WANG Yue-chao ,LI Xiu-fenLee TArpLEYpENG Shao-bing,DOU Fu-gen

1 Texas A&M AgriLife Research Center,1509 Aggie Drive,Beaumont,TX 77713,USA

2 National Key Laboratory of Crop Genetic Improvement/Key Laboratory of Crop Ecophysiology and Farming System in the Middle Reaches of the Yangtze River,Ministry of Agriculture and Rural Affairs/College of Plant Science and Technology,Huazhong Agricultural University,Wuhan 430070,P.R.China

Abstract Ratoon rice cropping is an important component of the rice cropping system in Texas and south Louisiana,USA,and expanded to Asian countries in 1970. Two field studies were conducted with widely planted rice (Oryza sativa L.) cultivars at Eagle Lake,Texas,USA to determine the effects of nitrogen (N) management in main (first) crop (MC) and ratoon (second) crop(RC) on RC yield. In 2012 and 2013,one cultivar (Presidio) was adopted to determine the effects of RC N management on ratoon yield and head rice yield. In 2016 and 2017,CL153,CL163 and CL272 in addition to Presidio were adopted to examine the effect of MC N management on ratoon yield and head rice yield. N applied at preflood after MC harvest considerably improved RC yield. Application of 99 kg N ha–1 at preflood after MC harvest was practically adequate for RC regrowth,development and approaching the yield potential for Presidio. RC could produce quite high average grain yields of 5.90 to 6.53 t ha–1 in 2012 and 2013,respectively. Main crop N rate only significantly affected MC yield;however,given N applied of 99 kg ha–1 at preflood after MC harvest,ratoon yield was not significantly affected by MC N rate. Neither the main nor ratoon crop N management had a significant effect on RC head rice yield. Considerable RC head rice yields(55–65%) were observed in all of the four cultivars and 4 years except for CL272 in 2016. These results indicat that without very high N fertilizer application,rice ratoon crop could produce a considerable grain yield and an expectative head rice yield. Rice ratooning could be a practical way to increase rice yields with the minimal input in south Texas and regions with a similar climate.

Keywords:rice,nitrogen management,main crop,ratoon crop,grain yield,head rice yield

1.Introduction

The practice of obtaining a second rice crop from tillers originating from the stubble of a harvested one is named(rice) ratooning,which has traditionally been an important component of rice cropping system in the southern USA approximately since 1927 (Anonymous 1963;Mengel and Wilson 1981;Douet al.2016). In this region where the production season is not long enough for a second rice crop from new planting,rice ratooning provides a unique opportunity to maximize the use of land,minimize the production costs,and increase rice yield and producers’income (Evatt 1966;Bollich and Turner 1988;Bond and Bollich 2006a). With the progress in rice breeding and improved management practices,shortage of labor,and predicted changes in climate (Ziskaet al.2018),use of rice ratooning has expanded around the world. For example,producers in Arkansas,USA,where traditionally only the main or first crop was planted,are recently utilizing ratoon crops. The practice of rice ratooning has also increased in China (Donget al.2017;Asensoet al.2019;Wanget al.2019) and Africa (Adigboet al.2012;Itabiyiet al.2016).Moreover,it has been reported that rice ratooning is more sustainable in terms of life cycle assessment (Firouziet al.2018) and energy consumption (Yuanet al.2019) than the first crop.

Nitrogen (N) is the most effective nutrient for promoting regrowth and development of ratoon tillers (Hsiehet al.1983;Chauhanet al.1985;Ichii 1988),and N applied immediately after harvest of the main crop is recommended(Anonymous 1963). Previous studies indicated that rice ratoon crops responded well to N fertilizer. Bahar and De Datta (1977) found that increased rate of N applied to the main crop increased grain yield in both main and ratoon crops. Mengel and Wilson (1981) reported that significant increase in ratoon yield was observed when N fertilizer was applied immediately after main crop harvest.It was also reported that rice ratoon crop produced the maximum yield with 99 kg N ha–1being applied after main crop harvest,and further increase in N rate (132 kg ha–1)did not increase yield in the Gulf Coast area,USA (Bond and Bollich 2006b). Turner and McIlrath (1988) examined the effects of N application rate and timing on ratoon yield in south Texas and found that N applied immediately after main crop harvest had a more evident and positive influence on ratoon yield than N applied before harvest. In ratoon rice production,N fertilizer cost is the greatest direct expense (Outlaw and Raulston 2014). Thus,improving N use efficiency of ratoon rice production will likely enhance the economical sustainability of rice production.

