Xiangwei Gong, Uzizerimana Ferdinand, Ke Dang, Jing Li, Guanghua Chen, Yan Luo,Pu Yang, Baili Feng*

State Key Laboratory of Crop Stress Biology in Arid Areas, Shaanxi Research Station of Crop Gene Resources & Germplasm Enhancement,Ministry of Agriculture,College of Agronomy, Northwest A&F University,Yangling 712100,Shaanxi,China

A B S T R A C T

1. Introduction

Proso millet (Panicum miliaceum L.), a minor grain crop, is planted mainly as a source of staple food and fodder in semiarid regions of China [1,2]. In recent years, with a readjustment of China's agricultural structure, minor grain crops are receiving increased attention. However, promoting the cultivation of minor grain crops and increasing yield are major problems. Selecting a reasonable and sustainable planting pattern for improving the efficient utilization of resources,protecting the environment, and meeting current food requirements are desired goals. We adopted an existing intercropping pattern and investigated its effect on crop yield.

As a planting pattern that involves growing two or more crops in the same space at the same time, intercropping maximizes available water, nutrients, and solar energy to increased crop yield relative to monoculture cropping [3,4].Intercropping has become increasingly popular, producing 15%-20%of the world's food supply[5].An ideal intercropping system can boost grain yield by 1.3 times relative to the yield of a single-crop system [6]. Intercropping can achieve complementary resource use and niche differentiation in space and time to optimize resource-use efficiency and crop yield simultaneously [7]. Intercropping systems produce a high land equivalent ratio (LER) on a limited land area and with minimal negative impact on the environment[8].One hectare of land under intercropping produced the same crop yield as 1.22 ha under monoculture [9]. Despite the benefits of intercropping, limited data are available on the use of resources such as nutrients, water, and light among two crops in time and space to promote photosynthesis, growth,and grain yield in intercropping systems.

The microclimate environment within the crop canopy,particularly with respect to light irradiation, is crucial to crop productivity and can result from either high interception of solar radiation or high light use efficiency [10]. Because intercropping systems include tall and short species, a substantial amount of light passes through the canopy of tall-statured crops to influence plant growth and development[11,12].Intercropped maize received high solar radiation by increasing photosynthetically active radiation (PAR) and radiation use efficiency (RUE) while decreasing sunlight reflectance by intercepting 3D light in maize-cabbage relay intercropping systems,instead of the planar light received by monocropped maize[13].In view of the combined influence of different spatial and temporal uses of radiation among component crops, intercropping shows greater radiation capture potential and utilization than monocropping. However, these light-environment changes may affect the morphology and physiological and ecological characteristics of plants [14]. There is a tradeoff between leaf photosynthetic capacity and light distribution in a canopy structure [15].Maize and soybean intercropping is a popular system worldwide, and intercropping advantages have been related to a high radiation interception of dominant species,especially C4species [14]. Intercropping of maize and cowpea received more solar radiation by increasing the total leaf area index of the component crops compared with sole cropping, thereby increasing light use efficiency[16].When other resources(e.g.nitrogen and water) are not limiting, PAR is the most vital resource for crop growth and development or yield in intercropping systems with tall-statured crops [13]. However,the quantitative relationships between yield and solar radiation in the canopy of an intercrop remain unclear.

Detailed study could reveal how a proso millet crop maximizes its efficiency of use of light resources to increase its yield. This alteration may increase leaf photosynthetic capacity and the translocation rate of vegetative organs to grain,thereby increasing proso millet productivity and LER.

The objectives of this study were as follows:(i)to evaluate PAR in the canopy and RUE and the photosynthetic characteristics of intercropping crops in several mixtures; (ii) to investigate variation in dry matter accumulation and translocation as well as productivity of proso millet in proso milletmung bean relay intercropping systems; (iii) to explain changes in yield from the perspective of light environment on the Loess Plateau of China.

