Chapter 5: Changes in Ecological Processes and Forest Conditions

"Ecosystems are dynamic entities whose basic patterns and processes were and are shaped and sustained on the landscape not only by natural successional processes, but also by natural abiotic disturbances such as fire, drought, and wind. – Collectively, these features influence the range of natural variability of ecosystem structure, composition and function."

Kaufmann et al. (1994)

This chapter describes the changes in ecological processes that occurred after the Mexican–American War in 1848 (the beginning of the Territorial Period) to the present. We discuss in particular–grazing and browsing; fire; demographics and land use; recreation; alteration of watershed processes, including erosion, stream course effects, and floods; introduction of exotic (non-native) plants; disturbance by forest insects and pathogens; timber harvest; succession; wildlife population dynamics; and air quality. How various ecosystem functions have changed within each forest community follows the descriptions of selected disturbances. The next chapter discusses how these disturbances have changed Southwestern forest ecosystems in terms of their biotic diversity, integrity and resilience, and ability to accommodate human needs.

Although climate affects forest conditions and may have shifted in the past 100 years, climate change on a global scale is beyond the scope of this assessment and not discussed.

GRAZING AND BROWSING

One of the earliest changes that occurred as a result of European settlement was the introduction of domestic livestock, beginning with the Spanish occupation. Juan de Onate brought sheep to the Rio Grande pueblos in 1598; cattle, horses, and goats soon followed. By 1880, cattle herds in Arizona and New Mexico were estimated at 172,000 head (Baker et al. 1988) and articles in Western Range reported that overgrazing was depleting the range. By 1890, cattle numbers in the Southwest increased to more than 1.5 million head; and additional large numbers of sheep were being grazed. The once lush grasslands were in danger of disappearing. The Governor of Arizona stated in 1893, "In nearly all districts, owning to overstocking, many weeds have taken the place of the best grasses" (Baker et al. 1988).

By the time forest reserves were proclaimed in 1891, ranchers had become accustomed to unregulated use of forest lands for summer range. At the turn of the century, there were so many livestock grazing the public lands that signs of range deterioration began to appear even in the "good years." In 1901, overstocking of sheep brought forest regeneration to a standstill. The forest floor in some places was "as bare and compact as a roadbed" (Baker et al. 1988):

"They [cattlemen and sheepmen] knew nothing of grazing capacity, and there was no fund of technical knowledge about forage management to rely on. Overgrazing could not readily be recognized until in an advanced stage. Thus, when the Forest Service came into being on February 1, 1905, the most complex problems facing southwestern foresters related to grazing rights and range management."

By 1912, livestock pressures had penetrated the most remote, timbered and mountainous areas. Theodore Rixon (Roberts 1965), one of the first foresters in the Southwest, portrays the dismal situation:

"At the beginning the mountains and heavily timbered areas were used but little, but as the situation grew more acute in the more accessible regions the use of these areas became more general and in course of time conditions within them were more grave than elsewhere... The mountains were denuded of their vegetative cover, forest reproduction was damaged or destroyed, the slopes were seamed with deep erosion gullies, and the water-conserving power of the drainage basins became seriously impaired. Flocks passed each other on the trails, one rushing in to secure what the other had just abandoned as worthless, feed was deliberately wasted to prevent its utilization by others, the ranges were occupied before the snow had left them. Transient sheepmen roamed the country robbing the resident stockmen of forage that was justly theirs."

Not only were there pressures from range livestock but also from domestic and feral horses and burros. For over 70 years, heavy grazing reduced community diversity and plant competition; as a result there were no fine fuels to carry surface and ground fires. Grazing, reducing competition from herbaceous species, allowed rapid growth of pinyon, juniper, and oaks in adjacent communities (Nabi 1978).

The numbers of range cattle and sheep in the Southwest peaked during or shortly after World War I and have since declined (Figure 5.1). About 1920 the numbers of range cattle reached about 1.6 million head in Arizona and in New Mexico. Currently, there are 0.5 million head in Arizona and 1.2 million head in New Mexico. Sheep numbers are only available after 1920 and for New Mexico which has a substantial commercial market. Sheep populations have experienced a steady and dramatic decline from 2.3 million head in 1920 to only 0.2 million head in 1996.

Figure 5.1 Estimated number of range cattle and sheep in Arizona and New Mexico since 1870. Values are derived from annual, January estimates provided by the USDA National Agricultural Statistics Service, 1996. Data for numbers of sheep in Arizona and number of sheep in New Mexico before 1920 are not available. Number of range cattle is determined by subtracting the number of dairy cows and cattle on feed from the total number of cattle. Estimates for number of cattle on feed are only available after 1930, about the time feed operations in the Southwest became significant.

Range management was one of the most difficult resource management situations for the Forest Service during its first five to six decades. Ecosystems grazed by domestic livestock, and to a lesser extent by wildlife, suffered significant damage. Only substantial investment could save some plant communities. In the early 1960s, the Forest Service began to gain some control over grazing. Since then, livestock numbers were reduced and more intensive management systems were initiated on many grazing allotments. Annual grazing reports document the decreases in permitted numbers of cattle and horses and numbers of sheep and goats in Arizona and New Mexico (Figure 5.2). Permitted numbers for cattle and horses in both states have declined by more than half their peak numbers in 1919. Sheep and goat permitted numbers are a small fraction of their 1919 levels.

Figure 5.2 Numbers of permitted livestock as cattle and horses and as sheep and goats in selected years from 1909 to 1992 for Arizona and New Mexico. Data provided by USDA Forest Service, Southwestern Region.

At the same time that livestock numbers have decreased, populations of wild ungulates have increased. Mule deer populations and range have increased with development of stock tanks and increases of woody vegetation (Davis 1982). By 1900, elk had been extirpated in Arizona; but they were reintroduced from 1913 to 1928 by transplanting animals from Wyoming. Elk populations reached such numbers by 1995 that the annual harvest in Arizona⁶ was 10,000 animals and in New Mexico⁷ 12,000 animals. In some areas, such as the Apache and Gila National Forests, livestock numbers have been adjusted to respond to resource damage; but overall impacts remain the same or increase as elk subsequently move into the area. In these areas, there is much public controversy concerning whether the land should be managed for cattle or for elk.

Grazing and browsing pressures in the Southwest have changed over time, and affected the forest ecosystem both directly and indirectly. Forest ecosystems have gone from light grazing pressure during the 1800s, to severe pressure for the first several decades of this century, to current levels that attempt to balance numbers with capacity. The result of this grazing history has been to reduce the amount and diversity of the forest understory and its ability to carry surface fires.

FIRE

Except for climate, fire probably had the largest single impact in shaping the ecology of the Southwest prior to European settlement. It continues today to be the greatest potential force controlling ecosystems. The historic fire regimes characterized in the previous chapter changed dramatically with the coming of European and American settlers. Livestock removed much of the grassy fuels that carried frequent, surface fires; roads and trails broke up the continuity of forest fuels and further contributed to reductions in fire frequency and size (Covington and Moore 1994b). Because settlers saw fire as a threat, they actively suppressed it whenever they could. Initially, fire suppression was very successful because of low fuel loadings; but without fires to consume them, fuels accumulated over time. By the early 1900s, fire exclusion began altering forest structure and fire regime (Covington et al. 1994b). Forests with historically frequent, low-intensity fires were those initially most affected (Arno and Ottmar 1993, Covington and Moore 1994a). Woodland, ponderosa pine, and drier mixed conifer forests shifted from a fire regime of frequent, surface fires to one of stand-replacing, high-intensity fires. Fire had already been infrequent but high-intensity in the spruce–fir forest, so suppression there had less effect (Covington et al. 1994a).

Fire suppression has contributed to the buildup of organic materials (fuels) on the forest floor. Logging added heavy fuels in the form of limbs, tree tops, and cull logs. In some areas, these heavy fuels have been removed by slash disposal (fuel treatment), prescribed fire, or firewood collection. The areas with the greatest fire hazard are those with the greatest fuel accumulations, such as stands never logged or logged without fuel treatment, and stands inaccessible to firewood collectors.

The disruption of natural fire regimes has decreased the diversity of stands across the landscape. Frequent fires have killed conifer seedlings encroaching into forest meadows. Fire exclusion permits this encroachment, and meadow acreage has decreased (Arno 1985, Gruell 1985). Establishment of young trees in older stands provides a fuel ladder for carrying fires into the canopy. With more stand-replacing fires, average stand age is reduced; the diversity inherent in old stands is lost.

Because of heavy fuel accumulations, fires that occur now are more intense and more difficult to contain. Certainly there are more, large fires. The number of fires burning more than 10 acres has increased each decade since the 1930's (Figure 5.3). The average size of fires since the 1970s has ranged from 14 to 16 acres per fire, double the average size of fires in earlier decades, 1940s to 1960s (Figure 5.4).

Figure 5.3 Average number of fires per year in the Southwest since 1930. Fires included are only those larger than 10 acres; annual data are reduced to an average for each decade or period. (USDA Forest Service 1996a).

Figure 5.4 Average number of acres burned per fire start in the Southwest from 1940 to 1987. (Swetnam 1988).

