The effect of livestock production on the environment and implications for cattle grazing as a conservation tool in Alberta


Stu Crawford


Humans have been herding livestock for a very long time.  Humans, like all other vertebrates, are incapable of hydrolyzing the beta 1-4 glucosidic linkage that binds glucose molecules together to form cellulose (Young, 1994).  As a result the vast majority of herbaceous plant roots and shoots are not digestible to us, even after prolonged cooking (Young, 1994).  Some herbivores have evolved complex symbiosis with certain microorganism that can digest cellulose.  And humans take advantage of that by using livestock to transform inedible plant matter into highly digestible and nutritious meat.

Humans have only been herding livestock in North America for 100 to 130 years, but we have been doing it for much longer elsewhere.  The Lapps herd reindeer (Rangifer tarandus) in the subarctic, the Tibetans herd yaks (bos grunniens) in the Himalayans, the Masai herd cattle on the Serengeti, and the MediterraneanÝs herd goats (Carpa hircus) on their hills (Young, 1994).  Some concerns with some of these traditional herding systems and other herding systems around the world will be reviewed.

One of largest pastoral areas is in the Northern Plateau of the Tibet Autonomous Region of China.  Goldenstein et al. (1990) recently published a study on this system.  The plateau is too high for agriculture and is home to approximately 500,000 nomadic pastoralists who herd yaks, goats, and sheep.  Each family has their individual household herd.  They move their livestock annually between two encampments, a summer camp and a September to December camp.  At each of these main camps the herds are moved around to various satellite camps during the year.  This traditional herding system has been sustainable for centuries, and appears to be very ecologically conscious.  Grazed areas on the plateau are more biodiverse than the ungrazed areas.

The communal herding system imposed by the Chinese destroyed this traditional herding system and was very detrimental to both the people and the environment.  Luckily it was abolished.  Recently, however, the Chinese government has decided that the traditional systems of herd management used there is irrational and leads to overstocking and deterioration of the rangeland and so they have implemented programs to reduce the herd sizes.  The nomadic community of Phala is supposed to reduce its herds by 20% because of the environmental degradation that was resulting.  Phala nomads disagree.  Contrary to what the officials claim, the herds are not increasing in size and there does not seem to be any sign of unsustainability.  The range degradation is only around the settlements were the officials live, not elsewhere.

            A similar situation is occurring in India.  The Gaddi are a pastoral community in the Himalayans in northwest India.  Government policy to curtail Gaddi grazing practices based on assumption of environmental degradation.  There does not appear to be evidence for this, and Gaddi grazing practices may indeed be sustainable (Saberwal, 1996;  but see Mishra and Rawat, 1998;  Saberwal, 1998).

The pastoralists in the dry open woodlands in north-west Namibia (95 mm/yr) herd goats and cattle.  There is currently a concern for the impacts of their communal landuse practices.  A recent study by Sullivan (1999) has shown the population dynamics of the vegetation to look stable.  There appears to be impact on a local scale, but not overall.  Climate change appears to be the main cause for changing vegetation.  Communal farming works for them because they have local resource practices that minimize impact while maximizing secondary productivity.  These practices supported successful herding economies before European interference.

In eastern Kenya, cover reductions from livestock grazing have little effect on erosion until the cover falls below 40%.  At this threshold erosion is a significant factor.  As the population increases in Kenya more efficient grazing practices may need to be adopted to prevent overgrazing, such as disciplined herding instead of uncontrolled grazing (Zobisch, 1993).

Succulent thicket in South Africa is an important center of endemic succulents and geophytes that is threatened by megaherbivore and goat grazing (Moolman and Cowling, 1994).  Goat grazing has greatly reduced cover and diversity of endemic geophytes and succulent shrubs.  Elephants have had more moderate effect, and at low densities have little effect.  Goats have eliminated about 50% of the endemics and still 90% of the succulent thicket used for goat grazing.  Alternatives to goat farming, such as game farming of elephants, could be encouraged to stop this trend (Moolman and Cowling, 1994).

One of worst threats in Patagonia, Argentina is desertification due to overgrazing.  Heavily grazed area turning into shrubland (Beekskow et al., 1995).   There is similar widespread concern for the effect of pastoralism on the ecosystems in the arid and semi-arid zones of Australia.  However, in a recent study James et al. (1997) found no change in biodiversity as they moved from the heavily grazed areas next to watering holes to ungrazed areas 15 km out,  an indication that the grazing levels may be sustainable.

            In Europe and Britian humans have destroyed the forests long ago and made room for heather bushes to grow.  In this way their beloved heath was formed.  Heath is very diverse and contains many endangered species of plants and animals, as well as good grouse habitat (Gardner et al., 1997).  Heath is treasured and the Europeans wish to preserve it, but changing agricultural practices have resulted in the heathland being able to develop into its climax forest community.

The cheapest way to maintain the heath is to graze it with sheep (Kottmann et al., 1985).  In Netherlands abandoned agricultural land is being grazed by sheep to rejuvenate the heathland and prevent woody encroachment (Bakker et al, 1983).  Grazing abandoned arable fields is found to increase their biodiversity after only two years (Gibson et al., 1987).  Moderate sheep grazing on formerly fertilized grassland near Oxford, England results in the maximum diversity of plants and invertebrates and the quickest recovery of the land to heathland (Treweek et al., 1997).  There are a few concerns with this method of restoration.  If grazed to heavily, the heath will turn into grassland (Bakker et al, 1983;  Gardner et al., 1997).  And in Derbyshire, England, sheep grazing will preserve the heather moorland but eliminate bilberry moorland so zonal variation in grazing needed to preserve diversity (Welch, 1998).

One of the greatest concerns with livestock production is the effect of cattle on the Great Plains of North America.  From the first introduction of domestic cattle to North America in the 1500Ýs the cattle population grew to 26 million by 1890 (Briske and Richards, 1994).  Still today, cattle graze a large proportion of the Great Plains.  What is the result of this massive alteration to the natural ecosystem of the Great Plains?

            The Great Plains have had a long evolutionary history of being grazed.  They have been home to large grazers since the glaciers retreated at the end of the Pleistocene

(Lauenroth et al., 1994).  Before then, for most of last 7 million years large grazers have roamed this area (Brown and McDonald, 1995). The grasslands are adapted to having large herds of bison and other ungulates grazing on them.  Bison grazing actually increases the biodiversity of tallgrass prairie (Kaiser, 1998).  But humans have changed the species of herbivore.  The change from bison to cattle may effect the ecosystem because the two species exert different pressures on the grassland that they graze.  Cattle put more pressure on lowlands and cool season grasses, and are more selective feeders (Lauenroth et al., 1994).  Cattle also browse more, with bison diets consisting of 90% grass while cattle diets consist of about 58% grass (Pieper, 1994).