2. Materials and methods

2.1. Experimental site

A field experiment was conducted in the Yulin Modern Agriculture Demonstration Garden of Yulin, Shaanxi (37°56′26″ N, 109°21′46″ E, and 1120 m above sea level) during the 2017 and 2018 growing season.The site has a temperate semiarid continental monsoon climate with a mean annual temperature of 8.3 °C and an annual mean precipitation of approximately 400 mm, which occurs mainly during the growing season of proso millet from June to September. The precipitation (mm) and mean air temperature (°C) during the proso millet growing season in the two studied years are presented in Fig. 1-A. The soil in the study field had a loesslike loam texture with pH 8.6, 6.2 g kg-1organic matter,0.61 g kg-1total N,39.6 mg kg-1available P,and 221.3 mg kg-1available K in the 0-20 cm soil layer.

2.2. Experimental design and treatments

The experimental design was a split plot with randomized blocks with three replicates.The proso millet cultivar‘Shanmi 1' and mung bean cultivar ‘Zhonglyu 8', both adapted to the local conditions,were selected.The treatments were as follows:(1)two rows of proso millet alternating with two rows of mung bean (2P2M); (2) four rows of proso millet alternating with two rows of mung bean (4P2M); (3) four rows of proso millet alternating with four rows of mung bean (4P4M); and (4) two rows of proso millet alternating with four rows of mung bean(2P4M). Sole proso millet (SP) and sole mung bean (SM) were planted as controls. Rows were oriented in the north-south direction.Fig.1-B shows a schematic view of the intercropping patterns. All rows, irrespective of crop, were planted 33 cm apart. Each experimental plot for three-strip proso millet and mung bean had an area of 30 m2(6 m×5 m).

Fig.1- Monthly rainfall and mean temperature from May to September in the 2017 and 2018 growing seasons(A)and schematic illustration of row layout of proso millet and mung bean in the experimental planting patterns(B).

In 2017, proso millet was sown on June 12 and mung bean on May 28, with harvesting dates of September 23 for proso millet and August 24 for mung bean. In 2018, the dates of sowing were June 10 for proso millet and May 25 for mung bean and the dates of harvesting were September 20 for proso millet and August 18 for mung bean. Urea and superphosphate were applied to the soil before planting.The application rates were as follows: N (150 kg ha-1), P2O5(100 kg ha-1), and K2O (75 kg ha-1). Disease occurrence was treated by chemical spraying as needed. Weeds were controlled by hand weeding, particularly in the early part of the growing season.

2.3. Measurements

2.3.1. PAR and LAI (leaf area index)

PAR and LAI in each row configuration of the intercropping systems were measured at anthesis (65 days after sowing) in 2017 and 2018. Anthesis for proso millet is the stage when vegetative and reproductive growth both occur. At this time,metabolic ability is the highest and growth is most exuberant[17,18]. Each treatment for the PAR and LAI of the intercrops was measured between 10:30 and 14:00 on a sunny day, and each measurement was duplicated three times at different heights with a linear AccuPAR LP-80 ceptometer (Decagon Devices, Pullman, WA USA). The ceptometer measures light across 80 sensors embedded along an 84 cm probe. The canopy was divided into the upper (two-thirds of canopy height), middle (one-third of canopy height) and lower (soil surface) zones according to the number of plant nodes. At least five readings of PAR and LAI were taken at each height from the different plots and measured from the bottom to the top in the vertical direction. The diurnal course of PAR was measured every 2 h on a sunny day from 8:00 to 18:00. Mean PAR and LAI between proso millet rows (border and middle rows)were calculated.

RUE(g MJ-1)was established as follows[19]:

where, ADM is the accumulated dry matter of the intercrops (g m-2) and I is the quantity of daily incident PAR (MJ m-2). Daily total radiation values were multiplied by 0.5 for the conversion of total radiation to PAR. F represents the fraction of the (intercepted photosynthetically active radiation) IPAR of the intercropped proso millet on certain days. Thus, F was used in the calculation of the cumulative IPAR of the intercrop. The RUE is the PAR use efficiency.