In the USA,rice producers’ income depends on not only grain yield (i.e.,rough rice yield) but also milling quality,especially the head rice yield on which the true price of a rough rice sample is based (Deliberto and Salassi 2011;Lymanet al.2013;Lyman and Nalley 2013). Published literature has well reported the effects of N fertilization on head rice yield of main crop. The effect of N management on head rice yield in conventional rice cropping system(one-harvest rice) also has been intensively reported(Jongkaewwattana and Geng 1991;Alcantaraet al.1996;Walkeret al.2005;Maet al.2009). However,information about the response of yield of ratoon crop and head rice to N management is limited (Mengel and Wilson 1981). The objective of this study was to determine the effects of extra N applied in main crop and N topdressing in ratoon crop on the yields of ratoon crop and head rice.

2.Materials and methods

2.1.Site description and general management practices

Field experiments were conducted at the Texas A&M AgriLife Research Station near Eagle Lake,Texas,USA (29°37´N,96°22´W). The soil at Eagle Lake was a Hockley silt loam(fine,smectitic,hyperthermic Typic Albaqualfs) with 19% silt,15% clay,0.7% organic matter,and pH 5.9. The contents of soil nitrate and ammonium ranged 5–7 and 4–7 g kg–1,respectively,during 2012–2017. Yearly amounts of rainfalls were 913,1 062,1 125,and 1 384 mm,respectively,in 2012,2013,2016,and 2017. Mean annual temperatures for 2012,2013,2016,and 2017 were 21.2,20.0,21.7,and 21.8°C,respectively (Fig.1). The cropping systems for both studies were 1-yr rice and 2-yr fallow for decades. Planting dates were April 9 (second planting,the first planting was severely damaged by birds),March 28,April 5,and March 22 in 2012,2013,2016,and 2017,respectively. Rice was dry seeded around 1 cm in depth using a planter with a seeding rate of 215 seedlings m-2. In our present study,45 kg P2O5ha–1and 45 kg K2O ha–1were applied at preplant.P fertilizer was Ca(H2PO4)2;H2O (triple superphosphate),with P2O5concentration of 45%;K fertilizer was KCl,with K2O concentration of 60%;N fertilizer was urea,with N concentration of 46.2%. Flush irrigation was applied until the 4-to 5-leaf growth stage and then a permanent flood(5–10 cm depth) was imposed and maintained until maturity.More detailed management information was reported by Douet al.(2016).

Two nitrogen fertilization studies were conducted in the field. Study I was conducted in a completely randomized design with four replications,with inbred cultivar Presidio as experimental material in 2012 and 2013. Concentrations of 49.5,69.3,39.6,and 39.6 kg N ha–1were applied at N management in main (first) crop (MC) at preplant,preflood,panicle differentiation,and late booting,respectively (sum of 198 kg N ha–1in the MC). Six ratoon crop nitrogen (RC N) management schemes were utilized,and the detail information is shown in Table 1.

Study II was also conducted in a completely randomized design with four replications,with inbred cultivars Presidio,CL153,CL163,and CL272 as experimental materials in 2016 and 2017. Five main crop nitrogen (MC N) rates,including 0 (MC N omission),165,198,231,and 264 kg N ha–1,were utilized. Concentrations of 25,35,20,and 20%of MC N fertilizers were applied at MC preplant,preflood,panicle differentiation,and late booting,respectively. All ratoon crops in Study II received 99 kg N ha–1at preflood after MC harvest.

Fig.1 The air temperature of Eagle Lake station,USA in 2012,2013,2016,and 2017.