The role of fire in specific communities is discussed further under "Resultant Changes in Forest Condition."

DEMOGRAPHICS AND LAND USE

It was not until the Territorial Period that both population and associated impacts on the environment increased dramatically. Population growth and resource extraction were spurred by construction of the transcontinental railroads. In the late 1870s and early 1880s, the long-awaited railroad arrived in the Southwest and linked it to population centers and commercial markets on the east and west coasts. Connections to the East promoted and expanded new industries, including sheep and cattle ranching, mining, and timber. The railroads supported population growth through homesteading and employment for construction and operations. The exploitation of minerals, grasslands, and forests as part of the new, commercial economy opened the Southwest to more intensive use than the preceding, subsistence economy had found possible or necessary. The new Southwestern settlers introduced changes to the economy that altered settlement and land use patterns, along with natural resources. The consequences were depletion of forage, degradation of riparian areas, and changes to water tables, forest communities, wildlife habitats, and wildlife populations (de Buys 1985, Bahre 1991). The centuries-old system of irrigation farming with acequias had changed little before the 1920s; frustrated by antiquated farming methods, the new settlers introduced "modern" farming approaches and techniques.

Within decades of the arrival of the railroad, the population of the Southwest increased rapidly. Population levels soon exceeded the carrying capacity of the land using only traditional technologies. To meet subsistence and economic needs, new practices in farming, irrigation, ranching, and timber harvesting were initiated. Resource extraction increased exponentially, not only for use within the Southwest but also for export throughout the United States. Consequently, resource demands exceeded the system's capacity of renewal and led to unsustainable practices such as–the excessive trapping of beaver from the early 1800s, overuse of forage by domestic livestock from the mid-1800s, and depletion of old-growth trees and forests in this century (Covington and Moore 1994a, Johnson 1994, Allen 1989, Cooper 1960, Dick-Peddie 1993). Extensive and unregulated mining led to degradation of upland slopes, riparian areas, and streams. Hydroelectric dams, over-harvesting, and loss of habitat impacted fisheries. Once-flowing streams were de-watered to satisfy irrigation needs.

Especially in the past few decades, the population of the Southwest has boomed and become more urban. The population of Arizona first exceeded that of New Mexico in the 1930s; since then Arizona has grown more quickly than New Mexico and by 1960 surpassed a million (Figure 5.5, data from de Gennaro 1990, Vest 1996, Bureau of Business and Economic Research 1994). If expected trends continue, by 2010 there will be two million people in New Mexico and three times as many in Arizona. The majority of residents live in the most urban counties of each state (those with the largest cities). Maricopa County (including Phoenix) has a population of 2,355,900; and Bernalillo County (including Albuquerque) has 650,000 residents.

Figure 5.5 Population growth in Arizona and New Mexico 1870-1990 and projected population for the year 2010. (de Gennaro 1990, Vest 1996, Bureau of Business and Economic Research 1994).

RECREATION

The transcontinental railroads not only opened the Southwest for resource extraction and settlement, but also opened a market for tourism. In the last quarter of the 19th century, the Santa Fe Railroad expanded its ridership through tourism to the Grand Canyon in Arizona and the Indian pueblos in New Mexico. In its efforts to encourage tourism of the Southwest, the railroad was aided and assisted by the Fred Harvey Company (Howard and Pardue 1996):

"The Fred Harvey Company and the Santa Fe Railway joined forces at an auspicious historical moment. Borrowing new techniques in marketing and advertising, the companies promoted the American Southwest as an exotic destination. The railroad, the travelers, and the indigenous communities of the region were all integral elements in a partnership that spanned more than three-quarters of a century."

Still, much of the recreation potential of Southwest remained undiscovered. Only the most spectacular sites that were easily accessible from the railroads attracted significant visitation. Recreation facilities and services in most of the region's arid landscape were sparse and primitive, simply because there was little demand before 1940 (USDA Forest Service 1980):

"Recreation use of the forest reserves grew slowly at first, then more rapidly as automobiles became numerous and roads penetrated further into what had previously been remote and inaccessible areas. General prosperity and more leisure time increased the human flow into the national forests, a flow which eventually became a flood."

Beginning in the 1970s, the Southwest was "discovered." The population expansion of the Phoenix and Tucson metropolitan areas was unparalleled anywhere in the United States. At the same time, the mystique of the "Land of Enchantment" and the "Santa Fe" style captured the attention of trend-setters throughout the country, and tourism to New Mexico exploded. Because of the mild weather, the Southwest also became a winter mecca for retired people as well as those with the freedom of long vacations. The Southwest became the nation's number-one target destination for the retired and recreational vehicle (RV) traveler, making it a year-round recreation area.

The Southwest has a diversity of high-quality recreation opportunities ranging from the primitive settings of the Gila Wilderness (first wilderness in the National Forest System) to the urban settings of the Salt River Chain of Lakes outside of Phoenix, Arizona. With the large increase in population and the attraction of the Southwest's climate, culture, and scenery, recreation use has increased tremendously over the past 20 years. Recreation use (as recreation–visitor–days, RVD) on National Forest System lands since 1965 has increased fourfold:

Developed Site Recreation

There has been an increase in demand for high-quality developments that include toilets, showers, lights, and reservation systems. Modern campers are seeking spaces designed for 40-foot RVs with individual hookups for sewage, water, power, and cable television. They also want trails, mountain bike paths, and interpretive nature walks. Their expectations on safety and security far surpass traditional offerings. The effects on forest health of this localized use include soil compaction, loss of vegetative cover, and soil erosion.

Wilderness

At the other end of the spectrum, there has been an increase in numbers seeking solitude in backcountry and wilderness areas. The idea, concept, and spirit of wilderness found their beginning in the Southwest. Heroes of the wilderness movement, Arthur Carhart and Aldo Leopold, worked in New Mexico and were influential in creating the first administratively designated wilderness, the Gila. The Forest Service in the Southwest now administers 52 designated Wildernesses with a combined total of 2,736,500 acres and an additional 175,112 acres in the Blue Range Primitive Area. These areas are ecologically diverse and range in size from 5,200 acres to 558,065 acres. Some are adjacent to the major cities of Phoenix, Tucson, and Albuquerque. These present complex management situations of protecting the resource from overuse, and resolving issues of non-conforming uses such as structures, illegal motorized access, overflights, and cultivation of cannabis. With more and more people using wildernesses there has been a noticeable effect on forest health including introduction of exotic weeds and an increase in human-caused fires. From 1992 through 1996, there were 18,747 human-caused fires in the Southwest (all ownerships) compared to 12,988 lightning-caused fires.

Heritage Sites

Of all the many tourist destinations in the West, few combine the landscapes and the romance of the Western experience as strikingly as the archaeological ruins of the Southwest. A recent study of heritage tourism by the New Mexico Office of Cultural Affairs (1995) indicates that visiting heritage sites and learning about American Indian culture ranked second only to scenic beauty, and higher than outdoor recreation, in reasons given for why people visit the state. Similar findings have been reported by Kaufman⁸ for Arizona. In a market overview of the Four Corners region, sightseeing at cultural and historic sites was noted as the number one purpose of trips to the area (Lillywhite 1994). In New Mexico alone, over 12 million visits are made each year to heritage sites and events, and the total economic impact of heritage resources is estimated to be $1.6 billion.

Interest in the heritage resources on national forests is increasing as people seek out opportunities to discover and explore Southwestern cultures in more remote, uncrowded settings. Over 50,000 heritage sites have been inventoried on national forests in Arizona and New Mexico. These include cliff dwellings, massive pueblo ruins, and a vast array of other historic and prehistoric remains. Heritage sites hold important cultural, educational, and scientific values for many people and contain clues for reconstructing prehistoric landscapes and past ecosystems. Many of these sites are threatened by natural and human forces, including erosion, vandalism, and the cumulative effects of visitation. Southwestern Indian tribes continue to express concern about the protection and appropriate use of these resources. Experience gained through programs like "Passport in Time" shows the public is eager for opportunities to assist in the preservation and study of heritage resources. Balancing the public's desire to experience the past with the exigency of protecting sites and the need to be sensitive to tribal concerns is an important challenge.

River Use

Another example of the public's attraction to the Southwest is reflected in the tremendous growth of whitewater river use during the past five years. Rafters, canoeists, and kayakers, both individually and through outfitter guide services, have discovered that the Southwest has miles of fine whitewater and attractive riparian habitats. As a result, river management needs to balance increased use with a need to protect the ecosystem on which the use is dependent. River use brings new health issues including disposal of human waste, water pollution, litter, and wildlife disturbance.

