Maize–soybean relay cropping increases soybean yield synergistically by extending the post-anthesis leaf stay-green period and accelerating grain filling

Yiling Li, Ping Chen, Zhidan Fu, Kai Luo, Ping Lin, Chao Gao, Shanshan Liu, Tian Pu, Taiwen Yong*,Wenyu Yang

College of Agronomy, Sichuan Agricultural University/Key Laboratory of Crop Ecophysiology and Farming System in Southwest, Ministry of Agriculture and Rural Affairs/Sichuan Engineering Research Center for Crop Strip Intercropping System, Chengdu 611130, Sichuan, China

ABSTRACT Relay cropping of Poaceae and Fabaceae promotes high yield and land-use efficiency by allowing a double harvest.However, it is difficult to increase yield synergistically because of the reduced photosynthetic abilities of legume leaves under the shade of graminoids.Leaf photosynthetic capacity in relay cropping systems is associated with ecological niche differentiation and photosynthetic compensation after restoration of normal light.We conducted a field experiment in southwest China in 2020–2021 to evaluate the effects of three cropping patterns: maize–soybean relay cropping (IMS), monoculture maize(MM), and monoculture soybean (SS), and N application levels: no N application (NN:0 kg N ha-1),reduced N (RN: 180 kg N ha-1), and conventional N (CN: 240 kg N ha- ).Compared to monocropping,relay cropping increased the stay-green traits of maize and soybean by 13% and 89%, respectively.Relay cropping prolonged the leaf stay-green duration in the maize and soybean lag phase by almost 4 and 8 days, respectively.Relay cropping maize (IM) increased the leaf area index (LAI) by 79.4% to 88.5% under NN and 55.5% to 148% under RN.Relay cropping soybean (IS) increased the LAI from 115%to 437% at days 40 to 50 after anthesis.IM increased yield by 65.6%.IS increased yield by 9.7%.HI and system yield were at their highest values under RN.In the relay cropping system, reduced N application extended green leaf duration, increased photosynthesis inside the canopy at multiple levels, ultimately increases soybean yield synergistically.

1.Introduction

Worldwide attention is focused on developing production methods to ensure food security for the constantly growing human population in the face of declining resources[1,2].With a continual reduction in arable land, increasing pollution of the environment,and shortage of land resources, increasing crop productivity and reducing the environmental footprint of cropping is a challenge for China and the world [3–5].

Intercropping is a widely used system that provides yield advantages and environmental benefits over monocropping [6].In particular,Poaceae and Fabaceae relay cropping has been found to stimulate shoot growth, biomass partitioning, and yield increases by allowing intercrops with seasonal and spatial niche differences to use more of the light falling on the crop canopy and of the soil nutrients and to benefit from ecological control of disease, pests, and weeds [7].

Green leaves are necessary for plant photosynthesis [8].Staygreen,a desirable trait in which both green leaf area and photosynthetic competence are retained after anthesis, especially during grain filling, is generally associated with higher grain yields than occur in its absence [9,31].In response to developmental age and to multiple internal and external environmental factors, leaves undergo physiological deterioration and degeneration including loss of chlorophyll and termination of photosynthate assimilation[10].Crops with functional stay-green genotypes or phenotypes exhibit delayed leaf senescence and increased grain accumulation after anthesis [11–15].Prolonging the green leaf area period has been shown to favor canopy photosynthesis and net assimilation rate (NAR) [16].The extended expression of photosynthesisassociated genes (rbcS and cab), the higher expression of Rubisco activase and soluble starch synthase in flag leaves,and the reduced pheophorbide a oxygenase activity in stay-green mutants suggest that extension of the green leaf area converts more sunlight to energy.Sufficient energy is a prerequisite for grain growth, and roots responsible for absorbing nutrients and water would benefit from more energy produced by shoots.The stay-green genotype was found to have more vertical root length and produce more lateral roots,supporting more water uptake to satisfy water demands during the grain-filling phase [9,17–19].Environmental factors,such as light and N fertilizer, may affect leaf stay-green.Red light promotes the absorption and utilization of nitrogen in plants to delay leaf senescence,and mild N starvation encourages biological nitrogen fixation in leguminous plants to preserve leaf photosynthetic ability [20,21].

