1. Introduction
The fall armyworm,
Spodoptera frugiperda (J.E. Smith) (Lepidoptera: Noctuidae), once considered endemic to North and South America, has become an invasive pest in Africa [
1]. It was detected for the first time in Sao Tome and Principe, Nigeria, Benin, and Togo in 2016 [
2]. To date, the presence of
S. frugiperda has been reported in more than 30 sub-Saharan countries [
3,
4]. It has been documented to feed on 353 host plants belonging to 76 plant families, mainly Poaceae, Asteraceae, and Fabaceae [
5].
In the absence of proper management methods,
S. frugiperda has the potential to cause maize yield losses of 8.3 to 20.6 million metric tons per year in 12 of Africa’s maize producing countries, which represents a range of 21 to 53% of the annual production of maize [
4]. The value of these losses ranged from US
$ 2.48 billion and US
$ 6.19 billion [
4]. In Benin,
S. frugiperda attacks mainly maize crops [
6]. Surveys carried out in 2016 by the Ministry of Agriculture, Livestock, and Fisheries of Benin revealed that over 395,000 ha of maize were damaged, resulting in a loss of 415,000 tons, or 30% of national production [
6]. Therefore, it represents a threat to the country’s food security and economy.
The main management methods used in America against
S. frugiperda are synthetic pesticides and genetically modified crop varieties [
7]. Several studies have indicated that
S. frugiperda is resistant to several insecticides such as pyrethroids, organophosphorus, and carbamates [
8,
9]. In addition, recent studies have shown resistance of
S. frugiperda to several genetically modified varieties of maize such as MON89034, TC1507, and NK603 [
10,
11,
12,
13]. Therefore, alternative methods that reduce the application of synthetic pesticides and that use botanicals and natural enemies are recommended in Africa [
1,
4]. Information on farmers’ knowledge and management practices are essential for developing appropriate management methods suited to farmers’ need [
14,
15,
16]. Farmers develop knowledge and management practices and have their own ideas on how to solve a given problem in the practical and economical ways [
17]. One of the main barriers to implementing a pest management program has been shown to be the lack of information about farmers’ knowledge, perceptions, and management practices [
18].
In the literature, two complementary approaches to the development and extension of technologies are known: conventional and participatory approaches. Some critics to conventional approach viewed it as a linear process of practical application of scientific knowledge [
19], where farmers’ knowledge could be overlooked in the development of technologies [
20]. African farmers are well-known as innovators and experimenters [
21]. The participatory approach addresses the limits of the conventional approach by considering farmers’ knowledge and involving farmers in the process of development and extension of technologies [
22,
23].
In this study, knowledge refers to what farmers know about the biology and ecology of
S. frugiperda. Perception refers to how farmers perceive
S. frugiperda attacks, the damage caused by this pest, and the effectiveness of the management practices they use. The identification of pests and the quantification of their damage by farmers could be quite different and less accurate than that of a trained expert. Nevertheless, they provide crucial information, as farmers make decisions based on what they think is the problem [
24]. Management practices used by farmers are the cumulative result of their knowledge and perceptions of the pest and depend on access to pesticides. These knowledge and perceptions are often specific to each region [
25] and influenced by many socio-economic factors, for example membership in a farmer organization [
26,
27].
Farmers’ knowledge and management strategies for
S. frugiperda have been poorly documented since its appearance in Africa. Thus far, the only study specifically focusing on farmers’ knowledge of
S. frugiperda has been carried out by Kumela et al. [
28] in Kenya and Ethiopia. Their results revealed farmers’ knowledge on
S. frugiperda infestation, damage, and development stages. The management practices of farmers, such as the use of synthetic pesticides, plant extracts, handpicking of larvae, and application of soil to maize whorls were also reported by the same study. However, specific information regarding natural enemies, host, and repellent plants known by farmers has not been reported, though these are important for developing sustainable pest management methods. Additionally, the effectiveness of the different management practices adopted by farmers has not been addressed. The objective of the current study was to improve the understanding of the behaviour of maize farmers regarding the invasion of
S. frugiperda in Benin. A national survey was carried out to assess farmers’ knowledge of the pest, their perceptions, and management practices. We hypothesize that farmers could develop effective management strategies against
S. frugiperda based on their knowledge and perceptions. This study will be useful to define the actions required for the sustainable management of
