Appropriate Planting Density Can Improve the Storage Root Yield and Commercial Features of Sweet Potato (Ipomoea batatas L.) by Optimizing the Photosynthetic Performance

Abstract

Planting density is a crucial factor in sweet potato output. However, the relationship among photosynthetic performance, yield, and storage root commercial features that respond to planting density is not well understood. We conducted a three-year field experiment with four planting densities (83,280 plants hm⁻², plant spacing 15 cm, D15; 62,520 plants hm⁻², plant spacing 20 cm, D20; 50,025 plants hm⁻², panting spacing 25 cm, D25; and 41,640 plants hm⁻², 30 cm, D30; 80 cm row space for all) to investigate the dynamic of photosynthetic performance, dry matter, yield, carbohydrate metabolism, and commercial features of storage root. The result showed that the highest yield was observed in the D20 treatment, and the yield increment was by 8.47–24.92% when compared to the D25 control treatment during the three growth periods. The observation can be attributed to the fact that appropriate planting density D20 can shape a good canopy structure to improve photosynthetic performance by significantly increasing IPAR, TPAR, light transmission, and extinction coefficient through different canopy levels. Hence, the Pn, Tr, Ci, Gs and WUE, and the chlorophyll fluorescence parameters were significantly improved. Eventually, promoting root sink development by up-regulating starch, fructose, glucose, and sucrose in storage roots, resulting in vigorous carbon flux from the source toward the root sink. Therefore, the optimal planting density D20 treatment increased individual plant yield and commercial features by increasing the number of storage roots, particularly medium-sized ones. Herein, we claim that optimizing the plant population density of sweet potatoes can be a good way to increase the yield and commercial features, and our results are great and important for improving the market value and profits of sweet potatoes.

1. Introduction

Sweet potato (Ipomoea batatas L.), being one of the most important staple food crops worldwide, is characterized as the sixth most widely cultivated crop globally [1]. It is widely consumed due to its exquisite flavor and health benefits, and China is the world’s largest producer of sweet potatoes [2,3]. On sweet potatoes, at a given planting density, the final yield is determined by the storage root number and storage root fresh weight of individual plants [4]. Increasing the storage root number per plant, on the one hand, can significantly increase the yield, but more importantly, it can improve the storage root appearance by reducing individual storage root fresh weight and ratio of length/diameter, which results in a better storage root commercial feature [5]. Hence, balancing the relationship between the storage root number and the storage root fresh weight of the individual plants is the key point to improve the final yield and commercial feature of sweet potato. At the canopy closure period, the sweet potato storage roots (root sink) had formed, and the storage root number was almost certain. Photo assimilation is the main component that deposits in those root sinks to enlarge the root size. Therefore, lengthening the photosynthetic duration is the optimal strategy to tap the yield potential of sweet potatoes. At a given population scale, the canopy ‘source capacity’ will reach a stable level. Hence, the key factor in increasing yield and storage root commercial feature strongly relies on the photo assimilation from the source whose expenditure would facilitate more storage root formation in the closure period and whose allocation to each root sink during the bulking period was more evenly. However, the underlying mechanism to balance the relationship between storage root number and storage root fresh weight of individual plants that are regulated by planting density is still unclear.

Planting density is an important agronomic component that influences crop output potential [6]. Individual competition among plants that responds to planting density mostly through canopy structure light absorption and root system nutrition uptake [7]. Planting density is one of the determining factors in crop interception of radiation by regulating the canopy structure. Intercepted photosynthetically active radiation (IPAR) reflects the level of light capture and uptake absorption. Canopy IPAR is extremely associated with the canopy structure, which is strongly linked to leaf physiological and morphological traits, such as leaf area index, leaf nitrogen content, specific leaf area, and canopy micro-environments like canopy temperature and humidity [8]. The technique to increase sunlight energy use efficiency is also strongly tied to proper canopy structure. In cereal crops such as maize [9] and wheat [10], an appropriate planting density has been shown to boost production by enhancing canopy leaf exposure to light.

