Skip to main content
ARS Home » Pacific West Area » Albany, California » Western Regional Research Center » Invasive Species and Pollinator Health » Research » Research Project #429088

Research Project: Management of Invasive Weeds in Rangeland, Forest and Riparian Ecosystems in the Far Western U.S. Using Biological Control

Location: Invasive Species and Pollinator Health

2020 Annual Report

1) Determine the host specificity, life cycle, and efficacy of new candidate biological control agents for invasive weeds of western rangeland, forest and riparian ecosystems, such as yellow starthistle, Russian thistle, Cape-ivy, and French broom. Subobjective 1.1: Determine feasibility of biological control of ice plant and other invasive weeds in the far western U.S. Subobjective 1.2: Determine host specificity, life cycle, and efficacy of new biological control agents of yellow starthistle, Russian thistle, French broom, and Cape-ivy. 2) Release and evaluate new biological control agents of invasive weeds in western rangeland, forest and riparian ecosystems, and evaluate previously released and adventive agents in the context of variation in weed genotype, climatic influences, and land management regimes, including the use of other control methods. Subobjective 2.1: Determine effect of plant genotype on efficacy of extant biocontrol agents of yellow starthistle. Subobjective 2.2: Determine distribution and impact of adventive or recently released insects on Dalmatian toadflax. Subobjective 2.3: Release and evaluate new biological control agents targeting arundo and Cape-ivy.

We will determine the current status of biological control of ice plant by surveying field sites for extant herbivores, including two soft scale species that feed on the leaves, and several parasitic wasps that were introduced to control these scales, over 40 years ago when ice plant was valued as an ornamental. We will determine the feasibility of biocontrol of ice plant and other candidate invasive weeds using new agents through surveys of land managers and other stakeholders, and by scoring weeds according to invasiveness, damage caused, and the likelihood of finding host-specific and efficacious biological control agents in their native ranges. These studies will take phylogenies of the weeds and related native plants into account to determine the feasibility of avoiding nontarget plant damage. We will determine the host ranges of new candidate biological control agents of yellow starthistle, Russian thistle, and French broom through overseas collection by collaborators and no-choice and choice tests in our quarantine laboratory. These studies will also evaluate the biology and impact of candidate agents on targeted weeds. We will determine the ability of the Cape-ivy moth to reproduce and feed on closely-related nontarget plants. Information from host range testing and other studies on new candidate agents will be used to submit applications to the USDA for permits for field release. We will determine the ability of previously-released biological control agents of yellow starthistle, including a seedhead-feeding weevil and a seedhead-galling fly, to damage, survive and reproduce on invasive western U.S. genotypes of yellow starthistle in relation to genotypes from the Greek native range where the agents were originally collected, and from western Mediterranean Europe, where yellow starthistle in the western U.S. originated. These studies will be conducted under no-choice and choice conditions in the greenhouse and in field plantings. New accessions of these agents will be collected from western Mediterranean Europe and evaluated for host specificity among close relatives prior to release. At field sites in southern and northern California, we will evaluate the ability of a leaf- and stem-feeding weevil to reduce invasive Dalmatian toadflax plant size and Dalmatian toadflax population size, and determine the degree of recolonization of invaded sites by native plants. We will release and evaluate the impact of a stem-galling wasp and a shoot- and root-feeding armored scale for biological control of the invasive giant grass known as arundo in the Sacramento-San Joaquin Delta and associated river watersheds, where arundo is impacting water resources. These studies will determine the effect of climate on wasp and scale establishment success. We will release and evaluate a shoot tip-galling fly for biological control of Cape-ivy at field sites along the California coast. Studies on arundo and Cape-ivy will include evaluations of agent dispersal within field sites, and of integrated biological-chemical control, in which herbicides will be applied and the ability of biocontrol agents to colonize and have impact on regrowth will be determined.

