Elsevier

Biological Control

Volume 135, August 2019, Pages 89-94
Biological Control

Direct herbivory by biological control agents, and the consequent disruption of plant-nutrient-feedback cycles, combine to reduce the invasiveness of Melaleuca quinquenervia (Myrtaceae)

https://doi.org/10.1016/j.biocontrol.2019.05.009Get rights and content

Highlights

  • Herbivore exclusion reveals direct and indirect impacts of biological agents.

  • Melaleuca quinquenervia was directly impacted via biomass consumption.

  • Exotic trees were indirectly impacted via disruption of plant-nutrient feedbacks.

  • Melaleuca quinquenervia is now less invasive during the early stages of invasion.

  • Knowledge gained is vital to the development of effective management programs.

Abstract

Herein we present the results of a 5-year herbivore exclusion experiment in which the suppression of two successful biological control agents revealed a nutrient feedback cycle that promoted the growth and dominance of the exotic tree Melaleuca quinquenervia. When herbivory was restricted, plants produced greater quantities of aboveground biomass, which transferred a greater volume of litterfall and a larger load of nutrients to the soil surface. The resulting organic layer on the surface of the soil decomposed more quickly releasing nutrients and resulting in a larger pool of total phosphorus in the surficial soil layer. These mineralized-soil-nutrients were available for uptake by the exotic plant, which produced a greater quantity of surficial roots in the restricted-herbivory plots creating a positive nutrient-feedback cycle to biomass production. Our data suggest that, in addition to the direct suppression of exotic plants through biomass consumption, biological control agents may provide indirect suppression of weeds by preventing the establishment or altering the magnitude and/or direction of an invasion-promoting feedback cycle.

Introduction

Melaleuca quinquenervia, an evergreen tree native to the coastal wetlands of eastern Australia, successfully colonized most freshwater communities in its introduced range in Florida where historically direct control via insect herbivores was largely absent (Center et al., 2012). Pre-productive stages of this species produce high root densities at multiple depths, independent of resource availability or the presence of competing vegetation (Lopez-Zamora et al., 2004). These root systems supported fast-growing, dense populations of this tree that ultimately formed closed-canopy, monotypic forests with massive aerial seed banks in both low- and high-resource habitats (Serbesoff-King, 2003). Today direct control of M. quinquenervia has been re-established via the intentional introduction of a series of monophagous insect herbivores as part of an integrated pest management program that began in the mid-1990s. Several of these insects are now widely distributed throughout the species introduced range and have been shown to control populations of the exotic tree on a landscape-level (Tipping et al., 2008, Tipping et al., 2009).

Several studies documented how invasions of M. quinquenervia altered plant community structure and nutrient storage and cycling before the introduction of direct controls (Martin et al., 2009, Martin et al., 2010, Martin et al., 2011; and Tipping et al., 2008, Tipping et al., 2009). Little is known, however, about the interaction of direct and indirect forces that promoted the species’ dominance during invasion. In this study, we measured above- and below-ground nutrient transfer, storage, and availability in experiment plots where herbivory from the biological control agents had been suppressed for 5-years. These plots offered a unique opportunity to investigate the mechanisms that promoted populations of M. quinquenervia during the early stages of invasion before the introduction and subsequent suppression provided by the insect herbivores.

Section snippets

Experimental design

The study site was located in southwest Florida as described in Martin et al., 2009. This experiment was established in 2002 as described in Tipping et al., 2009. Briefly, 3-meter × 3-meter experimental plots were established and consisted of even-aged pre-productive M. quinquenervia saplings, half of which were sprayed with water (“unrestricted-herbivory plots”) and half of which were treated with monthly-foliar-applications of an insecticide (hereafter referred to as

Results

Mean values (±S.E.) of standing biomass, annual litterfall, organic layer, and root biomass in the unrestricted-herbivory plots (hereafter, in recording the statistical data, abbreviated as “UR”) and restricted-herbivory plots (abbreviated as “R”) are presented in Fig. 1. The g m−2 total standing biomass (9580 ± 2030 R and 1920 ± 373 UR, t ratio = −5.09, df = 24, P = <0.0001), annual total litterfall biomass (238 ± 11.6 R and 51.5 ± 12.0 UR, t ratio = −7.68, df = 13.2, P = <0.0001), M.

Discussion

Fast-growing, high resource adapted plants may disproportionately benefit from the lack of the direct regulation of herbivory when introduced into new habitats (Blumenthal, 2006). Melaleuca quinquenervia significantly altered ecosystem structure and function in habitats throughout South Florida and, prior to the introduction of the biological control agents, there existed a well-documented positive feedback cycle between fire and M. quinquenervia that promoted the spread of this species into

CRediT authorship contribution statement

Melissa R. Martin: Conceptualization, Methodology, Validation, Formal analysis, Investigation, Resources, Data curation, Visualization, Project administration, Funding acquisition. Philip W. Tipping: Conceptualization, Methodology, Validation, Formal analysis, Investigation, Resources, Data curation, Visualization, Project administration, Funding acquisition. Michelle C. Mack: Conceptualization, Validation, Supervision. K.R. Reddy: Conceptualization, Validation, Supervision, Funding acquisition.

Acknowledgements

We thank E. Pokorny, K. Nimmo, D. Fitzgerald, F.G. Wilson, Y. Wang, A. Bochnak, S. Daroub, and V. Nadal. This work was partially funded by National Science Foundation, USA grant number 0504422, the Florida Exotic Pest Plant Council, and the Everglades Foundation.

References (21)

There are more references available in the full text version of this article.

Cited by (2)

  • A global review of target impact and direct nontarget effects of classical weed biological control

    2020, Current Opinion in Insect Science
    Citation Excerpt :

    However, because some agents disperse rapidly, keeping control plots free from attack can be difficult [27]. Insecticides or fungicides have been used successfully to this end [8,28•,29], but they also alter the control plot ecosystem by eliminating nontarget insects and/or soil fungi [30] and may have direct effects on the target weed [31]. Exclosure studies and other set design experiments do establish cause and effect relationships and can provide valuable insights into mechanisms of successful weed management and interactions of herbivory with other biotic factors, but they are often limited in duration and spatial scale (one or a few sites) [e.g. 5••,8,32].

1

Present address: United States Department of Agriculture, Natural Resources Conservation Service, 1400 Independence Avenue SW, Washington, D.C. 20250, United States.

2

Present address: Northern Arizona University, Center for Ecosystem Science and Society, P.O. Box 5620, Flagstaff, Arizona 86011, United States.

View full text