The chlorophyll fluorescence is a nondestructive evaluation of PSII reaction center activity in individual plants. Among these parameters, the YII measures light energy capture efficiency, which reflects the actual direct sunlight energy conversion efficiency of the PSII reaction center [11]. Fv/Fm, the maximal photochemical efficiency of PSII, determines the potential quantum efficiency of PSII [12]. The external light environment has a normal influence on the electron transport rate (ETR) [13]. The non-photochemical quenching (NPQ) and coefficient of photochemical quenching (qP) are two important parameters to represent the degree of openness of the PSII reaction center, with a higher qP resulting in increased electron transfer activity in PSII [14]. Leaf gas exchange plays an important role in photoassimilation and crop yield determination [15]. However, the individual plant under dense planting can significantly decrease the stomatal conductance (Gs) and intercellular carbon dioxide concentration (Ci), which has a negative impact on the photosynthetic system [16]. CO₂ concentration is strongly linked with net photosynthetic rate, which varies depending on species and environmental conditions [17]. Plants with low planting density have reduced chlorophyll content, a greater electron transfer rate, and ribulose-1,5-bi-phasphate carboxylase/oxygenase when compared to dense planting conditions [18]. The vigor of carbon flow to the sink is an important factor in determining sweet potato yield, which is determined by the source capacity, such as photo assimilation and transportation activity, as well as the sink strength, which is primarily determined by the number of storage roots per plant and the weight of each storage root. Starch, a major component of store roots, is produced by the sucrose-starch conversion process, which involves several enzyme activities. Previous studies have shown that boosting the enzymic activity of AGPase, SSS, SBE, and GBSS can promote starch production and deposition in the sink, leading to sink formation [5,19]. The distribution and accumulation of dry biomass in the root system is a critical component in defining sweet potato storage root production and development, as well as influencing the root-to-shoot ratio. A reasonable root/shoot ratio is heavily reliant on a balance of shoot and root development, which promotes sink creation and growth.

We hypothesize that optimal planting density not only shapes a reasonable canopy structure to improve sunlight distribution and individual photosynthesis but also coordinates individual plant development by enhancing the relationship between individual and population plant growth and development, resulting in higher yield. Furthermore, optimal planting density can enhance photo-assimilate to facilitate storage root formation, thus increasing storage root number, which sculpts a good storage root appearance by enhancing the uniformity of storage root fresh weight, ultimately improving commercial features. The objectives of this study were to investigate (1) the performance of photosynthetic characteristics, such as canopy sunlight distribution and individual photosynthesis responses to planting densities, (2) the relationship between photosynthetic assimilation to the root sink and photosynthetic characteristics during storage root bulking period, and (3) the relationship between storage root appearance and photosynthetic performance.

2. Materials and Methods

2.1. Experimental Materials

Fertilizers used in this experiment included urea (N 46%, Sinopec, Co., Ltd., Dongfang, China), calcium superphosphate (P 16%, SDIC Xinjiang Lop Nur Potassium Salt Co., Ltd., Hami, China), and potassium sulfate (K 52%, Guangdong Zhanhua Group Co., Ltd., Zhanjiang, China).

2.2. Experimental Design

Three-year field trials were conducted from November 2021 to March 2023, with the first round taking place on 3 November 2021 and ending on 21 February 2022 at Hainan University (20°06’ N, 110°33’ E). The second round started on 19 March 2022 and ended on 5 July 2022, while the third one began on 31 October 2022 and was completed on 5 March 2023, both were set in the Breeding Base of the Institute of Nanfan & Seed Industry, Guangdong Academic of Science at Yazhou District, Sanya City, Hainan province (18°21’ N, 109°9’ E). The soil physical and chemical properties at the 0–30 cm soil tillage layer were described in Table S1. The climate characteristics during the three-round experiments are detailed in Table S2. Sweet potato Yanshu25 (Yan-25), a popular type grown in China and known for its orange-fresh flavor, was chosen for this experiment. Vegetative terminal cuttings of around 25 cm in length that were virus-free were used. The field trials were arranged as a complete randomized block design (CRBD) for three replications; each plot area was 30 m² (6 m × 5 m). We set four planting densities, namely, D15 (83,280 plants hm⁻²), D20 (62,520 plants hm⁻²), D25 (50,025 plants hm⁻², as control), and D30 (41,640 plants hm⁻²). For all treatments, the plant spacing was 15, 20, 25, and 30 cm, with a row distance of 80 cm for all. Before sowing, each treatment plot received N 120 kg ha⁻¹, P 112 kg ha⁻¹, and K 240 kg ha⁻¹ as base fertilizer. Stem cuttings that were virus-free and grown with four stem inter-nodes were sown in the soil using the oblique planting method. Throughout the experiment, field management approaches were consistent with local high-yield field practice.

2.3. Sampling Method

The sampling times were arranged at 50, 70, 90, and 110 days after planting. Before sampling, the functional leaf (the 4th or 5th fully-expand leaf from the main stem terminal) had its SPAD (chlorophyll content), photosynthetic characteristics, chlorophyll fluorescence performance were measured, and plant population canopy photosynthetically active radiation (PAR) measured. Five representative plants were randomly selected at random from each treatment. Each plant was divided into the shoot and root systems for further analysis.

2.4. Physiological Analysis

2.4.1. Functional Leaf SPAD Value and Leaf Nitrogen Content

A hand-held chlorophyll meter was used to measure the functional leaves SPAD value (chlorophyll content) and leaf nitrogen content (TYS-4N, Beijing Jinkelida Electronic Technology Co., Ltd., Beijing, China). Nine functional leaves from each treatment were selected for measurement. Each leaf was measured twice at the same sites in the middle part of the leaf.