            A further concern is that some areas currently being grazed in North America, such as the Great Basin and the intermountain area, have not had any large ungulates since the Pleistocene (Pieper, 1994).  The plant communities in these places are not adapted to deal with large herbivores.

The effect of cattle grazing on plant communities in the grasslands of North America is very controversial.  According to Fleischner (1994) grazing decreases plant biodiversity and abundance in grasslands.  This directly contradicts Bork (2000) who maintains that grazing actually increases plant biodiversity and abundance.  See Figure 1 and Figure 2 (Bork, 2000 Pg 11 Fig 1, Fig 2).  A wide range of literature proposes mechanisms for the effect of grazing on plant biodiversity.  The most accepted hypothesis is presented by Pieper (1994).  Biodiversity increases with success ional stages as a disturbed area develops into a climax community.  Grazing moves the grassland back along the success ional pathway, thus decreasing biodiversity.  However, in many cases, such as tallgrass or fescue prairies, the penultimate successional community is more diverse than the climax community (Laycock, 1994).  Light grazing can move the plant community back one successional stage and thus increase biodiversity.

            Some empirical data contradicts this.  In northern shortgrass steppe in Colorado ungrazed grassland was found to be more similar to freshly disturbed grassland than heavily grazed grassland was (Lauenroth et al., 1994).  In this case grazing is not moving the community back along the successional pathway.

A more recent hypothesis is the threshold model.  A community can be globally stable, where under any disturbance it will always proceed along the same successional pathway to the same climax, or it can have multiple stable states, and once a threshold is reached the community will proceed to a different climax (Laycock, 1994).  See Figure 3 (Laycock, 1994, Fig 2 pg 269).  The grassland therefore doesnÝt deteriorate in a linear fashion from grazing, it moves through a series of levels with increasing pressure as each new threshold is reached (Blackburn and de Haan, 1999).  It seems that most mesic grassland communities have a global stable state while most arid grasslands have multiple stable states (Laycock, 1994).

Grazing pressure may be necessary to maintain a plant community in a biodiverse steady state.  Parsons et al.Ýs (1991) mechanistic model of the dynamic interaction between plants and animals determines the stability of plant species in a continuously grazed mixture.  It shows that a mixture is intrinsically unstable, and the advantaged species will take over and form a monoculture.  Only if advantaged species is preferred species will a stable mixture of plant species arise.

Grazing caused biodiversity changes in grasslands can also be predicted with island biogeography (Olff and Ritchie, 1998).  Grazing could reduce plant competition and increasing regeneration sites, thus increasing the rate at which new species colonize an area.  It could also increase the rate at which species are extirpated from an area.  When new species are established faster than they go extinct biodiversity will increase.  See Figure 4 (Olff and Ritchie, 1998 Fig 3).  Grazing induced changes in biodiversity can then be scale dependent, increasing diversity on small scale by increasing local colonization but decrease diversity on a larger scale by reducing the community to only grazing tolerant species.  See Figure 5 (Olff and Ritchie, 1998 Fig 1).

            An opposite trend was shown empirically by Willms et al. (1988) on fescue in Alberta.  In moderately grazed pastures cattle selectively graze due to palatability differences in grass, then maintain this trend because young grass tastes better.  This results in highly stable patches of overgrazed and undergrazed pasture, increasing the coarse-grained heterogeneity of the landscape without necessarily increasing small scale biodiversity.

There are several empirical studies on the effect of grazing on plant communities in the Great Plains.  Within the Great Plains the shortgrass prairies seem to be less effected by grazing, perhaps because they evolved under heavy grazing by bison (Laycock, 1994).  Grazing seems to help maintain the climax community by keeping out invaders (Lauenroth et al., 1994).  Heavily grazed shortgrass steppe can sustain cattle consuming 60% of its aboveground production, similar to the Serengeti and among the highest herbivory rates of any ecosystem (Lauenroth et al., 1994).

In northern shortgrass steppe in Colorado light grazing regimes (cattle utilizing 20% of the biomass of aboveground production) resulted in increased plant cover (Lauenroth et al., 1994).  Even long-term heavy grazing (60% utilization) increased total plant cover although it decreased cover of the more palatable species and decreased biodiversity because it removed native and invading opportunistic species.  Overall grazing had little effect, with heavily grazed plots indexing 0.1 dissimilar (WhittakerÝs Index) from ungrazed plots after 47 years (Lauenroth et al., 1994).  The blue grama grass dominated grassland seems to be in a stable state and resistant to change from grazing (Laycock, 1994).

            In nutrient poor, semi-arid Ponderosa pine-grassland in Arizona grazing can increase plant diversity (Rambo and Faeth, 1999).  Plant species richness was higher in grazed areas and total plant abundance did not change, although some species decreased.  As well, in shortgrass prairie in Texas species competition remained relatively constant with different levels of grazing (Lauenroth et al., 1994).

Some mixed grass prairies seem to be slightly less resistant to grazing.  In northern mixed prairie in North Dakota and Montana there was little change in species composition (WhittakerÝs Index of 0.190 and 0.148, respectively) after 41 years of moderate (35% utilization) grazing (Lauenroth et al., 1994).  But more effect was found in more productive areas of northern mixed prairie in South Dakota (0.663 dissimilarity after 27 years) and North Dakota  (0.418 dissimilarity after 40 years of 45% utilization) (Lauenroth et al., 1994).

            In rough fescue grassland in southwestern Alberta light grazing (1.2 Animal Unit Months per hectare) was found to have little effect on the plant community, while moderate grazing (1.6 AUM/ha) caused a reduction in fescue cover and heavy grazing (2.4 AUM/ha and 4.8 AUM/ha) had a substantial effect (Willms et al., 1985).  See Figure 6 (Willms et al., 1985, Fig 3).  Reduction in fescue cover resulted in increased biodiversity in the moderately grazed system (Laycock, 1994).

