Establishment of the wasp Tetramesa romana in northern California for biological control of arundo donax and the role of release plot manipulation

Authors: BITUME, ELLYN - Forest Service (FS); ROGERS, VALLE - University Of California Berkeley; Pratt, Paul; Goolsby, John; Moran, Patrick

Submitted to: Biological Control
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/7/2024
Publication Date: 3/14/2024
Citation: Bitume, E.V., Rogers, V.D., Pratt, P.D., Goolsby, J., Moran, P.J. 2024. Establishment of the wasp Tetramesa romana in northern California for biological control of arundo donax and the role of release plot manipulation. Biological Control. 192/105489. https://doi.org/10.1016/j.biocontrol.2024.105489.
DOI: https://doi.org/10.1016/j.biocontrol.2024.105489

Interpretive Summary: Arundo or giant reed, a giant non-native, invasive grass that grows as tall as 25 ft, is a major invader of shoreline habitats along arroyos, creeks, rivers and lakes in the southwestern U.S., including the Central Valley of northern California, where irrigated agriculture produces over $32 billion in crop value per year. Arundo consumes scarce water in this arid region, blocks access to water, impedes flood control, displaces native plants and animals and fuels wildfires. Scientists with the USDA-ARS in Albany, California released a black wasp (Tetramesa romana) that makes galls (tumors) in main shoot and side shoot tips of arundo. Female wasps lay eggs inside shoot tips, which then form galls. Larvae hatch from eggs and feed inside the gall and chew exit holes when they emerge as adults. These exit holes make it easy to determine if a field site has the wasp, and allow estimation of how abundant they are. The wasp was released at 11 sites in the Sacramento and San Joaquin River watersheds of northern California between 2017 and 2020. At six of the sites, small plots of arundo (about 7 x 7 ft) were either cut to the ground, mowed or 'topped' at about chest height, or left uncut in the spring, and wasps were released later in the summer onto regrowth shoots in the gorund-cut or topped plots and onto uncut shoots in those plots. One year later, in 2018, exit holes made by emerging wasps were up to 26-fold more abundant in plots that had been cut to ground, and 6-fold more abundant in topped plots, than in uncut plots, suggesting that pre-cutting of plots prior to release led to more fresh shoot tips being available for the wasp. In an additional plot manipulation study at two sites, some plots were 'double-cut'-cut to ground in the spring and then the regrowth topped in the summer-while others were 'single-cut'-cut to ground only. Exit holes of the wasp were up to 11-fold more abundant in the double-cut plots. After it initially (in 2018) appeared that the wasp was established at only two of nine sites started in 2017, by 2023 the scientists found the wasp to be present at 10 of the 11 total sites, based on evidence from dissections of shoots at seven sites, two-minute surveys of exit holes and immature galls taken at many points per site at all 11 sites, and yellow sticky trapping of adult wasps at these sites. The wasp T. romana is now widely established on arundo in northern California and holds promise for biological control of arundo. Manipulation of the release plots by plot cutting improved local establishment of the wasp.

Technical Abstract: Arundo donax is an invasive giant grass of riparian systems in the southwestern U.S.A., including the water resource-critical Sacramento and San Joaquin River watersheds of northern California. Between 2017 and 2020, a shoot tip-galling wasp, Tetramesa romana, was released at 11 sites in these two watersheds and in the Sacramento-San Joaquin Delta. Prior to release, shoots in plots were manipulated by cutting to ground, topping to 1 m height or leaving uncut. One year after 2017 release at nine sites, exit holes made by emerging adults were found at two sites. Exit hole density per main shoot length was 26-fold higher on regrowth shoots in ground-cut plots than in uncut plots, and 6-fold higher than in topped plots. Counts of exit holes and galls indicated establishment of T. romana at seven sites by 2021, including one in each of the watersheds and five sites in the Delta. An additional plot manipulation study at two sites found that exit hole density was 11-fold higher in plots that were double-cut (cut to ground and regrowth topped) than in single-cut plots (cut to ground only). By 2023, sampling and dissection of shoots, combined with counts of exit holes and galls and trapping of adult T. romana with sticky traps, indicated establishment of T. romana at 10 of 11 sites. This study demonstrates the use of multiple methods to verify agent establishment, density and dispersal, and illustrates the utility of physical manipulation of host plants to improve establishment outcomes.