2.4.2. Photosynthetic Characteristic

Seven functional leaves from each treatment were measured at 1800 µmol m⁻² s⁻¹ of light on a sunny morning from 9:00 a.m. to 12:00 p.m. The CO₂ content in the leaf chamber was kept constant at 400 µmol L⁻¹. Leaf photosynthetic rate (Pn), stomatal conductance (Gs), transpiration rate (Tr), and intercellular carbon dioxide concentration (Ci) were measured using a portable photosynthesis system (LI-6400, Li-Cor, Lincoln, NE, USA). Furthermore, water use efficiency (WUE) was formulated as follows [20]:

WUE = Pn/Tr

2.4.3. Chlorophyll Fluorescence Measurements

A PAM-2500 High-performance field and laboratory chlorophyll fluorometer from Heinz Walz GmbH Co., Ltd., Nürnberg, Germany, measured the maximum quantum efficiency of PS II (Fv/Fm), the quantum yield of photosystem II (Y(II)), photo-chemical quenching (qP), non-photochemical quenching (NPQ), and electron transport rate (ETR) [21].

2.4.4. Canopy Photosynthetic Measurement

During the canopy closure period, the SUNSCAN Canopy Analysis System (Delta-T Device Co., Ltd., Burwell, UK) measured intercepted photosynthetically active radiation (IPAR) (measure site: 10 cm above the canopy), transmitted photosynthetically active radiation (TPAR) (measure site: 10 cm above the ground), and leaf area index (LAI). The extinction coefficient (k value) was computed as [22]:

K = −ln (TPAR/IPAR) × LAI⁻¹

2.4.5. Agronomic Traits

The fresh weight, number of branches, main stem length, average branch length, and stem base diameter, as well as storage root diameter and fresh weight, were measured or calculated. The plant organs were collected and steamed at 105 °C for 30 min to gain constant dry weight at 80 °C. Then the Root/shoot ratio = Root dry biomass/shoot dry biomass × 100%; Organ dry biomass allocation rate = Organ dry biomass/total plant dry biomass × 100% was calculated.

2.4.6. Storage Root Yield, Yield Component, and Commercial Traits

According to the grading standards of Villordon (2013) [23] and Si (2022) [24], the roots whose diameter reached 1.00 cm above are storage roots. At harvest time, all storage roots were dug out. The average storage root number per plant, average storage root weight, and yield were calculated. The storage roots of five representative plants of per plot were measured, and their length/diameter ratio (L/D) has been used to define shape in agricultural goods, with an L/D ratio of 1 indicating that the object is circular. The uniformity of storage root weight is expressed by the coefficient of variation (CV), CV = standard deviation/average value. The smaller the CV, the better the uniformity of storage root weight. At last, according to the actual production in China in recent years, the storage roots (diameter of storage root 2.00 cm, weight of storage root more than 50.00 g) were classified into large commercial storage root (250–500 g root), medium-sized commercial storage root (100–250 g root) and small commercial storage root (50–100 g root), and the number of commercial storage root of each grade was investigated manually.

2.4.7. Storage Root Carbohydrate Content

The procedure of carbohydrate content determination was optimized. A 250 mg fresh root sample was taken from the −80 °C refrigerators and immediately placed in 5 mL distilled water and extracted by 80 °C water bath for 30 min in a test tube. After cooling to room temperature, the sample was centrifuged at 8000× g rpm for ten minutes under 4 °C. The supernatant was used to determine the contents of starch, sucrose, glucose, and fructose according to the analysis kits produced by the Nanjing Jiancheng Bioengineering Institute (To acquire detailed information, please consult via http://www.njjcbio.com/ (accessed on 1 July 2014)). Each treatment was tested three times.

2.4.8. Real-Time Quantitative PCR Analysis

The frozen storage root total RNA was extracted according to the manual of the Plant Total RNA Isolation Kit Plus (Foregene, RE05024, Chengdu, China). The first-strain cDNA generated by (MonScriptTM RTIII Super Mix with dsDNase (Two-step), Wuhan, China) and real-time quantitative PCR were performed in a 20-µL reaction volume containing 1 × MonAMPTM ChemoHS Qpcr Mix (Monad Biotech, Wuhan, China). Quantitative analysis was conducted by ABI QuanStudioTM⁵ System under standard mode. The 2^−ΔΔCt method calculated relative expression levels from three biological replicates that used the β-actin gene as a housekeeping gene. The primers used for qRT-PCR are listed in Table S3.

2.5. Statistical Analysis

One-way Anova determined statistical analysis with a significant level at p < 0.05 tested by Duncan. The statistical analysis was performed by SPSS 19 and Excel 2019, and the figures were designed by GraphPad Prism software (Version 8.4.2 for Windows).

