previous chapter
Compatibility of Biofuel Production with Wildlife Habitat Enhancement1
next chapter

Compatibility of Biofuel Production with Wildlife Habitat Enhancement[1]

John Disano, Bertin W. Anderson, Julie K. Meents, and Robert D. Ohmart[2]

Abstract.—A stand of native cottonwood trees (Populusfremontii ) with hedges of quail bush (Atriplexlentiformis ) would attract high avian densities and diversities. Densities and diversities of birds and rodents reached above-average levels for riparian vegetation in the lower Colorado River valley within two years from planting on two experimental plots and within one year on a third plot. The rapid growth rate of native trees and the acceptance of revegetated areas by wildlife, in conjunction with the current demand for wood as fuel, suggests that the two objectives are compatible and the latter can be economically productive.


Since 1977 we have been studying the feasibility of reintroducing native riparian vegetation along the lower Colorado River. Previously, there was virtually no information about the environmental conditions necessary for growth or survival rates of native plant species used in revegetation. We have also studied the densities and diversities of wildlife species associated with the reintroduced vegetation.

The objectives of this report are to discuss: 1) environmental conditions which lead to greatest productivity of trees; 2) the economic potentials of tree farming; and 3) a planting and harvest rotation design which would be compatible with wildlife enhancement.

Factors Affecting Growth and Survival of Trees

We planted about 2,000 trees of four species in February 1979 on a riparian zone dredge-spoil site along the Colorado River about 16 km. south of Blythe (Riverside County), California. The ensuing discussion is based on generalizations drawn from data collected from growth of those trees. Details of this study have been presented elsewhere (Anderson and Ohmart 1981a). Anderson and others will present details concerning the growth of cottonwood (Populusfremontii ) elsewhere in these proceedings (Anderson etal . 1983).

Importance of Tillage

Tillage is defined as breaking up and mixing the soil. In this case tillage was provided with a power auger or backhoe at each tree planting site. When planted in sandy soil, cottonwood and willow (Salixgooddingii ) trees grew to an average height of 5 m. within 680 days of planting if the saplings were provided with tillage to a depth of 3 m. (fig. 1). With no tillage, growth averaged less than 2 m. The advantage of deep tillage is that it permits rapid root penetration to the water table. The effect of deep tillage was not pronounced during the first six months, but thereafter saplings with deep tillage rapidly outgrew those without tillage (fig. 1). Tillage also affected survival of the trees. Among 112 trees planted with no tillage, 43 (38%) died by the end of the second growing season; among 772 trees provided with tillage to 3 m., 20 (<3%) died. Tillage to 3 m. was the single most important factor affecting growth and survival in our study.

Other Factors Affecting Growth


Competition from other vegetation, which invaded tree planting sites as a result of irriga-


Figure l.
Effect of tillage on growth of cottonwood
trees at three time intervals after planting in
sandy soil along the lower Colorado River.

tion, seriously affected growth and survival of trees. Among 77 willow and honey mesquite (Prosopisglandulosa ) trees planted with tillage to 3 m. which received moderate to severe competiton from Bermuda grass (Cynodon dactylon ), Russian thistle (Salsolaiberica ), and smotherweed (Bassiahyssopifolia ), 35 (45%) had died after two growing seasons, and growth of the survivors was significantly less than in areas where trees had little competitive interference.

Control of weeds potentially involves one of the greatest expenses associated with tree farming. This can be minimized if trees are planted in areas where the surface soil is very sandy, at least to a depth of 1 m. Competing species have difficulty becoming established because of the high temperatures and extremely dry conditions typical of surface sand.

Soil Density

Tree growth decreased as soil density increased below a depth of 1 m., even when tillage was to 3 m. and competitive vegetation was absent (Anderson and Ohmart 1981a). As the soil included more clay, tree mortality also increased. We recommend that tree farming not be attempted in dense soils, especially those containing significant amounts of clay.

Length of Irrigation Period

Growth rates were maximum and survival approached 100% when planting was in sandy soil, tillage was to 3 m., and irrigation was continued for 150 days at 30–40 1. of water per day. As soil density and competition from other vegetation increased, longer periods of irrigation were necessary (ibid .).

Avian Use of Riparian Vegetation

Value of Trees

In our studies of avian use of riparian vegetation along the lower Colorado River, we found that high avian density and species richness were consistently associated with stands of cottonwood and willow trees (Anderson and Ohmart 1981a, b; 1983). This was especially true for insectivorous bird species (fig. 2). Doves were attracted more to stands of mesquite trees (fig. 2). Frugivorous birds were virtually absent in stands of cottonwood and willow, but were primarily associated with honey mesquite. This, however, was not because of any intrinsic value of mesquite iself. Honey mesquite is parasitized by mistletoe (Phoradendron californicum ) to a greater extent than other species of trees along the lower Colorado River. Frugivorous birds eat mainly mistletoe fruit (Anderson and Ohmart 1978).