S. frugiperda.
3. Results
3.1. Socio-Economic Profile of Farmers
The surveyed farmers were mostly men (83.2%), and their household had an average of nine people. The number of years of experience in maize production averaged 19.5 years. About 61.2% of farmers were illiterate. They practised agriculture as their main activity (92.6%). Livestock production (e.g., poultry, goat, sheep, cattle, or pigs), food processing, and trade were their secondary activities. About 35% of them belonged to a farmer organization and had contacts with research or extension services. Six-point five percent of farmers had received training in crop pest management (
Table 2).
3.2. Farmers’ Knowledge and Perceptions of S. frugiperda Attacks
Most farmers (91.8%) recognized the damage of
S. frugiperda on maize crop. The majority (78.9%) of them were able to identify the pest during its larval stage (
Table 3). Farmers (88.6%) observed the activities of
S. frugiperda in their maize fields (
Table 3). They recorded the first attacks of
S. frugiperda in 2015 and 2016. They perceived
S. frugiperda as a new pest. Currently, there is no name in local languages to specifically refer to
S. frugiperda.
For most farmers, maize plants were more vulnerable from the 1st to the 4th week after planting, and attacks were more severe during periods of light rain (
Table 3). The majority of farmers saw an increase in the spread of
S. frugiperda attacks in 2018 compared to the previous year. About 32% of the farmers believed that
S. frugiperda caused more damage compared to the other maize pests they encountered in their fields (
Figure 2). These include Formosan termites (
Coptotermes formosanus), cob borers (
Mussidia nigrivenella), maize leaf rollers (
Marasmia trapezalis), grasshopper (
Zonocerus variegatus), pink stalk borer (
Sesamia calamistis), and corn leaf aphid (
Rhopalosiphum maidis).
Chi-square tests indicate that there was a relationship between farmers’ knowledge and perceptions of
S. frugiperda and their socio-economic characteristics. Farmers’ knowledge of the pest was significantly associated with education level (
p = 0.003), main activity (
p = 0.006), membership in a farmer organization (
p = 0.024), and contacts with research or extension services (
p = 0.001). In addition, their perception of damage was significantly associated with membership in a farmer organization (
p = 0.001) and contacts with research or extension services (
p = 0.001) (
Table 4).
3.3. Cropping Practices
Two main cropping systems were used for maize production: single-cropping in rotation with other crops (cotton, cowpea, cassava, soybean, and groundnut) and intercropping with cassava, groundnut, cowpea, and sorghum. Overall, single-cropping was the most common cropping system (67.6%) (
Table 5). More than half of the farmers applied mineral fertilizers to the maize plants, on average 134.6 kg/ha of NPK (Nitrogen, Phosphorus, and Potassium) and 75.4 kg/ha of urea.
Traditional varieties refer to local varieties cultivated by farmers over generations while modern varieties refer to improved varieties developed by research centers. About 58.7% of the farmers grew traditional varieties of maize (
Table 5). Farmers who grew modern varieties of maize used more mineral fertilizers and intercropping. In addition, farmers who grew traditional varieties of maize used fewer mineral fertilizers and practised more single-cropping in rotation with other crops.