3. Results

3.1. Canopy Spectrum Parameter

The findings of a field investigation conducted over a period of two years revealed comparable patterns in light transmission, IPAR, TPAR, and extinction coefficient. The initial increase in these parameters was followed by a decrease as the planting density decreased. Notably, the highest value was observed in the D20 treatment, which was significantly higher than the other treatments (p < 0.05; Figure 1). And the average increment of IPAR, TPAR, light transmission, and extinction coefficient value in D20 treatment plants was by 3.55%, 9.63% and 26.77%; 22.03%, 25.03% and 84.61%; 30.12%, 30.25%, and 39.49; 6.45%, 11.86%, and 2.32%, respectively, when compared to the D15, D25, and D30 treatment plants.

3.2. Leaf Nitrogen Content, Chlorophyll Content, LAI, and LA Characteristics During Bulking Period

A reduction in planting density resulted in a significant increase in leaf nitrogen content, SPAD, and leaf area (p < 0.05; Figure 2A,B,D). The LAI value was found to be greatest in the D20 treatment (p < 0.05; Figure 2C), and the average increment in D20 treatment was by −8.48%, −2.85%, and 31.30%, 18.19%, 7.73%, and 13.37%; 10.28%, 4.7% and 14.20 when compared to the D15, D25 and D30 treatment during the bulking period, respectively.

3.3. Leaf Gas Exchange

As compared with the control D25 treatment, the D20 treatment significantly increased the Pn (p < 0.05; Figure 3A,B) and WUE (p < 0.05; Figure 3I,J), and the average increment of Pn and WUE was 9.25%, 5.19%, and 3.65%; −3.29%, 21.02%, and 12.39%, respectively, during the bulking period. However, Ci and showed insignificant differences (p > 0.05; Figure 3C,D). The Gs were significantly higher in D20 treatment during 50–70 days after planting as compared to the rest of the treatments, while there was no significance among treatments at 90 days after planting (p < 0.05; Figure 3E,F). In contrast, Tr has a similar trend in the two-year experiments. In detail, plants in dense planting (D15 and D20 treatment) had a comparable effect on Tr and were significantly higher than plants in reduced planting density (D25 and D30 treatment) during 50–70 days after planting; however, the opposite effect was detected at 90 days after planting (p < 0.05; Figure 3G,H).

3.4. Chlorophyll Fluorescence Parameters

There was no discernible variation in the Fv/Fm analysis in response to planting density (p > 0.05; Figure 4A,B). Interestingly, the YII and ETR under the plants in dense planting D20 treatment were significantly higher than the plants in the other three treatments during the storage root bulking period (p < 0.05; Figure 4C,D,I,J). Compared to the D25 treatment plants, the average increment of YII and ETR was by 9.90%, 4.66%, and 0.85%, 19.06%, 4.66%, and 7.61%, respectively, during the bulking period. In addition, during the 50–70-day period following planting, the qP parameter value under dense panting plants (D15 and D20 treatment) was significantly greater than that of the plants in low plant density (D25 and D30 treatment), whereas the opposite tendency was observed at 90 days after planting (Figure 4E,F). Furthermore, no discernible variation was seen in the plant NPQ analysis between the D15 and D20 dense planting treatments, and the values of NPQ parameters gradually decreased with the decrease in planting density (Figure 4G,H).

3.5. Above-Part Development

The field studies found a comparable effect on sweet potato above-part development (Figure 5). During the storage root bulking period, decreasing plant population density resulted in substantial increases in leaf fresh weight (Figure 5A), above-part fresh weight (Figure 5B), main stem length (Figure 5C), average stem length (Figure 5D), branch number (Figure 5E) and stem base diameter (p < 0.05, Figure 5F).

3.6. Dry Matter Accumulation and Allocation

Reduced planting density led to significant increases in the above-part dry weight, storage root, and total dry matter weight of individual plants (p < 0.05; Figure 6A,D). In contrast, the plants under dense planting D20 treatment had a comparable effect on dry matter allocation to each plant storage root and shoot as the D30 low planting density treatment (p < 0.05; Figure 6B,E). The D20 treatment plant had a significantly larger root/shoot ratio than other treatments due to a higher allocation of dry matter to storage roots and lower allocation to shoots (p < 0.05; Figure 6C,F). The average increment of root/shoot ratio in D20 treatment plants was by 27.39%, 14.75%, and 9.26%; 112.5%, 27.27%, and 3.84%; 132.5%, 16.67%, and 10.14% during the bulking period when compared to the D15, D25, and D30 treatment plants, respectively.

