Effects of plant density and nitrogen rate on cotton yield and nitrogen use in cotton stubble retaining fields

2021-06-24 09:26:06WANGShihongMAOLiliSHlJialiangNlEJunjunSONGXianliangSUNXuezhen
Journal of Integrative Agriculture 2021年8期

WANG Shi-hong,MAO Li-li,SHl Jia-liang,NlE Jun-junSONG Xian-liangSUN Xue-zhen

1 Agronomy College,Shandong Agricultural University,Tai’an 271018,P.R.China

2 Liaocheng Academy of Agricultural Sciences,Liaocheng 252000,P.R.China

3 Dezhou Academy of Agricultural Sciences,Dezhou 253000,P.R.China

Abstract Increasing nitrogen (N) rate could accelerate the decomposition of crop residues,and then improve crop yield by increasing N availability of soil and N uptake of crops. However,it is not clear whether N rate and plant density should be modified after a long period of cotton stubble return with high N rate. This study seeks to assess the effects of N rate and plant density on cotton yield,N use efficiency,leaf senescence,soil inorganic N,and apparent N balance in cotton stubble return fields in Liaocheng,China,in 2016 and 2017. Three plant densities 5.25 (D5.25),6.75 (D6.75) and 8.25 (D8.25) plants m-2 and five N rates 0 (N0),105 (N105),210 (N210),315 (N315),and 420 (N420) kg ha-1 were investigated. Compared to the combination used by local farmers (D5.25N315),a 33.3% N reduction and a 28.6% increase in plant density (D6.75N210) can maintain high cotton yield,while a 66.7% N reduction at 6.75 plants m-2 (D6.75N105) can only achieve high yield in the first year. Biological yield increased with the increase of N rate and plant density,and the highest yield was obtained under 420 kg N ha-1 at 8.25 plants m-2 (D8.25N420) across the two years under investigation. Compared to D5.25N315,N agronomic efficiency (NAE) and N recovery efficiency (NRE) in D6.75N210 increased by 30.2 and 54.1%,respectively,and NAE and NRE in D6.75N105 increased by 104.8 and 88.1%,respectively. Soil inorganic N decreased sharply under 105 kg N ha-1,but no change was found under 210 kg N ha-1 at 6.75 plants m-2. N deficit occurred under 105 kg N ha-1,but it did not occurr under 210 kg N ha-1 at 6.75 plants m-2. Net photosynthetic rate and N concentration of leaves under N rate ranging from 210 to 420 kg ha-1 were higher than those under N rate of 0 or 105 kg N ha-1 at all three densities. The findings suggest that D6.75N210 is a superior combination in cotton stubble retaining fields in the Yellow River Valley and other areas with similar ecologies.Keywords:leaf senescence,N application rate,N use efficiency,seed cotton yield,soil apparent N surplus

1.lntroduction

Due to the intense competition for arable land from food crops,cotton planting regions have sharply decreased in China,which poses great challenges to the country’s cotton production (Dai and Dong 2014). Hence,increasing cotton yield per unit area seems an inevitable option to address the aforementioned problem. Cotton yield can be increased by optimizing N fertilizer management and plant density (Aliet al.2007;Donget al.2012). In recent years,it is compulsory to return crop residues to soil to avoid air pollution and soil degeneration,according to Chinese government’s regulation. It is necessary to increase N rate to promote straw decomposition in the field with crop residues.However,N losses also increase with the increase of soil available N induced by extra N fertilizer addition and crop residue N after long term residue return,thereby threatening soil and environment quality such as surface water pollution and greenhouse gas emissions (Hodges 2002;Cuiet al.2010;Alan and John 2013). The amount of N application reduction is important in fields with crop residue return,but inadequate N supply of soil leads to soil N depletion (Liet al.2017). Therefore,appropriate N rate and plant density without causing yield reductions are essential for improving N use in cotton stubble retaining fields.

