This research was conducted from June 2021 to January 2022 at a the farmer field located in the lowlands of Pa′bundukang village, South Bontonompo sub-district, Gowa Regency, South Sulawesi Province, Indonesia, located at 5.234783°S and 119.223958°E. This area has a dominant clay soil type, a small part of dusty clay at 4 meters above sea level. When the study was conducted, this area had an average rainfall with an intensity of 318.56 mm (Central Bureau of Statistics, 2022), average temperature 27.37\u00B0C, duration of sunlight: 6.13 h/day, and humidity 82.69% (Indonesian Agency for Meteorological, Climatological and Geophysics, 2022).

The experimental design used a Randomized Block Design Split Plot Model with 2 factors. The first factor in the main plot is the planting model, which consists of (1) a tile planting model (70 cm × 40 cm) and (2) a legowo planting model (2:1) (50 cm × 35 cm) × 90 cm. The second factor is the type of variety as a sub-plot consisting of 7 varieties: (1) Srikandi Kuning, (2) Pulut Uri, (3) Provit A1, (4) Anoman, (5) Srikandi Ungu, (6) Bisma, and (7) Sukmaraga. The number of treatment combinations is 14, each repeated 3 times. Treatment testing (1) Tile planting model + yellow Srikandi variety, (2) Tile planting model + Pulut Uri variety, (3) Tile planting model + Provit A1 variety, (4) Tile planting model + Anoman variety, (5) Tile planting model + Purple Srikandi variety, (6) Tile planting model + Bisma variety, (7) Tile planting model + Sukmaraga variety, (8) 2:1 legowo planting model + Yellow Srikandi variety, (9) 2:1 legowo planting model + Pulut Uri variety, (10) 2:1 legowo planting model + Provit A1 variety, (11) 2:1 legowo planting model + Anoman variety, (12) 2:1 legowo planting model + Srikandi Ungu variety, (13) 2:1 legowo planting model + Bisma variety, and (14) 2:1 legowo planting model + varieties Sukmaraga. Each treatment was planted on a plot measuring 3 m × 5 m with 2 seeds per planting hole. Fertilization was carried out using NPK compound fertilizer (15:15:15) with a dose of 300 kg.ha⁻¹. Fertilizer application was carried out on plants aged 15 and 30 days after planting using a ditch with a distance of 5 cm from the plant growth hole.

The parameters observed included plant height, leaf width, level of leaf damage, level of S. frugiperda population larvae, O. smaragdina weaver ant predator population, cob height, 50% flowering age, and seed yield. Observations of the parameters of the level of cob damage and seed yield were observed at the age of 105–115 days after planting. Observations of these agronomy parameters were carried out conventionally by following the guidelines for observing the morphology of corn plants (IBPGR, 2013). Observations of the S. frugiperda population larvae and the O. smaragdina population were calculated based on the average population of insect species on each plant in the same treatment plot. Observations of the level of leaf damage were calculated using the following formula (Fattah et al., 2020). I = ∑ i = 0 z ( n 1 × v 1 ) Z × N × 100

I: Intensity of damage

n_i: the number of leaf with a v_i scale

N: Number of leaves observed.

Z: The higher v_i.

Scale value v_i:

0: no damage on leaves

1: leaf damage >0–20%

3: leaf damage >20–40%

5: leaf damage >40–60%

7: leaf damage >60–80%

9: leaf damage >80–100%

The research data were analyzed using Ms. Excell, Minitab V. 20, and R Studio V. 4.2.2 applications. All observation data related to agronomic performance, S. frugiperda population, O. smaradigna population, percentage of leaf damage, and percentage of cob damage were analyzed using factorial analysis of variance and LSD test at 0.05 probability. Correlation analysis using the pairwise Pearson method and biplot analysis of genotype ranking patterns based on 4 main data groups, namely S. frugiperda population, percentage of leaf damage, rate of cob damage, and seed yield.