3.7. Genes Expression Level on Regulating Carbohydrates Metabolism in Storage Root

The expression levels of SuSy (Figure 7A), SSS (Figure 7B), AGPase (Figure 7C), SBE1 (Figure 7D), GBSS (Figure 7E), and SPS (Figure 7F) were found to exhibit a significant increase initially, followed by a decrease in planting density. The highest levels were observed under the D20 treatment, and this difference from the other treatments was statistically significant (p < 0.05; Figure 7). The expression levels of the genes mentioned above in the D20 treatment were, overall, roughly 20–80% greater throughout the store root bulking period than in the other treatments in the storage root sink (Figure 7).

3.8. Carbohydrate Content on Storage Root During Bulking Period

The starch and fructose content was first increased and then decreased with the decrease in planting density, and the maximum value was observed in dense planting D20 treatment plants, which had a significant comparison to other treatments (p < 0.05, Figure 8A, D). Furthermore, the starch content in each treatment steadily increased as the growth period proceeded; the sucrose content grew dramatically as the planting density fell and peaked under the control D25 treatment (p < 0.05, Figure 8B). Furthermore, the glucose content was significantly increased under the D20 treatment compared to the control during 50–70 days after planting but decreased at 90 days after planting in both years (p < 0.05; Figure 8C).

3.9. Storage Root Development Characteristics During Bulking Period

In both years, decreasing planting density resulted in a substantial increase in storage root width and fresh weight (p < 0.05; Figure 9). Additionally, each treatment gradually rose with the growing period (Figure 9).

3.10. Storage Root Characteristics at Canopy Closure Period

Data reveals that planting density had a significant effect on sweet potato storage root characteristics during the canopy closure period (

Table 1. Effect of planting density on sweet potato SR onset characteristics at canopy closure period.

Years	Treatment	Average SR Diameter (mm)	Average SR Fresh Weight (g)	Young SR Number (2 < Φ < 5 mm)	Developing SR Number (5 < Φ < 20 mm)	Mature SR Number (Φ > 20 mm)	SR Number
2021 (Haikou)	D15 †	5.14 c ‡	1.35 c	2.33 a	2.33 c	-	4.67 b
	D20	7.84 b	2.60 b	1.67 b	4.00 a	-	5.67 a
	D25	8.07 b	4.87 a	0.67 c	3.67 ab	-	4.33 bc
	D30	10.62 a	4.70 a	0.67 c	3.00 b	-	3.67 c
2022 (Sanya)	D15	18.99 b	18.96 b	-	3.33 a	3.67 b	7.00 a
	D20	21.60 a	26.74 a	-	2.33 bc	4.33 a	6.67 a
	D25	21.11 a	27.74 a	-	2.00 c	4.00 ab	6.00 b
	D30	22.67 a	28.87 a	-	2.67 b	4.00 ab	6.67 a
2023 (Sanya)	D15	20.95 c	31.14 c	-	0.67 c	4.67 b	5.33 c
	D20	20.80 c	32.90 b	-	2.33 a	4.67 a	7.00 a
	D25	23.27 b	36.42 a	-	1.67 b	4.67 a	6.33 b
	D30	24.66 a	34.09 ab	-	1.67 b	3.67 b	5.33 c

Note: SR, storage root, † D15, 15 cm plant distance and 83,280 plants ha⁻¹; D20, 20 cm plant distance and 62,520 plants ha⁻¹; D25, 25 cm plant distance and 50,025 plants ha⁻¹; D30, 30 cm plant distance and 41,640 plants ha⁻¹. The row distance was set at 80 cm for all. ‡ Values followed by different letters present significant differences (p < 0.05) among planting density treatments.

). As compared with the control D25 treatment, the D20 treatment significantly increased the storage root number, mainly increasing the mature ones during the 2022 and 2023 growth seasons with an increment of 8.25% and 27.24%, respectively. And an increment of 8.99% of developing ones during the 2021 field experiment (p < 0.05; Table 1). Additionally, the average storage root diameter and fresh weight were dramatically increased with planting density deceased (p < 0.05; Table 1).

3.11. Storage Root Yield, Yield Component, and Appearance Quality at Harvest Period

A three-year field experiment showed that planting density had the same influence on storage root yield, yield component, and appearance quality (

Table 2. Effect of planting density on storage root yield, yield component, and appearance quality.