N application rates affect cotton growth and yield formation by influencing leaf senescence and N acquisition from soil (Houet al.2007;Donget al.2012). Limited N supply capacity decreased cotton yield by photosynthesis reduction and premature senescence of cotton plants (Donget al.2012),which also caused soil N deficiency because the soil N supply could not satisfy plants’ N demand (Devkotaet al.2013;Liet al.2017). Overuse of N usually leads to luxury absorption of N in plants and delays in maturity (Hodges 2002),which also decreases N use efficiency (Daiet al.2015b). High N supply caused severe environment impacts through leaching,volatilization,N2O emission from fertilized croplands,while decreased N use efficiency (Di and Cameron 2002;Juet al.2009;Cuiet al.2010;Cameronet al.2013),which sharply decreased China’ crop production by 60-70% in the 1960s and 20-30% in the 2010s (Zhanget al.2015). According to the recommended N rate (270 kg ha-1) and plant density (5.25 plants m-2) without cotton stubble return in the Yellow River Valley,China (Liet al.2017),local farmers apply 315 kg N ha-1at 5.25 plants m-2in the fields with cotton stubble return. Cotton stubble,an easily accessible organic material,has been returned to soil to improve soil fertility by increasing N uptake of cotton and reducing soil N losses in recent years (Maoet al.2019),since crop residues affect soil texture and fertility through soil nutrients (Zhanget al.2009;Malhiet al.2011). Increasing plant density with a moderate reduction in N rate is an effective approach to improve N use efficiency (Luoet al.2018). While there is research on cotton stubble return to coastal saline soil,none focused on the modification of N rate and plant density when cotton stubble is returned to clay soil.

Increasing N rate or plant density up to a certain level increases cotton yield,however,a further increase of N rate or density can lead to opposite results,which have been documented without considering the sustainable N supply in soil (Houet al.2007;Daiet al.2015a). Liet al.(2017) found that the highest yield and N recovery efficiency (NRE) were achieved in a medium soil applied with 270 kg N ha-1which meets the N balance between cotton demand and soil N supply without the addition of cotton stubble. Luoet al.(2018) reported cotton yield and N use could be mediated by a moderate reduction in N rate at high density without cotton stubble return. Therefore,it is essential to determine the optimal combination of N rate and plant density with high yield and N use efficiency in long term cotton stubble return from the perspective of maintaining the N balance between soil supply and crop demand. The objectives of this research were to quantify the effect of N rate and plant density on cotton yield,N use efficiency,leaf senescence,soil inorganic N content,and N balance in fields with cotton stubble return,and to provide a scientific basis for sustainable soil and plant management.

2.Materials and methods

2.1.Experimental design

The field experiments were conducted at Cotton Experimental Base of Shandong Agricultural University located in Chiping District (36°31´N,116°16´E),Liaocheng City,Shandong Province,China,from 2016 to 2017. The climate of the study area is temperate and monsoonal. The mean daily weather data (temperature,precipitation and solar radiation) during the cotton growing season are presented in Fig.1.The experiment was carried out after 7 consecutive years of cotton stubble return. All plots were used for cotton breeding with 5.25 plants m-2for 7 years using the same management before the experiment. Only cotton was planted,and cotton stubble was crushed and incorporated into 0-40 cm soil layer at the end of November each year in the experiment. The soil was clay soil with pH 7.7,organic matter 13.7 g kg-1,total N 0.91 g kg-1,available P 26.5 mg kg-1,and available K 189.2 mg kg-1in the 0-20 cm soil depth,and it was collected before cotton sowing in 2016. All soil nutritional status was consistent in each plot before the experiment in April,2016.

Fig.1 The weather data (temperature,precipitation and solar radiation) during cotton growing seasons of 2016 (A and C) and 2017 (B and D).

Cotton (Gossypium hirsutum) was planted in late April and harvested in late October every year. Liaomian 6,a conventional cotton cultivar,was used in the field. A split-plot design with three replications was used for the study each year. The main plots were in three plant densities (5.25,6.75 and 8.25 plants m-2),referred as D5.25,D6.75and D8.25,while N (in the form of urea) rates (0,105,210,315,and 420 kg N ha-1) were the subplots,referred as N0,N105,N210,N315,and N420. Each plot contained 10 rows of cotton;each row was 10 m long with an inter-row spacing of 0.80 m. The combination N rate of 315 kg N ha-1at 5.25 plants m-2is used by local farmers when returning cotton stubble to soil.

