We used a thulium-doped 50 W fiber laser with a wavelength of 2 µm with a collimated beam (Ø: 2 mm) manufactured by Futonics Laser GmbH, Katlenburg-Lindau, Germany. The laser was placed within a steel box (68 cm × 68 cm × 68 cm) with a door with a metal interlock ( Figure 1 ). On laser activation, the door locks automatically to avoid risk of laser exposure.

The target organisms were placed approximately 40 cm below the laser head and exposed to increasing dosages of laser energy (from 0 to 23.9 J mm^-2)

Tenebrio molitor were bought from the company InsektOrama A/S, Herning, Denmark. Adults and half of the delivered larvae were used for the first experiment and lasered two days after the delivery. The rest of the larvae were reared to produce an adequate number of new adults, larvae (to repeat the experiment) and pupae (for the first experiment and its repetition). When larvae were developed, they were moved to a round open plastic tray (10 cm high; Ø: 29 cm) and placed in a climate cabinet (at 20°C ± 2°C in darkness). The bottom of the tray was covered with rolled oats (3 cm high) necessary for the larvae’s rearing. Additional oats were added when necessary to ensure a continuous supply of feed. Four to five slices of fresh potato (Ø: 3−4 cm) were added to the tray every second day to ensure water supply. The tray was checked daily, and as soon as a pupa was observed, it was immediately separated from the larvae and placed in a new, similar tray. Adults were isolated from pupae in the same way to avoid cannibalism. Adults were producing eggs, creating a new generation of larvae required to repeat the experiment. All insect stages were kept and reared under the same conditions until they were lasered.

Ten individuals each of larvae, pupae, and adult T. molitor were exposed to the laser beam for 0, 1, 10, 20, 50, 100, and 500 ms corresponding to 0, 0.016, 0.16, 0.32, 0.80, 1.59, and 7.95 J mm^-2. Ten individuals of each developmental stage of Tenebrio molitor were treated with each dose with 4 replicates (10 × 4 individuals). The larvae were retained in a groove in a small piece of wood during the treatment. The pupae were placed on a piece of paper, and the adults were placed in a glass tube (length: 40 mm; Ø: 18 mm) to prevent them escaping the treatment ( Figure 1 ). The laser was aimed at the dorsal surface, approximately halfway along the length of the body, but because they were alive and able to move some were hit in other locations. Afterwards, the insects were carefully moved to four plastic transparent boxes with lids (12 cm × 12 cm × 4 cm) (10 individuals per box) immediately after lasering. The lids were partially (1/3 of the lid) covered by a net (holes: 1 mm × 1 mm) ensuring gas exchanges.

The bottom of the boxes was covered with approximately 1 cm of rolled oats and slices of fresh potato (Ø: 3−4 cm) was added as feed and water supply. Additional rolled oats and fresh potato was added when necessary to ensure a continuous supply of feed and water. The number of live and dead individuals were counted 8 and 15 days after the laser treatment. Damages and deformities were noted.

Adalia bipunctata, produced by EWH BioProduction ApS, Tappernøje, Denmark was bought via Horticoop Scandinavia A/S, Hinnerup, 143 Denmark. The beetles were kept in transparent plastic containers (28 cm × 20 cm × 22 cm) with lids, partially (2/3) covered by a net (holes: 1 mm × 1mm) allowing gas exchange, and kept in a climate cabinet at 21°C with 12 hours light from 8 a.m. to 8 p.m. The beetles were fed with honey diluted with water (ratio 1:10) during the experimental periods. The diluted honey was supplied via pieces of filter papers (3 cm × 3 cm). Four to five new papers were placed in the box with the beetles every day. The beetles were kept and reared in the same conditions until being lasered.

For each laser treatment, ten beetles were exposed to a laser dose. Each beetle was placed in a plastic tube (length: 50 mm; Ø: 5 mm) during the irradiation preventing the beetle from escaping. We aspired to hit the beetle in the center of their dorsal surface, but they were able to move a little and therefore some were hit in other places. The laser dosages were 0, 1, 10, 20, 50, 100, and 500 ms corresponding to 0, 0.016, 0.16, 0.32, 0.80, 1.59, and 7.95 J mm^-2. After exposure, each beetle was moved to a plastic container together with the nine other beetles receiving the same dose and kept exactly the same way as the pretreated adults. The mortality rate was recorded over 15 days. There were four replicates for each dose (total 4 ×10 beetles). Two independent experiments were done.

Statistical analyses were done using the statistical program R´ (R Core Team, 2021). All data sets were initially analyzed using a dose-response model. If no dose-response trend was present in the data, the lowest observed adverse effect level (LOAEL) and the no observed adverse effect level (NOAEL) were found as the lowest dose showing a significant effect compared to the control group and the highest dose with a non-significant effect compared to the control group, respectively.

In all earthworms experiments except one, the mortality rates of the worms living in clay, sandy, and organic soil exposed to laser heating were not significantly different from the controls ( Supplementary Table S1 ). Consequently, the NOAEL was the highest dose of 23.9 J mm^-2 and the LOAEL could not be estimated. For E. crypticus, the highest laser dose of 23.9 J mm^-2 was the only dose significantly higher than the control (p=0.0185), and accordingly the LOAEL, making 15.9 J mm^-2 the NOEAL ( Supplementary Table S1 ).