Years	Treatment	SR Diameter (mm)	SR Length (cm)	L/D Ratio	Average SR Weight (g)	CV (%)	Average SR Number Per Plant	Yield (kg/hm2)	Yield Increment (%)
2021 (Haikou)	D15 †	36.95 c ‡	9.98 bc	2.7	104.56 b	5.64	3.57 d	29,763.09 b	14.41 #
	D20	38.75 b	9.69 c	2.5	111.13 b	1.77	4.52 a	32,497.89 a	24.92
	D25	39.39 a	10.24 b	2.6	127.26 a	11.87	4.16 b	26,013.56 c	-
	D30	39.15 a	11.74 a	3.0	133.97 a	5.49	4.00 c	24,154.95 d	−7.14
2022 (Sanya)	D15	39.04 b	10.15 a	2.6	88.50 d	7.69	5.26 c	38,795.60 b	3.46
	D20	39.81 b	9.15 b	2.3	110.81c	2.90	6.19 a	42,883.33 a	14.13
	D25	39.72 b	9.53 ab	2.4	131.66 b	6.71	5.71 b	37,494.90 b	-
	D30	42.71 a	9.82 ab	2.3	170.61a	7.64	5.01 d	35,592.04 c	−5.34
2023 (Sanya)	D15	51.64 b	14.31 b	2.7	174.65 c	5.55	5.44 b	47,801.67 b	3.69
	D20	50.19 b	15.15 ab	3.0	200.59 b	3.65	6.41 a	50,025.00 a	8.47
	D25	51.31 b	15.92 a	3.9	215.37 ab	4.52	5.99 ab	46,120.27 b	-
	D30	53.52 a	15.87 a	3.0	235.26 a	6.33	5.63 b	45,416.67 b	−1.51

Note: † D15, 15 cm plant distance and 83,280 plants ha⁻¹; D20, 20 cm plant distance and 62,520 plants ha⁻¹; D25, 25 cm plant distance and 50,025 plants ha⁻¹; D30, 30 cm plant distance and 41,640 plants ha⁻¹. The row distance was set at 80 cm for all. ‡ Values followed by different letters present significant differences (p < 0.05) among planting density treatments. # Compared with the control.

). The dense planting D20 treatment resulted in significantly higher storage root yield compared to other treatments, likely due to a greater quantity of storage roots (p < 0.05; Table 2). In addition, the yield increase over three years with D15, D20, and D30 treatment was by 14.41%, 24.92%, −7.14; and by 3.46%, 14.13%, −5.34%; and by 3.69%, 8.47%, −1.51%, when compared to D25 control, respectively. Furthermore, D20 treatment improved storage root appearance quality by reducing length (p < 0.05; Table 2), L/D ratio, and the CV of individual root weight. In contrast, the storage root diameter and average storage root weight were significantly increased with the decrease in planting density (p < 0.05; Table 2).

3.12. Storage Root Commercial Characteristics at Harvest Time

A three-year field experiment found that decreasing planting density led to significant increases in commercial storage root weight and large storage root number (p < 0.05;

Table 3. Effect of planting density on storage root commercial characteristics at harvest period.

Years	Treatments †	Commercial SR Weight (g)	Commercial SR Number Per Plant	Large SR Number	Middle SR Number	Small SR Number	Commercial Rate (%)	Commercial SR Yield (kg hm−2)	Increment (%)
2021 (Haikou)	D15	122.61 b ‡	2.33 c	0.33 c	1.67 b	0.33 a	65.71	19,558.23 b	−0.51 #
	D20	129.91 b	3.00 a	1.00 b	2.00 a	0.00 b	74.95	24,367.17 a	23.35
	D25	158.45 a	2.33 c	1.33 a	1.00 d	0.00 b	75.45	19,697.51 b	-
	D30	162.41 a	2.67 b	1.33 a	1.33 c	0.00 b	70.37	16,998.00 c	−16.16
2022 (Sanya)	D15	98.89 d	3.00 c	1.00 b	2.00 b	0.00 c	64.32	24,954.02 c	29.60
	D20	117.89 c	4.67 a	1.00 b	3.00 a	0.67 a	80.08	34,342.08 a	33.46
	D25	136.87 b	4.00 b	1.67 a	2.00 b	0.33 b	68.48	25,676.50 b	-
	D30	201.50 a	3.00 c	1.67 a	1.33 c	0.00 c	70.79	25,198.06 bc	−1.95
2023 (Sanya)	D15	179.46 c	4.33 c	0.00 d	4.33 a	0.00 a	81.78	39,092.21 b	0.92
	D20	203.59 b	5.33 a	1.00 b	4.33 a	0.00 a	84.39	42,216.09 a	8.96
	D25	232.03 a	4.67 b	1.67 a	3.00 b	0.00 a	83.99	38,736.41 b	-
	D30	242.80 a	4.33 c	1.67 a	2.67 c	0.00 a	79.37	36,047.21 c	−6.94

Note: † D15, 15 cm plant distance and 83,280 plants ha⁻¹; D20, 20 cm plant distance and 62,520 plants ha⁻¹; D25, 25 cm plant distance and 50,025 plants ha⁻¹; D30, 30 cm plant distance and 41,640 plants ha⁻¹. The row distance was set at 80 cm for all. ‡ Values followed by different letters present significant differences (p < 0.05) among planting density treatments. # Compared with the control.