The Srikandi Kuning variety has physical characteristics of weight of 1,000 seeds around 275 g, yellow seed color, semi-pearl seed shape, harvest age of 105–110 days, weight of 1,000 seeds around 275 g and resistance to corn stem borer (Table 1). The weight of 1,000 seeds is lower than the results of research by Sirappa and Nurdin (2010), the Srikandi Kuning variety has a weight of 1,000 seeds around 285.53 g and the number of rows per cob around 14.40 rows. While Pulut Uri has the characteristics of white seed color, the number of rows per cob is 14–16 rows, the weight of 1,000 seeds is 356 g, the harvest age is around 85 days, and is somewhat resistant to downy mildew disease (Jamidi et al., 2022).

Bisma variety has a harvest age of 96 days, the number of rows per cob is 14.13, the color of the seeds is yellow, the weight of 1,000 seeds is 307 g, and it is somewhat resistant to lodging. The Sukmaraga variety has a harvest age of 105–110 days, the number of rows per cob is 15.33, the color of the seeds is dark yellow, the weight of 1,000 seeds is 270, and it is adaptive to acidic soil. According to Etika et al. (2016) the weight of 1,000 seeds of the Srikandi Kuning variety is 200.89 g, Bisma is 170.11 g, and the number of rows per cob of Srikandi Kuning is 14.33 rows and Bisma is 11.78 rows.

The results showed a significant effect of planting model treatment on yields (6.63, p ≤ 0.01) but did not significantly affect the agronomic performance of corn plants. Separately, differences in variety types significantly affected agronomic performance such as plant height (168.78–219.30 cm, p ≤ 0.001), leaf area (6.60–7.85 cm, p ≤ 0.01), flowering age (45.33–49.00 dap, p ≤ 0.01), and cob height (83.42–119.32 cm, p ≤ 0.01) (Table 1). Factors that affect plant performance and yield are variety and its interaction with the growing environment. According to Uberti et al. (2023) state that some varieties can only grow well and produce high yields in a supportive environment. Morphological traits are essential in achieving optimum seed yield (Xu et al., 2024). Furthermore, Dias et al. (2019) stated that differences significantly influence variations in agronomic performance of corn plants in varieties and distance between plants.

In plant height plant performance, the highest value was in Sukmaraga (219.30 cm), and the lowest was in Srikandi Ungu (168.78 cm). The most expansive leaf area was Srikandi Kuning (7.85 cm), and the lowest was Srikandi Ungu (6.60 cm). The highest flowering age was Bisma (49 daps), and the lowest was Pulut Uri (45.33 daps). The highest cob position was in Bisma (124.08 cm), and the lowest was Srikandi Ungu (83.42 cm). The legowo 2:1 planting pattern gave a higher yield (7.89 t ha⁻¹) compared to the farmer’s planting model (5.38 t ha⁻¹) (Table 1). Plant and cob height and cob ratio affect changes in plant seed yield (Liu et al., 2022). A plant height of around 240–300 cm can optimize harvest yields and avoid the risk of lodging (Qin et al., 2016). The ability of the quantity and quality of plant variety production results from plant breeding, including the selection of plant characters with smaller leaf angles and large leaf orientation values (Liu et al., 2022; Zhao et al., 2022).

Plant arrangement with a 2:1 legowo planting pattern can avoid the effects of the canopy structure of corn varieties with wide leaf angles on reducing sunlight intensity at the bottom of the canopy. The vast planting distance between every two rows of plants (90–100 cm) is expected to increase light penetration to the base of the plant stem in each row. This study showed a significant increase in production using the 2:1 legowo planting pattern of 46% from the tile planting pattern (Table 2). An increase in plant population of 57.8% has the potential to increase seed yields by 11.2% (Shao et al., 2024). Using the right varieties with the right planting distance innovation is a practical strategy to increase corn production (Assefa et al., 2018; Liu et al., 2022). The upright plant type with a smaller leaf angle can potentially increase plant density, leaf orientation values, and optimum production results (Liu et al., 2021; Yang et al., 2021). This character increases radiation intensity in the lower canopy and increases light interception (Perez et al., 2019). The results of our study showed a significant effect of higher seed yields (7.89 t ha⁻¹) in the 2:1 legowo planting pattern compared to the Tegel planting pattern (5.38 t ha⁻¹) (Table 2). This result is influenced by the increase in population and the suppression of pest attacks on leaves and cobs. Suppression of the main pests of corn plants, such as S. frugiperda, significantly affects the final results of cultivation. The higher the suppression of the S. frugiperda pest population, the greater the potential yield can be obtained (De Groote et al., 2020).

