Reducing maize yield gap by matching plant density and solar radiation

LlU Guang-zhou,LlU Wan-mao,HOU Peng,MlNG Bo,YANG Yun-shan,GUO Xiao-xia,XlE Ruizhi,WANG Ke-ru,Ll Shao-kun

1 Key Laboratory of Crop Physiology and Ecology,Ministry of Agriculture and Rural Affairs/Institute of Crop Sciences,Chinese Academy of Agricultural Sciences,Beijing 100081,P.R.China

2 The Key Laboratory of Oasis Eco-agriculture,Xinjiang Production and Construction Corps/College of Agronomy,Shihezi Univerisy,Shihezi 832000,P.R.China

Abstract Yield gap exists because the current attained actual grain yield cannot yet achieve the estimated yield potential. Chinese high yield maize belt has a wide span from east to west which results in different solar radiations between different regions and thus different grain yields. We used multi-site experimental data,surveyed farmer yield data,the highest recorded yield data in the literatures,and simulations with Hybrid-Maize Model to assess the yield gap and tried to reduce the yield gap by matching the solar radiation and plant density. The maize belt was divided into five regions from east to west according to distribution of accumulated solar radiation. The results showed that there were more than 5.8 Mg ha–1 yield gaps between surveyed farmer yield and the yield potential in different regions of China from east to west,which just achieved less than 65% of the yield potential. By analyzing the multi-site density experimental data,we found that the accumulated solar radiation was significantly correlated to optimum plant density which is the density with the highest yield in the multi-site density experiment (y=0.09895x–32.49,P<0.01),according to which the optimum plant densities in different regions from east to west were calculated. It showed that the optimum plant density could be increased by 60.0,55.2,47.3,84.8,and 59.6% compared to the actual density,the grain yield could be increased by 20.2,18.3,10.9,18.1,and 15.3% through increasing plant density,which could reduce the yield gaps of 33.7,23.0,13.4,17.3,and 10.4% in R (region)-1,R-2,R-3,R-4,and R-5,respectively. This study indicates that matching maize plant density and solar radiation is an effective approach to reduce yield gaps in different regions of China.

Keywords:maize,yield gap,yield potential,matching density and radiation

1.lntroduction

Maize (Zea maysL.) is the most productive food crop(Andelkovicet al.2017),however,the current levels of productivity cannot meet the food demand for meeting the need of growing global population (Rayet al.2013; Baileyet al.2019). An important yield limiting factor is sub-optimal plant density (Alfaroet al.2008; Al-Naggaret al.2015;Trachselet al.2016),which is 50×103–60×103plants ha–1on average in China and is lower than the average plant density of 75×103–83×103plants ha–1in the United States(Menget al.2013). In China,because of the lower plant density,the average grain yield is about 6.0 Mg ha–1and the average highest recorded grain yield (yields obtained by agronomists under optimal condition) is 16.7 Mg ha–1.Compared to the maize yields in the United States,the average and highest attainable yields are much lower in China (Chenet al.2012; Menget al.2018). Although higher yields have been reported in some areas,they are still lower than the estimated yield potential (Merloset al.2015).

Yield potential is defined as the yield of a crop can achieve when grown in an environment to which it is adapted and free from nutrients and water limitations and pests and diseases stresses (Evans and Fischer 1999; Cassmanet al.2003;Menget al.2013; Senapati and Semenov 2020). Normally,yield potential is based on model simulations but can also be obtained from yield maximization experimental plot or farmer field (Lobellet al.2009). Crop models are good tools to provide reasonable estimates of yield potential by using historical climate data (Grassiniet al.2011). The difference between yield potential and the actual yield represents the yield gap (Cassmanet al.2003). Crop yield gaps may exist due to various limiting factors,such as the biotic,abiotic and climate (Kassieet al.2014; Bezaet al.2017; van Oortet al.2017) and often results from sub-optimal crop management practices (Senapati and Semenov 2020). Usually,due to the climatic risks and some negative environment factors,it is difficult to close a full yield gap (Lobellet al.2009;van Ittersumet al.2013). However,by evaluating the yield potential and yield gap,factors limiting the yield can be identified and strategies developed to reduce the gap(Naabet al.2004; Bhatiaet al.2008). Bridging yield gap will be important to achieve food security (Menget al.2013;Senapati and Semenov 2020).