Tallgrass is more susceptible to changes from grazing.  Tallgrass plants are adapted for light competition, not grazing, and are thus more effected by grazing (Lauenroth et al., 1994).  In tallgrass, heavy grazing moves the community to a mid-grass or shortgrass community (Laycock, 1994).  Experiments in Oklahoma have shown a fairly large effect of grazing on tallgrass community structure, with a decrease in biodiversity with grazing (Lauenroth et al., 1994).  More recent studies in the same area have shown some biodiversity increases with grazing (Lauenroth et al., 1994).  As well, periodic manipulations of grazing, mowing, or burning may be required to maintain the climax tallgrass community and keep out invaders (Laycock, 1994).

These studies and others indicate that even just within the Great Plains some plant communities are very sensitive to grazing while others are not.  Much of this variation can be explained by precipitation levels.  Lauenroth et al. (1994) has shown a positive correlation with increasing mean precipitation levels and thus increasing productivity of the plant community, and increasing effect of grazing on that community.  See Figure 7.  As a grassland gets more productive it is increasingly effected by grazing.

Milchunas et al. (1988) developed a model to explain global variation in the effect of grazing on grasslands using the evolutionary history and mean precipitation of the area.  A longer evolutionary history of grazing is predicted to result in less of an effect of grazing on species diversity.  More productive grasslands should be more effected by grazing.  If the grassland is adapted and is productive then there should be a quick change in dynamic balance of the community to a suite of species adapted for grazing as opposed to competition for light.  See Figure 8 (Milchunas et al., 1988, Fig 3).

Milchunas and Lauenroth (1993) tested this hypothesis with a worldwide 236-site data set.  As the model predicted, they found that aboveground net primary productivity, evolutionary history, and level of consumption were three factors that explained species response to grazing.  And within reasonable levels of grazing, the geographical location was more important than level of consumption.  But they found a trend of areas with longer the evolutionary history of grazing having greater species dissimilarity under grazing.  This may be explained if a new suite of grazing-adapted species moves into an area when it is being grazed.

Besides just effecting the species composition of the plant community, grazing also effects the productivity of each species.  It is a general hypothesis that defoliation reduces the productivity of a plant (Fleischner, 1994), but there is a large body of literature to suggest that this is not always true.  In shortgrass prairie in Colorado it was found that in some cases light or moderate grazing actually increased the aboveground productivity of the plants (Lauenroth et al., 1994).  In their worldwide survey, Milchunas and Lauenroth (1993) found an increase in aboveground net primary productivity under grazing in some situations.  In a more controlled experiment, Semmartin and Oesterheld (1996) found that mowing large patches of grass (removal of 53% of the aboveground biomass) resulted in a positive effect on the grassÝ net primary productivity.  In controlled experiments with western wheatgrass defoliation was found to increase photosynthetic rates (Briske and Richards, 1994).  See Figure 9.

Plants and grazers have co-evolved, and plants have developed adaptive tolerance to compensate.  Defoliation causes numerous changes in the physiological growth processes of the plant phytohormones released by roots increased N and chlorophyll levels and result in more efficient CO2 use (Briske and Richards, 1994).  Changes in activity levels of symbiotic soil organisms in rhizosphere that make more nutrients available to the host (Manske, 1998).  Grazing also alters the microclimate of the plant community by changing light transmission, moisture relations, and temperature (Manske, 1998).  Herbivory can be always determental to the growth of a plant, but because of these factors the determental effect can be compensated for by compensatory growth.  The plant may overcompensate, resulting in increased growth at moderate levels of consumption (McNaughton, 1983).  See Figure 10 (McNaughton, 1983, Fig 1, in Briske and Richards, 1994 pg 163).

A common problem in grassland communities is the encroachment of woody species.  In many places in North America shrublands, woodlands, and forests have expanded within the last 100 years and replaced what was once grassland (Archer, 1994).  Many factors, such as increased atmospheric CO2, climatic fluctuations, fire suppression, and removal of native herbivores may be determining this vegetation change (Archer, 1994).  Livestock stocking rates above certain point also have a significant effect, in arid ecosystems in particular grazing favors woody encroachment (Archer, 1994). 

In the intermountain sagebrush region heavy livestock grazing decreases perennial grasses and increases woody plants and introduced annuals (Miller et al, 1994).  A similar effect can be seen in the sagebrush-grass communities of the Great Basin, as well as in shrub-grass communities dominated by saltbrush, shadscale, or winterfat (Laycock, 1994).  In these cases the dominance of woody plants causes the community to cross a threshold after which shrub-driven successional processes begin to predominate, and the community moves towards a new woodland steady state instead of climax grass community (Laycock, 1994).  Sagebrush steppe in Idaho damaged by heavy grazing in the spring has still not recovered after 50 years of livestock exclusion (Bork et al, 1998).  West (1999) also notes that grazing damaged sagebrush steppe reached a new threshold with more shrubs and annuals in the understory instead of perennials

            Livestock grazing effects woody plants in other areas as well.  In the ponderosa pine and mixed conifer forests of western interior US, what was historically widely spaced, fire-tolerant trees underlain by dense grass swards has developed into dense stands of more fire and disease sensitive species during the last 100 years (Belsky and Blumenthal, 1997).  This has been attributed in part to livestock grazing (Belsky and Blumenthal, 1997).  Also, in fescue-wheatgrass prairie in Montana grazing has increased unpalatable shrubs and forbs (Laycock, 1994).  In some areas grazing has had an opposite effect.  In northern mixed prairie in Wyoming livestock grazing prevents aspen encroachment (Laycock, 1994), and in tallgrass prairie in Oklahoma grazing prevented trees from moving in (Lauenroth et al., 1994).

            Livestock grazing can have other effects on the grassland ecosystem.  Root biomass may be effected.  It is conventional lore that grazing reduces root growth (Crawly, 1983).  However, Lauenroth et al. (1994) found root biomass to increase with grazing.  Light grazing on rough fescue in southern Alberta resulted greater root growth (Laycock, 1994).  Milchunas and Lauenroth (1993) found no consistant effect of grazing on root biomass in the 236-site data set that they surveyed.

Livestock grazing also effects nutrient cycling.  Herbivory increases decomposition and returns nitrogen to the soil as urine (Pieper, 1994).  Moderate grazing of mixed grass in North Dakota resulted in greater litter decomposition and nitrogen mineralization than either heavy grazing or no grazing (Manley et al., 1995).  Grazed mixed grass prairie in Wyoming had more carbon and nitrogen in the top three inches of soil after 11 years (Manley et al., 1995).

