Invasion of fall armyworm led to the succession of maize pests in Southwest China

Zezheng Fan , Yifei Song, Shengyuan Zhao, Kongming Wu#

1 Agricultural Information Institute, Chinese Academy of Agricultural Sciences, Beijing 100081, China

2 Guangdong Laboratory for Lingnan Modern Agriculture, Guangzhou 510642, China

3 State Key Laboratory for Biology of Plant Diseases and Insect Pests, Institute of Plant Protection, Chinese Academy of Agricultural Sciences, Beijing 100193, China

Abstract The invasive fall armyworm Spodoptera frugiperda (J.E.Smith) invaded Asia in 2018, colonizing the tropical and southern subtropical regions as well as migrating with the monsoons into Northeast Asia during spring and summer.This has resulted in widespread infestations, with significant impacts on maize production in various Asian countries.Previous studies have shown that the invasion of this pest can alter the species relationships of maize pests, but the actual impact on maize pest management is still unclear.This study investigated the changes in maize pest occurrence and pesticide use in the annual breeding areas of S.frugiperda in Yunnan Province and the Guangxi Zhuang Autonomous Region of China during 2017–2021, based on surveys and interviews with small farmers in maize production.The results showed that S.frugiperda has emerged as the dominant species among maize pests after invasion and colonization, replacing traditional pests such as Ostrinia furnacalis, Spodoptera litura, Agrotis ypsilon, and Rhopalosiphum maidis.The variety of pesticides used for maize pest control has changed from chlorpyrifos, lambda-cyhalothrin, and acetamiprid to emamectin benzoate-based pesticides with high effectiveness against S.frugiperda.Furthermore, the frequency of maize pest chemical applications has increased from an average of 5.88 to 7.21 times per season, with the amounts of pesticides used in summer and autumn maize being significantly higher than in winter and spring maize, thereby increasing application costs by more than 35%.The results of this study clarified the impact of S.frugiperda invasion on maize pest community succession and chemical pesticide use in tropical and south subtropical China, thereby providing a baseline for modifying the regional control strategies for maize pests after the invasion of this relatively new pest.

Keywords: Spodoptera frugiperda, annual breeding region, sweet waxy corn, economic evaluation

1.Introduction

The fall armyworm (FAW),Spodopterafrugiperda(J.E.Smith) (Lepidoptera: Noctuidae), is a major migratory pest native to tropical and subtropical America (Luginbill 1928; Sparks 1979).In America, the annual breeding region ofS.frugiperdaranges from Florida and southernTexas in the north, to La Pampa in Peru, Chile, and northern Argentina in the south, with the migration of adults to high-latitude regions and reproductive damage being accomplished through a multigeneration seasonal migratory pattern (Machadoet al.2007; Meagheret al.2008).In 2016,S.frugiperdainvaded West Africa,spreading rapidly throughout sub-Saharan Africa in only two years (Rwomushanaet al.2017).In May 2018,S.frugiperdainvaded Karnataka in the southern part of India (Sharanabasappaet al.2018), and it then successfully crossed the Red Sea before expanding from eastern Africa into Asian countries such as Yemen and India (Padheeet al.2019).Spodopterafrugiperdathen migrated from Myanmar to China in December 2018 before completing its colonization of southern China in 2019 (Wu 2020; Sunet al.2021).So far,S.frugiperdahas invaded more than 70 countries worldwide, becoming a significant ecological issue that is affecting agricultural development and the environment.After the invasion of Africa byS.frugiperda, maize yields in countries such as Ghana, Zambia, and Zimbabwe suffered losses up to 9.4–67%, potentially leading to millions of dollars in economic losses each year (Dayet al.2017; Wossenet al.2017; Baudronet al.2019; Kumelaet al.2019).After the invasion of Asia, the area of maize affected amounted to 3.9% in India in 2019, with yield reductions of 1.2–9% and up to 51% in individual fields (Naviket al.2021); and pesticide consumption on maize increased after the invasion of this pest (Deshmukhet al.2021).Spodopterafrugiperdadamage has occurred in 50 provinces of Thailand, with maize infestations up to 100% in six severely affected provinces resulting in yield losses of 4.62 million tons along with economic losses of 130–160 million USD (FAO 2019).In China,S.frugiperdawas found across 1.125 million ha in 2019 and caused damage to 160,000 ha even after prevention and control,therefore posing a critical threat to maize production(MARA 2019).

