The invasion since the 1920's by exotic saltcedar (Tamarix ramosissima) has caused enormous damage to native plant and animal communities in riparian ecosystems of the western United States. Conventional controls (manual, mechanical, herbicidal, and fire) are expensive and may damage the natural vegetation and must be applied repeatedly. An analysis of the damage caused and beneficial values of saltcedar (both economic and ecological), and natural enemies potentially available for introduction, indicate that biological control has a high potential for success, and without the disadvantages of the other control methods.
Damage by saltcedar includes the displacement of the extremely valuable cottonwood/willow, seepwillow baccharis, and other native plant communities often by dense monotypic thickets of saltcedar. Many wildlife species (especially birds) are unable to utilize saltcedar because its small fruits and seeds, lack of insects, and unpalatable foliage provide little or no food, and its density and structure is unsuitable as cover or for nesting. Also, it uses great amounts of groundwater and lowers water tables, causing springs to dry up and plants to perish. It increases soil salinity and is highly susceptible to fires, both of which kill associated intolerant cottonwoods and other plants. Several endangered species, especially birds and fishes, are severely impacted and common species are made more rare. Also, it causes sedimentation and narrowing of channels, increases flooding, and interferes with recreational usage.
Saltcedar has only minor beneficial values, mostly as ornamental shrubs and for honeybee pollen and nectar. Some wildlife species utilize it for cover or nesting (especially the white-winged dove) in the absence of their native habitat that saltcedar has displaced, and feed elsewhere or on insects from other nearby plants.
The potential for successful biological control is very high. The taxonomic isolation of Tamarix in the Old World has promoted the evolution of many host specific insects: 26 genera (with over 200 species) are restricted to developing entirely or in major part on the genus Tamarix. The taxonomic isolation of saltcedar in the Western Hemisphere (no species of Tamarix or of the family Tamaricaceae are native) implies a very small risk that introduced insects might attack non-target plants. The somewhat beneficial athel (Tamarix aphylla) is distinct from the weedy species and the insects considered for introduction do not damage it.
Our cooperators in France, Israel, Turkmenistan, Kazakhstan, and China have conducted preliminary testing on 21 insect species, including 10 species being tested in quarantine at Temple, TX. Two of these have been recommended by the APHIS Technical Advisory Group for the Introduction of Biological Control Agents of Weeds (TAG) of the United States Department of Agriculture, Animal and Plant Health Inspection Service (USDA-APHIS) for field release, pending approval of an Environmental Assessment; these are the mealybug Trabutina mannipara from Israel and the leafbeetle Diorhabda elongata that occurs from China to Turkey. Other promising species under study are two other Trabutina species; two weevils, Coniatus sp. and Corimalia sp.; two gall midges, Psectrosema spp.; two foliage feeding moths, Ornativalva sp. and Agdistis sp.; two psyllids, Crastina sp. and Colposenia sp., another leaf beetle, Cryptocephalus sp.; and a scale insect, Adiscodiaspis sp.
The research protocol and methodologies of biological control of weeds are well understood and strong safeguards are in place to minimize risk to non-target plants. The 130 years of experience worldwide (762 projects in 55 countries against 118 weed species), in North America (control agents released to control 34 weeds since 1945) and Hawaii (control agents released to control 20 weeds since 1902) have demonstrated a high degree of safety and a ca. 33% rate of complete or substantial control to date. The objective of biological control is to reduce the weed below the threshold of important damage; the method has never eradicated a target weed. Attack on non-target plants has been rare, especially during the past 30 years since strict safeguards have been in place, and such attack has always been of minor importance.
At present, we are preparing a Biological Assessment for consideration by the U.S. Fish and Wildlife Services that evaluates possible impacts on endangered species in southwestern riparian areas that are infested with saltcedar, especially on the southwestern subspecies of the willow flycatcher. If an opinion of no jeopardy is reached, then an Environmental Assessment (EA) can be prepared by APHIS, which will satisfy the requirements of the National Environmental Policy Act (NEPA)). If APHIS reaches a finding of no significant impact (FONSI) on the EA, then permits for release can be issued.