Grazing can have a profound effect on soil water, influencing infiltration by soil compaction and evapotranspiration by defoliation.  Because in Ohio the ground is wet and easily damaged in late winter, winter grazing there causes increased runoff and 15 times more erosion than controls (Owen et al., 1997).  Lauenroth et al. (1994)

Found increased water runoff on heavily grazed as opposed to lightly grazed northern mixed grass.  Naeth and Chanasyk (1995) found that heavy stocking rates on Alberta foothills fescue increased surface soil water and decreased soil water depth.  Lighter grazing treatments had less of an effect.

The cryptogamic crust can also be negatively impacted by grazing (Pieper, 1994).  Memmott et al. (1998) found the cryptogamic crust of arid sagebrush-grass rangeland in Idaho to be damaged by spring and summer grazing.  This layer, made mostly of moss and lichen, reduces water and wind erosion, increases water absorbtion and seedling survival, and fixes nitrogen.  Heavy grazing for 10 days reduced the cryptogamic crust by 62% if done in the spring and 48% if done in the summer.  See Figure 11 (Memmott et al., 1998, Fig 1 and 2).

Changes in the grassland ecosystem effect the animals that live there.  Grazing often occurs in patches, increasing the heterogeneity of the habitat (Laycock, 1994).  Besides increasing the diversity of habitat structure, grazing can alter the composition of the vegetation, increase productivity of selected species, and increase the nutritive quality of forage (Severson and Urness, 1994).  Although potentially damaging, these effects can be beneficial to some wildlife species.

Cattle are potentially competitors of other ruminants, but cattle grazing can be beneficial to other ruminant species.  It was found that in Utah mule deer tend to browse on shrubs while cattle graze on grasses and forbs (Severson and Urness, 1994).  Together they provide more balanced use of all forage components, resulting in a more stable community.  In Oregon cattle change plant composition to be more favorable to deer, increasing bushes and Sandberg bluegrass (Severson and Urness, 1994).  Increases in sagebrush from cattle grazing can result in good habitat for pronghorns (Severson and Urness, 1994).  Short term, high intensity grazing in late spring allows for regrowth of tillers and increases protein and phosphorus contents in foliage, thus increasing the quality of elk forage (Severson and Urness, 1994).

Fleischner (1994) maintains that microbiotic species richness decreases with grazing.  In ponderosa pine-grassland in Arizona insect abundance was decreased 4 to 10 times in grazed plots while insect diversity remained constant (Rambo and Faeth, 1999).  Grazing increased grasshopper abundance and decreased the abundance of other insects in experiments done in Oklahoma (Fleischner, 1994).  Grasshoppers were more abundant on lightly grazed than heavily grazed grassland in Colorado, and grasshopper species composition was changed with grazing in experiments done in South Dakota (Fleischner, 1994).  According to Laycock (1994) below-ground arthropods, scavenging arthropods, and grasshoppers generally increase in numbers with grazing, and beetle numbers decrease.

            In riparian systems grazing increases chub and sucker populations while reducing trout popoulations (Severson and Urness, 1994).  Grazing can be determental for wandering garter snake habitat and lizard habitat (Fleischner, 1994).  Milchunas et al. (1998) looked at the effect of grazing on plants, lagomorphs, rodents, birds, above and belowground macroarthropods, microarthropods, and nematodes.  They found the animal populations to be much more effected than the plant populations.  Bird and aboveground macroarthropod populations seemed to be effected the most.  See Figure 12 (Milchunas et al., 1998, Fig 3).  In California grassland habitats, out of 252 bird, mammal, and amphibian species, 52 were helped, 171 not effected, and 29 harmed by grazing (Severson and Urness, 1994).  Birds and reptiles tended to be favored by grazing, while mammals and amphibians either harmed or not effected.

Grassland birds have coevolved with grazers and every species has different preference for height and density of grass and do well under different levels of grazing (Dale and Prescott, 2000).  See Figure 13 (Dale and Prescott, 2000,  Fig 1).  Bock et al. (1984) found more birds in grazed areas in the summer and no differences in total bird abundance in the winter, but different bird species were effected differently.    Generally, more xeric birds are commoner in grazed areas (Bock et al., 1984).  Mountain plover, horned lark, and McCownÝs longspurs prefer heavily-grazed prairie, chestnut-collared longspurs, western meadow larks, and lark buntings prefer lightly-grazed prairie, and  BairdÝs sparrow, Sprague pipits, and savannah sparrows prefer ungrazed prairie (Laycock, 1994).

In riparian areas in Oregon, insectivorous birds preferred grazed sites while herbivorous and granivorous birds preferred ungrazed sites (Fleischner, 1994).  Small overgrazed areas within pastures create good habitat for bob white (Severson and Urness, 1994).  Increases in sagebrush as a result of grazing creates good habitat for sage grouse (Severson and Urness, 1994), but too much sagebrush can be detrimental for sage grouse habitat as well (West, 1999).  Laycock (1994) claims that in tallgrass moderate grazing increases number of and diversity of breeding birds.

            Rodents can be effected by the structure of their habitat as well.  Black-tailed prairie dogs, and thus black-footed ferrets, prefer heavily grazed areas (Severson and Urness, 1994).  On shortgrass prairies heavy grazing pushes out black-tailed jackrabbits and has no effect on white-tailed jackrabbits (Laycock, 1994).  Desert cottontails prefer the moderately grazed areas (Laycock, 1994).  Small mammal communties in tallgrass and montane grasslands are effected more by grazing than small mammal communties in shortgrass or bunchgrass (Grant et al, 1982).

            Within a grassland ecosystem, certain areas may be effected more than others.  Lowland areas seem to be effected more than upland areas (Lauenroth et al., 1994).  And riparian communities seem to be especially sensitive to livestock grazing, trampling, and deffication.  In the riparian zone grazing can effect the shape of the stream channel, the structure and stability of the soil, and the shape and quality of the water column (Fitch and Adams, 1998).  Increased runoff and sedimentation rates increase the temperature of the water and the amount of suspended nutrient and sediment in the water and alter timing and volume of the water flow (Fitch and Adams, 1998).  This lowers the water table and summer stream flow, increase summer water temperatures, and increases winter icing (Armour et al., 1994).  This in turn effects the aquatic and riparian vegetation, fish, and wildlife species (Fitch and Adams, 1998).