Interspecific competition due to biological invasions is a primary factor affecting the stability and diversity of species in an ecosystem.Furthermore, community succession can occur when certain species dominate the competition (Dennoet al.1995).Divyaet al.(2021)demonstrated that the invasion ofS.frugiperdain India had contributed to a decline in the population of another native Lepidopteran pest,Sesamiainferens(Walker).In China,S.frugiperdahas been found to occur in combination withHelicoverpaarmigera(Hübner),Spodopteraexigua,S.litura, andMythimna loreyi(Duponchel) pests in maize fields, although it possesses a stronger competitive advantage (Guoet al.2019; Chenet al.2020; Songet al.2021).Before the invasion ofS.frugiperda, maize crops in the border areas of Southwest China had long suffered from a variety of maize pests, includingO.furnacalis,Paralipsagularis,R.maidis,S.litura,S.exigua, andA.ypsilon, among others (Lvet al.2013; Lin and Lu 2016; Yuet al.2018).Following its invasion,S.frugiperdahas shown a marked competitive advantage compared to locally occurring pests such asS.lituraandH.armigera(Song and Wu 2020).During winter,S.frugiperdaprimarily infests maize and sugarcane fields in tropical and south subtropical regions of southern China and neighbouring Southeast Asian countries, before gradually moving into several regions of southwestern, central, and northern China with the East Asian and Indian monsoons in the following spring (Wuet al.2019, 2021a, b; Yanget al.2021a).In addition to maize,S.frugiperdafeeds on potatoes (Zhaoet al.2019), peanuts (Heet al.2020), wheat (Xuet al.2019; Liet al.2020), barley, oats, and millets (Zhaoet al.2020) as primary hosts.As annual breeding regions forS.frugiperda, Yunnan and Guangxi both cultivate maize year-round, so maize is present in fields throughout the year, alongside nearly 13 million acres of maize that are planted in the bordering countries of Myanmar, Laos, and Vietnam (DAYN 2022), so a huge base of pest sources can be found both inside and outside the country.Once the population density ofS.frugiperdahas increased in these areas, the populations will spread rapidly to other regions, undermining the prevention and control ofS.frugiperdanationwide.Therefore, it is necessary to clarify the occurrence and damage patterns ofS.frugiperdain the tropical and south subtropical regions to contain the infestation at its source while preventing the migration to the primary maize producing regions in the north.

This study investigated the changes in pest population succession and the control patterns ofS.frugiperdaas a pest of maize in the annual breeding regions of Yunnan and Guangxi, China, both before and afterS.frugiperdainvasion, while also investigating the changes in on-farm management measures adopted by farmers in response to theS.frugiperdainvasion.The results of this study provide a scientific basis for the prevention and control of the spread ofS.frugiperdaat the national level.

2.Materials and methods

2.1.Climatic and cultivation characteristics of the survey areas

The locations selected for this survey included Ruili(97°31′–98°02′E, 23°38′–24°14′N) and Jinghong(100°25′–101°31′E, 21°27′–22°36′N) in Yunnan Province,as well as Guigang (109°11′–110°39′E, 22°39′–24°02′N)and Chongzuo (106°33′–108°06′E, 21°36′–23°22′N) in Guangxi, which are within the annual breeding region ofS.frugiperdain China.A field investigation focusing on the occurrence of maize pests, along with a questionnaire survey of maize farmers in 22 villages across the four cities, was conducted between March 2021 and October 2021 (Fig.1; Table 1).The background information on climate, precipitation, crop production and other factors in the four cities are listed in detail in Table 1.