Table 1 Application time and rate of ratoon crop nitrogen (RC N)managements at Eagle Lake,USA in 2012 and 2013 in Study I1)

Cultivars utilized in the study were of good ratoon potential and commonly planted in Texas,USA. Water management and weed control were adapted according to the Texas Rice Production Guidelines (Way and Cockrell 2008). Different N treatments within the same cultivar had quite similar times to developmental stages of main and ratoon crop in both years. Therefore,only the number of days from emergence to specific developmental stages averaged across all N treatments are listed in Table 2. At main and ratoon crop maturity,a plot combine was used to harvest each plot individually for assessment of rice grain yield. The main crop cutting height (stand stubble height after main crop harvest) was about 25 cm. The harvested rice grains were weighed,analyzed for moisture content and the reported grain yields had been adjusted to a uniform moisture content of 120 g water (H2O) kg-1before statistical analysis. The head rice yield (%) was measured by a laboratory rice mill (Zaccaria PAZ/1-DTA,Brazil).

2.2.Data analysis

Data were analyzed for each study separately,using the mixed linear model procedure (PROC MIXED) (SAS 9.4,2015,North Carolina,USA). The effects of RC N management in Study I and MC N rate and variety in Study II were treated as fixed effects or class variables. Year was treated as a random variable. Main and ratoon crop yields were analyzed using a repeated statement,and differences between fixed effects were separated by the least square means at the 5% probability level.

3.results

3.1.Effects of ratoon crop nitrogen management on grain yield and head rice yield

Since there were no significant interactions between experimental years and the effects of ratoon crop nitrogen management on RC grain yield,RC head rice yield or TC grain yield,the data of RC and TC were analyzed for the main effect of ratoon crop nitrogen management (Table 3).Ratoon crop nitrogen management did not significantly affect RC head rice yield (P=0.720);however,significant effects on the grain yield of RC and TC were observed (P＜0.001).The RC N application significantly increased the grain yields of RC (96–116%) and TC (25–30%) over nitrogen omission(Table 4). With the RC N rate of 99 kg ha–1,rice RC produced grain yield of 5.90 t ha–1. With the further increase in RC N rate,however,only relative small increase (≤10.7% or≤0.63 t ha–1) in RC grain yield was observed,compared with the RC N rate of 99 kg ha–1. In addition,splitting RC N application did not benefit RC grain yield.

Table 2 Rice development stages of main and ratoon crops in Study I and Study II

Table 3 The ANOVA of rice main crop (MC),ratoon crop (RC) and total crop (TC,main plus ratoon crops) grain yield and head rice yield (HRY) of MC and RC at Eagle Lake,USA in Study I

Table 4 Rice ratoon crop (RC),total crop (TC,main plus ratoon crops) grain yield and ratoon crop head rice yield (RC HRY) at Eagle Lake,USA in 2012 and 2013 in Study I1)

3.2.Effects of main crop N rate on grain yield and head rice yield

Both cultivar and MC N rate significantly affected the MC grain yield (P＜0.001),but no significant interaction of cultivar by MC N rate was observed (P=0.590,Table 5). Averaged across MC N rates,Presidio produced the lowest MC grain yield in both years,and the highest ones were observed with CL163 in 2016 and CL153 in 2017 (Tables 6 and 7). For all cultivars in both years,MC N applications significantly enhanced MC grain yields over MC N omission (P＜0.05).However,the differences in the MC grain yield among different MC N rates (165,198,231,and 264 kg ha–1) were relatively small and inconsistent for all varieties in both years.

Cultivar had a significant effect on the RC grain yield(P＜0.001),while neither effect of MC N rate (P=0.234) nor interactive effect of cultivar by MC N rate (P=0.996) was significant (Table 5). Presidio and CL163 produced higher RC grain yields than CL153 and CL272 in 2016,however,the differences between cultivars were insignificant in 2017(Tables 6 and 7).

