Until biocontrol organisms have become naturalized in North America, human intervention will remain the only way to prevent re-invasion of a site by saltcedar. However, there is a cost-effective means for reducing the rate and extent of re-invasion; specifically, the rapid establishment of a functional native ecosystem after saltcedar removal.
Ewel (1987) discussed the defining properties of functional ecosystems, among them sustainability, resistance to invasion, nutrient retention, and biotic interactions. These properties are largely unique to healthy, functional ecosystems, and may be used to judge the success of habitat restoration (St. John 1990). Resistance to invasion is of considerable interest in the context of saltcedar control. If we can establish restored ecosystems that resist re-invasion by saltcedar, we can maintain habitat quality and reduce control costs by reducing the frequency of human intervention.
We can focus our restoration objectives by understanding the mechanisms of resistance to invasion. A number of processes have been set forth in the ecological literature that can partially explain resistance to invasion. No single process is sufficient in itself to explain the resistance of almost all native vegetation types to invasion by exotic plant species. Several mechanisms contribute, with their importance likely variable between different kinds of ecosystems.
Allelopathy (the chemical interference of plants with other plants) no doubt plays a role, but it is likely that the most important and most widely-applicable mechanisms are related to the unique nutrient dynamics of a healthy, functional ecosystem. The term "ruderal" applies to plant species that invade disturbed sites. Typical ruderals are herbaceous annuals, but many exotic perennials, including saltcedar, fit the definition within the context of the shrub and tree growth forms. Salient characteristics of ruderals include fast growth rate, copious seed production, widespread seed dispersal, and an ability to grow on severely disturbed sites.
Rapid growth rate, a key factor in the ability of ruderals to preempt disturbed sites before native vegetation can return, is in large measure dependent upon the availability of mineral nutrients in water-soluble forms (Grime and Hunt 1975; Chapin 1980). Mineral nutrients at disturbed sites are quickly transformed to soluble forms by processes related to removal of the original vegetation and mechanical actions. Processes that accelerate mineralization include cessation of biological uptake, loss of biological inhibitors, and increased soil moisture, aeration, and temperature (St. John 1988).
An additional factor that favors ruderals is the loss of mycorrhizal fungi and the other beneficial soil organisms that mediate ecosystem function. Mechanical disturbance greatly reduces or eliminates native inoculum of these important symbionts, effectively preventing the quick return of native species that rely upon them heavily. Ruderals tend not to require mycorrhiza (St. John 1988). The result of mechanical disturbance is a site that lacks the symbiotic organisms required for ecosystem functionality, but rich in the soluble forms of mineral nutrients that allow ruderal species to achieve very rapid growth rates.
It is important to note that although nutrient ions are less abundant in the soil solution of functional ecosystems, the total stock of nutrients, including that in organic soil fractions, detritus, microbial biomass, and plant tissue, may be much higher than in disturbed ecosystems. If, as is likely, the key factor in resistance to invasion is lack of soluble ions in the soil solution, the credit for that in a functional ecosystem belongs to the intense exploitation of the soil volume by roots and the hyphae (fungal filaments) of mycorrhizal fungi. Nutrients absorbed by roots and hyphae are effectively removed from the soil, either by translocation into plant tissues or by incorporation into detritus and the microorganisms that process it.
The soil of a healthy ecosystem is pervaded not only by roots, but by the fungi that form mycorrhiza. These fungi link the cells of the fine roots with nutrient sources in the soil. The fungi that form arbuscular mycorrhiza (AM), the most common type, have a very wide host range. The network of mycelium in healthy ecosystems appears to interlink the roots of almost all native species in the plant community (Brundrett 1991). Most native species need to be mycorrhizal, and most ruderals do not. Saltcedar can become mycorrhizal (Bethlenfalvay et al. 1984), but in my experience it is only weakly mycorrhizal. Isolated plants may not become mycorrhizal at all.
These observations suggest that, like other ruderals that sometimes become mycorrhizal, it benefits little or not at all from the symbiosis. A few mycorrhizal native plants are not enough to do the job; to achieve ecosystem functionality, the soil must be thoroughly permeated by a network of mycorrhizal hyphae. This hyphal network is detrimental to ruderals, even species that sometimes become mycorrhizal (Francis and Read 1994). The suppressive effect is more than simple nutrient removal, and appears to require contact between the mycorrhizal network and the roots of the ruderal plants. The hyphal network has been the subject of extensive research in the past decade (Brundrett 1991), and its importance in ecosystem function is irrefutable. The challenge in control of exotics is to establish a functional ecosystem, and with it the hyphal network, during the short interval between removal of saltcedar and its re-invasion from off site.
The rules for establishing a mycelial network are for the most parts the rules for successful and cost-effective habitat restoration. The important components of the plan are to make the soil favorable for the growth of native plant roots and their mycorrhizal fungi, introduce propagules of both plants and beneficial microorganisms, and take action against any weed invasions that threaten to overwhelm the native vegetation. The most important factors in making favorable soil conditions are:
Unfortunately, the weeds will be growing while the mycorrhizal network becomes established. It takes time to fill the soil with roots and hyphae, and any deep-rooted species, including saltcedar, that become established before the mycorrhizal network is in place are unlikely to be affected by it. Some weed control is often required during this race for control of the soil volume. Mycorrhizal inoculum can be placed in the root zone by broadcasting and incorporating or by injecting with specialized machinery.
