Predicting phototaxis of almond moth, Cadra cautella (Walker) using ANN models: insights for wavelength and intensity as key factors.

Background: The almond moth, Cadra cautella (Walker), is a significant pest of stored products globally, causing severe damage and contamination. This insect was reported to have attraction towards light and this phenomenon can be exploited for its management. Our study examined the phototactic response of male and female C. cautella to seven different wavelengths (250, 330, 410, 470, 530, 580, and 680 nm) of light across three light intensities (60, 80, and 100 lx) using light emitting diodes (LEDs).

Results: Wavelength and intensity had a marked impact on attraction, with 410 nm at 60 lx eliciting the strongest response. Shorter wavelengths generated higher attraction rates (250 to 410 nm), while increased intensities typically dampened the response (>80 lx). An imbalance in the attraction of male-to-female ratio negatively influenced both sex ratio and fecundity. Correlation analyses indicated that wavelength significantly affected attraction, and F1 emergence was strongly and negatively correlated with sterility and number of eggs laid. These results emphasize the critical role of wavelength in regulating C. cautella behavior. Moreover, the artificial neural network (ANN) model (2-13-1 topology) effectively predicted insect attraction, with a low chi-square and root mean square error (RMSE), and correlation coefficients of 0.90013, 0.94986, and 0.94155 for training, validation, and testing, respectively.

Conclusion: Our study results provide valuable insights into designing eco-friendly light traps to manage C. cautella, reducing the reliance on chemical pesticides and promoting sustainable pest control. The ANN models demonstrated strong predictive capability, and enhancing light optimization for pest management. © 2025 Society of Chemical Industry.