Table 1 The background information for the four survey cities

Fig.1 The locations of the four survey areas in Yunnan Province and the Guangxi Zhuang Autonomous Region, China.

2.2.Survey methodology and content

In this study, sample villages and farmers were randomly selected for the survey.Subsequently, the village committees surveyed provided statistics on maize farming, from which the farmers were selected according to the variations of the village farming scale.The method for sampling maize farmers was a combination of random sampling and recommendations of “typical cases” by the local government, while the field ownership areas of the surveyed farmers were in the range of 0.01–3 ha (the average area was approximately 0.47 ha).The farmers in each city completed 150–500 questionnaires, with a total of 1,200 questionnaires being completed across the four cities, 1,115 of which were valid.The questionnaire interviews were primarily conducted through face-to-face interviews, including surveys of the farmers cultivating farmland and household interviews with farmers.The questionnaires were designed with questions relevant for the purpose of this study, with each household being interviewed for 30–40 min.

2.3.Variable calculation method

The pest occurrence percentages were calculated by surveying each farmer’s fields.If the field was not infested, the value was assigned as 0; if the field was infested by FAW, the value of FAW was assigned as 1,and the rest as 0; if the field suffered from other pests,then the value of other pests was assigned as 1 and the rest as 0.The percentages of occurrence of different pests were calculated independently.

Spodoptera frugiperdaoccurrence percentage in farmer fields=Number of farmers affected by FAW/Number of total surveyed farmers

The percentage of the main pesticide used was calculated by determining the pesticides used by each farmer, and calculating the proportion of each pesticide used in the local sample.

2.4.Data analysis

Multiple comparisons using one-way ANOVA with Duncan’s new multiple range method were conducted for the frequency of pesticide application, pesticide application cost per hectare, seed cost per hectare,fertilizer cost per hectare, and crop production per hectare between different years.

To further validate the results of the multiple comparison analysis and to evaluate the effects ofS.frugiperdainvasion on on-farm management practices and crop yields, as well as how agricultural management practices were changed specifically, ordinary least squares (OLS)was also conducted here.To empirically isolate and test the effect ofS.frugiperdainvasion on the outcome variables, the models were based on an econometric model of farmer application behaviour (Pingali and Carlson 1985; Huanget al.2003), using the survey data collected.The model was estimated in the form of:

wherei,p, andtdenote farmer, plot, and year variables,respectively, whileYiptdenotes the outcome variable representing either the number of applications, the cost per hectare of application (CNY ha–1) or fertilizer input (CNY ha–1) applied by farmerion plotpin yeart.Meanwhile,Fiptis a dummy variable that is 1 if the plot suffered fromS.frugiperda, or 0 otherwise.Xiprepresents a vector of observable plot and householdlevel covariates, and since farm household characteristics directly influence farm production behaviour, it can explain unobserved household-level heterogeneity by introducing household fixed effects, thereby controlling household fixed effects, while also covering a range of plot characteristics to control the observed and unobserved characteristics of specific plots.Tjindicates year effects, in order to measure trends in farm management practices using a series of year dummy variables(including four years in total, 2018, 2019, 2020, and2021, with the control year being 2017) as explanatory variables.Pitindicates price effects to include the influences of movements in pesticide prices, seed prices,and fertilizer prices.Vjindicates village fixed effects to control for different agroclimatic conditions, as well as prevailing market and economic conditions in the village.Furthermore,εiptis the parcel and household-specific error term.The parameters to be estimated wereα, β, δ, σ,ηandφ; however, the primary concern was to examine the impact and extent ofS.frugiperdainvasion on farm management practices, which isβ.

All data analyses were processed using SPSS 26.0 and Stata SE 16.