Laycock (1994) maintains that moderate grazing in riparian areas of northern Nevada has no significant effect in bird communities while Lauenroth et al. (1994) claims that sometimes light grazing in riparian areas creates better habitat by opening up the area, which has some benefit for waterfowl.  However, in a recent study cattle use of cottonwood riparian areas decreased species diversity and total abundance of vegetation and birds (Saunders and Hurly, 2000).  And in a review of 136 studies of riparian areas Belsky et al. (1999) found that in every case the effects of cattle on the riparian ecosystems were detrimental.  Livestock have damaged 80% of riparian ecosystems in western United States (up to 90% according to Armour et al., 1994).  This is unfortunate, as riparian habitat is the most productive habitat in western North America and provides essential wildlife habitat (Fleischner, 1994).  Riparian ecosystems are also important to prevent fragmentation, they maintain linkages between upland, floodplain, and aquatic ecosystems (Elmore and Kauffman). 

The riparian damage extends beyond where the cattle actually enter the ecosystem.  Belsky et al. (1999) found that cattle spend 5 to 30 times more time in riparian areas than elsewhere in their pastures.  During this time each cow puts an average of 8.5 L of shit and 7.5 L of piss into the river each day (Gary et al., 1983).  As a result of this downstream water had higher nutrient and sediment load, higher flood levels, and  lower levels during reduced flow (Belsky et al., 1999).  In a 90 square mile area in Ontario encompassing 300 livestock farms with average herd densities of 267 cattle per square mile, effluent from the headwaters showed bacterial and chemical characteristics similar to domestic sewage (Thornley and Bos, 1985).  But in an experiment with more reasonable stocking rates Gary et al. (1983) found that suspended solids, nitrate, and ammonium did not significantly increase in the stream after cattle were allowed free access to it.

To not damage the riparian ecosystems, they must be managed to preserve plant species important for function and to stop overland flow, and not be used when banks are unstable (Fitch and Adams, 1998).  To do this a rancher may control the cattleÝs access to the stream, and the timing, frequency, and intensity of the grazing (Fitch and Adams, 1998).  One way to do this is to completely exclude the cattle from the riparian area.

On humid pasture in Ohio fencing cattle away from river instead of allowing them continuous access to the river reduced sediment loss by 40% and reduced stream sedimentation by 50% (Owens et al, 1996).  After a four year exclusion in a study by Dobkin et al. (1998) bird populations started to recover and it appeared to be a fairly quick recovery for the riparian area to return to its natural state.  In a study done in Utah by Platts and Nelson (1985) exclosure from a previously heavily grazed area dramatically improved riparian vegetation, streambanks, and stream channel conditions after nine years. 

            Exclosure may not always be necessary.  Livestock damage to riparian areas can be site specific and depend on characteristics of that area (Clark, 1998).  More damage is arid areas, as in the humid temperate zones cattle spend less time in riparian areas (Clark, 1998).  Compatibility of livestock with riparian ecosystem varies with conditions (Clark, 1998).  Marlow et al. (1987) found that cattle grazing after early August when the banks is stable because of drier soil has little effect on the extent of channel alteration.  It is the combination of high flow, moist banks, and cattle grazing that leads to increased bank erosion.

            Riparian systems with high natural stress, bentonitic soils, and high erosion potential, can take less grazing stress.  Low natural stress areas with sandy loam soils and low gradients can take more (Elmore and Kauffman, 1994).  Riparian ecosystems extremely complex, the management strategy depends on the particular system (Elmore and Kauffman, 1994).

A general outline of Elmore and KauffmanÝs (1994) outline for managing riparian systems is as follows.  Total exclusion is the most effective, but other strategies work.  Grazing the riparian area separately from upland areas allows for more efficient management.  Only grazing in the early growing season when grass more palatable than woody plants or winter grazing to prevent herbivory during the growing season can result in lighter riparian use.  Fall or summer grazing both reduce willow populations, so rotation between fall and summer grazing or rotation with rest years is better.  The best management strategy depends on the specific case, but a grazing strategy can be tailored to each situation.  Generally winter grazing is better than early season grazing which is better than late season grazing which is better than rotational grazing.  Other grazing regimes are to be avoided.   Other suggestions for riparian areas are to graze different types of livestock near streams (Armour et al., 1994) or to select or train livestock to spend less time there (West, 1999).

            The above review of the literature on the effect of cattle grazing on the grasslands of North America indicates the complexity of the subject.  Generally it can be said that grazing does have an effect on the ecosystem, and this effect varies greatly with the geographic location of the area being grazed and how the grazing is being managed.  Mesic grassland communities that evolved under grazing tend to be resistant to change from grazing, and can recover relatively quickly.  On arid or semi-arid rangeland overgrazing can cause rapid degradation, and these rangelands may reach a steady state and not quickly recover.  More humid tallgrass plains are effected more by grazing pressure but recover quickly if allowed to rest.  Some areas, such as ponderosa pine forests and riparian areas seem inherently delecate.

            The effects of grazing pressure are complex.  Grazing effects plant biodiversity and abundance, basic ecosystem functions, cryptogamic crusts and microbial populations, and diversity and abundance of animal species.  Individual grazing regimes can benefit certain populations but no one grazing strategy benefits all populations.  Grasslands are not static communities were there is an ideal state.  They are dynamic and have different suites of species adapted to different grazing regimes.  A grazing regime that maximizes the heterogeneity of the landscape will probably conserve the most biodiversity and be the most sustainable.  Traditional continuous grazing strategies may not be effective for this.  Shorter duration grazing is more adaptable and may allow for ranchers to maintain sections of land under different grazing regimes with no large area being seriously overgrazed while still remaining economical.  As well, more delicate areas, such as riparian areas, have to be managed separately.

            Some feasible grazing strategies proposed by Bryant et al. (1989) are as follows.  In three herd-four pasture rotation systems, where one pasture in four is rested for four months a year, implemented in Texas cattle gains were maintained and range condition was maintained or slowly improved.  Yearlong deferred rotation programs with five pastures and monthly rotations maintained good animal performance and range condition.

In Oklohoma it was found that if cattle took no more than 40% during the growing season, then take 20 ˝ 40% could be taken during the winter with no ill effect.  Fall or winter grazing is often less damaging, as plants are not injured during the growing season and the soil is more stable in delicate areas such as riparian areas.

            Seasonal concerns can be taken into account to minimize effects on selected wildlife species.  In North Dakota waiting until third or fourth week of May avoids disturbing nesting waterfowl (Severson and Urness, 1994).  Delicate riparian areas can be grazed responsibly using the procedures outlined by Elmore and Kauffman (1994) and reviewed above.