). However, compared to the control D25 treatment, the commercial storage root number, especially medium storage root number, and the D20 treatment significantly increased commercial e storage, particularly medium storage root number and yield (p < 0.05; Table 3). Moreover, commercial storage root production increased by 8.96–33.46% in three years compared to D25 treatment (p < 0.05; Table 3).

3.13. Correlation Analysis

The correlation analysis found that the storage root number per plant was positively correlated with the yield. At the same time, storage root fresh weight had a negative correlation with the storage root number and yield (

Table 4. Pearson correlation analysis between yield components and photosynthesis characteristics.

	SRN †	SRFW	Yield	IPAR	TPAR	Pn	Gs
SRN	1.00	−0.751 ***	0.793 ***	0.791 ***	0.593 **	−0.596 **	0.018 ns
SRFW	−0.751 *** #	1.00	−0.682 ***	−0.938 ***	−0.618 **	0.540 *	0.324 ns
Yield	0.793 ***	−0.682 ***	1.00	0.764 ***	0.658 **	−0.393 ns	0.010 ns
	Ci	Tr	Fv/Fm	qP	ΦPSII	ETR	NPQ
SRN	0.506 *	−0.233 ns	0.389 ns	−0.546 *	−0.705 ***	−0.500 *	0.571 **
SRFW	−0.316 ns	0.429 ns	−0.637 **	0.730 **	0.901 ***	0.685 **	−0.631 **
Yield	0.317 ns	−0.175 ns	0.549 *	−0.648 **	−0.793 ***	−0.676 **	0.578 **

Note: † SRN, storage root number per plant; SRFW, storage root fresh weight. ^{# ns} indicated no significant difference; * and ** and *** indicated significant differences at p < 0.05 and p < 0.01, and p < 0.001, respectively.

). The parameters of canopy IPAR, TPAR, Ci, Fv/Fm, as well as NPQ have significant positive correlations with yield and storage root number (Table 4). Additionally, a significant positive correlation between the storage root fresh weight and Pn, Gs, Tr, qP, ΦPSII, and ETR parameters was observed (Table 4).

4. Discussion

Selecting a reasonable planting density can provide a good canopy structure by optimizing the spatial distribution of plants and the structure, ventilation, and light transmittance of the canopy, thus contributing to better light interception and utilization [25,26], which resulting in an improvement of leaf gas exchange and Chlorophyll performance of the individual plant [27]. In this study, we found that the D20 treatment can better shape a canopy structure for optimal light distribution on different canopy layers when compared to the other treatments. Therefore, the IPAR, TPAR, light transmission and extinction coefficient K values were significantly increased (Figure 1). Similar findings were also recorded in maize under different planting densities [28]. Leaf gas exchange measurement is a good way to estimate yield [15]. Otherwise, chlorophyll fluorescence is an important approach to evaluate the light-use efficiency of photosynthesis in plants [29,30,31]. The SPAD value of functional leaves, leaf nitrogen content, leaf area, and leaf area index are important parameters for determining leaf vigor and leaf senescence, which are significant factors influencing crop photosynthesis [32]. In the present study, the result showed that the Pn and WUE in D20 plants were significantly higher than other treatments (Figure 3). This result indicated that plant density has a great effect on leaf stomatal conductance (Figure 3), and optimal plant population density may improve stomatal traits to increase net photosynthetic rate as well as water use efficiency [33]. The leaf pigment, leaf area size, and leaf nitrogen content were linearly related to the photosynthesis and adjusted to respond to the unfavorable environment of high density and exhibited a negative effect with planting density (Figure 2) [34]. However, we found that D20 plants had a significantly higher LAI than other treatments in the whole growth period, especially during the 70–90 days after planting period. That can be concluded that LAI is the main factor in determining the photosynthetic performance and final yield via capturing more light to promote photosynthesis, thus increasing the light use efficiency [35]. Higher maximal PSII quantum yield (Fv/Fm), effective PSII quantum yield (YII), and electron transfer rate (ETR) indicated that the photosynthetic performance of leaves was better [36]. Otherwise, photochemical quenching (qP) and non-photochemical quenching (NPQ) indicated the degree of openness of the PSII reaction center [14]. In the present study, we found that the Fv/Fm was no statistical difference among the treatments (Figure 4), this indicated that there was no shading stress under the field conditions. The (YII) and ETR were significantly increased in D20 treatment plants, while the qP in dense planting (D15 and D20 treatment) was significantly higher than the plants in D25 and D30 treatment plants during 50–70 days after planting period. However, the NPQ has no significant difference among treatments. This indicated that the PSII flux absorbs more energy for the photochemical process in D20 treatment plants, but the non-photochemical quenching process was limited and eventually increased the light use efficiency. The results were consistent with maize [32,37], wheat [38], and soybean [39] when responding to varying planting densities.