Similar effects of cultivar and MC N rate on the TC grain yield were observed as for the MC grain yield (Table 5),and the interaction of cultivar and MC N rate on the TC grain yield was insignificant (P=0.831) as well. Averaged across MC N rates,Presidio produced the lowest TC grain yield,while CL163 performed well with respect to the TC grain yield in both years. MC N application significantly enhanced the TC grain yield over MC N omission (P＜0.05),except for four treatments (CL272 with 264 kg N ha–1in 2016,Presidio with 165 kg ha–1in 2017,and CL163 with 165 and 198 kg ha–1in 2017).

Table 5 The ANOVA of rice main crop (MC),ratoon crop (RC) and total crop (main plus ratoon crops) grain yield and head rice yield (HRY) of MC and RC at Eagle Lake,USA in Study II

Table 6 Grain yield and head rice yield (HRY) of rice main crop (MC) and ratoon crop (RC),and total crop (TC,main plus ratoon crops) grain yield at Eagle Lake,USA in 2016 in Study II

Both cultivar and MC N rate had a significant effect on the MC head rice yield (P＜0.001),and also significant interaction of cultivar and MC N rate was observed (P＜0.001;Table 5).Differences in the MC head rice yield among different MC N rates were inconsistent across cultivars and years.

Cultivar had a significant effect on the RC head rice yield(P=0.002),but MC N rate did not significantly affect the RC head rice yield (P=0.646;Table 5). No significant interaction of cultivar by MC N rate was observed (P=0.179). MC N application did not significantly improve the RC head rice yield over MC N omission,except for CL272 with 165 and 231 kg N ha–1in 2016 (P＜0.05).

4.Discussion

The effect of ratoon crop nitrogen fertilizer management on rice ratoon crop yield was examined using a widely planted rice inbred cultivar Presidio at Eagle Lake in 2012 and 2013(Study I). N applied immediately after main crop harvest was recommended in Texas (Anonymous 1963). It was also reported that N applied after main crop harvest significantly increased ratoon crop grain yield in Louisiana (Mengel and Wilson 1981) and Texas (Turner and McIlrath 1988;Bond and Bollich 2006b). In our present study,rice ratoon crop yield responded well to N fertilizer (Table 4). N application after main crop harvest approximately doubled rice ratoon crop yield over N omission. However,the differences in ratoon yield among three N rates (99,132 and 165 kg ha–1)were relatively small. Bond and Bollich (2006b) concluded that rice ratoon crop produced the maximum grain yield with 99 kg N ha–1applied immediately after main crop harvest for several cultivars. Also,given N fertilizer of 99 kg ha–1applied at preflood,there was no huge enhancement in ratoon yield when extra N (33 and 66 kg ha–1) was applied at 2 wk after main crop harvest. These results indicate that one dose of 99 kg N ha–1applied at preflood after main crop harvest was almost adequate for the ratoon crop regrowth,development and production.

Table 7 Grain yield and head rice yield (HRY) of rice main crop (MC) and ratoon crop (RC),and total crop (TC,main plus ratoon crops) grain yield at Eagle Lake,USA in 2017 in Study II