Biological control is expected to reduce the abundance of saltcedar by up to 75-85% but will probably require the introduction of several agents over several years. Control at a given site may require 5 to 10 years and control may not be satisfactory in all areas. As saltcedar is controlled, the native vegetation is expected to gradually return, though some areas may already be too saline for any but salt-tolerant species. Ultimately, saltcedar should be reduced to an uncommon (or common), but not an abundant, member of the plant community. The insects introduced to control saltcedar will also be used as food to some extent by wildlife, especially by insectivorous birds.
Biological control of weeds, as used herein, is defined as the planned use of undomesticated organisms (usually insects or plant pathogens) to reduce the vigor, reproductive capacity, or density of weeds. This definition is similar to that used by Harris (1988). This definition excludes cultural controls (grazing management, crop rotation, transgenic manipulation of crops, etc.) and natural control (the action of organisms without human direction). Biological control of weeds is a tried and proven method. The first recorded planned attempt was against prickly pear cactus (Opuntia vulgaris) in 1836. The scale insect Dactylopius ceylonicus from Brazil had been accidentally introduced into northern India in 1795 and controlled the cactus. In 1836, this scale insect was introduced into southern India and in 1865 into Sri Lanka (Ceylon) as the first deliberate attempt at biological control of a weed. It provided complete control of the cactus over vast areas in both India and Sri Lanka (Julien 1992). A program was initiated in South Africa in 1913 and in Australia in 1914 where D. ceylonicus also gave excellent control of this cactus. Since then, numerous insects have been introduced into Australia to control other species of cacti and other weeds, many of which have been successful (Julien 1992).
A similar program was developed in Hawaii beginning in 1902 to control the exotic lantana (Lantana camara), an ornamental shrub from Mexico, which had become a serious weed of rangelands. Lantana is a difficult weed to control because of its many cultivars and their differing susceptibility to insect attack. However, excellent to acceptable control was eventually achieved in most areas after the introduction of 22 species of insects from the native range of lantana from Mexico to Argentina. The same insects have been used to control lantana in several other countries, though not always with the same success (Julien 1992). The first project on biological control of weeds in mainland North America was in California to control common St. Johnswort or "Klamath weed" (Hypericum perforatum), an introduction from Europe. Spectacular and almost complete control was achieved in California within 12 years of initial releases of the leaf beetle (Chrysolina quadrigemina) imported from Europe (Holloway and Huffaker 1951), via Australia. Additional insects were later introduced to provide better control in other areas of the Northwestern U.S. (McCaffrey et al 1995; Rees et al 1996). These initial successes stimulated great interest in biological control of weeds in several countries after World War II. Through 1992, control had been attempted for 118 weed species in 55 countries, for a total of 762 weed-insect per country projects (Table 1). Several projects have been initiated recently but have not had sufficient time to achieve success.
To date, introduced insects have been released to control 34 weed species in the continental United States and Canada. Control has been complete or substantial (at least in large areas) on 10 of these, with success appearing imminent on several others (Table 2). In Hawaii, control agents have been released to control 20 weed species, with substantial or complete success with 9 of these (Goeden 1978; Funasaki et al 1988; Julien 1992). (Table 3).
Several reviews have been published describing the theory and/or current progress of projects: in the United States (Huffaker 1957, 1959, 1964; Holloway 1964; Goeden 1978; DeLoach 1991, 1997), for aquatic weeds (Andres and Bennett 1975; Brezonik and Fox 1975; Buckingham 1994; Center et al 1995), for native weeds (DeLoach 1995), for crop weeds (Charudattan and DeLoach 1988), in the western United States (Nechols et al 1995; Rees et al 1996), in Canada (Kelleher and Hulme 1984), in South Africa (Hoffman 1991), in New Zealand (Cameron et al 1989), internationally (Andres et al 1976; Julien 1992; Schroeder 1983), by the use of pathogens (Charudattan and Walker 1982; Templeton et al 1979); in a series of international symposia on biological control of weeds (Freeman 1978; Delfosse 1981, 1986, 1990; Delfosse and Scott 1995; Moran and Hoffman 1996).
Return to Workshop Home Page, return to the table of contents for the biological control paper, or continue on to the Research Protocol and Safeguards
For information on the outcome of this workshop or integrated weed management in the Pacific Region (Region 1), U.S. Fish and Wildlife Service, Portland, OR, contact: Scott_Stenquist@fws.gov