Keeping Pest Populations Lower for Longer: Connecting farms and natural systems

Abstract

Native vegetation (NV) in agricultural landscapes provides a range of ecosystem services such as, erosion control, carbon storage, and habitat for wildlife beneficial arthropods, which contribute, to controlling insect pests. However, in Australia, evidence of impact is lacking, as well as knowledge of the mechanism for this pattern. This research builds on results from previous projects, which investigated the role of native vegetation for supporting beneficial insects and seeks to measure the impact of these landscape features on pest suppression and link the outcomes with vegetation management guidelines. The aim of the study was twofold, to understand the combined effects of semi-natural area and insecticide spraying on different predators, and to evaluate importance of these predators for biocontrol service provision in time and space in cotton production systems.

To achieve this we completed an analysis of an extensive data set provided by a commercial consultant, several field, and a glasshouse experiment. Results from these contributed to: - Identifying patterns in pest and beneficial populations, relationships with pest thresholds, and how these relate to landscape structure and - Determining whether pest populations stay lower for longer and have a slower rate of increase as a function of distance (e.g. near vs far) and landscape structure (e.g. amount and location of natural areas, diverse habitats, crop and non-crop at 1 km, 1.5 km and 2 km from cotton fields).

Collecting data at the landscape scale is expensive and time consuming. However, pest counts are regularly conducted by commercial consultants and agronomists and represent extensive and rarely available information. Obtaining and analysing a large data set from a commercial consultant was very valuable and allowed us to quantify the impact of semi-natural areas on biological control in relation to management activities. For example, results showed that the probability of mirids being present in cotton are likely to be less and remain so for longer in fields where there is 20% or more semi-natural habitat at a 3 km radius. Spraying reduced the probability of mirids in a crop. However, the likelihood of mirids increasing after spraying insecticide is greater where there is less native vegetation. There is also a relationship between the size of the cotton field and proportion of native vegetation and probability of mirids. For example, the time between sprays increases with increasing native vegetation in small fields. Similarly, the likelihood of mirids being present in cotton increases as field size increases but where there is also less native vegetation. Although the results support that, the presence of semi-natural areas could result in fewer sprays being required, a comprehensive cost benefit analysis was not included in this study.

Results from predation experiments showed that there are specific functional groups of predatory insects, depending on their spatio-temporal activity (diurnal vs. nocturnal and ground vs. canopy dwelling) that contribute to pest control in cotton crops. The results showed that there are not only differences in pest and beneficial numbers, and that their impact is not only related to landscape structure but is also related to management of insect pests, i.e the rate in increase of beneficial insects is reduced after the application of insecticides.

Pollination and biocontrol are ecosystem services provided by mobile arthropods and are generally affected by similar factors. Thus, they are likely to be simultaneously enhanced (by e.g. increasing semi-natural area surrounding crop fields) or reduced (e.g. insecticide application) by certain practices. Trade-offs are evident from the interactions between pollination and biocontrol effects on yield. However, results indicate that they cannot, and might not, need to be simultaneously delivered to enhance cotton yield, but that one service can

act as insurance when the other service is reduced. That is when biocontrol is low (i.e. pest numbers are high) pollination can contribute to yield and reduce boll loss caused by mirids and other pests. For cotton production systems to take full advantage of these services the findings underline the importance of considering and integrating multiple functions (pollination and biological control), within field management and surrounding landscape features simultaneously.

The results support the concept that native vegetation in conjunction with the judicious use of insecticides can keep insect pests lower for longer. Together with recommendations, they provide further evidence of: i) the importance of semi-natural vegetation in facilitating ecosystem services specifically how ii) the negative impact that insecticides have on these services and iii) the proportion and scale of semi-natural vegetation required to measure a significant benefit. Furthermore, while these results show areas of semi-natural vegetation have positive outcomes for insect pest management in cotton, it remains to be demonstrated whether managing existing native vegetation by reducing weeds, grazing and increasing vegetative diversity could increase the benefits reported here. The next step will be to incorporate these findings into insect pest management strategies including adjusting and testing pest thresholds and understanding if these results are transferable to other cotton pests. Doing this would give consultants and growers a broader range of options to maximise biocontrol and reduce insecticide applications to promote cotton production landscapes that are productive, profitable, and environmentally sustainable.