Tephritid fruit flies below the radar: How sub-detectable populations persist and spread in California
Early, sub-detectable phases of population spread are a critical but often overlooked and little understood aspect of invasion biology. Nonnative tephritid fruit flies in California are a key example. Long-term spatiotemporal detection data collected by the California Department of Food & Agriculture shows that 17 nonnative tephritid species across four tropical genera have been found in California since the 1950s. Previous research examining this unique dataset demonstrated that many of these tephritid species, often aggressive agricultural pests outside of their native ranges, have established in the California and are spreading slowly at ultra-low densities.
In order to explore the deeper dynamics of these subdetectable populations, we developed a simple spatial model to overcome obstacles associated with analyzing spatiotemporally rare data. The model aims to reveal potential continuity between rare detections in space and time by buffering around detection points within a moving three-year window. First, we tested the efficacy of the model. We then asked how patterns of detection and recurrence change over time at both large (state-wide) and local (Los Angeles area) scales.
Our preliminary analysis indicates scale-dependent patterns of persistence and spread underlying detections of tephritid fruit flies. We initially focused on Bactrocera dorsalis, the Oriental fruit fly. At the state-wide scale we found Bactrocera dorsalis to be spreading slowly, consistent with results from previous research. At smaller scales, we compared model outputs from real data to that of simulated data, and results suggest that initial spread and density increases are followed by a period of stabilization. Though populations are highly patchy in space and time at small scales, they maintain strong regional persistence. Lastly, we found that this approach provided a much deeper insight to population dynamics than considering detections without buffering and for single years at a time.
We intend to expand this analysis to other tephritid fruit fly species for interspecies comparison. Further analysis will reveal whether these spatial patterns of detection are correlated with distribution of host plants and climatic variables. These findings have important implications for both invasion theory and management, especially for early phases nonnative species spread behavior. Further, this project demonstrates the importance of simple, transparent models that can be easily interpreted by scientists, policy makers, and the public.