We also examined the main crop N rate effect on rice grain yield with four widely planted rice inbred cultivars in 2016 and 2017 (Study II). Our results indicated that rice main crop yield,rather than ratoon yield,was significantly improved by main crop N application (Tables 6 and 7).Our results showed that the rice main crop growing under extreme deficiency of N still produced a comparable ratoon yield,which does not seem to agree with previous studies stating that a good crop stand in main crop was essential in producing a good ratoon crop (Bahar and De Datta 1977;Ichii and Sumi 1983;Turner and Jund 1993). This may be partially due to the relatively low ratoon crop yields in 2016 and 2017,with a maximum at 4.85 t ha–1. Our multiple-year field trials at both sites indicated that nitrogen application rate at 99 kg ha–1was sufficient to achieve the ratoon crop yield potential. The relatively low ratoon yield in both years might be due to other reasons including unfavorable weather conditions. However,more field observations including rice physiological measurement,detailed yield component analysis and nutrient uptake analysis may help clarify this inconsistence between our results and others. If our ratoon crop yield level was at 6 t ha–1like in 2012 and 2013,we may see the importance of good crop stand in main crop for the ratoon crop yield. As concluded above (Study I),the possible reason explaining the discrepancy was that the N rate of 99 kg ha–1applied at preflood after main crop harvest compensated for the ratoon yield decrease caused by N deficiency of rice main crop. A previous study conducted by Turner and McIlrath (1988) suggested that N fertilizer applied immediately after main crop harvest had a more evident and positive influence on ratoon crop yield than N applied before harvest. Our results also indicated that appropriate N rates increased rice main crop yield,while the overdose might not benefit grain yield and farmers’ net profit. Furthermore,excessive application of N fertilizer not only decreased N use efficiency but might also cause serious environmental problems (Penget al.2006;Juet al.2009;Guoet al.2010).In summary,given N applied after main crop harvest at 99 kg ha–1,the main crop N rate affected rice main crop rather than ratoon crop yields,however,extremely high main crop N rate benefited neither rice main nor ratoon crop yield. It will be effective and reliable to determine the effect of main crop N rate on rice ratoon crop yield with and without N applied after main crop harvest. In this way,not only the main effects but also the interactive effect of rice main and ratoon crop N can be identified.

It was remarkable that,averaged across 2 yr of 2012 and 2013 for Presidio,a quite high ratoon crop yield of 5.90–6.53 t ha–1(except for ratoon crop N omission) that was approximately equal to three quarters of the main crop yield was observed in this study (Table 4). Similarly,high ratoon yields of 5.83 and 5.71 t ha–1were observed with elite local cultivars in central China by Donget al.(2017)and Chenet al.(2018),respectively. These results imply that rice ratoon crop possesses high yield potential (about 6 t ha–1). However,ratoon yield varied significantly across years (Tables 4,6 and 7),especailly with only 2.45–2.62 t ha–1(except for main crop N omission) being recorded for Presidio in 2017. These results indicated that there was a lack of stability in rice ratoon crop yield performance. For rice ratoon crop,besides the environment,the nutrient condition of the main crop stubble might be an important factor explaining the instability of ratoon yield. For example,it was reported that nutrient condition of the main crop culms during ripening stage might contribute to initial ratoon growth and development,and eventually to ratoon yield (Bollichet al.1982;Turner and Jund 1993). Douet al.(2016) also reported a significant correlation between main crop yield and ratoon crop yield.

Besides grain yield,the head rice yield was closely associated with rice producers’ income (Lyman and Nalley 2013). We also examined the head rice yield of ratoon crops. Our results indicated that neither main crop N rate nor ratoon crop N management affected ratoon crop head rice yield. Similarly,it also had been reported that ratoon crop N had no effect on ratoon crop head rice yield in the same region (Mengel and Wilson 1981). The possible reasons may include 1) environmental conditions during rice ripening stages and 2) harvesting moisture. Head rice yield of 55% was regarded as the expectative level and industry standard on which monthly long-grain rough rice prices were based (Lyman and Nalley 2013;Childs and Skorbiansky 2018). In our present study,high ratoon crop head rice yields (55–65%) were observed for all cultivars except for CL272 in 2016 (Tables 4,6 and 7),ensuring that farmers would receive an acceptable rough rice price for ratoon crop,except for a great reduction with CL272 in 2016. Head rice yield was examined only for several inbred cultivars without hybrid ones in our present study.Recently,however,hybrid cultivars which typically outyield inbred ones in both rice main and ratoon crops also have been widely adopted in rice ratooning in south Texas (Douet al.2016). Despite rapid adoption of hybrids in the midsouth USA,their economic advantage is still being debated largely due to inconclusive evidence that the hybrid milling quality was substandard relative to that of inbred cultivars(Lyman and Nalley 2013). Therefore,it will be meaningful and necessary to investigate ratoon crop head rice yield of hybrid cultivars and their responses to main and ratoon crop N management.

