Integration of GIS and spatial analysis with monitoring data of major pests for the purposes of area-wide management

To incorporate area-wide fruit fly monitoring data and its corresponding environmental variables, from various sources into a geographic information system (GIS), in order to manipulate, explore, analyse and spatially display the data. The rationale is investigate the spatial distribution of fruit flies on an area-wide basis, in terms of the main geographic factors driving their spatial distribution. This will lead to decision-making tools for area-wide fruit fly managers.

Using a geographic information system (GIS), spatial analysis and machine learning to analyse and explore fruit fly trap and associated geographic data on a local and area-wide scale.

• Ceratitis capitata (Medfly) and Ceratitis quilicii (Cape fly) share the same resources in a heterogeneous orchard environment.
• Fruit fly alternative hosts play a major role in the overall fruit fly populations in orchard
• Spatial analysis and machine learning proved to be very useful additional tools to analyse large fruit fly trapping datasets.
• Fruit fly population drivers are area-
• Rainfall seems to play a major role in the EGVV in determining the spatial distribution of fruit fly hot and cold spots, where hot spots occur in warm dry areas, while cold spots occur in cold and wet areas.
• Fruit fly hosts in urban areas contribute to early season fruit fly population pressures in commercial orchards.

This study provides the deciduous fruit industry of the Western Cape with a toolset to quantify the spatio-temporal distributions of C. capitata, on an area-wide scale, but also on orchard level. The information gained improves our understanding of how C. capitata spatio-temporal distributions are driven by stable geographic variables within heterogeneous fruit producing regions. This information will assist the deciduous fruit industry in improving its management of

1. capitata, by more precise application of fruit fly management actions. It is recommended, that this research be extended to include more precise and up to date pest monitoring data, and that an effort should be made to establish a real-time weather database for the Western Cape. Furthermore, real-time data on the cultivar phenology is also important. Although this type of data is easily obtainable on an orchard level, the challenge is to obtain this information on an area-wide scale, for entire production regions. Having access to real-time C. capitata trapping data, weather data, cultivar information and cultivar phenology, would enable the establishment of real-time seasonal spatio-temporal distribution models (Sciarretta & Trematerra 2014), not only for C. capitata, but many other pests, whose spatio-temporal distributions are mainly driven by these factors. Although there is currently a coordinated approach toward management of C. capitata in the Western Cape, data capturing and data handling are still aspects which can enjoy more resources, as most management decision-information is locked up in the data.