Half of each N was applied basally before cotton planting (at the end of April),and the other half was side-dressed at early flowering,in both years. All plots received 75 kg ha-1P2O5as calcium superphosphate and 120 kg ha-1K2O as potassium sulfate on local practice during sowing. Other management practices,including plant topping,irrigation,insect and weed control,and chemical control with plant growth regulators,were conducted according to local agronomic practices.

2.2.Sampling and measurements

Seed cotton yield was determined from 20 plants per plot randomly tagged at maturity excluding border rows. At the open boll stage,five plants were cut from the soil surface,air-dried for 20-25 days and weighted as the biological yield.Plant samples were milled and screened through a 0.5-mm screen in a sample mill,and mineralized using H2SO4-H2O2,after which the plant N concentration was determined using the Kjeldahl method. Harvest index was calculated as the ratio of seed cotton yield to biological yield.

Soil samples were collected at 20 cm intervals until 60 cm depth from each plot at harvest from 2015 to 2017 by using an auger. Soil NO3--N and NH4+-N were extracted with 0.01 mol L-1CaCl2and were analyzed using an AA3-AA001-02E auto-analyzer within 48 h of collection. Net photosynthetic rate (Pn) of the functional leaves (the 3rd leaf from the main stem terminal) was measured between 09:00 and 11:00 on cloudless days when ambient photosynthetic photon flux density exceeded 1 500 μmol m-2s-1,using an LI-6400 Portable Photosynthesis System (Li-Cor,Lincoln,NE,USA) 135 days after sowing (DAS).

The utilization of applied N by cotton was evaluated in terms of N agronomic efficiency (NAE) and N recovery efficiency (NRE). NAE was calculated using the following equation (Liet al.2017):

whereYtis the total yield in the fertilized plot (kg ha-1),Y0is the total yield in the control plot and N is the total amount of N applied (kg ha-1).

NRE from N rate was calculated using the following equation:

whereNfis the N uptake from the fertilized plot (kg ha-1) andNcis the N uptake from the control plot (kg ha-1).

Soil apparent N surplus (ANS,kg ha-1) was calculated by the following equation:

whereNresidualis the soil mineral N stock (kg ha-1) at harvest time,Ninitialis the total amount of soil mineral N stock at the last harvest time and cotton stubble-N return to soil,Nappliedis the N application rate,andNuptake(kg ha-1) is the amount of N absorbed by cotton plants.

2.3.Data analysis

All statistical analyses were performed by the Univariate General Linear Models procedure of SPSS 20,in which plant density,N rate and year were treated as fixed factors,while block (replicate) was entered as random factors. The factor block was nested within year. Least significant differences (LSD) were used to separate treatment means at the 5% statistical significance level. The relationship between N rate and seed cotton yield was simulated using the Nonlinear Regression Model procedure of SPSS 20.

3.Results

3.1.Cotton yield and harvest index

Seed cotton yield was significantly affected by the main effect of plant density,N rate and their interaction over the two seasons. The highest seed cotton yield was observed under D5.25N315,D6.75N105and D6.75N210in the first year,but occurred only under D5.25N315and D6.75N210in the second year (Table 1). Boll density and boll weight varied with plant densities and N rates (Fig.2),while lint percentage was not affected by plant density or N rate (not shown). Boll density increased significanlty by applying more N and raising plant density (Fig.2-A and B). The effect of N rate and plant density on boll weight was similar to that on seed cotton yield. The differences were found in 2016 that under D6.75N210and D5.25N315boll weight was more than under D6.75N105. However,no change in boll weight occurred among D5.25N315,D6.75N105and D6.75N210in 2017 (Fig.2-C and D).