Progress Report
This is the final report for project 2030-22000-027-00D, which has been replaced by bridging project 2030-22000-031-00D. For more information see the annual report for the new project. Initiation of a biological control (biocontrol) program requires long-term investment and significant up-front costs. It is necessary to first prioritize which weeds to target with biological control, discover candidate biocontrol agents, and begin risk assessment testing. These research activities were addressed under Objective 1. Sea fig ice plants smother native plants and alter soils. Under Sub-obj. 1.1, 20 field sites were sampled in California for the presence of two accidentally-introduced soft scale insect species, which were found at only two sites, indicating that new biocontrol agents are needed. In 2019, crystalline ice plant was added to the survey at the request of stakeholders, and was found to be free of damage. Under Sub-obj. 1.1, a weed biocontrol prioritization process for the western United States was developed in collaboration with scientists from Australia. The first phase included a census of 187 weed scientists and land managers across 10 western U.S. states, leading to a list of 33 key weeds. The second phase used the expertise of biocontrol scientists to assess program feasibility and likelihood of success. Nine weeds were identified as low priority and 24 as moderate or high. Six weeds were characterized as having both high biocontrol feasibility and greatest environmental benefits of successful biocontrol. A report was prepared for stakeholders. The host range, biology, and impact of candidate agents were evaluated for ongoing programs. Yellow-starthistle is a widespread invasive weed in California rangelands. Under Sub-obj. 1.2, the host range of the flower- and seedhead-feeding weevil, Larinus filiformis, was studied in field experiments with cooperators in Cappadocia, Turkey in 2015 and in Thessaloniki, Greece in 2019, to test risk to 10 nontarget plant species. Results were inconclusive because of low survival of plants and/or low weevil attack. Another field experiment was started in Bulgaria in 2020 but the pandemic delayed work. In October 2019, a field release permit was obtained for the root- and rosette-feeding weevil, Ceratapion basicorne. A colony from Greece was screened for contaminants and transferred to a greenhouse. Russian thistle invades rangelands and agricultural areas in the western United States. Under Sub-obj. 1.2, larvae of the stem-boring moth, Gymnancyla canella, from Sicily, were compared to those from France and found to be similar biologically. In a laboratory test, ultraviolet (UV)-A light simulating sunlight did not reduce reproduction of the leaf-feeding mite, Aceria salsolae, on two non-target native plants, although reproduction on Russian thistle was 3 to 55-fold higher. Field host specificity testing with cooperators in Italy in 2018 determined that this mite does not pose a risk to nontarget plants. French broom invades pastures and forests in the Pacific states. Under Sub-obj. 1.2, the shoot tip-galling and seed-feeding weevil Lepidapion argentatum, was found to complete multiple generations per year between March and August. Galling reduced seedling height 55%, canopy width 29%, and growth rate threefold in a greenhouse test. The suitability of seven non-target native California lupine plants for the leaf-feeding psyllid Arytinnis hakani was examined over three generations. The psyllid reproduced on several native lupines but did not reduce growth, while reducing survival and growth of French broom by 50% or more. Cape-ivy smothers native plants and obstructs waterways in coastal riparian habitats in California. Under Sub-obj. 1.2, evaluations of the leaf and stem-mining moth, Digitivalva delaireae, on two native non-target plants (waterleaf ragwort and rayless ragwort) indicated no larval survival. On arrowhead ragwort, a few larvae did complete development on the non-target plant alone. Moth pupae and technology for their rearing were provided to the Commonwealth Science, Industry and Research Organization (CSIRO) in Australia to initiate their Cape-ivy biocontrol program. Medusahead is an annual grass that invades grazing lands and displaces forage species, and is a close relative of wheat. A high level of host specificity is required for an acceptable biocontrol agent. Surveys of Central and Eastern Europe by an ARS scientist in Reno, Nevada, led to discovery of a leaf-feeding mite (Aculodes altamurgiensis). Under Sub-Objective 1.2, the mite was reared on medusahead in the ARS quarantine lab in Albany, California, but was also able to develop on wheat. Field surveys of medusahead in California led to the discovery of the mite as being already present on medusahead in the United States. Research on the release of new weed biocontrol agents, their establishment, efficacy, and integration into weed management was completed under Objective 2. Under Sub-obj. 2.1, manipulation of dormancy requirements of the root- and rosette-feeding weevil, Ceratapion basicorne, enabled production of a second generation on yellow starthistle in the fall. The world’s first field release of this agent was made in California in April 2020. Also under Sub-obj. 2.1, yellow starthistle plants from California, northern Greece (where three previously-released biocontrol agents were collected) and southern France and Spain (source of invasive starthistle) were planted in field plots, and exposed in greenhouse tests to these agents. Scientists tested the hypothesis that the seed-galling fly, Urophora sinuraseva, would achieve highest densities on plants from Greece, whereas two others, the seed-feeding fly, Chaetorellia succinea, and the flower-feeding weevil, Eustenopus villosus, would be less selective. The seed-galling fly showed no differences in abundance based on plant origin in either field gardens or greenhouse tests. The weevil achieved 40% higher densities when confined on Greek compared to Californian plants, but occurred at low abundance in field tests. The seed-feeding fly occurred at 30% higher density in California compared to northern Greek plants in 2015, and in greenhouse tests achieved 50% higher densities on California plants. Variability in subsequent tests, however, indicate that the three insects can reproduce on starthistle from a wide range of geographic origins. Dalmatian toadflax invades rangelands and natural areas in the western United States, but it has been reduced by a stem-mining weevil, Mecinus janthiniformis, that was released in several U.S. states. Under Sub-obj. 2.2, this weevil was released at a state park in southern California. The weevil population exploded during the next three years, reaching 100% stem infestation levels. Within five years, density of Dalmatian toadflax stems decreased over two-fold, demonstrating the weevil’s efficacy. Scotch broom is invasive in forests and pastures in the Pacific states. A mite (Aceria genistae) that causes galls and deforms growth, arrived in California in 2014. Under Objective 2, mite dispersal was tracked and climate-matching models were used to predict range. Laboratory and field tests determined that the mite did not induce gall formation on native or exotic plant relatives. Gall development caused the weed to replace galled leaves, but this resulted in fewer seed pods. Reproduction of an accidentally-introduced psyllid, Arytainilla spartiophila, was reduced two-fold by drought stress in a greenhouse. The psyllid reduced plant growth in both the greenhouse and field. The psyllid thus exerts impact in combination with drought. Arundo grass invades riparian habitats throughout the southwestern United States. Under Sub-obj. 2.3, surveys of nine release sites in the Sacramento and San Joaquin River watersheds of northern California confirmed strong establishment of the shoot tip-galling wasp, Tetramesa romana, at two sites. A few wasps were found at four additional sites. Chemical control was found to be compatible with biocontrol in unsprayed plots. Wasps galled young main shoots, a sign of impact. Collaborative studies with an ARS scientist in Edinburg, Texas, in the original wasp release area of the Lower Rio Grande Basin, showed that the wasp reduced live arundo biomass by 44%, and diversity of other plants increased two-fold. At Texas sites containing a second agent, the root- and shoot bud-feeding armored scale Rhizaspidiotus donacis, live biomass decreased an additional 50%. Establishment of the scale was confirmed at eight sites in the two watersheds in California. In a greenhouse study, wasps from Texas produced 30% more offspring in cages with enhanced compared to ambient humidity, while wasps from a population in southern California showed no difference. Low humidity at some California sites may thus be an important limiting factor. Also under Sub-Objective 2.3, the shoot tip-galling fly, Parafreutreta regalis, was released as the world’s first biocontrol agent against Cape-ivy, beginning in 2016. Over 80 releases were conducted at 18 field sites along the California coast, mostly in field cages, but in 2019 a technique involving planting of galled plants was tested. Establishment was confirmed at four sites, including two State Parks, at which density of shoot tip galls increased 40-fold between 2019 and 2020. Low densities of a parasitic wasp that also feeds on native flies was found in galls at one site. Melaleuca trees were one of the worst invaders of the Florida Everglades. The leaf- and stem-feeding weevil, Oxyops vitiosa, was released in Florida, but it was unknown if the insect would locate small melaleuca seedlings. In collaborative studies with ARS scientists in Ft. Lauderdale, Florida, under Objective 2, natural dispersal of the weevil was examined. Smaller trees did not escape damage, demonstrating the ability of this weevil to exert biocontrol impact.

