32].
To investigate the effect of SR on the polyphenol content of leaves, polyphenols were extracted from leaf segments cut to 1 cm 2 with 50% aqueous methanol and measured for absorbance at 765 nm using the Folin-Ciocalteu method previously reported [33,34]. The effects of SR were analyzed using the "765 nm absorbance × dilution ratio" as the phenol content index per leaf area.

Statical Analysis
Statistical analysis was performed using the statistical analysis tool in Microsoft Excel 2019. The mean value and standard error of the opposing elements were obtained using each solar radiation rate (position of the shelf) as a conditional element was indicated (Figures 2, 4 and 5). Analysis of variance (ANOVA) was also applied to the resulting data. The significance of the differences between the condition elements was determined by a t-test between the two elements assuming unequal variance, and groups of p < 0.05 significant differences were indicated by different letters in the figures. Figures 3 and 9 show the correlations between the conditional elements and the mean values of opposing elements using the approximate lines obtained by the least-squares method and the coefficient of determination (R 2 ). Figures 7 and 8 showed an approximate line and a correlation coefficient (R) obtained by the least-squares method for the correlation between the conditional element and the opposing element.

Summary
This study showed that reducing solar radiation stress by using a three-layered cultivation method of sweet potato increased the yield 4.4 times higher than conventional cultivation methods. In addition, to clarify the effects of solar radiation stress, we investigated the pigments, polyphenols, and shapes involved in the protection of leaves from UV stress. As the result, it was found that when sufficient CO 2 and fertilizers were available, the leaf size tended to decrease and the polyphenol content, lightness, and yellowness increased under intense solar radiation.
From these results, it can be concluded that sweet potatoes, which have low light saturation points and photosynthetic yields under conventional farming methods, block some ultraviolet and visible light with polyphenols and pigments to mitigate strong photooxidation, and reduce leaf size so as to not absorb extra light.
Furthermore, from these analyses, it was concluded that the increase in photosynthetic yield by the multilayer cultivation method was due to the increase in light absorption due to the three-dimensional increase in leaf area and the relaxation of UV stress.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.