Solar radiation is a primary determinant of crop yield potential,therefore,increasing radiation use efficiency is crucial to improve the grain yield (Wilsonet al.1995; Liuet al.2013; Houet al.2014; Denget al.2015). The improvement of modern maize grain yield is highly depended on the increasing plant density,which can intercept solar radiation optimally (Lambertet al.2014; Al-Naggaret al.2015;Testaet al.2016; Gouet al.2017). It has been reported that solar radiation significantly affects the optimum maize plant density (Iizumi and Ramankutty 2015; Xuet al.2017;Yanget al.2019). The Chinese maize belt (97.5–135.1°E,21.1–53.6°N) covers a wide range of longitude (nearly 38°)and latitude (nearly 33°) (Menget al.2016),which results in significant differences in solar radiation between the western and eastern regions. This large difference in solar radiation results in differences in plant densities and grain yields (Li and Wang 2008; Xuet al.2017; Yanget al.2019). Therefore,optimizing the solar radiation and plant densities will be of great significance for increasing maize yield and reducing the yield gap between farmers’ yield and yield potential in different regions of China.

The objectives were to 1) estimate yield gaps from east to west in Chinese high yield maize belt by using surveyed farmers’ yields and yield potential simulated by Hybrid-Maize Model; 2) examine the relationship between the accumulated solar radiation and the optimum plant density through analyzing field data from 20 experimental sites;and 3) calculate optimum plant densities and yield gaps reduction. This study will provide new approach to reduce the yield gaps by matching the solar radiation and maize growth (plant density) in different regions covering different solar radiation intensities.

2.Materials and methods

2.1.Division of regions

2.2.Database description

Field experimentsAccording to the regions described above,field experiments were conducted in nine consecutive years from 2009 to 2017 at 20 locations between longitudes of 89°34´E and 129°36´E in China,as Keshan,Zunhua,Zhangjiakou,Mudanjiang,Taonan,Nongan,Gongzhuling,Tonghua,Kangping,Zhangwu,Shenyang,Dandong,Tongliao,Huhhot,Yulin,Yinchuan,Wuwei,Jiuquan,Qitai,and Yili. An elite single-cross maize cultivar namely Zhengdan 958 (ZD958) which is the most widely cultivated cultivar in China was used in each experimental site. A randomized complete block design was used at each site,with four replicates. All experimental sites used four to six common planting densities:45×103,60×103,75×103,90×103,105×103,and 120×103plants ha−1. Each plot was 15 m in length and 6.5 m in width and consisted of 10 rows with a row spacing of 0.65 m at all sites. Seeds were sown by hand at a soil depth of about 5 cm. In each location,fertilizer was applied at the rates of 126–500 kg N ha–1,0–207 kg P2O5ha–1and 0–112 kg K2O ha–1,depending on the soil fertility status of a site and following the recommendations used in the previous studies (Houet al.2012; Liuet al.2015; Xuet al.2017). To avoid water stress,irrigation was applied at locations with low precipitation. Weeds,diseases and insect pests were well controlled at all locations.

The sowing,silking and maturity dates were recorded at each location; silking date was recorded when 60% of the ears showed silk emergence,while maturity was recorded when kernels had reached physiological maturity as the black layer appeared. After physiological maturity,all plants in the central four rows of each plot,representing an area of 13 m2,were harvested manually to measure the grain yield. Grain yield was determined at 14% moisture content,as tested using a portable moisture meter (PM8188,Kett Electric Laboratory,Tokyo,Japan).

Annual mean solar radiation during the growth duration at each experimental site over the years was obtained from the National Meteorological Information Center (http://data.cma.cn/) of China Meteorological Administration (CMA).The distance between the meteorological stations and the experimental sites varied from 3 to 39 km.

Highest recorded yieldThe highest recorded yields were collected from the published studies (Chenet al.2012;Menget al.2013; Liuet al.2017) which were obtained by the agronomists in different regions with sufficient inputs regardless of the economic costs and environmental risks.Total of 15 data points representing the highest recorded yields in the five regions were obtained.

Actual density and grain yieldData of actual plant densities and grain yields (farmers’ plant densities and yields) were collected through surveys of the constant 24 farmers in different experimental sites during 2009 to 2016.All the data points in each region were averaged to represent the average actual plant density and grain yield in a given region (Minget al.2017).

Yield potential simulationThe Hybrid-Maize Model was used to simulate the yield potential (Houet al.2013),which has been tested and widely used in the United States(Grassiniet al.2011),South Asia (Timsinaet al.2010) and China (Houet al.2014; Menget al.2014; Liuet al.2015).It is a process-based model that can simulate maize yield potential under growth conditions that are non-limiting (i.e.,nutrient,water) and free from pests and disease. The model has been reported to be reasonably accurate for estimating maize yield potential (Yanget al.2004,2006). Model inputs include weather data (solar radiation,the maximum and minimum temperatures),sowing and harvest dates and plant density. In the present study,the model was used to simulate yield potential by inputting the history weather data and plant density.

Yield gap calculationTo compare the differences between different yield levels,following two kinds of yield gaps were determined:

1) Yield gap between modeled yield potential and average farmer’s yield was calculated as:

Mother knew I d need the shirt as a reminder33 that a sense of humor, spiced with love, is one of the most important ingredients in a happy marriage. In a pocket was a note: Read John 14:27?29. I love you both, Mother.