1. Release of new biological control agent of yellow starthistle. Yellow starthistle is a spiny toxic invasive weed that has invaded over 19 million acres of rangeland in the western United States, costing an estimated $1.4 billion in California alone. It reduces forage for livestock, decreases desirable vegetation, and increases the risk of wildfire. Although six species of insects have been introduced that attack the flower heads to reduce seed production, there is still a need for an agent that attacks roots and the immature rosettes (leaves). In rigorous laboratory tests in the United States, and field tests in the Mediterranean native range, ARS scientists at Albany, California, and their foreign collaborators, demonstrated that the rosette weevil, Ceratapion basicorne, poses no significant risk to nontarget native wild plants or to native or non-native crops. Consequently, a release permit was issued, and the first release of the weevil as a biological control in the world was made in California in April 2020. Widespread establishment of this weevil should reduce yellow starthistle densities.

2. Establishment of a new biological control agent of Cape-ivy. Cape-ivy is a vine-like perennial weed that has invaded sensitive coastal stream bank, forest and scrub habitats along the California coast, and it is also invasive in Hawaii, Australia and southern Europe. Cape-ivy smothers native herbs, shrubs and trees and can clog coastal streams that supply agriculture. ARS scientists in Albany, California, released a fly from Cape-ivy’s native range in South Africa that makes tumor-like galls in Cape-ivy's shoot tips. The fly is the world’s first biological control agent against Cape-ivy. Releases were conducted at 18 field sites in California between late 2016 and 2019. The fly has established large populations at four sites, including two California State Parks, with gall abundance increasing 40-fold since early 2019. Continued dispersal and galling will decrease Cape-ivy shoot tip abundance and growth.

Review Publications
Erickson, K., Pratt, P.D., Rayamajhi, M.B., Horvitz, C. 2020. Seedling maturation drives spatial variability in demographic dynamics of an invader with multiple introductions: insights from an LTRE analysis. Biological Invasions. 22:2185-2203.
Hogg, B.N., Moran, P.J., Smith, L. 2019. Relative performance and impacts of the psyllid Arytinnis hakani (Hemiptera:psyllidae) on nontarget plants and the target weed Genista monspessulana (Fabales:fabaceae). Environmental Entomology. 48(3):524-532.
Hogg, B.N., Moran, P.J. 2020. Combined effects of drought stress and psyllid herbivory on the invasive weed Scotch broom, Cytisus scoparius. Entomologia Experimentalis et Applicata. 168(3):209-220.
Tofangsazi, N., Hogg, B.N., Portman, S.L., Pratt, P.D. 2019. Tritrophic interactions between an invasive weed (Lepidium latifolium), an insect herbivore (Bagrada hilaris), and a plant pathogenic fungus (Albugo lepidii). Environmental Entomology. 48(6):1317-1322.