2.3.2. Photosynthetic characteristics Net photosynthetic rate (Pn) was measured using a Li-6400 portable photosynthesis system (LI-COR Inc., Lincoln, NE,USA) equipped with a LED leaf chamber. The leaves were selected for measurement of photosynthetic parameters at anthesis. All measurements were made from 9:00 to 11:00 under a CO2concentration of 400 μmol mol-1and leaf chambertemperatureof25 °C.Aleaf wasplaced in a 6 cm2chamberatphotonflux densities of 1400,1200, and 1000 μmol m-2s-1for the upper, middle and lower canopies, respectively. At least five readings of the photosynthetic parameters were taken at each treatment from the different plots.

The chlorophyll content (SPAD value) was determined using the Minolta SPAD-502 chlorophyll meter (Minolta Camera Co.Ltd.,Osaka,Japan)in accordance with the method described by Abdelhamidg et al. [20]. The measurement was conducted five times for each leaf from each plot, and the mean was calculated as the SPAD value of the given leaf.

2.3.3.Illuminance, air temperature, and relative humidity

Illuminance was measured using a ZDS-10 illuminometer(Jiading Automation Instrumentation Co., Ltd., Shanghai,China), and air temperature and relative humidity were measured using a DHM2 psychrometer(Fengyang Instrument Industry & Trade Co., Ltd., Tianjin,China). All measurements were made from 12:00 to 14:00 on a sunny day at anthesis in 2017.

2.3.4.Dry matter

Stem, leaf, sheath, and spike samples of proso millet were collected at the heading, anthesis, filling and maturity stages in 2017. On each sampling occasion,five similar plants in the same row were selected randomly from the different plots and cut at the ground level. The samples were baked in an oven for 30 min at 105 °C. They were then dried to constant weight at 80 °C and weighed.

For determination of the recovery effect of the intercropped proso millet, the transportation amount or rate of dry matter in the vegetative organs and the contribution of vegetative organs to the grain using the following equations:

DTA ＝LDW-DWM;DTR ＝DTA/LDW×100%;GCR＝DTA/GDW×100%

where DTA (g plant-1) is the transportation amount of dry matter in the vegetative organs, LDW (g plant-1) is the largest dry weight of the vegetative organ, DWM (g plant-1) is the dry weight of the same vegetative organ at maturity, DTR is the transfer rate of dry matter in the vegetative organ (%), GCR is the contribution of vegetative organs to grain (%), and GDW (g plant-1) is the dry weight of the grain.

2.3.5.Yield and LER

Intercropped proso millet and mung bean were manually harvested and recorded according to the actual harvested area for each plot, whereas sole proso millet and mung bean were manually harvested and counted according to four rows in the middles of plots. Ten plants in each plot were selected randomly to determine grain yield per plant and 1000-kernel mass of proso millet and mung bean at harvest, respectively.Grain was air-dried for 2-3 weeks. LER was used as an indicator of land productivity and was calculated [21]as

where, Yspand Yipare the sole and intercropped proso millet yields, respectively, and Ysmand Yimare the sole and intercropped mung bean yields, respectively. An LER ＞ 1 signifies that the intercrop is more productive than the sum of the sole crops of its component species.

2.4. Data analysis

Means were compared by analysis of variance(ANOVA).SPSS 19.0 (SPSS Institute, Chicago, USA) was used. Differences between means were determined from the least significant difference(P ＜0.05).

Fig.2- Diurnal variation of average PAR on proso millet(A)and PAR at the soil surface(lower),one-third of canopy height(middle),and two-thirds of canopy height(upper)in five experimental planting patterns(B)at anthesis in 2017 and 2018.Each value represents the mean of five replications.The columns followed by different letters are significantly different at P ＜0.05.

3.Results

3.1. Effect of different intercropping patterns on canopy light distribution and utilization of proso millet

Fig.3- RUE in five experimental planting patterns at anthesis in 2017 and 2018.Each value represents the mean of five replications. Values labeled with different letters are significantly different at P ＜0.05.

The light environments of the experimental planting patterns are shown in Fig. 2-A. Diurnal variations of average PAR on proso millet were significantly higher than those on the monoculture, and the 2P4M showed the maximum PAR over the entire day.