YG1=Modeled yield potential–Average farmers’ yield (1)

2) Yield gap between modeled yield potential and the highest recorded yield was calculated as:

YG2=Modeled yield potential–Highest recorded yield (2)

2.3.Statistical analysis

All statistical analyses were performed using the SPSS 21.0 Software (SPSS Inc.,Chicago,Illinois,USA). Analysis of variance (ANOVA) was performed to test significance between different grain yields by the Duncan’s multiple range test at the 5% level of significance. Linear regression was used to test the relationships between plant densities and grain yield for calculating optimum densities at each experimental site. Then linear regression was used to examine the relationship between optimum plant densities and accumulated solar radiation across all the experimental sites.

3.Results

3.1.Yield potential and actual grain yield across different regions

As shown in Fig.1,the yield potentials and actual yields increased from R-1 to R-5,which coincided with the distribution of solar radiation from R-1 to R-5. The simulated yield potentials were significantly higher than actual yields in the R-1,R-2,R-3,R-4,and R-5 (P<0.05). There existed yield gaps of 5.8,7.9,8.9,12.4,and 14.3 Mg ha–1between the yield potential and actual yield (YG1) in the five regions,respectively. The farmers achieved 62.5,55.7,55.2,48.8,and 40.4% of the yield potential in the five regions,respectively. It is notable that the farmers in the regions with lower solar radiation achieved higher percentage of the yield potential,however,there was larger exploitable gaps in the regions with more abundant solar radiation.

3.2.Relationship between solar radiation and optimum plant density

The optimum plant density without water and nutrient stresses was significantly correlated with the accumulated solar radiation in the multi-site density experiments from east to west of China (Fig.2). The optimum plant density significantly increased with the increasing accumulated solar radiation of the whole growth period. Therefore,the optimum plant densities adapted to different radiations could be calculated using the linear equation (y=0.09895x–32.49,P<0.01).

3.3.Actual and optimum plant densities in different regions from east to west

Based on the equation (y=0.09895x–32.49,P<0.01)described in Fig.2,the optimum plant densities were calculated by using the average solar radiations in the five regions. As shown in Fig.3,the optimum plant densities were significantly (P<0.05) higher by 60.0,55.2,47.3,84.8,and 59.6% than the actual densities in the five regions,respectively. The actual densities were 62.5,64.4,67.9,54.1,and 62.7% of the optimum densities in the five regions,respectively. The exploitable plant densities were 33×103,32×103,34×103,59×103,and 50×103plants ha–1in the R-1,R-2,R-3,R-4,and R-5. Therefore,there was much higher density gap in the R-4 and R-5 with more abundant solar radiation.

3.4.Reducing yield gap by matching maize plant density with solar radiation

By increasing the actual densities to optimum densities,the simulated grain yield significantly increased with percentage of more than 10% in the five regions with the highest increase of 20.2% in the R-1 (Table 1). Based on these increasing ratios,the actual yield (farmers’ yield) could increase by more than 1.2 Mg ha–1in the five regions with the highest increment of 2.1 Mg ha–1in the R-4 only through increasing plant density. According to the YG1 in the five regions (Fig.1),the yield gap could be reduced by more than 10% in the five regions with the highest reduction of 33.7% in the R-1 (Table 1).

Fig.3 The actual and optimum plant densities in the five regions (R-1,R-2,R-3,R-4,and R-5) from east to west in China. Bars mean SD.

The highest recorded yields were obtained by the agronomists in different regions with sufficient inputs regardless of the economic costs and environmental risks.As shown in Table 1,the highest recorded yields increased from the R-1 to R-5,which coincided with the distribution of solar radiations. Their planting densities were 81.7×103,88.4×103,100.0×103,103.8×103,and 122.6×103plants ha–1in different regions,which were closed to those calculated in this study (Fig.3). According to the simulation of yield potentials,the yield gaps between modeled yield potential and the highest recorded yield (YG2) were 0.5,0.2,2.4,4.8 and 1.6 Mg ha–1. The highest recorded yields were 96.8,98.7,87.8,80.3,and 93.2% of the yield potentials in the five regions,respectively. This indicates that the actual yields can be increased by applying optimum plant density that we calculated through more precise field managements.