Although relay cropping of graminoids and legumes can increase land output and graminoid production,in maize(Zea mays L.) and soybean (Glycine max (Linn.) Merr.) or pea (Pisum sativum L.)relay cropping,little legume yield is realized.However,in recent years, southwest China has witnessed widespread adoption of maize–soybean relay cropping that results in yield gains for both crops.In addition to increased system yield, this relay-cropping technique has increased light energy and nitrogen fertilizer utilization rates [22–26].But the mechanism of the synergistic yield increase in relay cropping is still unclear, especially with respect to the relationship between leaf lifespan and the yield formation of the component crops.

In this study we investigated the roles of leaf stay-green duration, photosynthetic characteristics, and the accumulation and translocation of dry matter in maize and soybean to determine how these processes affect relay cropping system yields at several levels of N fertilization.In a two-year field experiment in Sichuan province,China,a relay-cropping area,we aimed to test the following hypotheses: (1) Leaves in relay cropping retain their green color and remain on the plant for a longer duration than those under monocropping.(2)The relay cropping system increases productivity by promoting a longer period of leaf greenness and faster grain accumulation rate.(3) Exploiting the contrasting nitrogen responses of maize and soybean will increase system yield while reducing N fertilizer application.

2.Materials and methods

2.1.Site description and plant material

From 2020 to 2021, field experiments (started in 2012) were conducted in Ren-shou county (30.07′N, 104.18′E), Sichuan province, China, where the climate is subtropical monsoon humid,with a mean annual temperature of 17.4 °C, precipitation of 1009 mm,and 1197 h of sunlight.Daily precipitation and temperature data from the two years of the spring maize–summer soybean growing season are shown in Fig.1.This experiment employed a compact maize variety (Denghai 605, provided by Shandong Denghai Seeds Co., Ltd.) and a shade-tolerant soybean variety (Nandou 12, provided by Nanchong Academy of Agricultural Sciences in Sichuan province).

2.2.Experimental design and management

The field experiment followed a two-factor split-plot design with three replicates.It comprised three levels of total N application: 0, 180, and 240 kg N ha-1, designated as no N (NN), reduced N (RN), and conventional N (CN), respectively, and three cropping patterns each: maize–soybean relay cropping (IMS) in which two maize rows (the ‘‘maize strip”) were alternated with two soybean rows (the ‘‘soybean strip”), monoculture maize (MM) and monoculture soybean (SS) (Figs.S1, S2).The maize density for both maize cropping patterns was 58,500 plants ha-1and the soybean density was 11,700 plants ha-1.Maize was sown on April 2,2020 and April 7, 2021 and harvested on August 1, 2020 and August 2, 2021.Soybeans were sown on June 7, 2020 and June 6,2021 and harvested on November 3, 2020 and October 31, 2021.

Both crops were fertilized with urea for N, calcium superphosphate for P, and potassium chloride for K.The P and K fertilizers were applied as base fertilizers at respectively 105 kg P2O5ha-1and 112.5 kg K2O ha-1for maize and 63 kg P2O5ha-1and 52.5 kg K2O ha-1for soybean.The N fertilizer for maize was applied in the RN and CN treatments in two ways: 72 kg N ha-1as base fertilizer and the remaining amount as topdressing.For IMS,the N topdressing for IM was combined with the soybean base fertilizer and strip-placed at a distance of 25 cm between maize rows and soybean rows [27].

2.3.Plant sampling

Data for maize and soybean were collected at days 0,10,20,30,40, 50, and 60 after anthesis (when more than half of plants had started flowering) each year.On each sampling occasion, five uniform and representative plants were selected from each plot.All leaves were collected and photographed for estimation of green leaf area parameters with ImageJ (version 1.53t, National Institutes of Health, Bethesda, MD, USA; Available at: https://imagej.nih.gov/ij/download.html).Leaves,stems,grain,and seeds were bagged separately and dried at 80°C for at least 3 days before weighing as dry matter.Whole plants from a 6 m-long strip were collected from each plot during harvest to determine yields when the water content of the grain or seeds was approximately 13.5%for weight calculation and yield per hectare estimation[28].Yield components were measured for a total of 30 plants with uniform and continuous growth.