Biological yield varied significantly by plant density,N rate and their interaction. The highest biological yield was observed under D8.25N420after two years’ experiment (Table 1). Harvest index value varied by plant densities and N rates. The highest value occurred under D6.75N210across the two years,indicating that more photosynthates were transferred to reproductive organs under this combination of N rate and plant density;while the lowest value was found at 8.25 plants m-2regardless of N rates (Table 1).

3.2.Plant N uptake at harvest and N use efficiency

N uptake of cotton at harvest was significantly affected by plant density,N rate and their interaction (Table 2). TheN0treatments had the lowest plant N uptake relative to other treatments with different N rates. The amount of N uptake varied among N rates and plant density in both years (Table 2). The combination D8.25N420absorbed more N than other treatments across two years.

Table 1 Seed cotton yield,biological yield and harvest index at three plant densities (5.25,6.75 and 8.25 plants m-2) and under five N rates (0,105,210,315,and 420 kg ha-1) in 2016 and 2017

NAE and NRE were significantly affected by N rate,plant density and their interaction. Both NAE and NRE showed a decreasing trend as N rate increased,with the mean values being 4.1,3.1,1.9,and 1.0 kg kg-1in NAE,and 35.3,28.4,22.7,and 17.7% in NRE,at 105,210,315,and 420 kg N ha-1rates,respectively (Table 2). NAE decreased with the increasing plant density after applying 315 or 420 kg N ha-1. NRE increased with increasing plant density,with the NRE at 6.75 and 8.25 plants m-2increasing by 25.4 to 33.1%,respectively,in comparison with that at 5.25 plants m-2in 2016. However,NRE at 8.25 plants m-2was lower than that at 6.75 plants m-2after applying 210 kg N ha-1in 2017,mainly because the amount of N absorbed by cotton decreased as plant density increased (Table 2). Under D6.75N105and D6.75N210,NAE and NRE were higher than under D5.25N315after two years’ experiment.

Table 2 Nitrogen (N) uptake at harvest,N agronomic efficiency (NAE) and N recovery efficiency (NRE) of cotton at three plant densities and under five N rates in 2016 and 2017

3.3.Pn and N concentration of leaves

Leaf senescence,as indicated byPnand N concentration of leaves at the late stage,was significantly influenced by N rate and plant density (Figs.3 and 4). Applying N increasedPnand N concentration of leaves in both years (Figs.3 and 4).Pnof leaves under the treatments of 105 and 210 kg N ha-1was between that under the treatment without N applied and that under the treatments of 315 and 420 kg N ha-1,at all densities in 2016 (Fig.3-A),butPnof leaves under 105 kg N ha-1decreased significantly in comparison to that under 210 kg N ha-1in 2017 (Fig.3-B).

Fig.3 Effects of plant densities (D) (5.25,6.75 and 8.25 plants m-2) and nitrogen (N) rates (0,105,210,315,and 420 kg ha-1) on net photosynthetic rate (Pn) of leaf on 135 days after sowing in 2016 (A) and 2017 (B). Bars mean SE (n=15). Small letters indicate significant differences in N rates (P<0.05).

N concentration of leaves under N rate from 105 to 420 kg ha-1was higher than that without N applied at 5.25 and 6.75 plants m-2,and N concentration of leaves after applying 210 to 420 kg N ha-1at 8.25 plants m-2was higher than that with 0 and 105 kg N ha-1in 2016 (Fig.4-A). However,N concentration of leaves under 105 kg N ha-1decreased in the second year,which was significantly lower than that under 210 kg N ha-1,at 5.25 and 6.75 plants m-2(Fig.4-B).

Fig.4 Effects of plant densities (D) (5.25,6.75 and 8.25 plants m-2) and nitrogen (N) rates (0,105,210,315,and 420 kg ha-1) on N concentration of leaves at the late stage in 2016 (A) and 2017 (B). Bars mean SE (n=3). Small letters indicate significant differences in N rates (P<0.05).

