Cropping systems alter the biodiversity of ground- and soil-dwelling herbivorous and predatory arthropods in a desert agroecosystem: Implications for pest biocontrol

In irrigation-maintained desert agroecosystems in northwestern China, cereal crop production occurs mainly as maize (Zea mays L.) monoculture (MM), along with wheat (Triticum aestivum L) monoculture (WM) and maize/wheat intercropping (MWI). However, our understanding of the role of different cropping systems in regulating the abundance and diversity of ground- and soil-dwelling insect herbivores (pests) and arthropod predators (natural enemies of pests), and thus cascading effects on biological pest control services is still poor. To address this important issue, we measured ground- and soil-dwelling insect pests and associated arthropod natural enemies in fields of MM, WM, and MWI using pitfall traps and hand-sorting in spring and summer at six sites in a desert agroecosystem in northwestern China. Each site contained three cropping systems with cultivation for at least five years. We also calculated the abundance- and richness-based ratios of ground- and soil-dwelling arthropod predators to insect herbivores (P: H ratios) as indices for assessing pest biocontrol services. For ground-dwelling communities, we found that MM harbored more abundant and diverse insect herbivores and insect predators than the other systems, whereas MWI harbored more abundant and diverse non-insect predators and had higher P: H ratios compared with the other systems across seasons. For soil-dwelling communities, we found that MM harbored more abundant and diverse insect herbivores than MWI in spring, whereas WM harbored less abundant and diverse insect herbivores than the other systems in summer. However, MWI harbored more abundant and diverse insect predators than the other systems, whereas WM harbored less abundant and diverse non-insect predators than the other systems in either season. Moreover, MM harbored more abundant and diverse non-insect predators than MWI in summer. The P: H ratios were higher in MWI than in MM and WM in spring, but did not differ across cropping systems in summer. Our results demonstrate that the magnitude and direction of the effects of cropping systems are highly season specific and vary across groups of ground- and soild-welling herbivores and predators. Furthermore, intercropping delivers greater pest biocontrol services than monoculture systems. These findings provide key insights towards developing more effective intercropping systems for pest suppression.