Study of visual responses of insect to artificial light at night

Most studies on the visual responses of insects rely on anthropocentric units of measurement. However, insects possess a visual system whose responses differ considerably from those of humans, potentially introducing biases in measurements and data analysis. The aim of this study is to develop a method to better understand insect visual responses to light stimulation. Several models were compared: correlated colour temperature (CCT), the blue percentage, and the percentage of each spectral band. These models used spectral data for LED lamps from various databases and insect visual responses came from published spectral sensitivity curves and electrophysiological measurements. The similarity coefficient, a scalar product between insect visual sensitivity and lamp spectral power distributions was calculated to quantify their spectral match.

The predictive performance of these models was assessed using 10-fold cross-validation. Statistical analysis showed that the model based on complete spectral distribution provided the best predictive performance (R² = 0.62), outperforming the models based on CCT (R² = 0.07) and blue percentage (R² = 0.08). These results indicate that complete spectral composition more accurately predicts insect visual responses than simplified descriptors.

A field experiment using UV-free LED light traps with different CCT (3500, 4500, 5500 and 6500 K) values assessed the importance of this component for insect vision. The results showed that CCT alone does not explain insect attraction, as only minor differences were observed among traps. Although these traps require further improvement, the spectrum of the lamp, does not appear to induce a behavioural response in insects, regardless of CCT.

This study represents a first step towards developing spatial planning strategies to preserve biodiversity. It may contribute to the design of lighting systems and management practices that reduce impacts on nocturnal insects while maintaining the safety of people in urban and peri-urban environments at night.