Hunters, Anglers, and Wildlife Viewers

The diversity of wildlife in the Southwest has attracted large numbers of hunters, anglers, and wildlife viewers. The Southwest serves as a major staging and stopover area for great numbers of migrating North American waterfowl. Only in the Southwest can four species of quail (Lophortyx) be found and hunted. Arizona and New Mexico are known for the last of the really great trophy elk as well as two kinds of bighorn sheep (desert and mountain). Other major game species are deer, black bear (Ursus americanus), pronghorn, mountain lion (Felis concolor), javelina (Tayassu tajacu), wild turkey, doves, grouse, and squirrel. In 1991, approximately 291,000 hunters spent 2.6 million days (8 hour days) pursuing game in Arizona and New Mexico (compiled data from various federal and state statistics). Fishing is an even more popular activity in the Southwest with 761,000 anglers and 6 million fishing days estimated for 1991 alone (same data sources as above). The major species taken are bass, trout, catfish, crappie, and sunfish. But the most popular activity, undertaken by 1.6 million people (9.2 million days in 1991), was observing and photographing the over 500 species of birds found in the Southwest.

Outfitter Guide Services

Outfitter and guide permits are issued for a variety of recreational pursuits. These include special events, horseback and llama-supported wilderness trips, jeep tours of backcountry areas, big game hunting, caving, whitewater rafting, and environmental awareness training. Many exotic types of dispersed recreation activities, which are provided by the private sector, such as bungee jumping, rock climbing, hang gliding, scuba diving, gold panning, and caving continue to gain in popularity, and pose unique, new management challenges. Major impacts of these newer uses tend to be concentrated in the same areas as the more traditional activities such as hiking, biking, off-highway vehicle ( OHV), water-oriented recreation, and horseback riding. Conflicts are on the rise in a region where there was once sufficient land capacity to accommodate all requests. The challenge is to find a sustainable balance.

Winter Use

Winter sports activities include destination, alpine ski areas such as Taos Ski Valley, Santa Fe Ski Area, Arizona Snow Bowl Ski Area and several Nordic ski centers. There are 10 alpine ski areas wholly or partially on national forest lands. In 1990, use on these areas totaled 620,000 RVDs with over a million ski visits. Ski resorts have significant, although local, impacts on the land. In addition, cross-country skiing, snowmobiling, and tubing attract many people.

Recreation Impacts

Many years ago, the numbers of people recreating on the national forests were so few and undemanding that they were little burden on the land. Now, people come in such great numbers and with such a wide variety of demands that their presence has brought significant impacts. Negative impacts include introductions of exotic weeds, increased human-caused fires, vandalism, littering, and stealing the artifacts of ancient human cultures. On the positive side, recreationists stimulate local economies and the tourism industry and provide numerous social and psychological benefits to themselves, their families and communities. Finally, as a result of their use and enjoyment of the forest, recreationists increase their awareness and possibly their involvement in forest health issues.

WATERSHED PROCESSES

Grazing and browsing, which reduced the ground cover of grasses and litter and reduced the frequency of fires, have also changed hydrologic cycles. Reduction in fire frequency has resulted in increased density of trees in most forest communities. Increased tree density has increased transpiration and interception of precipitation thereby making less water available for long-term stream flow. On the other hand, soil compaction caused by grazing of domestic and wild animals slows infiltration of surface water. Construction of roads increases runoff through interception and concentration of surface flows. These factors taken together can generate rapid runoff and powerful floods. Although total stream flow is reduced by increased evapotranspiration, periodic floods may generate peak flows greater than historic levels. Flooding occurs because water runs off the surface rapidly rather than being absorbed into the soil and released slowly to the watercourse. A flood may be powerful enough to remove the entire flood plain and its riparian vegetation. Floods may also cut deep, incised channels through old alluvial deposits. In a wet meadow, this channel cutting can lower the water table and drain hydric soils.

Impoundments and diversions on large rivers significantly modify the channel dynamics of erosion and deposition. The regeneration of cottonwoods and willows is dependent on natural floods to create seedbeds and moist conditions required for germination and establishment. Below large flood control dams, floods that would have supported this regeneration no longer occur.

Because arroyo formation in the late 19th century coincided with high livestock numbers, overgrazing has been portrayed as the primary cause of arroyos in the Southwest. However, cycles of arroyo cutting and filling have occurred repeatedly since prehistoric times (Cooke and Reeves 1976, Dean et al. 1985). Prehistoric arroyos are usually attributed to climatic variations. But, identifying the cause of arroyo formation in historic times has been difficult. Hastings and Turner (1965) and Cooke and Reeves (1976) conclude that arroyos probably arose from several interrelated environmental changes, including rainfall pattern, moisture regime and vegetation change. Although human-induced land disturbance may have played a role in arroyo formation, it was probably neither the sole nor primary cause. Apparently, various combinations of initial conditions and environmental changes can form similar appearing arroyos in different areas (Cooke and Reeves 1976).

INTRODUCTION OF EXOTIC PLANTS

In the early to mid 1800s, European and Asian immigrants to the Southwest brought from their native lands not only animals and material items but also various ornamental plants and plants of cultural significance. In their native environment, these plants had been subject to control by various insects and diseases not present in America. In their new home, these plants were carefully cultivated; and many found a suitable climate in which they could thrive even without care. By the early 1900s, the most aggressive species were out of control. Many were causing economic losses by competing with desirable plants, poisoning animals, hosting insect and disease agents, and altering various ecosystem attributes, such as fire regime.

Exotic weeds continue to invade rangelands, forests, and riparian ecosystems at an alarming rate. Control of infestations has been difficult, and the ecological consequences have been serious. The rapid expansion of exotic weed populations is a great deterrent, if not the greatest deterrent, to restoring native plant communities and re-establishing historic conditions. If exotic plants are not kept in check, long-term devastating effects to forest ecosystems can occur. The ecological effects include replacement of native plant species and reduction in ground cover which leads to loss of biodiversity, forage, habitat and scenic quality, and even soil productivity.

The Forest Service noxious weed strategy provides short-term direction for the containment, control, and management of noxious weeds through the Federal Noxious Weed Act of 1974, the 1990 Farm Bill Amendment, and USDA Departmental Regulation 9500–10. Species which are aggressive, difficult to control, toxic, parasitic, or hosts to serious disease or insect pests may be designated as noxious weeds. The Arizona Interagency Noxious Weed Committee has designated 37 native and exotic species as noxious weeds (Table 5.1); the New Mexico Interagency Noxious Weed Committee has designated 21 exotic species (Table 5.2).

FOREST INSECTS AND PATHOGENS

Forest insects, fungi, and parasitic plants continue to play important ecological roles in the re-structuring of forest communities (Holling 1992). The tree mortality that results directly or indirectly from their activity provides habitat, promotes nutrient cycling, drives plant succession, and contributes to biological diversity. Although their fundamental roles remain unaltered since historic times, the frequency, extent, or synchroneity of outbreaks may have changed for some of these disturbance agents. Insect and pathogen populations are ultimately limited by the availability of susceptible host trees. Therefore, changes in stand density, composition, and structure caused by fire suppression and logging would likely have affected outbreak patterns. Disturbance regimes of native insects and pathogens may also be affected by the introduction of exotics such as white pine blister rust (Cronartium ribicola). Detailed, standardized, regional reports of annual insect and disease conditions are available for only the past 25 years. Before then, reports⁹ of forest insect and disease activity were sporadic and emphasized situations where valuable economic or visual resources were threatened. In spite of limited information prior to 1900, there are good descriptions of more recent outbreaks and epidemics.

Bark Beetles

The known outbreak history of the roundheaded pine beetle in the Sacramento Mountains in southeastern New Mexico suggests an increasing trend during this century. At least six outbreaks have been reported in the Sacramento Mountains since 1937 (Bennett et al. 1994). From 1937 to the late 1960s, these outbreaks involved small acreages. In the early 1970s, an estimated 400,000 second-growth ponderosa pine trees were killed over 150,000 acres (Massey et al. 1977). This outbreak killed between 11 and 54 percent of ponderosa pines in sampled stands (Stevens and Flake 1974). Although the average diameter of ponderosa pine before and after the outbreak was unchanged, species dominance shifted to Douglas-fir and white fir. In the early 1990s, another outbreak of both the roundheaded pine beetle and the western pine beetle killed an estimated 100,000 trees over 87,000 acres. About the same time, two smaller yet still significant outbreaks of roundheaded pine beetle occurred in the Pinaleno Mountains of southeastern Arizona (Flake 1970, Wilson 1993); no prior outbreaks in the area are known. At least for these two areas, increases in outbreak size and frequency may have resulted from changes in stand conditions, especially increased density of susceptible host trees.

Elsewhere in the Southwest, large pine-beetle outbreaks have occurred only on the Kaibab Plateau (Colorado Plateau Province). A mountain pine beetle outbreak in this northern Arizona forest covered over 75,000 acres at its largest extent in the mid- to late-1970s. Otherwise, outbreaks in the pine forests have generally been less extensive and shorter duration. The susceptibility of the pine communities to bark beetles, however, continues to increase slowly over time so larger, more severe outbreaks may occur in the future.

In other pine forests and in mixed conifer forests, extensive, sustained outbreaks have been rare, even though stands there are at historically high levels of risk due to increased tree densities.

Large, spruce beetle outbreaks occurred in northern New Mexico, the Jemez Mountains during 1970s and the Pecos Wilderness between 1982 and 1985, and in Arizona, the White Mountains in the 1980s (USDA Forest Service 1976, Bennett et al. 1994). Infrequent, large outbreaks, however, are considered typical of natural disturbance regimes in spruce–fir forests (Veblen et al. 1994, Schmid and Frye 1977).