The important feature of the intercropping systems is the change in PAR spatial distribution to alter the planar light received by the sole system of two crops to 3D light intercepted by the intercropping systems. In this case, the PAR of the intercropping systems was significantly higher than that of the sole systems because of the little shade from the near proso millet, including the lower, middle and upper zones (Fig. 2-B). The order of PAR among the canopy levels was as follows:upper ＞middle ＞lower zones.The mean PAR intensity increased by 2.2%-23.4%, 19.8%-59.7%, and 61.2%-133.3% in the upper, middle and lower canopies (P ＜0.05),respectively,compared with SP.The 2P4M pattern showed the highest PAR among the intercropping patterns for each canopy structure.

The RUE fraction was related to LAI, crop height, and intercropping configurations. Across years and treatments,the systems showed the following order of RUE values:2P4M ＞4P4M ＞2P2M ＞4P2M ＞SP, indicating that the RUEs of the intercrops were higher than those of the sole crops(Fig.3).This difference may cause changes in the morphological and physiological characteristics of the proso millet leaves under different treatment conditions.

3.2. Effect of intercropping pattern on leaf growth and photosynthetic characteristics of proso millet

The row numbers significantly affected the LAI values of the whole canopy in the proso millet-mung bean relay intercropping systems. The order of LAI in the canopy structure was as follows: upper ＜middle ＜lower, owing to the lower leaf area density (Fig. 4). Across years and treatments, the mean LAI values of proso millet in 2P2M,4P2M, 4P4M, and 2P4M were respectively 1.39, 1.23, 1.52, and 1.67 times greater than that of SP.

The Pnvalues under the experimental intercropping patterns were high. The Pnvalues in the top, middle and lower layers of the proso millet canopy were greater by respectively 8.8%-32.5%,16.0%-46.3%,and 25.0%-114.4%than that of SP (P ＜0.05) (Fig. 5-A). SPAD showed the same results trend among the canopy structures. Across years and treatments, 2P4M showed the greatest capture among the four planting patterns:respectively 18.8%and 23.4%higher than SP in 2017 and 2018(P ＜0.05)(Fig.5-B).

Fig.4- LAI at the soil surface(lower),one-third of canopy height(middle),and two-thirds of canopy height(upper)in five experimental planting patterns at anthesis in 2017 and 2018.Each value represents the mean of five replicates.Values labeled with different letters are significantly different at P ＜0.05.

3.3.Effect of intercropping pattern on dry matter accumulation and translocation characteristics of proso millet

Significant differences were observed in response to different intercropping patterns at the same growth stage(Table 1).The aboveground dry matter weights under 2P2M, 4P2M, 4P4M,and 2P4M treatments were increased by respectively 24.7%,8.9%, 31.4%, and 39.9% for the stem; 22.4%, 9.0%, 31.6%, and 40.8% for the leaf; 30.8%, 15.9%, 44.4%, and 59.7% for the sheath; and 31.5%, 15.7%, 51.9%, and 70.0% from the spike,compared with SP at anthesis (P ＜0.05). However, except for spike dry weight,no significant differences were observed for the stem, leaf, and sheath between 2P2M and 4P4M at the maturity stage.

Intercropping pattern strongly influenced the dry matter distribution of aboveground tissue (Table 2), which occurred with the transformation of the growth centre. Intercropping can promote aboveground dry matter accumulation in vegetative organs,including the stem,leaf and sheath,which form a large photosynthetic source. The dry matter weight in the spike increased with the rapid growth of the plant when reproductive growth occurred,promoting grain enrichment.

At the heading stage, intercropped proso millet showed a dry matter content in the spike greater by a mean of 40.2%than sole proso millet. However, with growth, intercropped proso millet showed a low dry matter content in the spike,with means of respectively 9.0%, 7.0%, and 1.5% at the anthesis, filling, and maturity stages (P ＜0.05). In particular,the 2P4M treatment showed a high dry matter content in the spike, greater by 68.2%, 15.6%, 15.4%, and 2.0% at the four stages than SP.

Fig.5-The Pn(A)and SPAD(B)at the soil surface(lower),one-third of canopy height(middle),and two-thirds of canopy height(upper)in five experimental planting patterns at anthesis in 2017 and 2018.Each value represents the mean of five replicates.Values followed by different letters are significantly different at P ＜0.05.