3.4.Soil inorganic N content at harvest

Soil inorganic N content in the 0-60 cm soil layer at harvest increased with increasing N rates at three plant densities in 2016 and 2017 (Fig.5). Compared to 2015,soil inorganic N at harvest in 2016 and 2017 decreased under the treatments of 105 and 0 kg N ha-1but increased considerably under 420 kg N ha-1(Fig.5-A-C). No significant difference in soil inorganic N content was found after applying 210 and 315 kg N ha-1at 5.25,6.75 and 8.25 plants m-2in both years (Fig.5).Soil inorganic N under 0 or 105 kg N ha-1decreased in 2017,compared to 2016 (Fig.5-C).

Fig.5 Inorganic nitrogen (N) content of 0-60 cm soil profile at harvest after applying 0,105,210,315,and 420 kg N ha-1 from 2015 to 2017 at 5.25 (A),6.75 (B) and 8.25 (C) plants m-2. Bars mean SE (n=3). Small letters indicate significant differences in N rates (P<0.05).

3.5.Soil apparent N surplus

Soil apparent N surplus was found under the N treatments from 210 to 420 kg ha-1at all densities in both years (Fig.6-A and B) and under the treatments of 105 kg N ha-1at 5.25 and 6.75 plants m-2in 2016 (Fig.6-A). The apparent N surplus was significantly increased with the increase of N rate (Fig.6).Soil apparent N deficit was observed in all N0treatments at three densities across two years,implying that the supply of soil mineral N and stubble N could not satisfy cotton demand for N (Fig.6;Appendix A). N deficit was also found under D8.25N105in 2016 and under all treatments of 105 kg N ha-1in 2017,as a result of their much higher N uptake and lower N input than under other treatments (Fig.6).

Fig.6 Effects of plant densities (D) (5.25,6.75 and 8.25 plants m-2) and nitrogen (N) rates (0,105,210,315,and 420 kg ha-1) on soil apparent N surplus in 0-60 cm soil layers in 2016 (A) and 2017 (B). Bars mean SE (n=3). Small letters indicate significant differences in N rates (P<0.05).

4.Discussion

Our study provided new insights for understanding the response of cotton yield and N use in cotton field to N application rate and plant density in the context of cotton stubble return. N rate and plant density are two important factors affecting cotton yield,soil N supply and the capacity of cotton absorption and use of N (Bednarzet al.2000;Houet al.2007;Donget al.2010;Zhanget al.2012;Daiet al.2015a). Net mineralization was dominant when crop residues were removed,while there was an initial stage of net immobilization followed by a period of N remineralization due to the addition of crop residues,which would require more N supply to meet the demand of microbe and crop (Kumar and Goh 1999;Recouset al.1999;Dinget al.2011;Chenet al.2014;Huet al.2016). Increased amount of N fertilization increased the yield,but its overuse could cause serious environmental problems through leaching or gaseous emission when N supply exceeding crop demand for a long time (Daiet al.2015b;Zhuet al.2016). Conversely,insufficient N application would cause N deficiency in soil with neglecting N balance between input and output in a consequent application of low N rate. Therefore,soil N cycling and supply could be altered by returning cotton stubble to soil,which might differ from the soil N dynamics when cotton stubble was removed.This requires a re-assessment of N rate and plant density after a long-term cotton stubble return.