Value of Shrub-Like Vegetation

The term shrub-like refers to herbaceous vegetation such as Russian thistle, smotherweed, inkweed (Suaeda torreyana ), and quail bush (Atriplexlentiformis ). To examine the value of shrubs and shrub-like vegetation to birds, we compared three revegetation sites with varying shrub densities and composition. On one site shrub density was about one shrub per ha. Russian thistle and smotherweed reached densities of about 1,500 mature plants per ha. on a second site; quail bush and inkweed reached a combined density of 1,500 plants per ha. on a third site. All three areas had cottonwood and willow trees 3–7 m. tall, planted at a density of about 20 trees per ha. The combined density of all other tree species was less than one per ha.

Densities (number per 40 ha.) of various groups of birds on the revegetation sites were compared with the average bird densities in natural riparian vegetation along the lower Colorado River. These densities are expressed in standard units; thus, the average density for riparian vegetation is zero; positive and negative numbers indicate values above and below average, respectively.

When herbaceous vegetation was sparse, all avian groups except passerine granivores had below-average densities (fig. 3A). When herba-


Figure 2.
Number of times various vegetation variables were included
as a step in a significant multiple linear regression equation
(Y-axis). Maximum value possible for Y-axis is 25 (5 seasons
for 5 years). DVI—density of visiting insectivores; DPRI—
density of permanent resident insectivores; CW—number
of cottonwood/willow trees; HM—honey mesquite; SB—
screwbean mesquite; and SC—salt cedar per 0.4 ha;
FDD—foliage density and diversity at 0-0.6 m.

Figure 3.
Populations of various avian groups associated with various kinds and
densities of shrubs, other variables being equal. A. Very low density.
B. Shrubs abundant, primarily Russian thistle and smotherweed. C.
Shrubs abundant, primarily quail bush and inkweed. Avian densities
are expressed in standard deviation units where the mean was for all
riparian vegetation-types found in the lower Colorado River valley.
TOTAL SP—total species; VI—density of visiting insectivores;
PRI—density of permanent resident insectivores; GR—density of
passerine granivores; and GQ—density of Gambel Quail.

ceous vegetation consisted of Russian thistle and smotherweed, avian species richness and densities of passerine granivores and Gambel Quail (Lophortyxgambelii ) were above average, but densities of insectivores were about average (fig. 3B). When herbaceous vegetation consisted of quail bush and inkweed, all bird groups were above average, although visiting insectivores were near average (fig. 3C).

We conclude from these data that an area can be maximally enhanced for birds if cottonwood and willow trees, and shrubs such as quail bush and inkweed are planted. Dove densities would probably be somewhat below average, and frugivores would be absent unless honey mesquite with mistletoe was also present. Cavity-nesting bird spe-


cies, including woodpeckers, Lucy's Warbler (Vermivoraluciae ), Ash-throated Flycatcher (Myiarchus cinerascens ), and Wied Crested Flycatcher (Myiarchustyrannulus ) would be present in very low densities.

Profits from Farming

It has been shown that the trees grow rapidly and that tree farms could be attractive to wildlife, especially birds. By the end of the fourth growing season, trees grown in relatively sandy soil with tillage to 3 m. and competitive vegetation controlled will yield about one cord of wood per tree. In the Colorado River area one cord of wood sells for $80–$100 per cord (Fairbank 1980). We estimate a total production cost of $60 per tree, not including land rental or purchase. This estimate includes expenditures for management, labor and secretarial needs, moderate clearing and leveling requirements, irrigation system purchase, installation, and maintenance for 150 days, as well as harvest and transportation costs. An area planted with trees centered at 6 m., a 10% mortality rate, no harvest until the fourth year, which is then harvested at a rate of 40–60 ha. annually would yield a profit approaching $1 million in 12 years of operation if the wood was sold at $80 per cord (fig. 4). We believe our cost and mortality estimates are somewht high and the value per cord conservative. Our cost-income estimates are based on the design in figure 5.

Figure 4.
Profit estimates for a tree farm encompassing 160 ha. over
a 12-year period. See text for assumptions concerning
expenses and income. Mil = millions of dollars.

