The study does call that out.

The results from the analysis using the induced exposure study design, presented in Section 4, are generally intuitive. Vehicle colour effects on crash risk were greatest in daylight hours compared to twilight or dark times. This makes intuitive sense since use of vehicle lighting in twilight and at night reduces contrast differences between different vehicle colours. Use of daytime running lights or headlights during the day in Australia is not common hence visibility differences between vehicles would be left largely due to the colour of the vehicle.