2.4.Calculations

Leaf stay-greenness (SG, %) per plant after anthesis was calculated as [29].

Fig.1.Daily air temperature and precipitation during the maize and soybean growing period in Renshou in 2020 and 2021.

The contribution rate of post-anthesis dry matter translocation to grain (TCR) was calculated as [30].

where YIMis the yield of the relay-intercropped maize, YMMis the yield of monoculture maize, YISis the yield of the relayintercropped soybean,and YSSis the yield of monoculture soybean.

The leaf senescence rate of maize or soybean after anthesis was represented as the first derivative of leaf stay-greenness (SG):

where ΔSG is the change in SG and ΔT is the time step of dynamic calculation in the SG logistic curve.

The grain accumulation rate (for maize, GAR) or the seed accumulation rate(for soybean,SAR)after anthesis was represented as the first derivative of grain or seed dry matter (GDM/SDM):

2.5.Statistical analysis

Two-way ANOVA with Tukey’s post hoc test(P<0.05)was used to identify differences in LAI, NAR, and dry matter translocation characteristics among treatments.Origin Pro (version 2021, OriginLab Corporation, Northampton, MA, USA;) was used to simulate and plot the stay-greenness logistic curve.SPSS Statistics (version 26.0, IBM Corp., Armonk, NY, USA;) and SPSS Amos Graphics (version 26.0, IBM Corp., Armonk, NY, USA;) was used to construct structural equation model (SEM) for analyzing the influence of cropping pattern and N application level on yield.

3.Results

3.1.Leaf stay-green and photosynthetic characteristics

Relay cropping increased the leaf greenness of maize after anthesis.Relay-cropped maize (IM) had darker green leaves than monoculture maize(MM).NN significantly reduced leaf greenness in both the relay crop and monocrop (Fig.S3).At the beginning seed stage of soybean (R5), relay-cropped soybean (IS) displayed more luxuriant foliage growth than monocropped soybean (SS)(Fig.S4).On average, over the two years and two cropping patterns, leaf SG increased with N application level in maize but not in soybean.At maturity (day 50 after anthesis), compared to MM,IM significantly increased SG by 208% under NN, 109% under RN,and 73.3% under CN.Compared to SS, IS increased SG by 80.7%under NN, 87.3% under RN, and 98.6% under CN (Fig.2).

In maize, relay cropping increased LAI after 30 days postanthesis.Relay cropping significantly increased LAI by 79.4% to 88.5%under NN and 55.5%to 148%under RN across the two years.Increasing N application contributed to a pronounced increase in LAI at day 40 after anthesis, especially in IM, which rose from 1.07 under NN to 1.86 under RN averaged over the two years(Fig.3).At the blistering stage (R2), compared to IM, MM significantly reduced the relative chlorophyll content (SPAD) by 47.1%in middle leaves and by 88.7% in bottom leaves averaged over the two years (Figs.S6).

In soybean,relay cropping reduced LAI by 107% at days 0 to 20 after flowering, averaged across three N levels and two years.Increasing N application from NN to CN had little effect on the LAI of SS.On average, across the two years, relay cropping increased the LAI by 115% under NN, 328% under RN, and 437%under CN from days 40 to 50 after flowering,while monocropping resulted in an LAI near 0 (Fig.3).At the full seeding stage (R6), IS significantly increased Pnin the bottom leaves by 85.6% under NN, 65.2% under RN, and 43.0% under CN, averaged across the two years (Fig.S5).

3.2.Dry matter accumulation and partitioning characteristics after anthesis

In maize, relay cropping significantly increased NAR by 72.3%under NN and 42.5%under RN on day 10 after anthesis on average over the two years.Increasing N application led to a significant increase in DM, especially at day 50 after anthesis of IM, which increased by 51.6% from NN to RN averaged across the two years but only 6.0% from RN to CN.However, increasing N application reduced NAR, particularly at day 10 after anthesis of IM, which decreased by 35.4% from NN to CN averaged over the two years(Fig.4).Relay cropping increased the dry matter partition coefficient (MPC) of maize leaves, with MPC of leaves at day 10 after anthesis increasing by 5.8% under NN, 1.4% under RN, and 2.6%under CN averaged over the two years.N application increased the MPC of leaves but reduced that of grain over the two years and two cropping patterns (Fig.S7; Table S1).