Western Spruce Budworm

Although the frequency of western spruce budworm outbreaks has not changed in this century compared to the past, their spatial and temporal pattern has changed (Swetnam and Lynch 1989, 1993). The most recent outbreaks in northern New Mexico (Southern Rocky Mountain Province) have been more synchronous and therefore more extensive and perhaps more severe than previous outbreaks. Fire suppression, grazing, and harvesting preferences to remove pines have favored establishment of multi-storied stands of young shade-tolerant species that are preferred hosts for the budworm. These changes in forest composition and structure may well account for the observed changes in budworm outbreak patterns.

Root Decay Fungi

A recent survey of commercial timber-producing lands on six national forests in Arizona and New Mexico (Wood 1993) indicates that root diseases and associated pests are responsible for mortality of 34 percent of trees (> 5 inches dbh). Although it is impossible to determine whether the relative importance of root disease as a mortality factor has increased recently, it is likely that root disease fungi are responding to the greater abundance of host material (living and dead). Species conversion allows new opportunities for host-preferring fungi; increased tree density permits greater root contact; increased competition and insect activity further weakens trees; and more stumps provide additional and long-lasting sources of inoculum. The life histories of root disease fungi and bark beetles complement each other as a positive feedback system that could potentially lead to larger and more persistent outbreaks. The extent or distribution of root disease is also likely to expand to those areas where Douglas-fir and white fir have recently replaced ponderosa pine (Johnson 1994).

White Pine Blister Rust

White pine blister rust is caused by the fungus Cronartium ribicola, introduced in the Northwest about 1910. The fungus has spread across the coastal and northern states and has caused serious economic and ecological impacts on western white pine (Pinus monticola), sugar pine (P. lambertiana), and whitebark pine (P. albicaulis). In 1990, the rust was discovered in the Southwest on southwestern white pine (Hawksworth 1990); the rust probably first became established about 1970 near Cloudcroft, New Mexico. Surveys indicate the rust is now present throughout the Sacramento Mountains (Hawksworth and Conklin 1990) and adjacent Capitan Mountains. The fungus rapidly kills seedlings and saplings. Larger trees are initially flagged (infected branches killed); later, the bole is cankered and eventually girdled. Because southwestern white pine is highly susceptible and environmental conditions are especially suitable in the Sacramento Mountains, the epidemic is expected to severely impact the white pine population there. A large inoculum source in the Sacramento Mountains increases the threat to other white pine populations in the Southwest and northern Mexico. The loss of southwestern white pine from the mixed conifer forests of the Sacramento Mountains not only reduces species diversity but also may result in greater damage by other disturbance agents.

Dwarf Mistletoes

More than 2 million acres of national forest lands in the Southwest are currently infested with dwarf mistletoes (Johnson and Hawksworth 1985). Surveys of ponderosa pine forests conducted in the 1950s and 1980s suggest that the incidence of southwestern dwarf mistletoe in ponderosa pine may have increased in some areas due to past fire suppression and timber harvesting (Andrews and Daniels 1960, Maffei and Beatty 1988). Although basic principles of mistletoe control have long been known (Koristian and Long 1922, Pearson 1950), intermediate cutting practices that leave infected trees and infrequent use of final removal cuts and understory sanitation may have increased mistletoe abundance. Fire suppression may also have resulted in the retention of additional infected trees that serve as inoculum sources to the developing understory. Because mistletoe spreads rapidly from overstory to understory trees, improper use of uneven-aged management could also increase mistletoe abundance. Dwarf mistletoe is a native member of the plant community, and it provides numerous ecological benefits where host–pathogen levels are within a natural range. Unfortunately, present forest conditions are especially suitable for development of infestation levels not previously experienced in these forests. A consequence of greater infestation is an increase in the risk and severity of insect outbreaks and wildfire.

TIMBER HARVEST

Logging has been conducted in the Southwest for over 100 years. Major efforts began with the harvest of railroad ties and other products for construction of the transcontinental railroad in the 1870s and 1880s (Schubert 1974). During the early days, logs were removed from the forests by expensive rail transport. To make these operations economically feasible, 70 to 80 percent of the volume had to be removed (Schubert 1974). When trucks became available for hauling, lighter cuts became economical. To insure continued timber supplies until young trees could establish and grow to adequate size, harvests of large trees were further reduced by distributing the cut to two or more entries (Myers and Martin 1963). During this time, typical harvests removed one-third to two-thirds of the available volume (Myers and Martin 1963) and averaged about 50 percent (Schubert 1974). At these residual stocking rates, reproduction was good to excellent (Schubert 1974); so stem density increased while tree size and age deceased.

Logging in the Southwest has been practiced over many acres with various methods and associated activities. Harvest practices have ranged from high-grading which removed quality trees and left poor trees to clearcutting and group selection which established even-aged stands or clusters of vigorous trees. The result has been a number of different environmental effects, good or bad, that vary from site to site. Although not a universal practice, logging slash is commonly piled and burned. The effect of this treatment is to reduce habitat for small mammals and material that might have contributed to soil organic content (Harvey et al. 1987). Where this treatment was not used, habitat and soil organic material increased, but so also did fire hazard.

Timber harvest levels on National Forest System lands in the Southwest have been tracked since 1908. Harvest levels gradually increased through the 1950s and, under sustained-yield management, remained relatively flat through the 1980s:

There has been a steady decline in the amount of timber cut in the 1990s. The timber cut in 1996 was almost exclusively fuelwood:

The net annual growth rate of all size classes of saw timber in the Region, currently around 700 million board feet, is substantially greater than historic harvest levels (Johnson 1994).

Roads associated with logging have made human access to the forest easier. This has several effects, including improved fire fighting efficiency, increased recreational use, and easier access to fuelwood. Roads, however, also provide better access for poachers and may increase habitat fragmentation for some species. Improperly located roads can result in increased soil erosion, stream siltation, and meadow degradation.

Through the harvest of dead and dying trees, the number of snags may have been reduced in some areas. Snags are important habitat components for some wildlife species. On the other hand, logging has occasionally been used to reset succession to an early stage for the benefit of specific plants and animals adapted to these conditions.

Harvesting fuelwood in woodlands has resulted in removal of many of the larger trees in some areas. The combined effects of removal of large trees, intensive grazing, and fire exclusion have resulted in extremely dense stands in some forests. The dense cover of trees and grazing pressure have reduced the density of grasses and forbs, resulting in increased rates of soil erosion over large areas.

SUCCESSION

Plant succession and disturbance are now recognized as closely connected processes that together determine vegetation dynamics. Although succession can not be understood without reference to disturbance, many other sections of this assessment already discuss various ecosystem disturbances in the Southwest. This section focuses specifically on forest succession, the factors affecting it, and the resulting vegetation patterns and ecological consequences.

Definitions and Theories

Following a disturbance that kills or removes a significant portion of the dominant vegetation, succession is the recolonization and replacement of plant species that occupy and eventually dominate a site. Characteristic sequences of species (seres) develop. Early dominants modify environmental conditions, that affect subsequent immigration and reproduction. Succession continues until a regional and climatic climax develops. Various ecological mechanisms drive the process (McCook 1994), and succession may follow various pathways at varying rates and directions (Hagle and Williams 1995, McCune and Allen 1985). Although climax species are able to reproduce under conditions they create, further disturbance is common everywhere so a climax community may never develop on some sites (van der Maarel 1993). Succession is usually seen as species turnover, but it can also be described as changes in physiognomy or ecosystem processes. Forest succession characteristically proceeds through four physiognomic stages–stand initiation, stem exclusion, understory re-initiation, and old growth (Oliver 1981). As succession proceeds, ecosystem processes such as nutrient cycling progress through phases of exploitation and conservation; Holling (1992) integrates disturbance into this model by adding two short but critical phases, release and reorganization.

Succession is a fundamental process operating at a fine scale. Together with disturbance that can operate at a range of scales, these processes determine vegetation dynamics from single-tree canopy gaps to forest landscapes (Delcourt et al. 1983, Holling 1992). Succession is a gap and patch scale process because it is the result of various tree-to-tree interactions such as competition and allopathy (Shugart 1984, Watt 1947). Coarse-scale landscape patterns arise from the replication of these patches over large areas. Turner et al. (1993) describe a conceptual model that integrates disturbance and succession on both spatial and temporal scales. The temporal dimension is represented as the ratio of time between disturbance events and time required for recovery; the spatial dimension is represented as the ratio of area disturbed to total landscape area. Various regions on this spatial-temporal graph identify where the interactions of disturbance and succession lead to landscape stability, oscillation, or destruction. Veblen et al. (1994) provide a more concrete example of disturbance–succession interactions using observations from a spruce–fir forest. They describe a complex pattern of vegetation patches as the outcome of snow avalanches, fire, and spruce beetle acting on two tree species, subalpine fir and Englemann spruce, with very different survival and reproduction strategies.