In general, the non-linear model fitted the data well ( Figure 8 ). Model parameters are shown in Supplementary Table S5 . During the 15 days, some of the non-exposed beetles died. The mortality increased over time and with increasing dosages. Even 0.016 J mm^-2 affected the shape of the dose-response curve, and the beetle’s elytron became brownish in color ( Figure 9 ). A dose of 0.80 J mm^-2 severely harmed the beetles, and almost all beetles died during the 15 days. A dose of 7.95 J mm^-2 immediately killed all beetles burning significant holes in the beetles ( Figure 9 ).

There were no differences in mortality between the control group and worms exposed to all laser doses, with a single exception with E. crypticus in sandy soil at a dose of 15.9 J mm^-2. The spread of the heat in the soil from the laser beam depends on the water content and soil structure and composition. We cannot exclude that other conditions like a higher water content or other soil types would result in other mortalities. We consider a soil water content of 50% of the water capacity to be realistic in the early spring when weed control usually is conducted, but it depends on many factors in the field (e.g., variation in soil composition, precipitation, and evaporation). The earthworms in the tubes were living in a very small soil volume (<15 g) during the experiment mimicking the uppermost part of the soil profile. Although heat corresponding to 23.9 J mm^-2 was executed on a spot with a 2 mm diameter, which easily kills weed seedling, the exposure did not warm up the soil sufficiently to affect the mortality of the worms living lose to the soil surface within the seven days.

Tenebrio molitor is a large model insect. Larger insects, such as T. molitor, appear to be more resistant to the laser than smaller insects such as A. bipunctata. In general, the insects were all killed immediately at a laser dose corresponding to what would be appropriate for killing small weed seedlings (50 J mm^-2). We aimed to focus the laser on the middle of the body of the insects, but because the insects were able to move a little, there would be variations in where the laser hit. If we had focused the laser on the head or the rear part, the mortality might have been different.

We used a collimated beam in the experiments to precisely give the wanted dose independent of the precise distance to the target. However, if stones or other reflecting materials are hit with a collimated laser beam, the reflected beam may escape the target area, and humans or larger animals (dogs, hares, etc.) may be exposed, burned, or blinded. Therefore, the laser beam should not be collimated in laser weeding robots, but only be focused and concentrated on the meristem of the weed seedlings. If the laser then hits a reflecting material, the beam will be spread in a cone and the risk of harming the humans, animals and other plants would be significantly reduced due to the lower dose per area. That means that only insects or other organisms placed exactly in the focus point would receive the dose determined for the target plant. The further away from the focus point the lower the dose an organism would receive and the less harmful the exposure.

Some insects benefit the crop like ladybugs, spiders, and predatory beetles, as they can reduce the number of harmful insects (e.g., aphids (Aphididae) and rape beetles (Meligethes aeneus)). Some beneficial insects like ladybugs have characteristic colors and can easily be identified with recognition tools (Rakhmatulin et al., 2021). In principle, a laser-weeding robot could be programmed to recognize the different between beneficial and pest arthropods and kill the latter.

On the other hand, small dosages of laser energy could be considered to control harmful arthropods. Rakhmatulin (2021) used machine vision and showed that a low-cost device could be used to kill mosquitoes with a laser. The company Photonic Sentry (https://photonicsentry.com/) has introduced a laser mosquito neutralization technology to combating malaria. Flying insects are common vectors for the transmission of pathogens between crop plants (Heck, 2018). Mullen et al. (2016) presented proof of principle for an optical system capable of highly specific vector control using a combination of optical sources, detectors, and sophisticated software to search, detect, and identify flying insects in real-time, with the capability of eradication using a lethal laser pulse. They focused on two insect species: Diaphorina citri, a vector of the causal agent of citrus greening disease, and Anopheles stephensi, a malaria vector. There seems to be a great potential to use laser technology to protect people and crops from pestilent flying insects (Keller et al., 2020). Our experiments showed that laser dosages, which will not harm the plants, could significantly damage harmful insects.

Laser weeding seems to be a promising tool to replace or supplement herbicides and mechanical weed control, with only a small treatment area in contrast to other weed control measures. Unlike herbicide spraying, laser weeding leaves no chemicals in the field that may harm non-target organisms after the treatment. Herbicides may evaporate or leach to surface and groundwater and may expose the environment to short or long-term unwanted side-effects. Laser weeding only leaves the ash from the burned weed meristem in the field after the treatment, which may be taken up by the crop plants as fertilizer.

In contrast to laser weeding, mechanical weeding impact shallow living worms negatively (and potentially other soil organisms), as the weeding implements are passing through the top layer of the soil (Doran and Zeiss, 2021). Mechanical weeding also harms beneficial organisms on the soil surface, like spiders and predatory beetles (Michalko et al., 2019; Symondson et al., 2022). Therefore, laser weeding seems to have less negative impact on the environment that other weed control measures.