Fig.2.Leaf stay-greenness(SG)per plant in maize and soybean after anthesis under two cropping patterns and three N application levels simulated by logistic curves in 2020 and 2021 (R2 > 0.962).Solid lines represent relay cropping, dotted lines represent monocropping.IMNN, relay cropping maize with no N application; IMRN, relay cropping maize with reduced N application; IMCN, relay cropping maize with conventional N application; MMNN, monocropping maize with no N application; MMRN, the monocropping maize with reduced N application;MMCN,monocropping maize with conventional N application;ISNN,relay cropping soybean with no N application;ISRN,relay cropping soybean with reduced N application; ISCN, relay cropping soybean with conventional N application; SSNN, monocropping soybean with no N application;SSRN,monocropping soybean with reduced N application; SSCN, monocropping soybean with conventional N application.The ○or ●symbol represents the point of mean value.

Fig.3.Leaf area index(LAI)after anthesis in monocropping and relay cropping maize and soybean in response to three N application levels in 2020 and 2021.IMNN,the relay cropping maize with no N application;IMRN,the relay cropping maize with reduced N application;IMCN,the relay cropping maize with conventional N application;MMNN,the monocropping maize with no N application; MMRN, the monocropping maize with reduced N application; MMCN, the monocropping maize with conventional N application; ISNN, the relay cropping soybean with no N application; ISRN, the relay cropping soybean with reduced N application; ISCN, the relay cropping soybean with conventional N application;SSNN,the monocropping soybean with no N application;SSRN,the monocropping soybean with reduced N application;SSCN,the monocropping soybean with conventional N application.*,P<0.05 and **,P<0.01.The color of the asterisk corresponds to the color of the curve,representing significant differences between cropping patterns under the same N application level.Solid lines represent relay cropping, dashed lines represent monocropping, bar represents the standard error of the mean.

In soybean, DM showed slower growth in 2021 than in 2020.The NAR of IS increased by 79.4% compared to SS at day 50 after anthesis in 2021.Relay cropping showed a negligible effect on DM at maturity but reduced the MPC of stems by 36.1% under NN, 25.0% under RN, and 18.7% under CN averaged across the two years (Fig.S7).N application significantly decreased both the aboveground dry matter translocation rate(MTR) and the contribution of dry matter translocation to grain(TCR) from NN to CN by 81.1% and 49.0%, respectively(Table S1).

Fig.4.Dynamics of dry matter weight aboveground (DM) and net assimilation rate after anthesis (NAR) of maize and soybean under different cropping patterns and N application levels in 2020 and 2021.DAA, days after anthesis; IMNN, the relay cropping maize with no N application; IMRN, the relay cropping maize with reduced N application;IMCN,the relay cropping maize with conventional N application;MMNN,the monocropping maize with no N application;MMRN,the monocropping maize with reduced N application;MMCN,the monocropping maize with conventional N application;ISNN,the relay cropping soybean with no N application;ISRN,the relay cropping soybean with reduced N application;ISCN,the relay cropping soybean with conventional N application;SSNN,the monocropping soybean with no N application;SSRN,the monocropping soybean with reduced N application;SSCN,the monocropping soybean with conventional N application.Lines(solid and dashed lines)represent DM.Symbols(triangle, circle, square) represent NAR.Bar represents standard error of the mean.

3.3.Yield and yield components

In maize,relay cropping increased yield by 65.6%and grain number per cob(GN)by 51.3%under NN in 2021 while also increasing 1000-grain-weight (GW) from 301.02 g in monocropping to 316.56 g in relay cropping.Increasing N application increased yield,GN, and GW.Yield increased by 87.1%, system yield increased by 32.1%,GN increased by 81.0%, and GW increased by 9.3% from the NN to RN treatments.However,compared to RN,CN did not significantly increase yield,system yield,GN,or GW(Table 1).

In soybean, relay cropping increased yield by 9.7%, effective plant number per hectare(EP)by 10.6%,and seed number per plant(SN)by 17.5%on average across the two years and three N applications, with negligible effect on 100-seed-weight (SW).N application reduced the yield of IS by 32.7% under CN compared to NN,but did not affect the yield component of IS or SS (Table 2).