Competing theories by Clements and Gleason were proposed early this century to explain succession (see reviews by McIntosh 1980, McCook 1994, Cook 1996). Clements (1916) emphasizes the importance of competition and the effect of vegetation on modifying environmental conditions ("reaction"). He argues that succession proceeded unidirectionally to commonly occurring self-replicating stages characteristic of a region and climate (climax vegetation). He discounts the role of disturbance and thought that before settlement most sites supported climax vegetation. These ideas were repeated in many early textbooks and once widely accepted; but later evidence has accumulated to challenge this simple, deterministic explanation (Drury and Nisbet 1973). The alternative theory presented by Gleason (1926) recognizes greater significance of individual species differences (life history) and chance events (variation in seed source). With modification and elaboration, these ideas became incorporated into many subsequent theories and models of succession. Numerous studies confirm the importance of life history and disturbance on succession.

The differences between Clements (1916) and Gleason (1926) opened a debate on succession that has continued throughout the century. Tansley (1935) proposes that more than one climax may result on a given site. Watt (1947) identifies the importance of patches and disturbance cycles for creating vegetation mosaics. Egler (1954) notes that species did not always invade a site in relay but may be initially on the site and sequentially assume dominance. Pickett (1976) recognizes the importance of natural selection and disturbance. Connell and Slatyer (1977) propose three alternative mechanisms of species interaction for driving succession–facilitation, tolerance, and inhibition. Grime (1979) expands on explanations of succession due to life history characteristics and recognized three strategies– ruderal (in early stages), competitive (in middle stages), and stress–tolerant (at later stages). Cattelino et al. (1979) further refine the life history concept and integrated succession with disturbance regime; they demonstrate the possibility of multiple outcomes on a single site. Huston and Smith (1987) construct a general simulation model to explore species interactions and successional pathways. Expanding on the mechanisms of Connell and Slatyer (1977), Huston and Smith (1987) demonstrate how five successional patterns could emerge: sequential succession, divergence, total suppression, convergence, and pseudo-cyclic replacement. A number of books (e.g., Glenn-Lewin et al. 1992) provide a modern synthesis of successional theories that tend to be mechanistic (Shugart 1984) rather than holistic and to allow for non-equilibrium (chaotic) cycles (Oliver and Larson 1990) rather than require stable, deterministic outcomes.

Factors Affecting Succession and Resulting Patterns

The principal factors that determine the direction and rate succession are climate, site conditions (such as landform, elevation, and soil), and life history strategies. Various random influences (e.g., initial species mix and invasion opportunities) and disturbances also have a significant effect on succession (McCune and Allen 1985).

Climate, site conditions, and life strategies usually evolve slowly relative to successional rates, although there are exceptions. Climatic change has been relatively gradual throughout the present interglacial period; but future climatic warming could be more rapid (Delcourt et al. 1983). Pearson (1931) not only describes climate and forest types across the Southwest, but he also relates vegetation and soils. For example, he observes that substrates such as those with high clay content can be detrimental to pine reproduction and therefore limit the potential vegetation on a site. Because soils that develop in prairies are distinct from those that develop in forests, it is sometimes possible to determine that trees have recently invaded or retreated from an area. Life history strategies are the product of evolution and represent the integration of numerous physiological and morphological adaptions. Especially for long-lived trees in tightly connected ecosystems, life history attributes of a species probably change only very gradually most of the time. Rapid changes in genetics, however, can happen as when local populations are reduced to low numbers. Climate change, soil development, evolution, and human activities all provide a context for succession; sequences of species turnover that had occurred in one era may be impossible in the next.

Successional patterns reflect species accommodations to established environmental conditions and disturbance regimes. If the intensity, return interval, and area affected do not exceed limits imposed by rates of soil development, species longevity, and dispersal, even stand-replacing crown fires can be integrated into evolutionary and ecological systems (Pickett 1976, Odum 1969, Turner et al. 1993). This integration has occurred in fire-dependent systems such as northern coniferous forests (Wright and Heinselman 1973). New successional patterns, however, are established following novel, extreme (catastrophic or cusp) events (Jameson 1994). Introduction of new, non-native species (e.g., domestic livestock, weeds, insects and pathogens) may so re-sort species relations and environmental conditions that whole new vegetation emerges (see Sprugel 1991 for several examples). Southwestern white pine is a minor or major seral species in many mixed conifer habitat types of the Southwest (Moir and Ludwig 1979). But southwestern white pine is susceptible to blister rust. This stem disease is lethal to seedlings and saplings and has recently become established in southern New Mexico (Hawksworth and Conklin 1990). In some areas, southwestern white pine could even be extirpated. The effect in the Southwest of eliminating this seral species on successional pathways is unknown; but in the northern Rockies the ecological consequences of blister rust has been significant (Monnig and Byler 1992).

Succession and Ecosystem Trends

Ecosystem processes such as nutrient cycling are expected to respond with changes in the dominant vegetation of a site. Although it is quite evident that there are environmental and biotic differences between seral stages (if not, they wouldn't be recognized), there is a lack of agreement on which ecosystem trends are linked to succession and how they are linked. Odum (1969) categorizes ecosystem trends as community energetics, community structure, life history, nutrient cycling, selection pressure, and overall homeostasis. Some obvious and well-observed trends in the physical environment (especially light and moisture) result directly from changes in community structure as succession proceeds through the various physiognomic stages (Oliver 1981). Commonly observed trends are biomass accumulation, reduction of radiation below the canopy, and moderation of environmental extremes. Forest sites become more mesic as succession proceeds. Odum (1969) goes further in his contrast of young (early seral) and mature (later seral or climax) ecosystems. He suggests that production, growth, and quantity are associated with young systems and protection, stability, and quality are associated with mature systems. Observations of specific attributes such as ratios of production to respiration, species diversity, organism size, and nutrient conservation in actual ecosystems, however, do not always support the predictions (see Odum 1985). Even greater controversy arises from differences in explanations of the mechanisms responsible for observed trends. For example, DeAngelis and Waterhouse (1987) suggest that species stability is not fundamentally a property of ecological systems, but that an equilibrium state can emerge with extrapolation to sufficiently large spatial scales.

Succession in the Southwest

Many disturbance events like fires and insect outbreaks have not been allowed to run their natural courses over the past century. Consequently, they have become less frequent but more severe. Suppression of natural disturbances like fires and insects and deliberate removal of some late seral communities through logging may have resulted in an artificial over abundance of mid-seral communities.

Johnson (1994) reports that the area of both aspen and ponderosa pine decreased on national forests in the Southwest by more than 425,000 acres between 1962 and 1986. Most of the land that had been occupied by these early seral communities became part of the mixed conifer forest through succession. While this vegetation was succeeding to a later sere, average stand age decreased as harvest removed older trees and encouraged regeneration. The result has been a more homogeneous forest that lacks stands in the youngest and oldest classes.

In 1987, only 7 percent of national forest timberlands in the Southwest were nonstocked or in seedling and sapling age classes (Conner et al. 1990, Van Hooser et al. 1993). Reynolds et al. (1992) estimate that approximately 20 percent of timberlands would need to be in these condition classes to provide all age classes, including old growth, on a balanced, sustained basis.

RESULTANT CHANGES IN FOREST CONDITIONS

The changes in disturbance regimes and other forest processes listed previously have resulted in a transformation of forest conditions such as structure and composition. These changes in forest condition further contribute to changes in processes in a feedback cycle. Some of the structural changes that have been observed in Southwestern forests are listed below.

Forest Overstory/Understory Relationships

Structural diversity in the forests of the Southwest has changed considerably, including understory plants. Heavy livestock grazing not only removed the fine fuels needed to carry a fire but shifted the competitive advantage from the herbaceous understory to tree seedlings. This increased tree density within the forest and allowed tree expansion into meadows. As the large ponderosa pine and Douglas-firs trees were harvested, they were replaced by numerous seedlings which were not thinned by fire as in the past. Expanding coniferous thickets suppressed understory plants. Over large areas, important components of structural diversity, namely meadows, open-canopy and old-growth forests, have been converted to pine and fir thickets (Moir and Fletcher 1996).

The relationship between overstory density and understory productivity has been documented in numerous studies. Covington and Moore¹⁰ estimate the increase in overstory canopy density in the past century has reduced herbage production by 92 percent within some ponderosa pine stands. Moore and Deiter (1992) report on the relation between stand density index and understory productivity in a ponderosa pine forest on the Kaibab Plateau. Productivity of grasses, sedges, forbs, and shrubs decreased with stand density index; the rate of decrease was steepest at stand density index values from 0–500 (Figure 5.6). These results support earlier findings by Jameson (1967) for overstory–understory relations in ponderosa pine in northern Arizona and pinyon–juniper in northern and central Arizona. The understory within the ponderosa pine stands was 36 percent by weight Arizona fescue (Festuca arizonica) and 49 percent mountain muhly (Muhlenbergia montana).

Figure 5.6 Understory productivity of grasses, sedges, forbs, and shrubs by stand density index of ponderosa pine on the Kaibab Plateau, Arizona (redrawn from Moore and Dieter 1992).