The combinations D5.25N315and D6.75N210had the highest seed cotton yield across the two years,which is primarily attributed to their high reproductive allocation of dry matter and high boll occurrence (Table 1;Fig.2). However,soil mineral N content and apparent N surplus continued to increase after applying 315 kg N ha-1at 5.25 plants m-2in our study,which implied that soil N supply was excessive and increased potential risk of N leaching in this combination. Compared to D5.25N315,cotton yield was not decreased under D6.75N210(a 33.3% reduction in N rate and a 28.6% increase in plant density) with high harvest index and N use efficiency after two years’ experiment. It is an effective choice without sacrificing cotton yield if N rate reduces from 315 to 210 kg ha-1,and plant density increases from 5.25 to 6.75 plants m-2. However,the reduction of N input without causing N deficiency in soil and plant is crucial for sustainable production of cotton. Luoet al.(2018) suggested that the N rate could be decreased by 20% at higher density without yield reduction when cotton stubble was removed. However,much abscission of buds and bolls and early senescence would occur under limited N supplies due to fierce intraspecific competition for nutrient resource,leading to the lack of sustainable nutrient supply from soil. In our study,Pnand N concentration of leaves sharply reduced under limited N supplies (105 kg ha-1) during the late stage,which implied that early maturity could occur under insufficient soil N supply. Moreover,soil residual and inorgainic N under D6.75N105presented a decreasing trend over the two years,compared with those in pre-experimental soil,while no difference in soil residual N was found between soil under D6.75N210and pre-experimental soil (Fig.5). Furthermore,from the perspective of N balance,N deficit was found under D6.75N105in the second year,while soil apparent N surplus was observed in D6.75N210in both years (Fig.6). This finding implied that it is undesirable to reduce N rate from 315 to 105 kg ha-1when plant density increased from 5.25 to 6.75 plants m-2.This is becasue the N input could not satisfy cotton acquisition under D6.75N105,while under D6.75N210it could achieve the N balance between soil N supply and cotton demand.

N uptake of cotton plants increased with the increasing N rate and plant density,except under 210 kg N ha-1at 6.75 and 8.25 plants m-2in 2017. The reduction of N uptake indicated that the strong competition for limited nutrient and space at high plant density inhibited cotton growth and N uptake,which was consistent with our finding that NRE decreased with the increase of plant density varying from 6.75 to 8.25 plants m-2in 2017. Changes of N use efficiency to N rate and plant density will assist producers to make N management decisions that are beneficial to sustaining cotton and soil development (Rochester 2011;Zhanget al.2015). There was a reduction in NAE and NRE with the increase of N fertilization in this study. This result is in good agreement with several other studies (Rochester 2011;Donget al.2012;Stamatiadiset al.2016). Stamatiadiset al.(2016) reported that there was a negative relationship between N rate and N use efficiency,and Daiet al.(2015b) and Zhuet al.(2016) found that N loss was rapid when N input surpassed the crop assimilation capacity.

Our study found that high yield of cotton could be achieved by the modification of N rate and plant density after continuous years of cotton stubble return. Soil N surplus was excessive when N rate exceeded 210 kg ha-1,but soil N deficit occurred when N rate was 105 kg ha-1,which showed that the suitable N rate was in the range of 105-210 kg ha-1. Therefore,the sustainability of soil N should be considered in the case of increasing plant density and reducing the proportion of N application. We will continue to refine the reasonable N rate to guide cotton production in the next study.

5.Conclusion

Increasing plant density and reducing N rate can maintain high yield of cotton after a long term cotton stubble return.Compared to the treatment combination used by local farmers (D5.25N315),a 33.3% N reduction and a 28.6% increase in plant density (D6.75N210) obtained high seed cotton yield in both years,but a 66.7% N reduction and a 28.6% increase in plant density (D6.75N105) maintained high yield only in the first year. Compared to D5.25N315,D6.75N105greatly increased NAE and NRE,without sacrificing cotton yield in the first year,but led to early senescence with lowPnand N concentration of leaves and deficit in soil supply N in the second year. The apparent N surplus was positive when N application reached 210 kg ha-1,while the soil N supply was excessive after applying 315 kg N ha-1,leading to the decrease of N use efficiency at all densities. Compared to D5.25N315,D6.75N210had higher NAE and NRE with lower N input by quantifying leaf senescence and soil N balance at the same production level when returning cotton stubble to soil in this study.

Acknowledgements

The study was supported by the National Natural Science Foundation of China (31601253),the Natural Science Foundation of Shandong Province,China (ZR2016CQ20),the China Postdoctoral Science Foundation (2017M610438),and the Modern Agro-industry Technology Research System of Shandong Province,China (SDAIT-03-03/05).

Declaration of competing interest

The authors declare that they have no conflict of interest.

Appendixassociated with this paper is available on http://www.ChinaAgriSci.com/V2/En/appendix.htm


登录APP查看全文