Speaker
Description
Introduction
Weeds compete with crops for resources such as water, nutrients and light, causing an average yield reduction of 34 % globally (Oerke, 2006). However, weeds also have positive effects, such as providing food and shelter for arthropods (Selfors et al., 2018) and preventing soil erosion (Seitz et al., 2019). Weeds are not evenly distributed in the field, but form patches, so uniform weed management is often unnecessary (Castaldi et al., 2017). Mechanical weeding offers the possibility to reduce herbicides and is commonly used in organic farming. But it has drawbacks, such as damaging or killing crops by hoeing and increasing soil erosion and weed seed emergence after soil disturbance (Seitz et al., 2019). Site-specific weed management (SSWM) is the concept of adjusting management intensity according to the weed species and density distribution within a field. This technique can reduce the negative impacts of weed control. This work analyses approaches of site-specific mechanical weeding in maize in North-West-Germany.
Material and Methods
A field trial was conducted at the research station “Waldhof” of University of Applied Science Osnabrück, in the North-West-Germany, in 2021 and 2022. Two types of decision support were compared with uniform weeding. One is based on the weed cover (WC), and the other on the relative weed cover (RWC). The RWC is the ratio of weed cover to crop cover. For both treatments, three factor levels were tested: WC0.25; WC0.5; WC1.0; RWC0.1; RWC0.2; RWC0.4. Each treatment was repeated four times in randomised order. UAV-based multispectral cameras were used to distinguish between maize, weeds and soil. Different camera systems and algorithms were tested to optimise the workflow. Weed management was applied when the threshold of the treatment was exceeded. A uniform weeding was carried out with torsion harrow, followed by two site-specific hoeing applications. ANOVA with subsequent Tukey test (α = 0.05) was used to identify significant differences.
Results and Discussion
In 2021, higher maize yields and lower weed biomass were observed due to 106 mm more precipitation than in 2022. In both years, there was no difference in maize yield or weed biomass among the treatments. The thresholds of the first site-specific hoeing application were reached by all treatments. With the second site-specific hoeing, only 58 % of the area was hoed on average in 2021, while in 2022, 89 % of the area was hoed. These savings are similar to those reported by other studies (Castaldi et al., 2017; Niemeyer et al., 2024). In 2022, the smaller plants competed less with the weeds, resulting in higher weed growth in the early stage and more thresholds being exceeded. The non-conservative RWC0.4 treatment spared significantly less area, with -100 % and -66.6 % treated area in 2021 and 2022, respectively. For several of the treatments with low thresholds no significant differences to a uniform weeding were observed. The RWC treatments also considered maize growth. With better crop development more weeds could be tolerated and less area needed to be hoed. This study demonstrates the potential of SSWM and the RWC as weed control thresholds to enhance biodiversity and mitigate the negative effects of weeding.
References
- Castaldi, F., Pelosi, F., Pascucci, S., Casa, R., 2017. Assessing the potential of images from unmanned aerial vehicles (UAV) to support herbicide patch spraying in maize. Precis. Agric. 18, 76–94. https://doi.org/10.1007/s11119-016-9468-3
- Niemeyer, M., Renz, M., Pukrop, M., Hagemann, D., Zurheide, T., Di Marco, D., Höferlin, M., Stark, P., Rahe, F., Igelbrink, M., Jenz, M., Jarmer, T., Trautz, D., Stiene, S., Hertzberg, J., 2024. Cognitive Weeding: An Approach to Single-Plant Specific Weed Regulation. KI - Künstliche Intelligenz. https://doi.org/10.1007/S13218-023-00825-6
- Oerke, E.C., 2006. Crop losses to pests. J. Agric. Sci. 144, 31–43. https://doi.org/10.1017/S0021859605005708
- Seitz, S., Goebes, P., Puerta, V.L., Pereira, E.I.P., Wittwer, R., Six, J., van der Heijden, M.G.A., Scholten, T., 2019. Conservation tillage and organic farming reduce soil erosion. Agron. Sustain. Dev. 39. https://doi.org/10.1007/s13593-018-0545-z
- Selfors, L., Werts, P., Green, T., 2018. Looking beyond the jug: Non-chemical weed seedbank management. Crop. Soils 51, 28–53. https://doi.org/10.2134/cs2018.51.050
| Keywords | Biodiversity; Precision farming; Precision weed control; Sustainable weed management; Weed detection |
|---|