Productivity declined with overstory basal area and with percent canopy cover; the decrease was steepest from 0 to 40 sq. ft. per acre (Figure 5.7) or 0 to 20 percent canopy cover (Figure 5.8).

Figure 5.7 Herbage production of Arizona fescue and mountain muhly in relation to basal area of ponderosa pine in northern Arizona (redrawn from Jameson 1967).

Figure 5.8 Herbage production of Arizona fescue and mountain muhly in relation to canopy cover of ponderosa pine in northern Arizona, (redrawn from Jameson 1967).

Figure 5.9 Herbage production of blue grama and broom snakeweed in relation to canopy cover of pinyon and juniper in northern and central Arizona (redrawn from Jameson 1967).

Although these relations should apply generally throughout the Southwest, herbage production at any given forest density will vary from area to area depending on the site's capacity and history.

Variations in overstory–understory relations are due to differences in fire regime, grazing history, species composition, climate, and parent material. The decline in productivity displayed in Figures 5.6–5.9 may not have been as steep if these stands had been more frequently burned (especially, the ponderosa pine) and not as heavily grazed (both the ponderosa pine and pinyon–juniper). A variety of cool-season grasses are well adapted to the pinyon–juniper understory, and some species even increase in abundance as canopy cover increases. Clary and Morrison (1973) find that cool-season grasses such as mutton grass (Poa fendleriana), squirreltail (Elymus elymoides), Junegrass (Koeleria macrantha) and western wheatgrass (Pascopyrum smithii) were more abundant in northern New Mexico under the canopy of large alligator junipers than in open spaces. Pieper (1994) reports pinyon ricegrass (Piptochaetum fimbriatum) and New Mexico needlegrass (Stipa neomexicana) were positively related to canopy cover. During his 12-year study, end-of-season, herbaceous, standing crops varied from 200 to 1000 kg per ha; differences in July–August precipitation explained over 40 percent of the variation in crop levels. Jameson (1966) examines differences in the effect of cover by one-seed juniper on growth of blue grama at sites either derived from granite or from limestone. Canopy densities up to 13 percent on the granite site and up to 31 percent of the limestone site had either no effect or even a positive effect on blue grama productivity. These relatively low canopy densities may have benefited the understory by reducing evapotranspiration or ameliorating temperature; higher tree densities may have impacted understory growth by shading or various belowground effects.

Although little is known of how belowground processes may have changed since the Territorial Period, mycorrhizae, nutrient cycling, and carbon cycling are fundamental ecosystem processes that strongly interact with both the overstory trees and the understory vegetation. Perry et al. (1989) describe the ecosystem relation between plants and soils, especially soil microbes, as a self-generating positive feedback system. A large share of the primary productivity of plants is allocated to roots and symbiotic fungi that form specialized root-fungus structures called mycorrhizae (Allen 1991, Brundrett 1991). Most plants and all conifers studied in natural ecosystems are mycorrhizal (Perry et al. 1989). Mycorrhizae increase plant nutrients and water and provide protection against some pathogens (Klopatek 1995). Mycorrhizae are important for seedling establishment and growth. Nutrient cycling, especially nitrogen which is essential for plant growth, depends on various soil microbes. Organic material in the soil affects its texture and water-holding capacity (Harvey et al. 1987). The breakdown and release of carbon in the soil is effected through shredding by arthropods and other macroorganisms, cellulose digestion by fungi, and additional decomposition by other microbes. Grazing affects these soil processes by reducing the carbohydrates available to mycorrhizae, the litter input to the soil, and the mulching effect of ground cover. Fire affects these processes by lethal temperatures and rapid mineralization. Both grazing and fire can lead to excessive soil erosion or leaching. As with various aboveground processes, the effects of disturbances and belowground responses vary by forest community.

Oak Woodland

Harrington (1985) claims that summer burns repeated at frequent intervals reduces the density of Gambel oak. At least in the lower elevations, oak density has probably increased during the last century because of fuelwood harvesting and decreased frequency and severity of fire (McPherson 1992).

Studies in the oak woodlands of southeastern Arizona document a decrease in the number of trees, particularly large trees, an increase in mesquite and juniper in the lower elevations, and a decline in grasses, all of which are related, at least to some degree, to human factors (Bahre and Bradbury 1984, Hadley et al. 1991). Fuelwood harvesting in historic times was a major extractive activity. Over-grazing was also prevalent, and fire suppression in more recent times no doubt affected the composition of the oak woodlands (Propper 1992).

The normal result of increased tree density in forested ecosystems is reduced density and diversity of other plant species. In many woodland areas, the ability of understory grasses to control erosion is not offset by greater tree density. The result is increased surface erosion with permanent loss of productive capacity for the site. If allowed to progress, it may be impossible to replace the original plant community in less than geological time. If understory species become extinct, then the original community can never be replaced.

Coniferous Woodland

Pieper and Lymberry (1983) describe changes to the coniferous woodlands:

"Prior to widespread settlement of the Southwest, pinyon–juniper stands were more open and were confined largely to the rocky ridges or more level sites with shallow soils–tree densities have increased with subsequent invasion into adjacent grasslands."

The pinyon–juniper woodlands may not have been as confined to rocky shallow sites as this description suggests. The General Ecosystem Survey identifies vast areas of established coniferous woodlands on deep, productive soil of level sites (i.e., GES Map Unit 130). But the GES data also support the view that the extent of coniferous woodlands has increased. Pieper and Lymberry (1983) attribute this increase to lack of periodic fires, increased spread of seed by livestock, overgrazing and resultant reduction in competition by grasses, and a shift in climate favoring the woody species. These factors have accelerated succession to woody species, especially during wet years favorable for regeneration of shade-tolerant trees (Johnsen 1962). Heavy grazing led to a reduction in the numbers and intensities of fire that has resulted in a significant expansion of junipers (Wright 1990). Since the historic period, coniferous woodlands have generally become more dense and extensive, primarily by expansion into shrub–steppe and grasslands. As in the oak woodlands, coniferous woodlands have experienced reduced diversity of understory species, increased erosion, reduced productivity, and possible permanent changes in dependent plant and animal communities. Invasion of other plant communities by woodland species reduces biodiversity on a landscape scale.

In some areas, coniferous woodlands have been manually converted to grasslands. From 1960 to about 1970, there was a widespread but controversial effort to convert woodlands to grasslands. de Buys (1985) describes an incident near Pecos, New Mexico where the Forest Service cleared 13,000 acres of woodland and reseeded them to grasses. This action was warmly received by grazing permittees, but upset village residents who depended on the area for fuelwood. These reactions typify the contrast in attitudes over woodlands–some people value woodlands for aesthetic or utilitarian reasons whereas others view woodlands as a mere impediment to livestock grazing (Lanner 1977).

Grazing has affected cryptogamic soil crust present in some woodland areas. Cryptogamic soil crust has a stabilizing influence on the soil surface (Bailey et al. 1973, Anantani and Marthe 1974a, 1974b). The increased porosity typical of crusted soils improves infiltration of rainwater, thus reducing runoff and subsequent erosion (Booth 1941, Loope and Gifford 1972). Additional benefits are increased fixation of atmospheric nitrogen and improved soil water and nutrient levels (Durrell and Shields 1961). Trampling of cryptogamic soil crust significantly disrupts this process. Research by St. Clair et al. (1984) indicates that an intact or recovered crust enhances seedling development, regardless of the crust composition.

Ponderosa Pine

Striking changes have occurred in the ponderosa pine forests of the Southwest. Over large areas, the open structure of historic forests has been replaced by dense thickets of sapling and pole stands (Harrington and Sackett 1990). A disturbance regime of frequent, low-intensity fires has been replaced with one of stand-replacing, high-intensity fires.

These changes had their beginning in the intensive livestock grazing of the late 19th century (Faulk 1970). Much of the herbaceous vegetation could not survive the grazing pressure, and its coverage declined drastically. Because of a decrease in fine fuels, this vegetation decline led to a reduction in fire spread that preceded organized fire suppression by one or two decades (Touchan et al. 1996). Forestry practices in the early 1900s further reduced the spread of fires. Because virtually all Southwestern fire history studies (Weaver 1951, Dieterich 1980, Swetnam 1990, Allen 1989, Savage and Swetnam 1990) report a similar pattern, a reduction in fire frequency of the early 20th century was probably common throughout the ponderosa pine forests of the Southwest.

In addition to a reduction in fires, ponderosa pine forests also experienced increased logging and reduced growth of herbaceous understory. With less herbaceous competition, improved soil conditions for seed germination, and better seedling survival, ponderosa pine regeneration, particularly in wet years, increased dramatically (Cooper 1960). Before, densities had been 3 to 56 trees per acre (Covington and Moore 1992); by 1986 ponderosa pine density on National Forest lands was 294 trees per acre (Johnson 1994). Many of these trees occur as stagnant thickets of saplings and poles; fuels are at unprecedented levels (Biswell et al. 1973).