            But by far the most important point to follow for producing livestock with a low environmental footprint is to produce it extensively.  If grazed properly cattle can be produced with a low environmental impact, but barley and oats cannot.  It is useless to conserve the ecological integrity of native prairie while producing cattle on it if that land is subsequently plowed up to plant barley and oats to feed the cattle.  It seems possible to produce cattle on native grassland in an environmentally conscious manner, but only if they are actually produced on that native grassland.

The grasslands around Lethbridge are categorized as mixed-grass and dry mixed-grass by Alberta Environmental Protection (1997).  This is very similar to shortgrass steppe, with fairly low productivity and a long evolutionary history of grazing.  Drier grasslands such as Lethbridge are least susceptible to detrimental effects of mismanagement (Lauenroth et al., 1994), and are among the least responsive ecosystems to grazing in the world (Milchunas et al., 1998).  If appropriate grazing strategies are adopted, Lethbridge is an ideal place for grazing cattle.

            In the Canadian prairies we have lost 76.9% of our prairies (Alberta Environmental Protection, 1997), and that which is left is severely fragmented.  See Figure 14 (Special Places pg 19).  Most of endangered species are prairie species in Canada live on the prairies (Alberta Environmental Protection, 1997).  If our prairie ecosystems are to be conserved be have to act to save them.  The majority of the remaining native prairie in Canada is within the dry mixed-grass region of Alberta (Alberta Environmental Protection, 1997).  Luckily this is both an area that can be easily grazed in an ecologically sound manner and an area that we might have some power over.

Besides being preserved for its biodiversity, our native prairie should also be preserved as a carbon sink.  There is 2 to 3 billion tons of carbon in our grasslands and plowing it releases 35% of that (Janzen et al., 2000).  The predicted carbon loss from our soil if there is a 5 degree rise in temperature is much greater if the soil is cultivated (Lauenroth et al., 1994).  See Figure 15  (Lauenroth et al., 1994, Fig. 8, pg94).

No modern government can afford to buy large reserves of land to conserve biodiversity, and unless such reserves are managed in the right way they will become havens for predators, feral animals, and noxious weeds and be useless (West, 1999).  There are many areas of the world, such as the East Africa savannas, where people and wildlife coexist (Western, 1989).  The subsistence herding in these areas reduces wildlife production, but wildlife and livestock are complimentary to an extent and wildlife diversity in East Africa is still as great outside the parks as within.  And there is far more wildlife outside.

Fulltime ranching provides daily contact with the land over all seasons and thus provides a degree of attention and experience which agency personnel and sporadically interested environmentalists cannot hope to replace.  Encouraging range management in economically and ecologically sustainable ways is the key to biodiversity conservation on the prairies.  The best way to conserve our prairies and their biodiversity is to keep the rural people on the land using it as range land and give them economical and environmentally friendly ways to manage their range (Brown and McDonald, 1995).  Some species are helped and some hindered by grazing, but if we maintain diverse grazing strategies and preserve the few pristine ungrazed areas the way they are, a considerable heterogeneity in the grassland landscape can be maintained.

            By maintaining an environmentally friendly system of range management we are also maintaining sustainable food production, where the needs of the present are met without compromising the needs of the future (Thompson and Nardone, 1999).  Tillage agriculture uses large inputs of fossil fuel energy, irrigation water, seeds, fertilizer, and pesticides to transform ecosystems to provide food for humans and domestic animals (Brown and McDonald, 1995).  Pastoralism uses only small inputs of energy and materials to manage seminatural ecosystems so that their relatively native vegetation will produce forage for domestic livestock (Brown and McDonald, 1995).  Because of this rangeland agriculture is the oldest, most unobtrusive, mundane, environmentally friendly, fully sustainable form of agriculture known (Heitschmidt, 2000).

            Evidence indicates that if proper precautions are taken Alberta can produce beef cattle on native grassland in an ecologically sound manner.  As this appears to be the most environmentally sound manner of food production available in Alberta it is logical that this is how we should produce the majority of our food.  Presently almost all beef cattle in Alberta are grain fattened.  As grazing cattle is much more environmentally friendly than growing cereal crops, this practice should be eliminated.  A consumer demand needs to be developed for grass fattened beef.  A second important point is that many cattle in Alberta are grazed on domestic pasture.  By reclaiming this land to native grassland as well as can be done more prairie biodiversity can be preserved while maintaining some livestock production on that land.  In conclusion, by reclaiming native grassland and raising grass fattened cattle on it in an environmentally sound manner Alberta could maintain food production while maintaining its natural ecosystems and the biodiversity within them.



References cited


Alberta Environmental Protection, Natural Resources Service, Recreation & Protected Areas Division, and Natural Heritage Protection and Education Branch.  1997.  The Grassland Natural Region of Alberta.  Report prepared for the Special Places 2000 Provincial Coordinating Committee.


Archer, S.  1994.  Woody plant encroachment into southwestern grasslands and savannas:  Rates, patterns, and proximate causes.  In:  M. Vavra, W. A. Laycock, and R. D. Pieper (ed.) Ecological Implications of Livestock Herbivory in the West.  Society for Range Management, Denver, CO.


Armour, C., D. Duff, and W. Elmore.  1994.  The effects of livestock grazing on western riparian and stream ecosystems.  Fisheries, 19(9) : 9 ˝ 12.


Bakker, J. P., S. de Bie, J. H. Dallinga, P. Tjaden, and Y. de Vries.  1983.  Sheep-grazing as a management tool for heathland conservation and regeneration in the Netherlands.  Journal of Applied Ecology, 20 : 541 ˝ 560.


Beeskow, A. M., N. O. Elissalde, and C. M. Rostagno.  1995.  Ecosystem changes associated with grazing intensity on the Punta Ninfas rangelands of Patagonia, Argentina.  Journal of Range Management, 48 : 517 ˝ 522.


Belsky, A. J., and D. M. Blumwnthal.  1997.  Effects of livestock grazing on stand dynamics and soils in upland forests of the interior west.  Conservation Biology, 11 : 315 ˝ 327.


Belsky, A. J., A. Matzke, and S. Uselman.  1999.  Survey of livestock influences on stream and riparian ecosystems in the western United States.  Journal of Soil and Water Conservation, 54 : 419 ˝ 431.