In the intercropping systems, the four patterns showed a significant additive effect on the transfer of photosynthates from the stem, leaf and sheath to the grain (Table 3). On average, DTA from the stem increased by 47.6%, DTR by 16.6%, and GCR by 46.1% (P ＜ 0.05), compared with SP.Similarly, the DTA, DTR, and GCR of the intercropping systems were increased by 39.3%, 10.5%, and 39.2% from the leaf (P ＜0.05) and by 148.8%, 82.8%, and 155.3% from the sheath (P ＜0.05). 2P4M showed the highest DTA, DTR, and GCR values.

3.4. Effect of intercropping pattern on illuminance, air temperature, and relative humidity in proso millet canopy

Significant differences in illuminance, air temperature, and relative humidity in the proso millet canopy were observed in the lower, middle, and upper zones in response to intercropping pattern. This result can be attributed to the double effects of interspecific resource competition and crop row ratio setting (Table 4). The illuminance and air temperature of the intercropping systems significantly decreased in comparison with SP; they were 4.9%-19.4%and 2.8-11.0% lower in the upper zones, 9.2%-58.4% and 2.3%-9.7% lower in the middle zones, and 25.3%-70.5% and 2.0%-9.6% lower than SP in the lower zones (P ＜ 0.05).However, the opposite results were obtained for relative humidity, which was higher by 7.8%-22.1%, 5.6%-22.8%, and 9.0%-25.1% in the upper, middle and lower zones of the proso millet monoculture (P ＜0.05).

3.5. Effect of intercropping pattern on proso millet yield and LER

Proso millet yield and LER under relay intercropping and sole cropping are shown in Table 5. Row configurations significantly affected intercropping yield and LER, and all indexes in the table were greater in intercropping than in sole proso millet. The mean yields in 2P2M, 4P2M, 4P4M,and 2P4M were greater by respectively 17.1%, 6.8%, 20.1%,and 37.3% than that in SP in both seasons (P ＜0.05). The LERs for the intercropping patterns were all greater than unity (＞1), indicating that the cropping systems used less land but produced more grain than the corresponding sole plantings (Table 5). The maximum LER was 1.86 for 2017 and 2.22 for 2018 under the 2P4M treatment, whereas the minimum LER was 1.50 for 2017 and 1.71 for 2018 under the 4P2M treatment, suggesting that adding mung bean rows or reducing proso millet rows increased the productive use of land.

3.6.Relationship between PAR,RUE,Pn,and proso millet yield

Statistical analysis showed a positive curvilinear relationship(R2=0.5929-0.9127**)between PAR,RUE,Pn,and yield for the intercropped proso millet (Fig. 6-A-C.). The correlation coefficients were statistically significant(P ＜0.01),indicating that high PAR enhancing the photosynthetic characteristics of the leaves by increasing radiation use efficiency. This result also provides the best explanation why changing canopy structure by increasing PAR increases the yield of intercropped crops.

Table 1-Dynamics fn dry matter accumulation among organs of proso millet in 2017.