An increase in stand density alone does not describe the changes to the forest structure of Southwestern ponderosa pine. Changes in size class distribution and numbers of large, yellow pine are also important but controversial. Although different areas were surveyed or different size classes were used, there are a number of inventories that can be used to examine changes in forest size structure (Figure 5.10). Woolsey (1911) reports the results of inventories on three national forests, and Lang and Stewart¹¹ report a inventory of what is now the North Kaibab Ranger District. These inventories can be compared with data of Conner et al. (1990) and Van Hooser et al. (1993) as re-compiled by Woudenberg (Intermountain Research Station, personal communication).

Figure 5.10 Change in size-class distribution of ponderosa pine 1910 to 1985. The number of trees per acre are estimated from several forest inventories circa 1910 (Lang and Stewart unpublished on North Kaibab Plateau and Woolsey 1911 in northern Arizona) and circa 1985 (Conner et al. 1990 in Arizona and Van Hooser et. al 1993 in New Mexico). Size classes for the early inventories are small, 4 to 18 inches dbh; medium, 18 to 24 inches dbh; and large, greater than 24 inches dbh. Size classes for the later inventories are small, 6 to 18 inches dbh; medium, 18 to 23 inches dbh; and large, greater than 23 inches dbh.

Early surveys by Woolsey (1911) and by Lang and Stewart and recent surveys by Conner et al. (1990) and Van Hooser et al. (1993) include relatively large acreages of Southwestern ponderosa pine and use similar size classes. The Woolsey (1911) survey represents 1,888 acres of well-watered, rolling malpais flats with good stocking (Coconino National Forest), 5,920 acres on southeastern exposure, with dry soil, volcanic cinder, and light stocking (Tusayan Forest), and 128 acres on the Prescott National Forest. Lang and Stewart surveyed eight acres per quarter section across the North Kaibab Ranger District. These early surveys grouped trees 4 to 18 inches in a small-diameter class, trees 18 to 24 inches in the medium-diameter class, and trees over 24 inches in the large-diameter class. Conner et al. (1990) reports inventory totals for all forested lands in Arizona (survey 1985), and Van Hooser et al. (1993) reports inventory totals for New Mexico (survey 1986-1987). These recent surveys grouped trees 6 to 18 inches in the small class, 18 to 23 inches in the medium class, and trees over 23 inches in the large class.

Changes in the stand structure of Southwestern ponderosa pine from the early 1900s to the 1980s are illustrated by size-class distributions (Figure 5.10) using data from Woolsey (1911), Lang and Stewart, Conner et al. (1990), and Van Hooser et al. (1993). The most obvious change has been the increase in small-diameter trees (less than 18 inches dbh) from 17 trees per acre to 140 trees per acre. The number of medium-diameter trees (18 to 23 or 24 inches dbh) increased slightly from 5 to 6 trees per acre. The number of large-diameter trees (over 23 or 24 inches), however, decreased slightly from 3 to 2 trees per acre. Because of disparities in survey area and methods, no statistical significance can be placed on these differences, but they probably well reflect actual changes from a forest dominated by few but large trees to a forest of many small and few large trees.

Probably the largest effect on forest health in Southwestern ponderosa pine is due to the increase in the density of the trees. This effect is expressed in several ways:

1. Increased tree density reduces the abundance and diversity of understory plants.

2. Since most of the increase is in smaller trees, there is an increase in ladder fuels, so that crown fires, once rare in ponderosa pine, are now common.

3. Increased tree density increases susceptibility to bark beetles.

4. Dense, multi-storied stands provide suitable conditions for rapid spread and intensification of dwarf mistletoe.

5. Increased density results in lower water yields, which has a negative impact on riparian areas.

In addition to increased density, ponderosa pine forests have tended to become more uniform, with the loss of structural and compositional diversity. We do not have good inventories of the amount of land in old-growth condition, either prior to European settlement or now. However, many people believe that there is less old growth now than there was before logging began. The lack of a commonly accepted definition of old growth and a lack of information on pre-settlement conditions makes this difficult to verify. If true, it may indicate that in recent times there have been too many minor disturbances and too few major disturbances. A major disturbance could move an area into a condition where the trees would be both young and would consist of early seral species. Minor disturbances like selective harvest or thinning could result in the loss of only certain species or certain age classes. They could move an area from late seral to a middle seral condition or from a stand of old trees to a stand of middle-aged trees but fail to take it all the way back to an early seral condition or a very young age. The lack of early and late seral conditions and young and old stands could cause reduced populations of plants and animals that depend on these conditions, thereby reducing the biodiversity of the forest.

All of this indicates that attempts by managers to reduce disturbances like low-intensity fires and periodic insect outbreaks are at the heart of the unhealthy conditions existing in ponderosa pine forests today. The future health of these forests demands more disturbance, not less. However, these disturbances must reflect the historic variability in terms of frequency, intensity, and size to insure that the resulting forest structure and composition are sustainable.

Mixed Conifer

Although we do not have the detailed early inventories for mixed conifer forests that we have for ponderosa pine forests, it is reasonable to assume that many of the same types of changes have also occurred there. Whereas fires previously occurred at intervals of 5 to 25 years (Table 4.1), they now occur at much longer intervals and with much higher intensity. As in ponderosa pine, increased overall density and increased numbers of small trees contribute to the frequency of crown fires and the susceptibility to bark beetle attacks. An additional change in mixed conifer stands, however, is a trend toward larger numbers of late seral species such as white fir and Douglas-fir and reduced numbers of early seral species such as ponderosa pine and aspen. Increased tree densities and multi-storied stands of late seral species are very favorable conditions for attack by the western spruce budworm. Increasing the proportion of Douglas-fir in a stand also makes it easier for Douglas-fir dwarf mistletoe to spread and intensify. Loss of ponderosa pine and aspen from mixed conifer stands further reduces biodiversity. Increased overall density will also reduce the abundance and diversity of understory plants.

The 1962 and 1986 forest inventories for the Southwest indicate the extent of meadows within the mixed conifer forest. In some areas, there is evidence that these open montane grasslands are shrinking, with some small openings disappearing altogether. In the Jemez Mountains of northern New Mexico, for instance, more than 55 percent of the high elevation grasslands have been lost since the early 1900s. This loss is attributed to conifer encroachment caused by overgrazing and the removal of fire (Allen 1989). Over 40 percent of the species found in meadows are not found in adjacent mixed-conifer forests. The fragmentation of meadow habitats into smaller, more isolated parcels limits the ability of the ecosystem to support historic wildlife populations.

Spruce–Fir

The spruce–fir forest has always been an area of both rare but high-intensity disturbances and frequent patch-scale disturbances. Grissino-Mayer et al. (1995) reports that a severe fire every 300 to 400 years returns the spruce–fir of the Pinaleno Mountains to an early successional stage. The spruce beetle, high winds, and root disease can also rarely cause stand-replacing disturbances or patch-scale openings in the spruce–fir forest.

Due to the low product value and the difficulty of building roads, relatively little logging was done before 1950 in spruce–fir forests (Alexander and Edminster 1980). Because these forests were neither roaded nor cut, large portions were designated as Wildernesses. Spruce forests outside of reserved areas, however, have been extensively logged over the past 50 years. Because the natural disturbance regime of spruce–fir forests includes infrequent but high-intensity events like windthrow, fire, and spruce beetle epidemics, the current stand age-class structure of the spruce forest is probably not outside the historic norm.

Aspen

Johnson (1994) reports a 46 percent decline in aspen forest cover during the period from 1962 to 1986. This decline represents a substantial reduction in landscape biodiversity. The enormous aspen stands reported by Pearson (1931) no longer exist, and the number of smaller ones has decreased. Aspen is an early seral species that requires high-intensity disturbances such as stand-replacing fires or clearcut logging for successful regeneration. Because elk, deer, and livestock heavily browse young aspen sprouts and can destroy entire areas of regeneration, restoration of aspen is difficult. Some forests (e.g., Coconino and Apache–Sitgreaves) have resorted to installing very expensive fences capable of excluding not only cattle, but also deer and elk. If large fires were to occur in the mixed conifer forest, so much of the area may regenerate to aspen that animal damage would be widely dispersed and aspen could re-establish. This has not yet been demonstrated. Once established, aspen stands can usually persist for about 100 years before they deteriorate and are replaced by conifers.

Although aspen is little used in the Southwest for timber (except locally for house logs), aspen is an important ecological feature and recreation attraction. Aspen groves provide important foraging, nesting, breeding, and resting sites for over 200 wildlife species. Fall coloration by aspen draws numerous visitors to the Southwest, especially northern New Mexico. If current trends of reducing acreage of aspen continue, these values would be lost.

Riparian Wetlands

In alpine riparian systems, including montane cienegas, the process of trapping sediments moving over channel bottoms by native aquatic vegetation is important to their long-term stability. Under impacts such as livestock grazing starting in the late 1800s, native ungulate grazing, and the reduction of aquatic plants, high-elevation stream systems initiated a process of degradation and stream-channel down cutting. Introduced graminoids such as Poa, Agropyron, and Dactylis are able to invade riparian areas. But they are less able to withstand the various impacts of grazing, and their root systems are less able to prevent channel aggradation (Neary and Medina 1995).