Honey mesquite wood is valued at twice as much per cord as cottonwood and willow (ibid .), but honey mesquite trees grow much more slowly. Nonetheless, because of their economic value, honey mesquite trees would be an important species in a tree-farming operation.

After first cutting, trees will sucker at the root crown without additional irrigation if the water table is 4.6 m. or less from the surface. We assume that biomass production in the four years after the first cutting will be decreased by 50%. This may be a conservative estimate because we have obtained greater growth rates after trees have been harvested than from initial plantings. However, our data are for only one year. If this high growth rate continued for four years, the profits indicated here might also prove conservative. The main disadvantages of tree farming are the high initial costs and the relatively long period until first returns on the investment.

Rotation of Harvest

A large variety of planting and harvest schemes are possible. We present only one for the purposes of discussion.

It is desirable to develop a tree farm so that the entire area is not harvested at one time. This would increase the diversity of vegetation height and foliage density on the area. If habitat values for wildlife are to be enhanced, it is also desirable to plant shrubs. Planting shrubs introduces extra costs, but will greatly enhance the area for wildlife.

Shrubs also have practical value. For example, it is desirable to have fire lanes;


Figure 5.
A plan for planting and harvesting trees on a 160-ha.
plot. The plot is divided into eight subplots of 20 ha.
each. Arabic numerals represent year of planting; Roman
numerals represent the year of harvest; e.g., 1, IV, VII,
 and X in one block mean that the first block would be
planted the first year and harvested the fourth, seventh,
and tenth years after planting.

these can include roads bounded by hedges of quail bush. Quail bush is an evergreen species and is fire resistant. Once established, quail bush is a vigorous competitor and will reduce costs associated with controlling invasion by salt cedar (Tamarixchinensis ) and annual weeds (fig. 5). Dried annuals and the large volume of litter produced by the deciduous salt cedar greatly increase fire hazards.

The cottonwood community development envisioned includes planting trees on 38% of the area each of the first two years and 24% the third year. In the fourth year the first trees would be harvested (fig. 5). With this rotation there would be four-year-old trees 10–15 m. tall with relatively open areas, attractive to quail and doves. On our revegetation sites, trees of this height attracted, among others, Northern Orioles (Icterus galbula ) and breeding Yellow-billed Cuckoos (Coccyzusamericanus ), a species listed as endangered in California. The combination of trees and quail bush would attract above-average densities and diversities of birds in all seasons. If some trees four or five years of age were not harvested but were killed by girdling, the area would also become attractive to at least a few cavity-nesting species.


We wish to thank Jeannie Anderson, Susan M. Cook, Jane R. Durham, Dr. Julie K. Meents, and Cindy D. Zisner for editorial assistance. Marcelett Ector and Cindy D. Zisner typed the numerous drafts of the manuscript. Elaine Hassinger and Julie Huff prepared the illustrations. We are grateful to Dr. F. Aljibury, Les Ede, and Jule Meyer, University of California Agricultural Extension Service, Riverside California, for their advice and cooperation. Ronald Gass, Mountain States Wholesales Nursery, Phoenix, Arizona, kindly provided many trees. The work was jointly funded by the USDI Bureau of Reclamation and the USDI Fish and Wildlife Service Contract No. 7-07-30-V0009.

Literature Cited

Anderson, B.W., J. Disano, D.L. Brooks, and R.D. Ohmart. 1983. Mortality and growth of cottonwood on dredge-spoil. In : R.E. Warner and K.M. Hendrix (ed.). California Riparian Systems. [University of California, Davis, September 17–19, 1981.] University of California Press, Berkeley.

Anderson, B.W., and R.D. Ohmart. 1978. Phainopepla utilization of honey mesquite forests in the Colorado River valley. Condor 80:334–338.

Anderson, B.W., and R.D. Ohmart. 1981a. Vegetation management final report. In prepara- tion. USDI Bureau of Reclamation, Boulder City, Nevada.

Anderson, B.W., and R.D. Ohmart. 1981b. Agricultural final report. In preparation. USDI Bureau of Reclamation, Boulder City, Nevada.

Anderson, B.W., and R.D. Ohmart. 1983. Avian use of revegetated riparian zones. In : R.E. Warner and K.M. Hendrix (ed.). California Riparian Systems. [University of California, Davis, September 17–19, 1981.] University of California Press, Berkeley.

Fairbank, W.C. 1980. On biofuel for air conditioning: a preliminary evaluation of using mesquite for powering residential and small commercial air conditioning systems in the Colorado River Desert. 9 p. Cooperative Extension, University of California, Riverside, Calif.


previous chapter
Compatibility of Biofuel Production with Wildlife Habitat Enhancement1
next chapter