The treatment of the planting model, variety, and combination of both significantly influenced the intensity of corn cob damage (p ≤ 0.01). The variety treatment resulted in significant differences in the population level of S. frugiperda larvae (0.43–1.23, p ≤ 0.0001), the population of red ants O. smaragdina (1.03–2.33, p ≤ 0.0001), the intensity of corn leaf damage (12.32–25.23, p ≤ 0.0001) and the intensity of cob damage (8.80–16.17, p ≤ 0.0001) (Table 3). S. frugiperda is a migratory and colonial pest, causing severe damage to maize, rice, and other food crops (Nayyar et al., 2021; Rane et al., 2022; Yainna et al., 2022). S. frugiperda infestation and incidence and damage to plants are predominantly influenced by factors such as maize varieties (Santos et al., 2022), planting time (Mbaidiro et al., 2023; Tabu et al., 2024), and pesticide use (Briones Ochoa et al., 2023). Some maize varieties that are categorized as resistant to S. frugiperda show antixenosis and antibiosis resistance mechanisms (Pamidi et al., 2024). The level of leaf damage due to S. frugiperda attacks on plants is determined by the location and age of the plants. According to Liu et al. (2024) the young plants are preferred by S. frugiperda over old plants, and lowlands have higher levels of damage than highlands. Several countries in South Africa, such as Ethiopia, Kenya, and Zimbabwe, have higher levels of corn leaf damage due to S. frugiperda attacks found in lowlands (Kumela et al., 2018). Land with intensive weeding has lower S. frugiperda attacks than land with many weeds (Mutyambai et al., 2022).

Table 3 shows that the intensity of corn cob damage is lower in the 2:1 legowo treatment (11.58%) compared to the farmer’s method (13.09%). The combination of pest control and specific planting distance management methods significantly reduces overall plant damage, especially cob damage, and can increase yields (Kusi et al., 2024).

The lowest S. frugiperda population was found in the Bisma variety (0.53 heads) and the highest in Sukmaraga (1.23 heads). The intensity of corn leaf damage due to S. frugiperda attacks was lowest in Bisma (12.32%) and the highest in Sukmaraga (25.23%). The intensity of cob damage was lowest in Bisma (8.80%) and the highest in Sukmaraga (15.03%) (Table 3). Several corn varieties can recover from S. frugiperda pest damage at certain levels of damage with environmental support, such as sufficient rainfall and appropriate soil moisture levels (Abang et al., 2024).

This study showed a significant effect of the 2:1 legowo planting distance treatment with the farmer’s planting method on reducing the percentage of cob damage (Table 3). The effects of plant damage due to S. frugiperda pests can be reduced by arranging plant spacing (Yeboah et al., 2021). In addition to arranging plant spacing, arranging plants using the intercropping method with several plants from the Fabaceae family is also known to reduce the incidence and severity of corn leaf damage due to S. frugiperda (Wu et al., 2022; Soujanya et al., 2024), but not with plants from the Cucurbitaceae family (Baudron et al., 2019). Corn plant damage due to S. frugiperda can also be suppressed by utilizing the role of natural enemies (Afandhi et al., 2022; Keerthi et al., 2023).

