In March 1994, we submitted petitions to the TAG to release the mealybug Trabutina mannipara from Israel and the leaf beetle Diorhabda elongata from China. In December 1994, TAG recommended to APHIS that both could be released. Accordingly, APHIS-PPQ (Plant Protection and Quarantine) began preparation of the Environmental Assessment to obtain final approval and permits for release.
However, the southwestern subspecies of the willow flycatcher, Empidonax traillii extimus Phillips, was placed on the endangered species list in February 1995 (U.S. Fish & Wildlife Service 1995). In some areas of Arizona, the flycatcher is now nesting in saltcedar, since its natural nest trees, willows (Salix spp.), have been displaced by saltcedar. Some workers have hypothesized that the soil now may have become so saline in some areas as a result of dam construction and the salinifying effect of saltcedar itself, that if saltcedar were removed, other vegetation might be slow to return (Anderson 1995). We are now preparing a Biological Assessment for Section 7 consultation with the Fish and Wildlife Service on endangered species. Final approval of the Environmental Assessment awaits a resolution of the questions involving the flycatcher.
The willow flycatcher apparently does not nest in Tamarix below 625 m elevation and, at higher elevations, does so to a lesser extent and at a lower success rates than in native vegetation. The widespread replacement of native riparian vegetation by saltcedar is probably a major factor in the loss and modification of willow flycatcher habitat (U.S. Fish and Wildlife Service 1995). One factor thought to contribute to the low success rate of willow flycatcher nesting in saltcedar is the limited availability of desirable food (U.S. Fish and Wildlife Service 1995). It may feed on pollinating insects, but very few other native insects occur in saltcedar. The leafhopper, Opsius stactogalus Fieb., probably introduced from Asia with the original Tamarix importations, acts as a biocontrol agent (Liesner 1971), although without providing sufficient control; no evidence has been presented in the literature that the flycatcher feeds on it. The native Apache cicada (Diceroprocta apache) that is abundant in saltcedar and willow thickets in mid-summer is too large to be captured by the flycatcher. Other biocontrol insects that could be introduced in the future would provide more food for the flycatcher and other species of birds as well, thus making saltcedar a more valuable plant at the same time that native cottonwoods and willows were permitted to increase.
Some 50 endangered or threatened species occur in riparian ecosystems of the western United States (Anonymous 1995) that also are known to be infested with saltcedar. These include the southwestern willow flycatcher (see paper by Rob Marshall, this workshop) and five other bird species, the peninsular bighorn sheep, the amargosa vole, two species of reptiles and two of amphibians, 35 species of fishes, three plants, and one arthropod. In southern California, the rare peninsular bighorn sheep (proposed for listing federally as endangered) is threatened with extirpation in some areas because saltcedar causes springs to dry up and provides cover for its predators (Lovich and de Gouvenain, in press).
In addition, populations of numerous other species are depressed by the impact of the saltcedar invasion, although they are not yet listed as threatened or endangered. These include at least 12 species of birds - Gila woodpecker, gilded northern flicker, vermilion flycatcher, Arizona Bell's vireo, Sonoran yellow warbler, summer tanager, western yellow-billed cuckoo, elf owl, northern cardinal, and Crissal thrasher. Also, the brown-crested flycatcher and the yellow-breasted chat are strongly dependent on cottonwood/willow habitat (Hunter 1984, Hunter et 1987). In Death Valley National Monument, Tamarix trees caused water levels to fall or springs to dry up completely, reducing pupfish habitat and causing other wildlife to move. After Tamarix was removed, the water level rose and wildlife returned (Rowlands 1987, Loope et al 1988).
The degree to which present salinity levels may limit revegetation by some species (especially by salt intolerant cottonwoods) following control of saltcedar is a concern in some of the more saline areas (Anderson 1995). Soil salinity in bottomlands may have increased substantially during the past 50 years, but the amount is poorly documented; some areas apparently always have been saline. In a survey of 7,596 ha at four sites on the lower Colorado River, presently in monotypic saltcedar (80-100% cover), all are presently suitable for revegetation: 45% by screwbean mesquite that is relatively salt tolerant and by honey mesquite, and 45% by quailbush (Atriplex lentiformis (Torr.) Wats.); 10% was suitable for cottonwoods and willows that provide the best wildlife habitat but this is much more than presently survives there (Bureau of Reclamation 1995). Revegetation attempts during the 1970's and 1980's, using primarily cottonwoods, were not very successful because of high soil salinity, drought, improper planting, browsing by cattle and wildlife, insect attack, or lack of irrigation. However, revegetation using screwbean and honey mesquite (both valuable for some wildlife species), on the Gila River near Tacna, Arizona, was successful (Pinkney 1992) and revegetation using cottonwoods and willows has been successful in several areas (Briggs 1992, USDA-NRCS 1995).