Blackburn, H. W., and C. de Haan.  1999.  Livestock and biodiversity.  In:  W. W. Collins and C. O. Qualset (ed.).  Biodiversity in Agroecosystems.  CRC Press, New York.


Bock, C. E., J. H. Bock, W. R. Kenney, and V. M. Hawthorne.  1984.  Responses of birds, rodents, and vegetation to livestock exclosure in a semidesert grassland site.  Journal of Range Management, 37 : 239 ˝ 242.


Bork, E.  2000.  Plant community responses to grazing.  Proceedings of the Western Range Science Seminar, Lethbridge, AB, January 24, 2000.


Bork, E. W., N. E. West, and J. W. Walker.  1998.  Cover components on long-term seasonal sheep grazing treatments in three-tip sagebrush steppe.  Journal of Range Management, 51 : 293 ˝ 300.


Briske, D. D., and J. H. Richards.  1994.  Physiological response of individual plants to grazing:  Current status and ecological significance.  In:  M. Vavra, W. A. Laycock, and R. D. Pieper (ed.) Ecological Implications of Livestock Herbivory in the West.  Society for Range Management, Denver, CO.


Brown, J. H., and W. McDonald.  1995.  Livestock grazing and conservation on southwestern rangelands.  Conservation Biology, 9 : 1644 ˝ 1647.


Bryant, F. C., B. E. Dahl, R. D. Pettit, and C. M. Britton.  1989.  Does short-duration grazing work in arid and semiarid regions?  Journal of Soil and Water Conservation, 44 : 290 ˝ 296.


Clark, E. A.  1998.  Landscape variables affecting livestock impacts on water quality in the humid temperate zone.  Canadian Journal of Plant Science, 78 : 181 ˝ 190.


Dale, B., and D. Prescott.  2000.  Prairie birds and grazing:  Evolving together.  Proceedings of the Western Range Science Seminar, Lethbridge, AB, January 24, 2000.


Dobkin, D. S., A. C. Rich, W. H. Pyle.  1998.  Habitat and avifaunal recovery from livestock grazing in a riparian meadow system of the northwestern Great Basin.  Conservation Biology, 12 : 209 ˝ 221.


Elmore, W., and B. Kauffman. 1994.  Riparian and watershed systems:  Degradation and restoration.  In:  M. Vavra, W. A. Laycock, and R. D. Pieper (ed.) Ecological Implications of Livestock Herbivory in the West.  Society for Range Management, Denver, CO.


Fitch, L., and B. W. Adams.  1998.  Can cows and fish co-exist?  Canadian Journal of Plant Science, 78 : 191 ˝ 198.


Fleischner, T. L.  1994.  Ecological costs of livestock grazing in western North America.  Conservation Biology, 8 : 629 ˝ 644.


Gardner, S. M., S. E. Hartley, A. Davies, and S. C. F. Palmer.  1997.  Carabid communities on heather moorlands in northeast Scotland:  The consequences of grazing pressure for community diversity.  Biological Conservation, 81 : 275 ˝ 286.


Garry, H. L., S. R. Johnson, and S. L. Ponce.  1983.  Cattle grazing impact on surface water quality in a Colorado front range stream.  Journal of Soil and Water Conservation, 38 : 124 ˝ 128.


Gibson, C. W. D., T. A. Watt, and V. K. Brown.  1987.  The use of sheep grazing to recreate species-rich grassland from abandoned arable land.  Biological Conservation, 42 : 165 ˝ 183.


Goldenstein, M. C., C. M. Beall, and R. P. Cincotta.  1990.  Traditional nomadic pastoralism and ecological conservation on TibetÝs Northern Plateau.  National Geographic Research, 6 : 139 ˝ 156.


Grant, W. E., E. C. Birney, N. R. French, and D. M. Swift.  1982.  Structure and productivity of grassland small mammal communities related to grazing-induced changes in vegetative cover.  Journal of Mammology, 63 : 248 ˝ 260.


Heitschmidt, R. K.  2000.  Ecosystems, sustainability, and rangeland management.  Proceedings of the Western Range Science Seminar, Lethbridge, AB, January 24, 2000.


James, C. D., J. Landsberg, S. R. Morton, J. Stol, A. Drew, H. Tongway, and T. Hobbs.  1997.  Changes in biodiversity along gradients of grazing intensity in Australian rangelands.  Bulletin of the Ecological Society of America, 78 : 116.


Janzen, H. H., B. H. Ellert, and J. F. Dormaar.  2000.  Rangelands and global carbon cycle.  Proceedings of the Western Range Science Seminar, Lethbridge, AB, January 24, 2000.


Kaiser, J.  1998.  Bison prime prairie biodiversity.  Science, 280 : 677.


Kottmann, H. J., W. Schwoppe, and T. Willers.  1985.  Heath conservation by sheep grazing:  A cost-benefit analysis.  Biological Conservation, 31 : 67 ˝ 74.


Lauenroth, W. K., D. G. Milchunas, J. L. Dodd, R. H. Hart, R. K. Heitschmidt, and L. R. Rittenhouse.  1994.  Effects of grazing on ecosystems of the Great Plains.  In:  M. Vavra, W. A. Laycock, and R. D. Pieper (ed.) Ecological Implications of Livestock Herbivory in the West.  Society for Range Management, Denver, CO.


Laycock, W. A.  1994.  Implications of grazing vs. no grazing on todays rangelands.  In:  M. Vavra, W. A. Laycock, and R. D. Pieper (ed.) Ecological Implications of Livestock Herbivory in the West.  Society for Range Management, Denver, CO.


Manley, J. T., G. E. Schuman, J. D. Reeder, and R. H. Hart.  1995.  Rangeland soil carbon and nitrogen responses to grazing.  Journal of Soil and Water Conservation, 50 : 294 ˝ 298.


Manske, L. L.  1998.  Can native prairie be sustained under livestock grazing?  Proceedings of the Fifth Prairie Conservation and Endangered Species Workshop.  Pg 99 ˝ 108.  Provincial Museum of Alberta, Edmonton, AB.


Marlow, C. B., T. M. Pogacnik, and S. D. Quinsey.  1987.  Streambank stability and cattle grazing in southwestern Montana.  Journal of Soil and Water Conservation, 42 : 291 ˝ 296.


McNaughton, S. J.  1983.  Compensatory plant growth as a response to herbivory.  Oikos, 40 : 329 ˝ 336.