4. Discussion

4.1. Intercropping boosted proso millet yield

A yield advantage in intercropping can be achieved only when the component crops have resource use efficiency in the same time and space [22]. Prasad and Brook [23] reported that intercropping increases the total amount of radiationintercepted,owing to the rapid establishment of ground cover by the combined canopies of the component crops. An ideal intercropping system can increase the yield of one or two crops and has other advantages,including lowered inputs and minimized environmental impact [24]. In the present study,intercropping patterns were integrated into the alternative systems.The yield was higher under intercropping treatment than in other conventional planting patterns, across the two studied years (Table 5). The 2P4M pattern was applied as an intercropping system and increased grain yield by 20.7% and 53.9% in 2017 and 2018, respectively, compared with conventional monoculture,on the same base area.The mean LERs for the intercropping patterns were all greater than unity (＞1),and the maximum LER values were 1.86 for 2017 and 2.22 for 2018 under 2P4M treatment, whereas the minimum LER values were 1.50 for 2017 and 1.71 for 2018 under 4P2M treatment(Table 5),indicating that the cropping system used less land but produced more grain than the corresponding monocrops.This finding can be attributed to the large amount of PAR received (Figs. 6, 7), leading to improved photosynthetic characteristics for the whole canopy and increased transfer of aboveground dry matter to the spike from leaves,stems, and sheaths during the late-growth stage. These conclusions are similar to those of many previous studies on maize/soybean [25], maize/pea [26], and alfalfa/maize [27]intercropping systems.Liu et al.[28]reported that the red/farred ratio at the top of the soybean canopy is reduced by 17%-21% more than the PAR under maize/soybean intercropping compared with the monoculture, which leads to morphological changes such as increased internode lengths and reduced branching and reduction in grain yield. The results of the present study indicate that PAR is a key determinant of yield in plant intercropping systems because it improves the photosynthetic characteristics of the tall-statured crop.

4.2. Intercropping altered canopy light distribution and enhanced leaf photosynthetic characteristics of proso millet

Light is crucial for crop growth because it affects leaf morphological traits and leaf physiological traits [29]. A system's light environment can be changed by the different planting patterns of intercrops [30,31]. In the present study,the PAR of the intercropping systems was significantly higher than that of the monocrops(Fig.2),inducing a high amount of solar radiation to the tall-statured crop. In a relay-strip intercrop, two species can compete for light, and their competitive ability shifts with their niches [4]. Proso millet is a cereal and C4 crop that features high photosynthetic and Cgain activities. Thus, the high efficiency of light interception and utilization occurred in the intercropped proso millet as a result of reduced interspecies competition,and the PAR of the tall-statured proso millet in 2P4M was the highest among the intercropping patterns because of the border row advantage and spatial complementary advantage of intercropped proso millet plants (Fig. 7). These findings are similar to previously reported results [32,33]. Adequate illumination conditions improved field microclimate of the intercropped proso millet,resulting in low illuminance, air temperature and highrelative humidity (Table 4), which could prevent excessive transpiration of proso millet leaves,especially for the ecological environment in the dry areas of the Loess Plateau. The evaporation of water on the soil surface was greatly weakened[34], or a portion of the water lost from the soil evaporation channel was transferred to the channel of plant transpiration[35,36]. Thus, the increased PAR was mainly responsible for the high proso millet yields, given that the yield of intercropped proso millet under the intercropping patterns showed a positive curvilinear relationship with their PAR,RUE,and Pn(Fig.6).

Table 4-ANOVA results for the effects of different intercropping patterns on proso millet illuminance,air temperature,and relative humidity in different canopy levels in 2017.

Table 5-ANOVA results for the effects of intercropping pattern on proso millet yield,mung bean yield,and LER in 2017 and 2018.

Fig.6- Relationship between PAR(A),RUE(B),Pn (C)and yield of proso millet in 2017 and 2018.**Significant at P ＜0.01.

Greater productivity improvement of crop systems can result from either greater interception of solar radiation and high light use efficiency or a combination of the two [37]. In our experiment, the RUE of the tall-statured proso millet in the intercropping system was higher than that of the sole cropping system because of the increase in diffused light and less light saturation (Fig. 3). This result agrees with the findings of previous studies [19,30]. The main reason was that the greater light area and intensity from the marginal effect were directly stimulated under intercropping systems for combining tall and short species,resulting in an advantage of light energy utilization [13]. Moreover, this conclusion can be attributed to the LAI that was calculated based on leaf area density and land area.Across years and treatments,the mean LAI values of proso millet in 2P2M, 4P2M, 4P4M, and 2P4M were respectively 1.39, 1.23, 1.52, and 1.67 times higher than that of monoculture (Fig. 4), indicating that high LAI beneficially alters the distribution and utilization of PAR within plant canopies[38].