The highest population of red ant predators, O. smaragdina, was found in the Bisma variety (2.33 individuals) and the lowest in Sukmaraga (1.03 individuals) (Table 3). The Formicidae family of the Hymenoptera order is one of the natural enemies commonly found in corn fields (Perfecto, 1991). Most recently, there were 32 genera consisting of 47 species of ants that act as predators (Lutinski et al., 2024). Ecosystem management, including plant spacing and external input management, can affect the presence and population of weaver ants in cornfield ecosystems (del-Val et al., 2021). According to Sholahuddin et al. (2023) the use of the legowo row planting system in rice cultivation can increase the population of natural enemies such as the tomcat beetle Paederus sp. and the spider Tetragnatha sp.

S. frugiperda larvae are generally light brown in instar (Figure 1b) and attack plants on the top leaves in the vegetative phase (1c). According to Sumaryati et al. (2023) the S. frugiperda instar III larvae are light brown. S. frugiperda larvae attack corn plants from the vegetative to generative phases and attack in the vegetative phase that begins at the top of the plant (Figure 1c). According to Adi Pratama et al. (2020) the S. frugiperda larvae attack corn plants from the vegetative to generative phases and can cause leaf damage of around 60%. Furthermore, it is said that S. frugiperda larvae attack corn plants on the tops and appear to have holes, and usually, larval feces are present on the leaves.

The weaver ant O. smaragdina is one of the most effective predators in preying on various pests, including S. prugiperda pests on corn plants. These weaver ants have live on perennial plants such as soursop plants (Figure 2a) and rambutan plants (Figure 2b). According to Masram et al. (2023) the weaver ants O. smaragdina, from young insects or nymphs to adults, are reddish black or reddish brown in color and make their nests from fresh leaves at the top O. smaragdina ants are often found on plants that have broad and dense canopies such as plants from the Annonaceae group. Weaver ants O. smaragdina make nests by rolling up the leaves of the plants they use as their homes (Figure 2b).

Weaver ants search for food on annual plants such as corn around their habitat (Figure 2c). According to Ratri et al. (2017) the weaver ants O. smaragdina are insects that are very active in searching for food on various types of plants around their habitat and are pretty effective as predators or pest controllers because they can prey on numerous types of larvae and nymphs of green ladybug pests, leaf-eating caterpillars, and fruit-eating insects.

Based on the field observations, the level of cob damage indicates an interaction between the planting model treatment and the type of variety. Some varieties, such as Bisma, Anoman, Srikandi Kuning, Pulut Uri, Sukmaraga, and Srikandi Ungu, showed a greater level of cob damage in the use of the tile planting pattern used by farmers compared to the use of the 2:1 legowo planting pattern (Figure 3). In contrast, only the Provit A variety showed more significant cob damage in the 2:1 legowo planting pattern (Figure 3). This indicates that of the varieties tested, only the Provit-A1 variety is not recommended for use in the 2:1 legowo planting model. These results align with other observation variables, including leaf damage and increased seed production in the 2:1 legowo planting pattern (Table 2). The interaction between the treatment of planting models and varieties is thought to occur due to differences in the phenotype of the varieties tested, including morphological characters, response to the environment, and tolerance to extreme environmental conditions. Germplasm resources with resistant varieties greatly support the development of corn varieties resistant to S. frugiperda have been widely developed (Chiriboga Morales et al., 2021; Yao et al., 2023). In this study, the Bisma variety is a good variety for optimizing production that is correlated with suppressing leaf and cob damage due to S. frugiperda attacks due to the presence of resistant genes. The resistance properties of this variety have a mechanism for accumulating 33-kD tyrosine protease when responding to S. frugiperda larvae feeding (Pechan et al., 2000). In addition, the development of resistant varieties also includes resistance genes found in transgenic Bt corn. In particular, there have been many research results on unique corn resources, such as commercially grown transgenic fresh corn (Crubelati-Mulati et al., 2014; de Oliveira et al., 2018). Several insect-resistant variety resources have been extensively studied, but insect-resistant genes and insect-resistant proteins have not been comprehensively analyzed.