4.Discussion

The eastern and western regions of Chinese maize belt(97.5–135.1°E,21.1–53.6°N) cover a wide range of longitude and latitude (Menget al.2016),which results in a significant difference in solar radiation. This large variability of solar radiation leads to diverse yield gaps (Li and Wang 2008; Xuet al.2017; Yanget al.2019). The maximum yield gap (YG1) was 14.3 Mg ha–1in R-5 and the minimum yield gap was 5.8 Mg ha–1in R-2 (Fig.1). The farmers achieved 40.4–62.5% of the yield potential in these regions,which was higher than that reported by Menget al.(2013). Menget al.(2013) also reported the yield gap (between actual and recorded yields) of 7.6 Mg ha–1in China,which was lower than that in the R-2,R-3,R-4,and R-5,respectively. This may be due to the fact that different regions may have been covered in the two studies. Although in the present study,the farmers’ yield achievement in yield potential was high,it was still less than the 80% of that achieved in the United States (Grassiniet al.2011). Generally,the solar radiation was higher in the western region than in the eastern,which led to the differences in plant density and grain yield (Xuet al.2017; Yanget al.2019). However,in the eastern region with lower solar radiation,the yield gaps and yield potential were smaller. Yield gap in the R-5 with high solar radiation was larger,indicating a greater potential of yield improvement in this region where farmers achieved only 40.4% of the yield potential. Taking full advantage of the high solar radiation in west region can improve maize grain yield to reduce yield gap.

Consistent with our study,Menget al.(2018) reported that the maize plant densities in northwestern China are significantly higher than those of other regions. In the present study,we also found that the accumulated solar radiation was significantly related with optimum plant density (Fig.2),which coincided with some previous studies (Braconnier 1998; Hammadet al.2016). Xuet al.(2017) and Yanget al.(2019) also reported that high solar radiation could contribute to high grain yield by matching optimum plant density. Solar radiation significantly affected the optimum plant density (Iizumi and Ramankutty 2015),thereby resulting in different optimum plant densities in different regions (Xuet al.2017). When the solar radiation is matched with the plant density,the optimum plant density ranged from 89×103to 134×103plants ha–1in the five regions(Fig.3). However,the average actual densities were less than 70×103plants ha–1in these regions,which just achieved less than 70% of the optimum density and also lower than the average plant density of 75×103–83×103plants ha–1reported in the United States (Li and Wang 2010;Menget al.2013). Previous studies have shown that increasing plant density could effectively improve maize grain yield (Cuiet al.2013; Assefaet al.2016; Houet al.2020). According to the simulation of Hybrid-Maize Model,the grain yield could be increased by 10.9–20.2% through optimizing plant density (Table 1),which was lower than that reported by Menget al.(2013). This discrepancy may be due to that our results are based on the model and we covered different regions. By improving the optimum plant density,the maize population can better adapt to the solar radiation and improve the solar radiation use efficiency(Wanget al.2012),and ultimately increase high yield potential (Reynoldset al.2009; Hawkesfordet al.2013).

Based on the increasing ratios in different regions,the actual grain yield could increase 1.2–2.1 Mg ha–1in these regions,among which the highest yield increment was 2.1 Mg ha–1in the R-4. This indicated that the actual density was too low to match and take advantage of abundant solar radiation (Xuet al.2017). With the increase of grain yield by increasing plant density,the yield gap (YG1) could be reduced more than 10% in these regions with the highest of 33.7% in the R-1. However,in the R-5,the yield gap(YG1) was just reduced about 10% simply by increasing plant density which was already the highest density in the five regions. According to the highest recorded yields in the five regions,the recorded yields by the agronomists had achieved more than 80% of the yield potential with the highest achievement of 98.7%. The planting densities adopted in the fields with the highest recorded yields were closed to the calculated optimum densities in this study.This indicated that the precise management practices together with optimum density could increase maize grain yield more and reduce yield gaps further (Liuet al.2017),especially in R-3,R-4 and R-5. Therefore,by matching the solar radiation and plant density,maize grain yield can be improved,and yield gap can be reduced (Yanget al.2019). Further reducing the remaining yield gap,precise management practices and better hybrids together with optimum densities should be used to maximize the utilization of solar radiation (Liuet al.2019,2020).

Table 1 The yield increasing percentage,actual yield increment and yield gaps reduction by increasing plant density,the yield potential,highest recorded yields,and corresponding densities in the five regions (R),from east to west in China

5.Conclusion

Chinese high yield maize belt ranging from east to west exhibits different solar radiations. The sub-optimal crop management practices and adverse climactic factors lead to a large yield gap between actual grain yield and estimated yield potential. In this study,we report that maize grain yield could be increased by 10.9 to 20.2%by increasing plant density through matching the solar radiation in different regions. In addition,the yield gap could be reduced by 10.4 to 33.7% in these regions.Therefore,dense cultivation by matching maize growth and solar radiation is an effective approach to reduce yield gaps in different regions of China.

Acknowledgements

This work was supported by the National Key Research and Development Program of China (2016YFD0300110,2016YFD0300101),the National Natural Science Foundation of China (31871558) and the National Basic Research Program of China (973 Program,2015CB150401).