3.4. Damage Caused by S. frugiperda
The results showed that around 97.1% of farmers suffered from
S. frugiperda attacks in 2018. It confirms the presence of
S. frugiperda throughout Benin, despite the geo-climatic contrasts (
Table 1). About 93.9% of the maize fields were infested. The incidence of damaged maize plants per field was estimated to 58.9%, and the incidence of damaged maize ears was estimated at 50.4%. The estimated yield losses by farmers averaged 797.2 kg/ha of maize in 2018 (
Table 6). The farmers revealed that before the invasion of
S. frugiperda, they obtained an average maize yield of 1626 kg/ha. Thus, the yield losses caused by
S. frugiperda represented 49% of the average maize yield that farmers obtained before the invasion of
S. frugiperda.
The analysis of the severity of
S. frugiperda attacks was carried out by distinguishing the different parts of the attacked maize plants, in particular the leaves and ears. The majority of farmers reported that
S. frugiperda attacks were moderately severe on the leaves and ears (
Table 6). At the time of the attack, they found large elongated perforations on the leaves, whorl attack and damage ranging from 15 to 25% of maize grains.
Chi-square tests showed that there was a relationship between the severity of
S. frugiperda attack and the types of grown maize varieties. The severity of
S. frugiperda attack on the leaves and ears was significantly associated with types of grown maize varieties (
p = 0.001).
S. frugiperda attacks were more severe for modern varieties of maize. There was no significant relationship between the cropping systems, application of mineral fertilizers, and attack severity (
Table 7).
3.5. Knowledge of Natural Enemies, Host and Repellent Plants of S. frugiperda
About 16% of farmers stated knowing insects and birds that feed on S. frugiperda larvae. They identified the francolin (Francolinus bicalcaratus), the village weaver (Ploceus cucullatus), and the common wasp (Vespula vulgaris) as natural enemies of S. frugiperda. For 14% of farmers, S. frugiperda could destroy other crops besides maize. They said the pest could damage sorghum, cotton, cowpea, and tomato crops.
In addition, about 5% of farmers stated knowing plants that repel S. frugiperda. They identified yellow nutsedge (Cyperus esculentus), chan (Hyptis suaveolens), shea tree (Vitellaria paradoxa), neem (Azadirachta indica), tamarind (Tamarindus indica), and soybean (Glycine max) as repellent plants of S. frugiperda.
3.6. Farmers’ Management Practices
About 38% of the farmers surveyed used at least one control practice. The most common management method used by farmers was synthetic pesticides. Among farmers using at least one management method, 91.4% used synthetic pesticides, 1.9% used botanical pesticides, and 6.6% used other management practices. The wide range of synthetic pesticides used by farmers included Thalis 112 EC (emamectin benzoate and acetamiprid), Pyro FTE 472 (cypermethrin and chlorpyriphos-ethyl), Pacha 25 EC (lambda-cyhalothrin and acetamiprid), Lambda super 2.5 EC (lambda-cyhalothrin), and Emacot 019 EC (emamectin benzoate) (
Table 8).
The average effectiveness score for all of these synthetic pesticides is 5 out of 7. Thus, farmers believed that synthetic pesticides were relatively effective against
S. frugiperda. Among the synthetic pesticides, Emacot 019 EC is the one for which the average effectiveness score is 6 followed by Thalis 112 EC whose score is 5 (
Table 8).
The botanical pesticides used by farmers against S. frugiperda were usually made from neem leaves or seeds (Azadirachta indica), vernonia leaves (Vernonia amygdalina), pepper (Capsicum annuum), and ashes. Some farmers also added other raw materials such as soap, detergents, or petroleum to botanical pesticides. The average effectiveness score for all botanical pesticides is 6 and indicates that the botanical pesticides were effective against S. frugiperda.
Moreover, some farmers used other practices to manage S. frugiperda (7%). These included early planting, handpicking larvae, and the application of ash to the whorls of maize. The average effectiveness score for these practices is 3 and indicates that they were relatively ineffective against S. frugiperda.
Chi-square tests showed that there was a relationship between knowledge of the pest and the use of management practices (
p = 0.001) and between knowledge of the pest and the types of management practices used (
p = 0.002) (
Table 9). This result confirms that the implementation of management strategies by farmers was associated with their knowledge of the pest. The Chi-square tests also showed that the use of management practices by farmers and the type of management practices used were significantly associated with membership in a farmer organization and contact with research or extension services (
Table 9).