Correlation analysis showed that the population variable of weaver ants (O. smaragdina) was negatively correlated with the population of S. frugiperda (r = −0.76, p = 0.01), the scale of leaf damage (r = −0.76, p = 0.01) and the scale of corn cob damage (r = −0.58, p = 0.05). The observation variable of cob position was negatively correlated with the scale of corn cob damage (r = −0.72, p = 0.01) and positively correlated with plant height (r = 0.92, p = 0.001). The plant height variable was negatively correlated with the scale of corn cob damage (r = −0.56, p = 0.05). The scale of leaf damage was positively correlated with the scale of cob damage (r = 0.75, p = 0.01) (Figure 4).

The relationship between the population of weaver ants O. smaragdina and the population of pests S. frugiperda and corn leaf damage is negatively correlated, as shown in Figure 4 with r = −0.76, p = 0.01 (population of S. frugiperda) and r = −0.76, p = 0.01 (damage to corn leaves). This is in accordance with Karmawati and Wikardi (2020) when the population of weaver ants O. smaragdina is 1 per plant, the level of damage to cashew shoots is 14%, while when the population of weaver ants O. smaragdina is 2 per plant, the level of damage to cashew shoots becomes 9%. Weaver ants O. smaragdina are insects that like food that contains more protein than carbohydrates. According to Rezki et al. (2023) the weaver ants, in their food selection behavior, prefer beetle larvae (60.21%) compared to sugar feed which only reaches 32.45%. Furthermore, of the three types of feed for O. smaragdina weaver ants, tested chicken, sugar, and bread, the most commonly chosen by weaver ants as their food is chicken (Hasan et al., 2021). This proves that O. smaragdina weaver ants prefer food with a lot of protein over food with carbohydrates.

Biplot analysis aims to conduct selection based on the suitability of the seven varieties used with the planting method treatment used. The most ideal variety based on the observed parameters is the one that is located closer to the coordinates of the small circle with an arrow pointing to it. The results of the analysis show that to achieve optimal harvest results, the Bisma variety (G6) is more ideal when planted with a 2:1 legowo planting pattern, and Anoman (G4) is more ideal when planted with a tile planting pattern (Figure 5D). If both varieties are planted with the proper method (Bisma in a 2:1 legowo pattern and Anoman in a tile pattern). There is a chance for both varieties to experience a lower FAW pest population density and a lower percentage of leaf and cob damage (Figures 5A–C).

The S. frugiperda pest population positively correlates with the intensity of leaf and cob attacks. The higher the S. frugiperda population, the greater the chance of increasing attack intensity. According to Apriani et al. (2021) the a population of S. frugiperda of 4.47 per plant can cause damage to around 64.97%. The same thing was expressed by Dian and Wanta (2022) the lower the population of S. frugiperda, the lower the intensity of attacks; in this case, when the population of S. frugiperda was 12.40, the intensity of attacks caused reached 18.85% and when the population was 4.20, the intensity of attacks on the leaves became 9.88%. The population of S. frugiperda is negatively correlated with seed yields in corn. In other words, the higher the population of S. frugiperda, the lower the seed yield. According to Megasari and Khoiri (2021) that if the population of S. frugiperda reaches 0.2 per plant, it can result in a 5% decrease in seed yield, and at a population of 0.8 per plant, it can result in a 20% decrease in seed yield.

The coordinate points of the varieties in the middle (moderate) of the yield plot are G1 (Srikandi Kuning) and G5 (Srikandi Ungu) (Figure 5D). The coordinates of these two varieties are far outside the centroid of the S. frugiperda population plot (Figure 5A), and very close to the centroid of the percent damage to leaves (Figure 5B) and cobs (Figure 5C). These varieties can still be planted with the Legowo 2:1 (G5) and tegel (G1) planting patterns. Still, it is necessary to control S. frugiperda pests from the start because they can cause a high percentage of damage to these two varieties, even in low populations. This is indicated by the high rate of leaf and cob damage (Figures 5C,D) even though the S. frugiperda population is not significant (Figure 5A).