Relay cropping significantly increased the harvest index(HI)by 8.5% in maize and by 16.8% in soybean on average over the two years and three N applications.With the increase in the N application level, the HI increased and then decreased.Under RN, maize and soybean reached their maximum values of 0.67 and 0.49,respectively.The LER in the relay cropping system averaged 2.27 over the two years and three N application treatments, but was reduced by N application (Tables 1, 2).

3.4.Synergistic relationship between leaf stay-greenness and grain filling

In both maize and soybean,relay cropping delayed the onset of the maximum rate of leaf senescence and grain or seed accumulation (Fig.5).Similar to the growth curve, the curve of leaf staygreenness and grain or seed accumulation showed three phases of ‘‘lag–exponential–stationary” (Fig.2; Table S2).

In maize,relay cropping significantly prolonged the duration of leaf stay-green by almost four days in the lag phase and increased the grain accumulation rate (dGDM) by 12.5% in the exponential phase and 39.0% in the stationary phase on average over the two years and three N application levels.In both years, increasing N application was significantly negatively correlated with the duration of leaf stay-greenness in the lag phase.In contrast,N input significantly increased dGDM in IM, reaching a maximum under RN(Table S2).

In soybean, relay cropping increased the duration of leaf staygreen by almost 8 days in the lag phase and increased the seed accumulation rate (dSDM) by 28.0% in the lag phase, 55.5% in the exponential phase, and 31.8% in the stationary phase on average across the two years and three N application levels.In 2021, the seed dry matter(SDM)increased by 47.0%in the exponential phase and 53.0% in the stationary phase, on average, across the three N application levels (Table S2).However, there was a negative association between N application and dSDM,with increasing N application significantly reducing dGDM by 58.2% in the exponential phase and 31.0% in the stationary phase averaged across the two years and two cropping patterns.

4.Discussion

4.1.Post-anthesis leaf stay-green duration in maize soybean relay cropping system

This study supports the findings of previous studies that relay cropping can delay leaf senescence and improve photosynthesismore effectively than monocropping [32,33].Although the LAI of monocropped maize and soybean was slightly higher than that of relay cropping plants in the initial 0–20 days after anthesis, this advantage was short-lived.The use of monoculture led to a rapid decline in SG and LAI due to leaf senescence and loss(Figs.2,3,S4).

Table 1 Effects of year, N application, cropping pattern on yield and yield components in maize.

Table 2 Effects of year, N application, and cropping pattern on yield and yield components in soybean.

Relay cropping boosted the chlorophyll content in the middle and lower leaf layers of the maize and soybean plants (Fig.S6).In the relay cropping system, the ability of relay cropping to keep leaves green can be attributed to a main factor.In the two border rows within a maize or soybean strip, more light can be captured than in monocropping.After maize harvest, two soybean rows within a soybean strip become two border rows in relay cropping,receiving more solar radiation than the inner rows in monocropping after anthesis [34,35].

Although increasing N input resulted in an increase in the SG of maize leaves, the opposite trend was observed in soybean.The finding that SPAD and Pnof soybean did not increase under the CN treatment indicates that the increase in nitrogen fertilizer leads to redundant growth of soybean leaves, which were increasing in size and number (Fig.S4), decreasing the utilization of light and N by the canopy.Considering the low nitrogen utilization efficiency (NUE) in soybean, minimizing N input was a good choice in relay cropping system.Relay cropping is designed to compensate for soybean gain by nodules fixing N2from the air into NH4+.Nodules,as an energy-consuming organ in N2fixation,can balance the source–sink ratio of soybean and promote photosynthesis,allowing leaves to stay green for a longer period [36].Low N is advantageous to soil N availability, nodule development, and N2fixation in monocropped soybean.Low N also drives the migration of fixed N to maize in an intercropping system, contributing to a longer green leaf duration in the whole system [37–41].This may explain why there was no difference in maize yield between RN and CN treatments in a relay cropping system.In this case,maize has a competitive advantage, the ability to absorb N from the surrounding area to meet its growth requirements.The nitrogen fixed by soybeans helps to compensate for the yield gap resulting from reduced nitrogen application in maize.