3.Results

3.1.Changes in maize pest species dominance after the S.frugiperda invasion

Before theS.frugiperdainvasion, the controlled pests of sweet maize in Ruili primarily includedO.furnacalis,S.litura,P.gularis, andS.exigua.In 2021, the occurrence ofS.frugiperdaaccounted for 98.97% of the total maize pests after the farmers’ control measures,whereasS.liturawas the second most abundant.The less influential pests wereO.furnacalisandS.exigua,withS.frugiperdareplacingO.furnacalisandS.lituraas the dominant pest under local control.In recent years,A.ypsilon,O.furnacalis, andR.maidiswere the primary pests occurring in the maize fields of Jinghong.In 2021,the most prevalent pest of glutinous maize in Jinghong after farmer control wasS.frugiperda, with 96.36% of the total number of infested maize plants.The less impactful pests wereR.maidisandO.furnacalis, which accounted for 2.02 and 1.62%, respectively (Fig.2).

Next he disguised himself as an ordinary messenger, mounted a swift camel, and sped away to the place where the slave girl was to be handed over to him

Before theS.frugiperdainvasion, the controlled pests of maize in Guigang mainly includedO.furnacalis,A.ypsilon, andR.maidis.In 2021,S.frugiperdaaccounted for 94.78% of the total number of maize pests in Guigang.Furthermore, the most severe pest occurring in hybrid maize after farmer control wasS.frugiperda,while the less severe pests wereO.furnacalisandR.maidis.In recent years, the primary pests controlled in maize fields in Chongzuo includedO.furnacalisandR.maidis.However, in 2021, theS.frugiperdaoccurrence in Chongzuo reached 88.82% after control measures, replacingO.furnacalisandR.maidisas the dominant pests controlled in local maize fields (Fig.2).

The invasion ofS.frugiperdahas changed the dominant pest species of maize fields in the survey area,withS.frugiperdareplacing traditional pests such asO.furnacalis,S.litura,A.ypsilon, andR.maidis, and emerging as the dominant pest that poses a significant threat to local maize production.

3.2.Changes in maize pest control pesticides after the invasion

Farmers in the survey area tended to use chemical pesticides, while only a few farmers applied biological pesticides such asH.armigeranuclear polyhedrosis virus.The selection of pesticides is generally guided by local agronomy stations and pesticide retailers.Before theS.frugiperdainvasion, farmers commonly used pyrethroid pesticides to control target pests, and the most widely used was beta-cyfluthrin (Table 2).The other main types of pesticides were chlorpyrifos and chlorantraniliprole.

Table 2 Main insecticides used for control of the target pests by maize farmers during 2017–2019

In 2021, farmers used up to 35 chemical pesticides with various ingredients to controlS.frugiperda, with the most commonly applied being emamectin benzoate and emamectin benzoate+chlorfenapyr, emamectin benzoate+lambda-cyhalothrin, and emamectin benzoate+chlorpyrifos, among which emamectin benzoate accounted for the highest percentage ranging from 28.95% in Chongzuo to 53.35% in Ruili (Fig.3).According to the WHO classification of these pesticides,they are all moderately hazardous (WHO 2019).Although most pesticides can be found in the official guidelines forS.frugiperdacontrol, there are a few exceptions.For example, 13.36% of farmers in Jinghong apply the highly hazardous pesticides avermectins and chlorantraniliprole,alongside other moderately hazardous pesticides including acetamiprid in Jinghong, and lufenuron andchlorfenapyr in Guigang.

Fig.3 Main percentages of pesticide use chosen by maize farmers in 2021.EB, emamectin benzoate; EBL,emamectin benzoate+lambda-cyhalothrin; EBC, emamectin benzoate+chlorpyrifos.Some farmers used up to dozens of pesticides, but only the three main ones are shown in the figure.

Among the 12 insecticides with various ingredients used by farmers in Ruili to controlS.frugiperdain 2021,the use rate of insecticides with emamectin benzoate as the primary ingredient was as high as 55.35%; in addition,the proportions of farmers using emamectin benzoate+chlorfenapyr, emamectin benzoate, lambda-cyhalothrin,emamectin benzoate+beta-cypermethrin, and emamectin benzoate+lufenuron were 9.47, 8.23, 7.61, and 7.41%of farmers, respectively.Farmers in Jinghong used seven insecticides with various compositions to controlS.frugiperda, generally using two insecticides, with emamectin benzoate and emamectin benzoate+lufenuron being the most frequent.Furthermore, the usage rate of emamectin benzoate was as high as 30.77%, while that of emamectin benzoate+lufenuron was 18.62% (Fig.3).