Crop leaves are sensitive to the light environment, and photosynthetic capacity can be influenced by intercropping conditions [39,40]. In the present study, proso millet had competitive advantages in the above-ground part of the tall crops because it promoted plant growth by maximizing additional light. This phenomenon may be attributed mainly to the photosynthetic capacity, including Pnand SPAD,of the intercropped proso millet(Fig.5).The photosynthetic capacity of leaves depends on light intensity primarily because the photochemical reaction is characterized by the oxidation of water molecules in the optical drive, and the release of electrons is transmitted into the thylakoid cavity, eventually forming ATP. Therefore, the light environment is particularly important for photosynthesis.Similarly,in proso millet/mung bean intercropping systems, high PAR and RUE values could significantly promote the absorption of solar radiation within limits based on leaf morphological and physiological traits,thus improving leaf photosynthetic characteristics[41].

The architecture of the canopy, which is affected by crop densities, crop height, and row arrangement, is the deciding factor for crop intercepted PAR [42]. In the canopy layers, the order of PAR was as follows: upper ＞middle ＞lower among the canopy structure (Fig. 2-B). Our results agree with the findings of other authors, such as Awal et al. [43], who observed a consistent order in response to changes in PAR in the canopy structure. The significant differences under this condition can be attributed to two reasons. One is that the intercropped proso millet did not create a large amount of shade and illuminance from nearby plants, which could provide a high light energy basis for the improvement of PAR and field microclimate (Table 4). The other is that the light distribution in the canopy also affected Pnin the proso millet.Therefore,the high PAR and LAI in the intercropping systems resulted in the efficient use of light for canopy photosynthetic capacity.

4.3. Intercropping increased dry matter accumulation and translocation of proso millet

Light quantity (PAR) and quality (R/FR ratio) changes can be affected by the intercropping configuration and crop architecture and alter both morphology and growth of a crop plant.Dry matter accumulation, which is a comprehensive reflection of group quantity and individual quality,can affect grain yield response to the availability of resources[44].Therefore,a study under the intercropping systems on crop aboveground dry matter accumulation and translocation is highly necessary to obtain high crop yield [45]. The mechanisms of high productivity of cereal/legume combinations, which are considered as typical, were characterized as a result of interspecific interactions and facilitations [29]. In the present study,the intercropping of proso millet with mung bean significantly influenced the optimization of the dry matter accumulation and translocation of aboveground tissue (Table 1). The DTA,DTR, and GCR from the stem, leaf and sheath were significantly greater under the four intercropping treatments than under the control treatment (Table 3). the 2P4M treatment showed the highest effect in optimizing the dry matter translocation of aboveground tissue in proso millet because each row belongs to border rows,which did not create a large amount of shade for the nearby plants, and thus improved the transformation of the photosynthetic product [33].This result is consistent with the previous finding [46].that the dry matter accumulation and translocation of maize are greater in intercropping than in the monoculture Hence, improved agricultural practices that ensure large light area and maintain high amounts of nutrients can provide a favorable environment for improved crop yield by strengthening transfer amount, transfer rate, and contribution to grain yield of dry matter from the stem,leaf and sheath.

5. Conclusion

Intercropping increased the mean PAR in proso millet canopies and RUE in comparison with monoculture. The changes in the light environment increased the leaf photosynthetic capacity, including Pnand SPAD, of the whole canopy structure. An intercropping system with high light and suitable field microclimate is effective for increasing dry matter accumulation and contribution to the grain yield from the stem, leaf, and sheath compared with monoculture. Significant differences were observed in system productivity as a result of increased yield and LER.The intercropping systems showed more suitable light environments and greater resource utilization than the sole cropping systems, which may account for the yield advantage of intercropping. The optimum combination of the intercropping system on the Loess Plateau of China was 2P4M.

Declaration of competing interestNo conflict of interest exists in the submission of this manuscript, and the manuscript has been approved by all authors for publication.AcknowledgmentsThe research was supported by the Earmarked Fund for China Agriculture Research System (CARS-06-13.5-A26), National Natural Science Foundation of China (31371529), National Key Research and Development Program of China(2014BAD07B03), Shaanxi Province Key Research and Development Projects (2018TSCXL-NY-03-01), and Minor Grain Crops Research and Development System of Shaanxi Province(2009-2018).