Options include broadcasting over freshly ripped ground, then dragging with a timber or a piece of chain link fabric. The land imprinter (Dixon and Carr 1994) has been used in southern California to place seed, shape the soil, and inject mycorrhizal inoculum into the root zone (St. John 1996). Three local contractors now have machines equipped to inject mycorrhizal inoculum. In all cases it is important to protect the inoculum from unfavorable temperatures and other stressful conditions. Inoculation should be planned for a cool time of year, and the inoculum should be allowed to remain exposed on the soil surface for the shortest possible time, even in the most favorable weather. Other aspects of habitat restoration Bainbridge et al. (1993) have suggested that container stock is generally superior to seeds for desert restoration, and listed irrigation and plant protection methods suitable for restoration sites.
However, land imprinting (Dixon and Carr 1994) has been quite successful in some desert situations, and requires little or no care if initial conditions are favorable. The right mix of methods in many situations might be imprinting of the entire area, then application of pole cutting (Anderson and Omart 1985) and container planting methods in more intensively-managed "island" areas. These islands will serve as core areas from which a diversity of native plant species can spread outward over the years. Timing is always a critical factor is habitat restoration.
Just as a farmer must choose the right weeks to cultivate, plant, control weeds, and harvest, so the restorationist must choose the right time for each stage of his work. Too often the right season comes and goes while the project manager waits for funding or approval from a higher level in the bureaucracy. If farming were done this way, we would all be starving to death. Summary The key to replacing saltcedar with invasion-resistant native vegetation is to make the soil suitable for roots and beneficial microorganisms, then introduce both the organisms and the native plants, then make sure the site conditions favor natives rather than ruderal species. Mycorrhizal fungi are fundamental to ecosystem function, and must be a central focus of the restoration effort.
Anderson, B. W., and R. D. Ohmart. 1985. Managing riparian vegetation and wildlife along the Colorado River: synthesis of data, predictive models, and management. General technical report RM Rocky Mountain Forest and Range Experiment Station, United States, Forest Service. (120) p. 123-127. Paper presented at the 'Conference on Riparian Ecosystems and their Management: Reconciling Conflicting Uses,' April 16-18, 1985, Tucson, Arizona.
Bainbridge, D. A., N. Sorensen, and R. A. Virginia. 1993. Revegetating desert plant communities. P. 21-26 in: T. D. Landis. Proceedings, Western Forest Nursery Association. USDA Forest Service General Technical Report RM-221. Good summary of innovative methods for small scale desert restoration projects.
Bethlenfalvay, G. J., S. Dakessian, and R. S. Pacovsky. 1984. Mycorrhizae in a southern California desert: ecological implications. Can. J. Bot. 62:519-524. Brundrett, M. C. 1991. Mycorrhizas in natural ecosystems. Pages 171-313 in: A. Macfaydn, M. Begon, and A. H. Fitter, Eds. Advancies in Ecological Research. Vol. 21. Academic Press, London.
Chapin, F. S. III. 1980. The mineral nutrition of wild plants. Ann. Rev. Ecol. System. 11:233-260.
Dixon, R. M., and A. B. Carr. 1994. Land imprinting for low-cost revegetation of degraded land. Erosion Control 1:38-43.
Ewel, J. J. 1987. Restoration is the ultimate test of ecological theory. pp. 31-33 in: W. R.Jordan, M. E. Gilpin, and J. D. Aber (eds.). Restoration Ecology: a synthetic approach to ecological research. Cambridge University Press, Cambridge.
Grime, J. P., and R. Hunt. 1975. Relative growth rate: its range and adaptive significance in a local flora. J. Ecol. 63:393-422.
Francis, R., and D. J. Read. 1994. The contributions of mycorrhizal fungi to the determination of plant community structure. Pp. 11-25 in: A. D. Robson, L. K. Abbott, and N. Malajczuk (eds.). Management of mycorrhizas in agriculture, horticulture, and forestry. Kluwer Academic Publishers, The Netherlands.
St. John, T. V. 1990. Practical application of Ewel's criteria for a successful restoration. Pp. 376-380 in: H. G. Hughes and T. M. Bonnicksen (eds.). Restoration '89: the new management challenge. Society for ecological restoration, Madison, WI.
St. John, T. V. 1988. Soil disturbance and the mineral nutrition of native plants. Pp. 34-39 in: J. P. Rieger and B. K. Williams (eds.). Proceedings of the second native plant revegetation symposium, San Diego, California. Society for Ecological Restoration and Management, Madison, Wisconsin.
St. John, T. 1996. Specially-modified land imprinter inoculates soil with mycorrhizal fungi (California). Restoration and Management Notes. In press.
Return to Workshop Home Page or continue on to the next paper "Should Artificial, Chemical-Based Water Be Part of the Restoration Process? Options and Techniques."
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