Biocontrol has never eradicated a weed anywhere in the world in the more than 700 projects undertaken worldwide during the past 130 years (Julien 1992). Nearly always, the weed remains at least as a common, or sometimes abundant, member of the plant community. However, biological control has greatly improved the natural ecosystem and the native plant and animal communities in many projects, including several in the United States (including Hawaii) and Canada (Huffaker & Kennett 1959; Kelleher & Hulme 1984; DeLoach 1991, in press; Buckingham 1994; Rees et 1996). Biological control of weeds has never caused any significant damage to native plant communities, or to rare species, even in those few cases where a small amount of non-target feeding occurred on some plants (Turner et al 1987; Funasaki et 1988; Harris 1988; Turner & Herr 1996). In most cases, control has required 5 to 10 years, giving ample time for the natural vegetation to return.
We expect that the results of biological control, as currently perceived, will be a gradual reduction and thinning of saltcedar stands over a period of 5 to 10 years in any given area, resulting in a final control (reduction in biomass and/or canopy cover) of 75 to 85%. This degree of control will leave sufficient saltcedar to provide some habitat for wildlife if they choose it and also for honeybees. All of the athel will remain since the insects proposed for introduction will not damage it. We expect that each insect introduced will probably be more effective in some areas than in others, that some may fail to establish or will provide little or no control, and that several species must be introduced to obtain satisfactory control in most infested areas. When this level of control is reached, further introductions can cease, or additional control can be sought, depending on an analysis of the ecosystem needs at that time. The ongoing biological control program against leafy spurge is now achieving excellent control in some areas, after the introduction of 13 insects species over a 30-year period. We expect the biocontrol of saltcedar to be similar, but hopefully to progress more rapidly. We do not expect to eradicate saltcedar in any area but rather that it will remain as a permanent but not abundant or damaging member of the plant community.
We do expect biological control to result in a great improvement in ecosystem health, improved abundance and diversity of the native plant community, greatly improved habitat that will allow population recovery of many species of fish and other wildlife, including that of several endangered species, among them the southwestern willow flycatcher, the peninsula bighorn sheep, and several species of endangered fishes. The recovery over large areas is expected to be similar to the dramatic recovery following conventional controls in small areas in Death Valley National Monument (Rowlands 1987, Loope et al 1988) and in the Nature Conservancy's Coachella Valley Preserve, CA (Barrows 1993).
The degree of natural revegetation that will follow the reduction in saltcedar stands is difficult to predict. Many areas presently infested appear suitable for revegetation by desirable native plants such as quailbush, screwbean mesquite, willows, and even cottonwoods. Research by both the Bureau of Reclamation (1995) and Busch and Smith (1995) predicts that even a large part of the lower Colorado River valley, which is among the most damaged in the southwest, is suitable for revegetation. Some areas, especially along highly controlled rivers below dams, may have become so saline (because of the lack of periodic flooding) that only the more salt-tolerant species will return. We suggest that even these are unlikely to become worse than at present if biological control is successful, and might improve over time if saltcedar (which contributes a major part to the accumulating salinity) is removed. A recovery plan, including modification of these riverways so that periodic small planned flooding could be practiced, would greatly augment the effects of biological control in allowing revegetation by desirable native plants.
Anderson & Ohmart (1984), after intensive studies of the lower Colorado River over several years, concluded that a management plan to improve avian habitat should include three components: 1) above average foliage density and diversity and above-average numbers of cottonwood and willow trees, 2) mistletoe as food for frugivores, and 3) elimination of saltcedar. Biological control, even if highly successful, will not eliminate saltcedar but would greatly reduce its abundance and would allow for a substantial improvement in the other two factors.
Only three main options are available for the recovery of the native plant communities and wildlife populations in riparian areas damaged by saltcedar. The first option is the use of conventional (mechanical, herbicidal, fire, or manual) control methods. The first three methods are expensive and damage other non-target desirable plants, whose preservation is the objective of control. Manual control is effective and non-damaging but so expensive as to be practical only in small infestations. All methods require repeated applications to control the rapid re-infestation by saltcedar. Very large sums of money have been spent annually for many years to control a very small percent of the infested area, some of which is left barren for wildlife.
The second option is to do nothing. This will allow the present intolerable situation to persist, and probably to continue the extinction of one species after another over time. In fact, to do nothing will probably allow the situation to continue to worsen, as is now happening. Areas in China long infested with saltcedar appear far more saline than infested areas of the United States, indicating that salinity levels here could become much worse if saltcedar remains.
The third option is biological control. The history of past usage of this method indicates that it could be very successful with saltcedar. The relatively minor beneficial values of saltcedar, the taxonomic isolation of it from beneficial plants in North American, and the large number of control agents known overseas indicate that this method is almost ideally suited for control of saltcedar without damage to the ecosystem and at relatively low cost.
Return to Workshop Home Page, return to the table of contents for the biological control paper, or continue on to the Literature Cited
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