In Guigang and Chongzuo, the pesticides used for control changed from beta-cypermethrin, acetamiprid,and chlorantraniliprole to emamectin benzoate.Most of the pesticides used by farmers in Guigang have emamectin benzoate as the effective ingredient, with emamectin benzoate and chlorfenapyr being the two most frequently used and the usage rate of emamectin benzoate reaching 45.65%.Farmers in Chongzuo used eight types of pesticides with different ingredients to controlS.frugiperda, most of which included emamectin benzoate as the active ingredient, and emamectin benzoate was the most frequently used pesticide at a usage rate of 28.94%.

The economic benefits of cultivating sweet glutinous maize are relatively high, and under the high-input–high-output pattern, farmers respond to the damage fromS.frugiperdaby the high-frequency application of pesticides regardless of the cost, resulting in a dramaticincrease in pesticide usage.The frequency of application increased from 5.17 to 6.17 during winter (F=34.434,P<0.05) and from 6.01 to 7.44 during summer (F=46.719,P<0.05) in Ruili; while the frequency of applications increased from 4.94 to 6.01 during winter (F=16.106,P<0.05) and from 5.88 to 7.26 during summer (F=22.836,P<0.05) in Jinghong.The frequency of applications in Guigang increased from 5.02 to 6.20 during spring(F=24.762,P<0.05) and from 6.22 to 7.32 during autumn(F=15.840,P<0.05); and the frequency increased from 4.57 to 5.64 during spring (F=24.762,P<0.05) and from 5.42 to 6.78 during autumn (F=15.840,P<0.05) in Chongzuo.Significant seasonal discrepancies between all four cities were found regarding the magnitudes of the increases in application frequency (Fig.4).

Fig.4 Comparative analysis of pesticide application times of maize in 2017–2021.Season I, Ruili winter, Jinghong winter, Guigang spring, Chongzuo spring; Season II, Ruili summer, Jinghong summer, Guigang autumn, Chongzuo autumn.Data are mean±SE(Ruili, n=2,430; Jinghong, n=1,235; Guigang, n=1,150; Chongzuo, n=755).Different lowercase letters indicate significant differences in application costs per season between years in the same area (P<0.05).

From Table 3, the estimated coefficient forS.frugiperdainvasion is shown to be positive, after controlling for household fixed effects, village fixed effects, and price effects, and using 2017 as the base period for the control group.This indicates a positive correlation betweenS.frugiperdainvasion and application frequency.Therefore,S.frugiperdainvasion has led to a moderate increase in the application frequency for both seasons in the maize fields across the four cities since 2019.The magnitudes of the increases in the second season were stronger than the first season in Ruili, Jinghong, and Chongzuo, indicating that the extent of pest infestation was more severe in summer and autumn than in winter and spring.In contrast, the second season increased by less than the first season in Guigang, which has suffered in recent years from frequent droughts and the low economic efficiency of forage maize in the summer, causing the local farmers to contain pest control costs when planting in order to reduce the application frequency.

Furthermore, Table 4 also reveals that hiring labour and renting land had no significant effect on farm management practices.Meanwhile, distance from the home to the land did not affect farmers’ motivation to plant and had no significant effect on application frequency.There was also no significant effect of the role played by the agronomy station in pest control, which reflects the farmers’ confidence in the agronomy extension staff and whether they had been instructed by the agronomy extension staff.

Table 4 Model estimation results with different control variables

3.3.Changes in maize productivity costs and yields after the invasion

The productivity costs include pesticide application, seed,and fertilizer costs, of which pesticide application costs are the sum of the pesticide purchase and labour costs.Spodopterafrugiperdainfestation increased pesticide application costs by more than 35% per hectare of maize field per season.Some farmers increased fertilizer application to reduce the yield losses caused by pests,resulting in fertilizer inputs increasing by approximately 6%, due to more frequent applications and an increased labour burden.