Balancing the relationship between source and sink development of individual plants is an important factor in determining crop yield. The allocation and accumulation of dry biomass to the root sink is the main event that occurs during tuber crop yield formation and development [40,41]. In our present study, we found that the total dry weight of individual plants was significantly increased with planting density decrease (Figure 6A,D). However, the allocation of dry matter to the root sink was increased dramatically in high-density plants (Figure 6B,E), which resulted in a higher root/shoot ratio (Figure 6C,F). These results can be explained by the fact that plants with lower planting density had an over-grown shoot development with a higher leaf fresh weight, above-part fresh weight, which can consume large amounts of dry biomass to maintain shoot development (Figure 5A,B). Besides, the longer main and average stem length and more branch number in lower planting density treatment can hinder the photo-assimilation flow to the root sink (Figure 5C–E), eventually decreasing the root/shoot ratio (Figure 6C,F). This indicated that optimal planting density can improve the coordination between root and shoot, which promotes the carbohydrate flux to the sink organs.

Among these genes, the SuSy, SPS, SSS, AGPase, SBE, and GBSS play important roles in carbohydrate metabolism [42,43,44,45]. Sucrose is decomposed into hexoses by both the sucrose synthase and invertase pathways, and the former is the main source for fructose and glucose biosynthesis, which provides the substrate for starch biosynthesis [5,46]. Our results reveal that during the bulking period, the starch and fructose content in storage roots were significantly higher when treated with the D20 appropriate planting density than when treated with the D25 control (Figure 8A,D), and the glucose content in storage root was significantly increased during 50–70 days after planting but decreased sharply at 90 days after planting (Figure 8C); however, the sucrose showed a negative correlation (Figure 8B). That is because appropriate planting density could promote the process of sucrose-starch converse by up-regulating the expression level of the carbohydrate metabolism genes (Figure 7). These results were consistent with a previous study in which sweet potatoes were treated with different potassium levels [46].

Storage root number per plant and single storage root fresh weight are both significant factors in determining the yield and commercial features of sweet potatoes [5,24,47]. In the present study, we found that by increasing the number of storage roots per plant, an appropriate planting density D20 treatment substantially increased storage root yield compared to the control D25 treatment in both years by 8.47–24.91% (Table 2). Besides, appropriate planting density can also improve the appearance by reducing the length/width ratio and increasing uniformity on individual fresh weight of storage root (Table 2). As a result, the commercial features were improved by increasing the commercial storage root number, mainly the middle-sized ones (Table 3). Eventually, the commercial storage root yield significantly increased, and the increment was by 8.96–33.46% compared with the control D25 treatment in the three-year field experiments (Table 3). The increment of yield and commercial features were mainly attributed to the improvement of storage root formation during the canopy closure period (Table 1). In our present study, we found a substantial positive correlation between the number of storage roots and the yield. However, we observed that the fresh weight of the storage roots had an opposing influence on both the yield and the number of storage roots (Table 4). These findings were consistent with the previous study [5]. Furthermore, we found a strong correlation between the photosynthetic performance and the yield and yield components. The IPAR, TPAR, Ci, Fv/Fm, and NPQ showed a positive correlation with both yield and storage root number. On the other hand, Pn, Gs, Tr, qP, YII, and ETR exhibited a positive correlation with storage root fresh weight (Table 4). The findings suggest that the distribution of light throughout the canopy is a crucial element in determining the amount of storage roots. Planting density has a crucial role in influencing micro-climate at different levels of the canopy by controlling the distribution of sunlight. This, in turn, provides suitable atmospheric and soil conditions for the development of adventitious roots during the early growth stage [48]. However, further studies should be undertaken to validate our existing hypothesis. The correlation analysis also suggested that the photosynthesis capacity of individual plants is dependent on root welling. The efficiency of photosynthetic electron transfer and gas exchange are important factors in the deposition and accumulation of photo assimilation in the storage root. These factors contribute to the growth of larger storage roots.

5. Conclusions

Enhancing planting density is a critical factor in increasing agricultural yields. In the three-year field experiment, the yield increment of sweet potato was by 8.47–24.92% during the three growth periods through Haikou and Sanya city under appropriate planting density D20 treatment. The observation can be attributed to the fact that the optimum planting density can improve canopy structure, resulting in better sunlight distribution. Hence, the IPAR, TPAR, light transmission, and extinction coefficient in different canopy layers were significantly increased. As a result, improving individual plants’ photosynthetic performance by increasing the Pn, Tr, Ci, Gs, and WUE, as well as the chlorophyll fluorescence parameters. And eventually enlarge the sink strength by increasing the storage root number, particularly medium-sized ones. Therefore, improving the appearance of storage roots is accomplished by lowering the length/diameter ratio and maintaining a more uniform fresh weight distribution across roots of different sizes. These findings provide a solid foundation for the development of high-value sweet potatoes.