The intensification of agriculture with intensive use of inputs, simplification of the landscape, and reduction of semi-natural habitats, is known to contribute to the decline of arthropods. However, arthropods provide essential ecosystem functions and services in agroecosystems and for agricultural activities such as natural regulation of pests and weeds, and pollination. Increasing plant diversity in cultivated fields through the implementation of agroecological practices and infrastructures is a promising approach to promote the presence of arthropods and their associated ecosystem services. The overall objective of the thesis project was to design and experiment an innovative agroecological practice in maize cropping systems, which is easy to insert into existing cropping systems, and which provides an undisturbed habitat for ground-dwelling arthropods and supplementary floral resources for pollinators. The idea of the proposed practice was to capitalize on an already well known practice, the implementation of winter cover crops, by keeping a strip of a winter cover crop in the middle of the field during the whole maize cultivation period. Therefore, two research objectives were defined: i) to measure the impact of the practice on activity-density and diversity of ground-dwelling arthropods, on their dispersion in the cultivated area (spillover), and on the potential pest regulation in the adjacent cultivated area, and ii) to measure the interest of the practice for the conservation of wild pollinators in intensively cultivated landscapes.
Field experiments took place on 12 commercial fields of volunteer farmers in 2019 and 2020, in order to take into account the technical and regulatory constraints of farmers in a conventional arable cropping system. Different species groups were surveyed: communities of ground-dwelling arthropods (carabids, spiders, staphylinids and harvestmen), slugs (the main maize pest in the study area), and pollinating insects (bees and hoverflies). Spatio-temporal dynamics of the different natural enemy groups were analysed, predation rates measured with sentinel prey, and the composition of carabid and wild bee communities investigated. Results show that the undestroyed cover crop strips constitute reservoirs of biodiversity, in particular for carabids, spiders, and bees. In the cropped area, no clear effect of the distance from the strip was detected for the different groups of ground-dwelling arthropods, thus no indication of a potential spillover of arthropods into the crop. However, two main carabid species (Poecilus cupreus and Pterostichus melanarius) were more abundant in the vicinity of the strip (10 meters), but not inside the strip, indicating a potential phenomenon of aggregation of these species towards the strip. Predation rates were higher in the strip and seemed to decrease with increasing distance from the strip into the cropped area. Moreover, carabid and bee communities showed to be different from one habitat to another, as well as the distribution of their ecological traits. Thus, the strips can provide complementary habitat and resources for natural enemies and pollinators. Finally, the spontaneous plants of field margins appeared to be essential for oligolectic and less common bees.
The results of this thesis show that the conservation of a cover crop strip in the middle of cropped fields can be an effective practice for the conservation of beneficial arthropods in agricultural landscapes, but can also enhance ecosystem services such as pest regulation. Additionally, the results highlight the importance of preserving or even extending perennial semi-natural habitats such as field margins to contribute to biodiversity conservation in arable cropping systems and landscapes.