Fig.5 shows that the pesticide application cost per hectare in Ruili increased by 630.15 CNY during winter(F=80.301,P<0.05) and 840 CNY during summer(F=92.641,P<0.05); while the fertilizer cost per hectare increased by 397.8 CNY during winter (F=12.086,P<0.05) and 674.7 CNY during summer (F=21.520,P<0.05).The application cost per hectare in Jinghong increased by 526.95 CNY during winter (F=8.798,P<0.05) and 579.3 CNY during summer (F=9.223,P<0.05); while the fertilizer cost per hectare increased by 863.85 CNY during winter (F=2.623,P<0.05) and 775.05 CNY during summer (F=2.323,P<0.05).The application cost per hectare in Guigang increased by 381.75 CNY during spring (F=34.218,P<0.05) and 389.85 CNY during autumn (F=26.471,P<0.05); while the fertilizer cost per hectare increased by 202.65 CNY during spring (F=2.763,P<0.05).The lack of a significant increase in fertilizer cost during autumn may have been due to the fact that autumn was already characterized by lower rainfall and sunshine, so the farmers did not have high psychological expectations of yields due to drought conditions and would not increase their fertilizer application in the autumn to maintain yields (F=0.928,P>0.05).The application cost per hectare increased by 421.8 CNY during autumn in Chongzuo (F=21.843,P<0.05),while the fertilizer input increase per hectare was not significant.Although the infestation had a significanteffect on the application cost in spring, this impact was not significant across the year.

Fig.5 Comparative analysis of pesticide application costs of maize in 2017–2021.Season I, Ruili winter, Jinghong winter, Guigang spring, Chongzuo spring; Season II, Ruili summer, Jinghong summer, Guigang autumn, Chongzuo autumn.Data are mean±SE(Ruili, n=2,430; Jinghong, n=1,235; Guigang, n=1,150; Chongzuo, n=755).Different lowercase letters indicate significant differences in application costs per season between years in the same area (P<0.05).

As shown in Table 5, the seed cost per hectare of winter maize in Ruili ranged from 2,056.58 to 2,106.12 CNY, with no significant difference in seed cost per hectare between years (F=0.433,P>0.05).The seed cost per hectare of summer maize ranged from 2,096.62 to 2,103.53 CNY, with no significant difference in seed cost per hectare between years (F=0.007,P>0.05).Crop yields per hectare in Ruili ranged from 13,759.76 to 13,939.06 kg, with no significant differences in crop yield per hectare between years (F=0.577,P>0.05).Theresults of the multiple comparison analysis for seed cost per hectare and crop yield in the remaining three cities showed no significant differences between years, with all being consistent.Additionally, the regression analysis of the panel data was validated, with the results showing no significant effect fromS.frugiperdainvasion on either seed cost or crop yield.Compared to the maize yield losses reported in other countries in Africa and Asia that had also been affected byS.frugiperdainvasion, there were no serious reductions in maize yields after control measures in the study areas, so the actual losses of maize yields due toS.frugiperdadamage were less serious.

Table 5 Insecticide application and economic evaluation of maize production during 2017–2021

4.Discussion

Spodopterafrugiperdapossesses a highly competitive interspecific level; its ability to prey on other pests in similar ecological niches, as well as its predatory behaviour towards natural enemies such as syrphids,greatly enhances its interspecific competitiveness(Bentivenhaet al.2016, 2017; Songet al.2021).In addition, the characteristic life history ofS.frugiperdais another important reason for its dominance, as the larval stage ofS.frugiperdahides in the heart of the leaf (whorl)during the day and emerges at night to hunt for food,and it can infest maize throughout the growth period(Chapmanet al.1999; Songet al.2021).Laboratory and field trials conducted by our team in Yunnan have shown thatS.frugiperdacan outcompeteO.furnacalisat both individual and population levels (Songet al.2023).Our results also indicated thatS.frugiperdahas evolved into the dominant pest of maize fields in the annual breeding area.