Fig.5.Normalization of dSG and dGDM/dSDM under different cropping patterns and N application levels.dGDM/dSDM, first derivative of grain (for maize) or seed (for soybean)dry matter(GDM/SDM),described as the accumulation rate of grain/seed after anthesis;dSG,first derivative of leaf stay-greenness(SG),described as leaf senescence rate of maize or soybean;NN,no N;RN,reduced N;CN,conventional N;IM,relay cropping maize;MM,monocropping maize;IS,relay cropping soybean;SS,monocropping soybean.The associated statistics are presented in Table S2.

4.2.Dry matter accumulation, allocation after anthesis in a maize–soybean relay cropping system

High yields depend on high assimilation and assimilate remobilization[3,42–44].The peak NAR in maize–soybean relay cropping was higher than or similar to the NAR in the monocrops (Fig.4;Table S1), in agreement with the findings of previous studies[45–47].The NAR of relay cropping was higher than that of monocropping at day 20 after anthesis, especially under the NN treatment.This result was probably associated with the higher capture and use of solar radiation in relay cropping than in monocropping [48].Further experiments should be conducted to measure parameters such as photosynthetically active radiation and transmittance in these canopies.

The relay cropping for maize resulted in nearly equal MTR and TCR compared to monocropping, whereas relay cropping for soybean showed lower MTR and TCR than in monocropping.The relatively low rate of dry matter transport in relay-intercropped soybean might be attributed to the effect of the plant’s vascular system.A suite of phenotypic plasticity responses, including elongated petioles and slender stems (also known as shade avoidance syndrome), might reduce assimilate transport [26,49].

The higher HI of the relay cropping system may be attributed to the sustained grain accumulation between days 40 and 60 after anthesis, although the rate of grain and seed dry mass accumulation rate, as well as MPC during days 0–40 after anthesis, were slightly lower than those in relay cropping, and the rapid leaf growth during this period facilitated the accumulation and transportation of photosynthetic products to the grain.In contrast,monocropping allocated more photosynthates to the stalks and redundant branches (Fig.S4), and this tendency increased with the level of nitrogen application,which could account for the lower HI observed in monocropping (Table S1).

Other reports [25,50] have concluded that relay cropping with soybean increases the uptake of phosphorus.Phosphorus is associated with the production of ATP,PGA kinase activity,and transport of triose phosphate from chloroplasts.When phosphorus is insufficient, triose phosphate is synthesized in the chloroplast and is stored as starch instead of sucrose, thereby reducing grain-filling and seed-setting rates [51].

Crop growth is greatly influenced by climate.Temperature and precipitation alter the growth process of crops, especially soybeans.As shown in Fig.4, Table 1, and Table 2, there were differences in biomass and yield over two years.In 2021, heavy rainfall in July and August reduced NAR and grain-filling rate during the maize grain-filling period, resulting in a yield reduction in MM.Many monocropped soybean plants fell over, reducing the number of plants.In 2020, post-flowering heavy rainfall caused flower drop and leaf withering in soybeans, reducing photosynthetic production and seed filling (Fig.1; Table 2).

Fig.6.Structural equation model (SEM) for maize(chi-square=15.130, RSMEA=0.000,AGFI =0.825) and soybean (chi-square=0.508,RSMEA=0.000, AGFI=0.988).N,N application;SG,stay-greenness;LAI,leaf area index;NAR,net assimilation rate;DM,dry matter aboveground;GAR,grain dry matter accumulation rate;SAR,seed dry matter accumulation rate; MTR, dry mass translocation rate; GN, grain number per cob; SN, seed number per plant; GW, 1000-grain weight of maize; SW, 100-seed-weight of soybean;EP,effective plant number per hectare.Boxes represent variable names,and numbers in parentheses show the variance explained by this model(R2).The solid line indicates a significant positive relationship,and the dashed line indicates negative correlation.Numbers on arrows are standardized path coefficients and indicated the effect size of the relationship.Arrow width is proportional to the strength of path coefficients.A line with arrowhead indicates a putative causal link between the cause(base of the arrow) and effect (tip of the arrow).*, P < 0.05; **, P < 0.01, ***, P < 0.001.

