Monitoring the Phenology of Chromolaena odorata to Inform Management of an Incipient and Highly Invasive Species in Hawaiʻi

Abstract

The invasion of nonnative species has negative impacts on ecological processes and ecosystem services, and these impacts are being exacerbated by global trade and climate change. In Hawaiʻi, invasive species, along with associated biodiversity loss and habitat degradation, are the greatest threat to the archipelago’s endemic biota. In 2011, Chromolaena odorata (Devil’s Weed), a globally dispersed invasive species, was first detected in the Kahuku Training Area (KTA) on the Island of O’ahu. Known as one of the world’s worst weeds, C. odorata is an aggressive colonizer of disturbed environments that, once established, creates dense monotypic stands that prevent the growth and regeneration of other species. Since its discovery in 2011, C. odorata has spread to occupy ~1,042 ha in KTA. The objective of this study was to develop a C. odorata phenology monitoring program to investigate the correlation between observed phenophases, seed germination, and climate variables to inform integrated weed management (IWM). To address this objective, I monitored the phenology (i.e., phenophases or life cycle events) and plant condition of C. odorata every two weeks in KTA for 12 months in five study sites and recorded monthly precipitation and temperature from the closest weather station. In addition, I collected soil samples in each study plot monthly and monitored seedling emergence in the greenhouse over 12 months. Overall, I found that flowering occurred between November – February and fruit set occurred between February – April, with smaller flowering and fruiting events in May – June and June – July, respectively. Monthly precipitation and temperature had strong explanatory power for both overall plant condition and productivity-related phenophases (i.e., flower production and seed drop). In addition, a positive correlation existed between seedling germination and the presence of flowers. Based on this information, chemical and mechanical control should be conducted between August and October to reduce large flowering events beginning in November. Overall, the results of this study will allow for the adjustment and optimization of IWM practices for this species based on phenophases that are more susceptible to weed control methods, as well as informing the use of phenology in controlling and managing invasive species more broadly.