The excessive use of chemical pesticides has resulted in an increased economic burden while threatening human health and the ecological environment (Lewiset al.2016).After theS.frugiperdainvasion, the frequency of chemical pesticide applications in maize fields increased significantly compared to previous years, and in some areas the application frequency has grown to several times higher than that before the invasion (Yanget al.2021b).The results of our survey in four cities in Yunnan and Guangxi showed that the frequency of pesticide application and fertilizer costs of farmers increased significantly after theS.frugiperdainvasion, seriously affecting farmers’ motivation for production.The increase in fertilizer application may also lead to an exacerbation of pest infestation as the severity of the pest invasion continues to increase (Altieri and Nicholls 2003).For example, aphids are more common in crops with inorganic fertilizers than those with organic fertilizers (Moraleset al.2001; Altieri and Nicholls 2003).Therefore, the adoption of green and efficientS.frugiperdacontrol methods is urgently needed.

Before the 1990s, organophosphates, carbamates,and pyrethroids were the most popular pesticides in the Americas; however, the abuse of these pesticides has led to high resistance inS.frugiperda(Diez-Rodrigues and Omoto 2001; Saldamandoet al.2011; Carvalhoet al.2013; Okumaet al.2018; Gutiérrez-Morenoet al.2019).After the 1990s, Bt maize was widely grown in the Americas forS.frugiperdacontrol because of its outstanding performance in controlling these pests(Huanget al.2011; Fatorettoet al.2017; Huang 2021).Recently,S.frugiperdahas evolved resistance to Bt maize in America (Storeret al.2010; Omotoet al.2016; Chandrasenaet al.2018), with some farmers controllingS.frugiperdapopulations by intercropping,protective natural enemies, and maintaining biodiversity in the farm environment (Altieri 1980; Wyckhuys and O’neil 2007a).The control ofS.frugiperdain Africa has primarily been based on traditional control methods,such as physical control (e.g., hand-picking of eggs and larvae or the direct pulling or destruction of affected maize), chemical control (use of chemical and biological pesticides), and agroecological measures (crop rotation,intercropping, and regular weeding of farmland) (Harrisonet al.2019).Furthermore, farmers in some regions of Ethiopia and Tanzania also use parasitic wasps to controlS.frugiperda(Birhanuet al.2018; Sisayet al.2018),and some countries in Southeast Asia have borrowed the experiences ofS.frugiperdacontrol in America and Africa while advocating for integreted pest management methods to controlS.frugiperda(Upadhyayet al.2018;Hruskaet al.2019; Lamsalet al.2020).However,chemical control remains the primary tool used here.For example, farmers in Nepal use methomyl, cyfluthrin, and methyl-parathion to controlS.frugiperda, although this leads to soil fertility degradation and reductions in natural enemy populations (Bhusal and Chapagain 2020; Kandel and Poudel 2020; Khatriet al.2020).Previous studies from the eastern Himalayan region have shown thatS.frugiperdalarvae are attacked by entomopathogens,such as SpfrNPV,Beauveria,Metarhizium, and spiders,while hymenopteran wasps and heteropteran predators are the indigenous natural enemies ofS.frugiperda(Singhet al.2023).

During the initial stages ofS.frugiperdainvasion in China, emergency prevention and control problems were primarily solved by chemical-based integrated control methods to prevent the food security problems that would result from severe damage to crops such as maize and wheat.Our survey results showed that chemicalcontrol remains the primary option for farmers in the survey area considering the local agricultural production conditions and generally limited educational level of farmers.Although the use of chemical pesticides forS.frugiperdacontrol has been effective in the short term and has prevented significant losses, over-reliance on a single chemical pesticide will lead to rapid increases in pest resistance along with the re-emergence of pest populations.Furthermore, reliance on integrated management based on pesticide application is not sustainable, so long-term sustainable pest crop management is essential.