5.Conclusion

Our study summarized that N applied at preflood after main crop harvest considerably improved ratoon crop yield.Application of 99 kg N ha–1at preflood after main crop harvest was almost adequate for ratoon crop regrowth,development and yield production for Presidio. Main crop N rate significantly affected rice main crop;however,given N applied at 99 kg ha–1at preflood after main crop harvest,the rice ratoon crop yield was not significantly affected by main crop N rate. In addition,neither main crop N nor ratoon crop N had a significant effect on ratoon crop head rice yield. Rice ratooning could be a practical way to increase rice yield with minimal input in south Texas and similar climates.

Acknowledgements

This work was partially supported by the Texas Rice Research Foundation,USA.

Declaration of competing interest

The authors declare that they have no conflict of interest.

references

Adigbo S O,Olojede M O,Harris P J C,Ajayi O.2012.Ratooned lowland Nerica rice varieties as an option for triple cropping in inland valleys of derived Savannah in Nigeria.ExperimentalAgriculture,48,551–562.

Alcantara J M,Cassman K G,Consuelo M P,Bienvenido O J,Samuel P L.1996.Effects of late nitrogen fertilizer application on head rice yield,protein content,and grain quality of rice.CerealChemistry,73,556–560.

Anonymous.1963.Second crop rice– A significant factor in total rice production in Texas.RiceJournal,66,10–12,43.

Asenso E,Zhang L Y,Tang L M,Issaka F,Tian K,Li J H,Hu L.2019.Moldboard plowing with direct seeding improves soil properties and sustainable productivity in ratoon rice farmland in southern China.Sustainability,11,6499.

Bahar F A,De Datta S K.1977.Prospects of increasing tropical rice production through ratooning.AgronomyJournal,69,536–540.

Bollich C N,Turner F T.1988.Commercial ratoon rice production in Texas,USA.In:RiceRatooning.International Rice Research Institute,Los Baños,the Philippines.pp.257–263.

Bollich C N,Webb B D,Scott J E.1982.Variation in carbohydrate reserves from heading to maturity among rice varieties.Rice TechnicalWorkingGroup.Hot Springs,Ark.,USA.

Bond J A,Bollich P K.2006a.Effects of pre-harvest desiccants on rice grain yield and quality.CropProtection,26,490–494.

Bond J A,Bollich P K.2006b.Ratoon rice response to nitrogen fertilizer.CropManagement,5,1–5.

Chauhan J S,Vergara B S,Lopez F S S.1985.RiceRatooning.IRRI Res.Pap.Ser.vol.102.International Rice Research Institute,Los Baños,the Philippines.

Chen Q,He A B,Wang W Q,Peng S B,Huang J L,Cui K H,Nie L X.2018.Comparisons of regeneration rate and yields performance between inbred and hybrid rice cultivars in a direct seeding rice–ratoon rice system in central China.FieldCropsResearch,223,164–170.

Childs N,Skorbiansky S R.2018.RiceOutlook:U.S.Rough Rice Prices Expected to Drop in 2018/19.Publication RCS-18K.USDA Economic Research Service,Washington,DC.

Deliberto M,Salassi M.2011.Hybrid rice production possibilities in the southern rice producing region.Journal of American Society of Farm Managers and Rural Appraisers,2011,42–52.

Dong H L,Chen Q,Wang W Q,Peng S B,Huang J L,Cui K H,Nie L X.2017.The growth and yield of a wet-seeded rice–ratoon rice system in central China.FieldCrops Research,208,55–59.

Dou F,Tarpley L,Chen K,Wright A L,Mohammed A R.2016.Planting date and variety effects on rice main and ratoon crop production in south Texas.CommunicationsinSoil ScienceandPlantAnalysis,47,2414–2420.

Evatt N S.1966.High annual yields of rice in Texas through ratoon or double-cropping.RiceJournal,69,10–12,32.

Firouzi S,Nikkhah A,Aminpanah H.2018.Rice single cropping or ratooning agro-system:Which one is more environmentfriendly?EnvironmentalScienceandPollutionResearch,25,32246–32256.