4.3.Relationship between stay-green and system yield

The synergistic increase in maize and soybean yields is attributed to the sustained contribution of leaves to photosynthesis.But crop adaptability is influenced by ecological niche differentiation and environmental fluctuations, which ultimately determine yield increases.Fig.5 illustrates how the senescence curve of leaves and the grain-filling curve of relay cropping soybean have both shifted backward.This trend is due to the shade from maize suppressing soybean growth during the co-growth period,but after maize harvest,increased access to light resources facilitates growth soybean recovery [52,53].Consequently, relay-cropped soybeans exhibit a slower start and faster finish throughout the reproductive period than monoculture soybeans.The present study showed strong associations between post-anthesis leaf stay-green and yield formation across cropping patterns.The relay cropping system delayed leaf senescence and prolonged grain accumulation(Fig.5; Table S2).In fact, the more prolonged the lag phase of leaf stay-greenness, the earlier was the onset of grain formation and the greater the rate of grain accumulation, ultimately leading to greater grain yield.During the stationary phase of leaf area, vegetative organs cease to grow, and most of the photosynthates are allocated to grain production.Thus, retention of leaf area for an extended period confers higher yield potential [54].

SEM identifies three factors that contributed to yield formation in this study: leaf photosynthesis (SG, LAI, NAR), dry matter production (DM, GAR, MTR), and yield components (SN, SW, EP).For maize, relay cropping and N application prolonged SG and increased LAI, GAR, MTR and GN, and yield.Stay-greenness has a positive impact on DM (0.95**) and GAR (0.64**), while LAI contributes significantly to GW (0.59**) (Fig.6).This model suggests that stay-greenness promotes aboveground growth and the accumulation of photosynthetic products in grains.Increased grain number permits efficient export of assimilates from the leaves,helping to maintain source–sink balance and delay foliar yellowing[55].

For soybean, relay cropping greatly increased SG and NAR.LAI and NAR both contribute significantly to SAR (0.57**and 0.54***,respectively).In particular, SG plays a positive role in EP (1.46***)in such a way that soybean yield is increased.In contrast, LAI reduced EP (-1.37***).Monocropping soybean resulted in redundant foliage that weakened transmittance, causing severe senescence of lower leaves and pods.Compensatory growth of top leaves causes an increase in petiole length, higher pod position,and higher center of gravity, ultimately leading to post flowering plant lodging, which reduces EP and yield.However, the shadetolerant soybean variety chosen for this experiment can maintain a stable canopy structure in relay cropping systems.Relay cropping extended green leaf duration, increased photosynthesis inside the canopy at multiple levels, and promoted the morphological and physiological development of the stem, ultimately mitigating the soybean yield loss [56].

5.Conclusions

The maize–soybean relay cropping system achieved a synergistic yield increase of the two crops by increasing the leaf area, prolonging the leaf stay-green period,and accelerating grain filling.In maize, relay cropping resulted in increased leaf stay-greenness,LAI, grain accumulation rate, grain number, and yield.In soybean,relay cropping increased leaf stay-greenness, NAR, dry matter translocation rate, seed accumulation rate, effective plant number per hectare, and yield.Increasing nitrogen application led to redundant growth of leaves (LAI) in monoculture soybean, resulting in severe canopy closure and reduced photosynthetic capacity.Soybean yield was reduced by N application.Under reduced N application, the HI, LER and system yield were at their maximum.As maize yield was stabilized,soybean yield increased.In the relay cropping system, N input is utilized to its full potential, contributing to an increase in system yield as well as a savings in fertilizer.

CRediT authorship contribution statement

Yiling Li:Writing - original draft, data analysis and plotting.Ping Chen:provided overall guidance and some suggestions.Zhidan Fu,Kai Luo&Ping Linparticipated in manuscript revision.Chao Gao, Shanshan Liu & Tian Puparticipated in sowing and investigation.Taiwen Yong & Wenyu Yang:provided financial support.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This work was supported by the Special Fund for the Industrial Technology System Construction of Modem Agriculture (CARS-04-PS20), the National Natural Science Foundation of China(31872856, 31671625), and the National Key Research and Development Program of China (2021YFF1000500).We are thankful to all staff at Modern Agriculture Expert Courtyard in Ren-shou County, Sichuan province.

Appendix A.Supplementary data

Supplementary data for this article can be found online at https://doi.org/10.1016/j.cj.2023.05.011.