Saltcedar Management and Riparian Restoration

John P. Taylor

U.S. Fish and Wildlife Service
Bosque del Apache NWR Complex
P O Box 1246
Socorro, New Mexico 87801
Phone: 505-835-1828
FAX: 505-835-0314
E-mail: r2rw_bda@fws.gov

Rio Grande riparian areas historically comprised a mosaic of multi-aged forests, brushlands, meadows, and wetlands. Vegetative diversity was maintained through periodic flooding events which fostered patchy habitats with rich vertical structure. As with many southwestern river systems, habitat degradation has occurred through altered river hydrographs and constricted floodplains resulting from irrigation and flood control developments, and the introduction of exotic flora. Resource managers are frequently charged with restoring these habitats with little knowledge of methodology, costs or manpower requirements.

Much of this information is now available from riparian restoration work accomplished at the Bosque del Apache National Wildlife Refuge in central New Mexico. The single most important aspect of riparian restoration is site assessment at both the landscape and locallevel. A regional reconnaissance of riparian flora can provide insight into flora composition and structure which facilitates the selection of sites for restoration work. We use an adapted version of a vegetation classification system developed by Hink and Ohmart in 1984 for the Middle Rio Grande Valley which codes habitats based on woody species composition and structural hierarchy. This base map has been digitized into a Geographic Information System for easy access using Arcview 2 software. The information provides an idea of the degree of infestation by exotic species in a riparian community and the expanse of a particular habitat type. At this stage of planning, important issues such as fragmentation and fire management can be addressed.

Once a potential restoration site has been selected at the landscape level, local site suitability must be determined. Information critical to revegetation success includes soil texture, soil salinity, and depth to water table and must be obtained before any restoration work begins (Anderson 1989, Swenson and Mullins 1985). Techniques have recently been developed which can speed site assessment. Rapid salinity mapping can now be accomplished using remote sensing equipment (Sheets et al. 1994). Site assessment can also include the potential for controlled flooding to stimulate the natural regeneration of native species. If water and material resources are available to construct irrigation works, periodic controlled flooding can restore riparian system processes as well as successfully revegetate areas.

Techniques employed to control saltcedar have included various combinations of herbicide application, mechanical control and burning. Treatments have proven satisfactory primarily due to the flexibility in employing one or several of these techniques if and when required. Arsenal herbicide (American Cyanamid, 1994) has been used in both ground and aerial applications. Controlled burning is then used to eliminate aerial stems in preparation for revegetation. In some chemically treated sites, saltcedar plant populations have been reduced from over 7,000/ha to less than 50. Likewise mechanical control has resulted in a 99% reduction in saltcedar plant populations.

Mechanical control involves root plowing and raking using heavy equipment. Bulldozers pull large plows about 45 cm. below the ground surface, sheering root crowns from the remainder of the root mass. The root crown is the underground portion of the plant from which resprouts arise. Root crowns are then pulled from the ground using large rakes and then stacked for burning with front-end loaders equipped with brush rakes. The operation leaves an even surface which facilitates flooding or planting. Combinations of mechanical and chemical control have also been used. Decisions for the type of control can depend on soil type, equipment availability, time constraints, fire danger, environmental sensitivity, and the degree of infestation. Costs for saltcedar control range from less than $125/acre for herbicide/burn control to $395/acre for mechanical control. Regardless of the control measure employed monitoring should occur to assess control and cost effectiveness.

Revegetation using plantings relies on prescriptions based on accurate assessments of soil texture, soil salinity, and depth to water table. Generally, planted cottonwoods and willows perform best in sandy soils with low salinity and moderate water tables (Table 1).

Table 1. Salinity, Soil and Water Table Planting Requirements for Selected Riparian Species at Bosque del Apache National Wildlife Refuge, New Mexico

Species E. C. Level (dS/m) Soil Type H2O Table Depth (m)
Cottonwood < 1.0 - 2.5 Sandy - Loamy 1.5 - 3.9
Black Willow < 1.0 - 2.9 Sandy - Clay Loam 1.2 - 3.1
New Mexico Olive < 1.0 - 2.5 Sandy - Loamy < 1.2
Skunkbush Sumac < 1.0 - 2.5 Sandy - Loamy < 1.2
Silver Buffaloberry < 1.0 - 2.5 Loamy - Clay Loam < 1.2
Screwbean Mesquite 3.0 - 7.99 Clay Loam - Clay < 1.2
Wolfberry 3.0 - 7.99 Sandy - Loamy < 1.2
Four-Wing Saltbush 8.0 - 13.99 Sandy - Loamy <1.2 - 1.95

Rapid salinity mapping can generally determine whether an area can be successfully revegetated. If salinities are low, saltcedar clearing can proceed. After clearing we establish 0.2 hectare grids over the area to be revegetated and sample from the center of each grid. Contour maps are then prepared forming data layers on which prescriptions for planting are based. Often, an elevational map is prepared which is used to delineate flood zones if controlled flooding is utilized to stimulate natural regeneration at lower elevations. Planting crews are then provided grid sheets with planting prescriptions for the field. Cottonwoods and willows are established using the dormant pole planting technique.

Long slender poles with 5-7 cm. butt diameters are harvested from natural river sites or obtained from commercial nurseries and placed in holes augured to the water table using commercial production auger equipment. Seedlings can be established using passive water harvesting techniques or by using drip irrigation. All planting occurs during winter months when plant materials are dormant. Plant survival rates range from 50-80% for cottonwood and willow poles and about 40% for seedlings.

Trees currently approach 16 meters in height. Controlled flooding can also be used to revegetate areas through natural regeneration following saltcedar control. Typically, an area is surface flooded, and water is gradually removed creating mudflats. Aerially dispersed woody riparian seeds adhere to the moist substrate, quickly germinate, and mature. Low soil salinity and proper drawdown rate are critical to the development of a desirable native riparian community. Proper flooding and drawdown sequences can be obtained from historic river hydrographs. Plant densities have averaged 2 cottonwood seedlings/m2 using this methodology. Revegetation costs have averaged $621/acre for plantings while costs for natural regeneration are associated with controlled flooding developments. As with saltcedar control, monitoring should be an important part of revegetation.

Literature Cited

American Cyanamid Company. 1994. Arsenal herbicide technical manual. 319 pages.

Anderson, B. W. 1989. Research as an integral part of revegetation projects. Pages 413-419 in Proceedings of the California Riparian Systems Conference, Berkeley, CA.

Hink, V.C., and R. D. Ohmart. 1984. Middle Rio Grande biological survey. Final report to the U.S. Army Corps of Engineers, Albuquerque, N.M. 193 pages.

Sheets, K.R., Taylor, J.P. and J.M.H. Hendrickx. 1994. Rapid salinity mapping by electromagnetic induction for determining riparian restoration potential. Restoration Ecology. 2:242-246.

Swenson, E.A., and C.L. Mullins. 1985. Revegetating riparian trees in southwestern riparian floodplains. Pages 135-138 in Proceedings of the first North American riparian conference, April 16-18, Tucson, Arizona.


Return to Workshop Home Page or continue on to the next paper "Afton Canyon: A Saltcedar Success Story."

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

Conference proceedings hosted by the National Invasive Species Information Center