Guo J H,Liu X J,Zhang Y,Shen J L,Han W H,Zhang W F,Christie P,Goulding K W T,Vitousek P M,Zhang F S.2010.Significant acidification in major Chinese croplands.Science,327,1008–1010.

Hsieh C F,Lin C Y,Lai T S.1983.Effects of nitrogen fertilizer and delaying harvest on the recovery rate of ratoon rice.BulletinTaichungDistrictAgricultureImprovemerntStation,7,38–44.

Ichii M.1988.Some factors influencing the growth of rice ratoon.In:RiceRatooning.International Rice Research Institute,Los Baños,the Philippines.pp.41–46.

Ichii M,Sumi Y.1983.Effect of food reserves on the ratoon growth of rice plant.JapaneseJournalofCropScience,52,15–21.

Itabiyi O V I,Adebowale A A,Shittu T A,Adigbo S O,Sanni L O.2016.Effect of ratooning process on the engineering properties of Nerica rice varieties.QualityAssuranceand SafetyofCrops&Foods,8,21–31.

Jongkaewwattana S,Geng S.1991.Effect of nitrogen and water management on panicle development and milling quality of California rice (OryzasativaL.).JournalofAgronomyand PlantScience,167,43–52.

Ju X T,Xing G X,Chen X P,Zhang S L,Zhang L J,Liu X J,Cui Z L,Yin B,Christie P,Zhu Z L,Zhang F S.2009.Reducing environmental risk by improving N management in intensive Chinese agricultural systems.Proceedingsof theNationalAcademyofSciences of the United States of America,106,3041–3046.

Lyman L,Nalley L L.2013.Economic analysis of hybrid rice performance in Arkansas.AgronomyJournal,105,977–988.

Lyman N B,Jagadish K S V,Nalley L L,Dixon B L,Siebenmorgen T.2013.Neglecting rice milling yield and quality underestimates economic losses from hightemperature stress.PLoS ONE,8,e72157,

Ma Q,Zhang H C,Dai Q G,Wei H Y,Huo Z Y,Xu K,Yin C Y,Hang J,Zhang S F,Zhang Q.2009.Effects of nitrogen application rate and growth-development type on milling quality injaponicarice.ActaAgronomicaSinica,35,1282–1289.(in Chinese)

Mengel D B,Wilson F E.1981.Water management and nitrogen fertilization of ratoon crop rice.AgronomyJournal,73,1008–1010.

Outlaw J L,Raulston J.2014.Rice productin economics.In:TexasRiceProductionGuidelines2014.College Station,TX.pp.89–96.

Peng S B,Buresh R J,Huang J L,Yang J C,Zou Y B,Zhong X H,Zhang F S.2006.Strategies for overcoming low agronomic nitrogen use efficiency in irrigated rice systems in China.FieldCropsResearch,96,37–47.

Turner F T,Jund M F.1993.Rice ratoon crop yield linked to main crop stem carbohydrates.CropScience,33,150–153.

Turner F T,McIlrath W O.1988.N fertilizer management for maximum ratoon crop yields.In:RiceRatooning.International Rice Research Institute,Los Baños,the Philippines.pp.187–194.

Walker T W,Martina S W,Gerard P D.2005.Grain yield and milling quality response of two rice cultivars to topdress nitrogen application timings.AgronomyJournal,98,1495–1500.

Wang Y C,Zheng C,Xiao S,Sun Y T,Huang J L,Peng S B.2019.Agronomic responses of ratoon rice to nitrogen management in central China.FieldCropsResearch,241,107569.

Way M O,Cockrell J (eds).2008.TexasRiceProduction Guidelines2008.Texas A&M AgriLife Extension Service Publication,College Station,TX.

Yuan S,Cassman K G,Huang J L,Peng S B,Grassini P.2019.Can ratoon cropping improve resource use efficiencies and profitability of rice in central China?FieldCropsResearch,234,66–72.

Ziska L H,Fleisher D H,Linscombe S.2018.Ratooning as an adaptive management tool for climatic change in rice systems along a north-south transect in the southern Mississippi valley.AgriculturalandForestMeteorology,263,409–416.