Memmott, K. L., V. J. Anderson, and S. B. Monsen.  1998.  Seasonal grazing impact on cryptogamic crusts in a cold desert ecosystem.  Journal of Range Management, 51 : 547 ˝ 550.


Milchunas, D. G., and W. K. Lauenroth.  1993.  Quantitative effects of grazing on vegetation and soils over a global range of environments.  Ecological Monographs, 63 : 327 ˝ 366.


Milchunas, D. G., W. K. Lauenroth, and I. C. Burke.  1998.  Livestock grazing:  Animal and plant biodiversity of shortgrass steppe and the relationship to ecosystem function.  Oikos, 83 : 65 ˝ 74.


Milchunas, D. G., O. E. Sala, and W. K. Lauenroth.  1988.  A generalized model of grazing by large herbivores on grassland community structure.  The American Naturalist, 132 : 87 ˝ 106.


Miller, R. F., T. J. Svejcar, and  N. E. West.  1994  Implications of livestock grazing in the intermountain sagebrush region:  Plant composition.  In:  M. Vavra, W. A. Laycock, and R. D. Pieper (ed.) Ecological Implications of Livestock Herbivory in the West.  Society for Range Management, Denver, CO.


Mishra, C., and G. S. Rawat.  1998.  Livestock grazing and biodiversity conservation:  Comments on Saberwal.  Conservation Biology, 12 : 712 ˝ 714.


Moolman, H. J., and R. M. Cowling.  1994.  The impact of elephant and goat grazing on the endemic flora of South African succulent thicket.  Biological Conservation, 68 : 53 ˝ 61.


Naeth, M. A., and D. S. Chanasyk.  1995.  Grazing on soil water in Alberta foothills fescue grasslands.  Journal of Range Management, 48 : 528 ˝ 534.


Olff, H., and M. E. Ritchie.  1998.  Effects of herbivores on grassland plant diversity.  TREE, 13 : 261 ˝ 295.


Owens, L. B., W. M. Edwards, and R. W. van Keuren.  1997.  Runoff and sediment losses resulting from winter feeding on pastures.  Journal of Soil and Water Conservation, 52 : 194 ˝ 197.


Owens, L. B., W. M. Edwards, and R. W. van Keuren.  1996.  Sediment losses from a pastured watershed before and after stream fencing.  Journal of Soil and Water Conservation, 51 : 90 ˝ 94.


Parsons, A. J., A. Harvey, and I. R. Johnson.  1991.  Plant-animal interactions in a continuously grazed mixture.  II. The role of differences in the physiology of plant growth and of selective grazing on the performance and stability of species in a mixture.  Journal of Applied Ecology, 28 : 635 ˝ 658.


Platts, W. S., and R. L. Nelson.  Stream habitat and fisheries response to livestock grazing and instream improvement structures, Big Creek, Utah.  Journal of Soil and Water Conservation, 40 : 374 ˝ 379.


Rambo, J. L., and S. H. Faeth.  1999.  Effect of grazing on plant and insect community structure.  Conservation Biology, 13 : 1047 ˝ 1054.


Saberwal, V. K.  1996.  Pastoral politics:  Gaddi grazing, degradation, and biodiversity conservation in Himachal Pradesh, India.  Conservation Biology, 10 : 741 ˝ 749.


Saberwal, V. K.  1998.  Degradation and environmental conservation:  Response to Mishra and Rawat.  Conservation Biology, 12 : 715 ˝ 717.


Saunders, E. J., and T. A. Hurly.  Birds, creeks, and cows:  The influence of cattle grazing on birds and vegetation in cottonwood forests along the Oldman River.  Proceedings of the Western Range Science Seminar, Lethbridge, AB, January 24, 2000.


Semmartin, M., and M. Oesterheld.  1996.  Effect of grazing pattern on primary productivity.  Oikos, 75 : 431 ˝ 436.


Severson, K. E., and P. J. Urness. 1994.  Livestock grazing:  A tool to improve wildlife habitat.  In:  M. Vavra, W. A. Laycock, and R. D. Pieper (ed.) Ecological Implications of Livestock Herbivory in the West.  Society for Range Management, Denver, CO.


Sullivan, S.  1999.  The impacts of people and livestock on topographically diverse open wood- and shrub-lands in arid north-west Namibia.  Global Ecology and Biogeography, 8 : 257 ˝ 277.


Thompson, P. B., and A. Nardone.  1999.  Sustainable livestock production:  Methodological and ethical challenges.  Livestock Production Science, 61 : 111 ˝ 119.


Thornley, S. and A. W. Bos.  1985.  Effects of livestock wastes and agricultural drainage on water quality:  An Ontario case study.  Journal of Soil and Water Conservation, 40 : 173 ˝ 175.


Treweek, J. R., T. A. Watt, and C. Hambler.  1997.  Integration of sheep production and nature conservation:  Experimental management.  Journal of Environmental Management, 50 : 193 ˝ 210.


Welch, D.  1998.  Response of bilberry Vaccinium myrtillus L. stands in the Derbyshire Peak district to sheep grazing, and implications for moorland conservation.  Biological Conservation, 83 : 155 ˝ 164.


West, N. E.  1999.  Managing for biodiversity of rangelands.  In:  W. W. Collins and C. O. Qualset (ed.).  Biodiversity in Agroecosystems.  CRC Press, New York.


Western, D.  1989.  Conservation without parks:  Wildlife in the rural landscape.  In:  D. Western and M. C. Pearl (ed.) Conservation fot the Twenty-first Century.  Oxford University Press, New York.


Willms, W. D., J. F. Dormaar, and G. B. Schaalje.  1988.  Stability of grazed patches on rough fescue grasslands.  Journal of Range Manaement, 41 : 503 ˝ 508.


Willms, W. D., S. Smoliak, and J. F. Dormaar.  1985.  Effects of stocking rate on a rough fescue grassland vegetation.  Journal of Range Management, 38 : 220 ˝ 225.


Young, J. A.  1994.  Historical and evolutionary perspectives on grazing of western rangelands.  In:  M. Vavra, W. A. Laycock, and R. D. Pieper (ed.) Ecological Implications of Livestock Herbivory in the West.  Society for Range Management, Denver, CO.


Zobisch, M. A.  1993.  Erosion susceptibility and soil loss on grazing lands in some semiarid and subhumid locations of eastern Kenya.  Journal of Soil and Water Conservation, 48 : 445 ˝ 448.