Before theS.frugiperdainvasion, China had accumulated decades of practical experience in controlling lepidopterous migratory pests, such asH.armigera; therefore, the control ofS.frugiperdashould consider the experiences and lessons learned from those other pest control processes (Wu 2020).In India, the most commonly used insecticide againstS.frugiperdais emamectin benzoate and nearly two rounds of insecticides are used to manageS.frugiperda(Deshmukhet al.2021).TheS.frugiperdathat has invaded China is resistant to organophosphorus and pyrethroid pesticides.To postpone the development of resistance to highly effective pesticides such as emamectin benzoate, promoting the green prevention and control ofS.frugiperdais necessary.Farmers should be encouraged to apply bio-pesticides such asBacillusthuringiensisfor control,thereby reducing the use of chemical pesticides such as emamectin benzoate.Moreover, natural enemies such asTrichogrammapretiosum(Riley) andTelenomusremus(Nixon), which are indigenous to China, can also be used to prevent and control the spread ofS.frugiperda.Meanwhile, it is necessary to establish a monitoring and early warning system, which can use insect radar to monitor the sources of migrating insects abroad and the migration dynamics of the internal populations.Based on the results of migration monitoring, control resources can be deployed promptly to achieve source control in the annual occurrence regions, efficient interception in the migration transition regions, and precise control in the key prevention regions.In addition, Bt maize has shown high efficacy against lepidopteran pests such asS.frugiperda.Compared to non-Bt maize, Bt-Cry1Ab DBN9936 and Bt-Cry1Ab/Cry2Aj Ruifeng125 maize reduced lepidopteran pests by 61.9 to 97.3%,respectively, while avoiding yield losses by 16.4 to 21.3% and reducing mycotoxin contaminates by 85.5 to 95.5% without synthetic insecticides (Yanget al.2023).Bt-(Cry1Ab+Vip3Aa19) DBN3601T maize affected byS.frugiperdahad lower leaf damage scores and damage incidence than non-Bt maize, while having a significantly suppressive effect onS.frugiperdapopulations and damage in Yunnan (Zhaoet al.2023).Spodopterafrugiperdaadults primarily lay eggs on Bt-Cry1Ab C0030.3.5 and Bt-(Cry1Ab+Vip3Aa) DBN3601 maize compared to wheat, sorghum, grain, peanut, and soybean crops, so Bt maize can be used as a trap crop to protect other crops (Heet al.2021).Therefore, based on the experiences of managingS.frugiperdaresistance to Bt maize in the USA andH.armigeraresistance to Bt cotton in China (Li and Wu 2022; Tabashniket al.2017), implementing a sustainable management strategy is possible forS.frugiperdain the tropical south and subtropical regions with Bt maize as the core for achieving the source control ofS.frugiperda.

5.Conclusion

Our results showed that after theS.frugiperdainvasion,the occurrence of traditional maize pests such asO.furnacalis,S.litura,A.ypsilon, andR.maidistended to decrease, whileS.frugiperdareached 88.82–98.97%,becoming the dominant pest of maize.In response to this invasive species, farmers have significantly increased the frequency and dosage of chemical pesticide applications,with the frequency of applications being significantly higher in summer and autumn than in winter and spring, while the cost of applications increased by more than 35%.Furthermore, the variety of pesticides used has changed from chlorpyrifos, cyfluthrin, and acetamiprid to emamectin benzoate, which is highly effective againstS.frugiperda.In addition, farmers have also promoted the recovery of pest-infested maize by increasing fertilizer application,which has also increased production costs.The results of this study clarify the effects ofS.frugiperdainvasion on maize production in the tropical and southern subtropical regions of Southwest China, while providing a basis for adjusting the regional control strategies for maize pests.

Acknowledgements

This work was supported by the Lingnan Modern Agriculture Project, China (NT2021003) and the earmarked fund for China Agricultural Research System(CARS-02).

Declaration of competing interests

The authors declare that they have no conflict of interest.

Ethical statement

All applicable international, national and institutionalguidelines for the care and use of animals were followed.