Non-native species invasions can disrupt pollination services and result in negative impacts on native plant reproduction and genetic diversity. Such impacts are particularly acute for oceanic islands, which are well known for high endemism and unique biological diversity, but also particularly susceptible to invasions. Non-native invasive predators (NIPs) consume animal pollinators and can reduce pollinator populations, possibly eliminating entire pollinator guilds. This research examined the impacts of NIPs on pollination and native plant reproduction in a tropical dryland ecosystem in Hawaii. The project team combined field observation, experimental manipulation, and laboratory analysis to examine interactions between eight focal native plants (endangered and common), insect pollinators (native and non-native), and NIPs.

The principal objective of the research was to determine the impacts of NIPs on insect pollinators and pollinator services in an invaded tropical dryland ecosystem in Hawaii. NIPs consume animal pollinators and, by doing so, reduce pollinator populations and possibly eliminate entire pollinator guilds. Loss of pollination services due to NIPs likely is an important, although poorly understood, factor in both native plant conservation and management for long-term sustainability of native island ecosystems. In this study, the project team (1) identified current insect pollinators and pollination effectiveness for focal native plant species, (2) examined diets of study site NIPs to identify direct interactions with insect pollinators, (3) applied common NIP control methods to experimentally determine their effectiveness at both reducing NIP populations and NIP impacts on pollinator-plant interactions, and (4) analyzed the data using Bayesian hierarchical models to identify the direct and indirect relationships between NIPs, insect pollinators, and native plants. This project was carried out at the U.S. Army Pōhakuloa Training Area (PTA) on Hawaii Island. Pollination is a critical ecosystem process and this study aligns with the Department of Defense’s (DoD) ecosystem-based approach to managing its training lands with the goal of ensuring that native species and their associated habitat are protected and restored.

In order to gain a more complete understanding of current pollination services in invaded landscapes, the project team combined field observation, experimental manipulation, and laboratory analysis to examine interactions between native plants, pollinators (native and non-native) and NIPs. The project team established three primary technical objectives for the project:

1. Identify current pollinators for focal native plant species and measure pollination effectiveness in an invaded tropical dryland ecosystem in Hawaii.
2. Examine diets of NIPs to determine interactions with pollinators.
3. Assess the impacts on pollination services with regard to common management actions used by DoD natural resource managers to control non-native species.

Results from the study have several management implications. First, current pollination differs for endangered versus common plant species in the study system. Pollinator management and conservation are likely important for managing endangered plant species. Second, pollinator interactions are localized in time and space, and can vary by time of year. It may be necessary to perform observations over repeated seasons and across multiple locations, in order to capture the spatial and temporal fluctuations. Third, NIPs can affect pollinator-plant interactions, and in most cases negatively. Controlling rats and ants likely would increase pollinator frequency, and potentially lead to more out-crossing for plant species. Finally, predator control treatments can be moderately effective at suppressing predator abundance or activity; however, the treatments require continuous effort. There is temporal and spatial variation in predator dynamics that must be taken into account when undertaking suppression.

The project team investigated the relationship between NIPs, insect pollinators, and native plants (both endangered and common species) at PTA. They identified the direct interaction between insect flower visitor species and focal native plant species through flower visitation observations (Technical Objective 1). They confirmed the direct interaction between the study site NIPs and insect flower visitors through diet analyses (Technical Objective 2). Through NIP suppression, they analyzed the indirect effect of NIPs on plant pollination and found significant negative relationships between NIPs and pollinators (Technical Objective 3).

PTA land managers, specifically the Natural Resources Office staff, were regularly updated on the research by way of field activity reports (approximately every 2 to 4 weeks). The project was included under External Stakeholders in the draft 2017 PTA Integrated Natural Resources Management Plan (INRMP), which states that “The findings will guide future land management decisions involving invasive predators at PTA.” As part of the technical transfer, the project team created a User’s Guide which includes options for identifying insect visitation to flowering plants and determining plant breeding systems; a brief guide to common insect pollinator taxa in dryland Hawaii; efficacy of predator suppression methods, cost analyses, and alternative suppression methods; comparison of diet analysis methods; propagation of our focal plant species in the greenhouse; and results from the research study. The project team provided a workshop based on the User’s Guide in August 2019 to PTA Natural Resources Office staff as well as to other land managers and interested parties.