The floodplains–plains riparian systems of the Southwest have suffered more from human activities than other riparian associations. The floodplains portion is typically found on older, meandering river systems with extensive flats. One manifestation of human activity is the explosion of exotic shrubby species such as Russian-olive in the northern river sections and saltcedar in the southern parts of the region and northwestern New Mexico. The gallery forests that covered the floodplains riparian systems were composed of large trees and were often close to human settlements. Trees were initially cut for fuel and shelter. Riparian trees were also cut to clear land for agriculture and settlement. The natural resiliency of riparian vegetation might have eventually resulted in the restoration of the former forests, if it had not been for the impoundment of the streams and rivers.

The hydrology of dammed rivers was altered in ways that thwarted some of the reproductive mechanisms of native riparian species. Wells and drains dropped water tables and reduced the number and extent of phreatophytic habitats. Not only were flow rates below dams reduced, but annual flooding of benches and terraces bordering channels was eliminated. This drastically curtailed the major reproductive mechanisms for cottonwoods. The decline in the number of native trees and the altered available water regime on rivers and streams, coupled with livestock grazing, have eliminated some wet meadows and set the stage for the establishment of exotics such as Siberian elm (Ulmus pumila), Russian-olive, and saltcedar (Dick-Peddie 1993). Costs of eradication, if desirable, are high. For example, removal of all the saltcedar along the Lower Colorado River and restoration of native vegetation have been estimated at between $45 and $450 million (USDI Bureau of Reclamation 1992).

These impacts and others, such as the removal of beavers from many waterways, have caused an overall deterioration of riparian ecosystems. Those wildlife species dependent on riparian ecosystems that are currently at reduced population levels are at risk for further declines in population as threatened, endangered, or sensitive species (TES). The influence of brown-headed cowbird (Molothrus ater) parasitism to many small songbirds continues to be pervasive due to readily detectable nest sites caused by the lack of dense riparian vegetation. Insectivorous bird species that prefer slow-moving insect-rich boggy areas will continue to decline due to the lack of vegetation and beaver activity. The majority of species currently on the Regional Forester's Sensitive Species list for the Southwestern Region are riparian-dependent species or species which thrive in healthy riparian ecosystems.

WILDLIFE POPULATION DYNAMICS

In the past 150 years, significant changes have occurred to wildlife populations of the West. In general, some species have been extirpated, some have increased in abundance, and some new species have been introduced. Despite the lack of data, there is consensus that as resource management intensified, ecosystems were simplified, habitats were fragmented, and animal populations declined (Harris 1984).

Virtually all of the game animals are at significantly increased levels compared to the 1890s. Wildlife population estimates are difficult to obtain and are seasonally dependent. Populations in the spring, after the peak of reproduction and prior to juvenile mortality, can be several times greater than populations after hunting and winter mortality. Wildlife estimates for Arizona are estimates from the Arizona Game and Fish Department for post-hunt adults; estimates for New Mexico are from the New Mexico Department of Game and Fish. Since populations fluctuate yearly with variations from site to site, population estimates at this broad scale should be used in the context of understanding overall trends through time, not site-specific analysis.

Large predators were aggressively persecuted until as recently as 1960. Gray wolves (Canis lupus), mountain lions, and grizzly bears (Ursus arctos) were largely extirpated from most Western states. Mountain lion populations have rebounded in most Western states since 1970. Black bears and mountain lions are managed through regulated hunts in New Mexico and Arizona. Black bear numbers were estimated to be 770 in New Mexico in 1926 and increased under hunting regulation to about 3,000 by the early 1960s (Ames 1967). Information from ongoing research indicates that the current New Mexico population may be 4,000 to 6,000 animals. The black bear population in Arizona has apparently remained stable over the last 10 years and is estimated at about 2,500 animals.

Thomas and Toweill (1982) report on the history of elk populations. Unregulated hunting, including market hunting, reduced elk populations to low numbers by the 1890s; a few years later, elk were extirpated from Arizona. Tragically, Merriam's elk (Cervus elphus merriami), a subspecies found in southern New Mexico and Arizona, became extinct by 1900. Rocky Mountain elk were extirpated from northern New Mexico around 1909. Reintroduction efforts began in 1910 and continued through 1966 (Gates 1967). After 1900, following a four-decade period of low or no legal hunting, elk populations began to recover. Today, elk numbers in some areas appear higher than during any period in recorded history (Irwin et al. 1993). The 1994 estimate of elk in Arizona was 29,000 adults. In New Mexico, elk numbers were estimated to be 712 in 1926 and grew to about 11,000 to 12,000 in the early 1960s (Ames 1967). The current elk population in New Mexico is estimated at 30,000 to 50,000 adults.

Deer populations are more variable in number. Irruptions occur periodically in which large populations are observed that subsequently crash. Causes for these irruptions were varied and probably interdependent (Caughley 1976, Pengelly 1976, Mitchell and Freeman 1993). Causes include such factors as predator control, protection from hunting, drought, disruptions of the natural fire regime, logging, and competition with cattle. In Arizona, from extreme low numbers in the 1890s, deer peaked about 1960. After an extended drought in the 1970s, they peaked again in the mid-1980s. The current population is again at the relatively low level of the 1970s. 1994 Arizona population estimates for mule deer and white-tailed deer (Odocoileus virginianus) are 120,000 and 82,000 respectively. In New Mexico, mule deer reached a low population around 1926 at an estimated 41,000 animals. Numbers increased until about 1960 at an estimated 318,000 animals (Ames 1967). Mule deer populations fluctuate greatly due to the influence of weather on reproduction and survival. Post-hunt population numbers for all age classes have probably fluctuated between 235,000 to 300,000 animals during the past 10 years.

In the late 1890s, big game seasons were closed for bighorn sheep and pronghorn in Arizona. At this time, mountain bighorn sheep were estimated at no more than 1,500 animals in Arizona. By the early 1900s, mountain bighorn sheep were extirpated from New Mexico (Sands 1967). Restoration of bighorn sheep began in 1932 in New Mexico and continues to the present day. Currently, five populations exist totaling an estimated 570 animals, with the majority located in the Pecos Wilderness. Desert bighorn sheep in the San Andres and Big Hatchet Mountains in New Mexico reached a low in about 1955. Transplants have been made to the Peloncillo, Hatchet, and Ladrone Mountains between 1979 and 1993 from the Red Rock captive population. Total numbers in 1994 are estimated at 360 animals. The 1994 estimate of bighorn sheep in Arizona is 6,500–about 6,000 desert bighorn sheep and about 500 mountain bighorn sheep.

In the Southwest, pronghorn numbers reached a low in the early 1900s. At that time, there were an estimated 1,200 animals in New Mexico (Larsen 1967). The transplant program begun in 1937 has been responsible for reintroducing pronghorn to most suitable habitats in New Mexico. The 1993 New Mexico population estimate is 36,000 animals (Gleadle 1994), and the 1994 Arizona estimate is 7,800. Pronghorn populations are subject to great fluctuations due to changes in habitat quality.

Javelina in New Mexico received protection as a game species in 1937 (Donaldson 1967). Numbers were then estimated to be 400 and 3,500 by 1964 (Ames 1967). It has been noted that their distribution has been expanding through natural dispersal to the San Andres and Guadalupe Mountains. Current numbers are unknown but may be increasing. The Arizona estimate for javelina in 1994 is 38,000 animals.

Wild turkey populations can fluctuate widely from year to year. In New Mexico, turkey numbers reached a low by about 1926 and were estimated to be 21,000 statewide. The population was estimated to be about 25,000 in 1964. No current statewide estimate has been made. In Arizona, there are currently between 30,000 and 40,000 turkeys.

Another influence that altered Southwestern ecosystems was the removal of beaver by aggressive fur trapping. Tens of thousands of beaver were removed from riparian areas. Their removal encouraged channelization and reduced the hydrologic effect of pooling. The loss of beaver affected native fish such as the Apache trout (Salmo apache) and Rio Grande cutthroat trout (Salmo clarki), and birds dependent upon insect-rich boggy conditions. It also narrowed the riparian corridor due to decreased water infiltration. In many areas of the Southwest, beaver have not returned to many historic waterways.

Numerous activities of management and resource use have had detrimental effects on populations of native fish. Poisoning "undesirable" fish and stocking popular game species such as the rainbow trout (Salmo garidneri), reduced native populations and introduced competitors and predators. Mining, timber harvesting, and livestock grazing have all affected water quality and riparian habitat. In the warm and sunny Southwest, riparian vegetation is especially important for cooling and shading forest streams. The native fish most sensitive to these changes include the Apache trout, Gila trout (Salmo gilae), Little Colorado spinedace (Lepidemeda vittata), spikedace (Meda fulgida), loach minnow (Tiaroga cobitis), and Sonora chub (Gila ditaena).

Each forest type and its respective successional stages have specific wildlife communities associated with them. As cover type and/or successional stages change, there are corresponding changes in the composition of wildlife communities. The BISON-M habitat relationships database was queried to analyze the numbers of birds and mammals in different cover types and successional stages to evaluate effects of vegetation trends. Seventy-three percent of birds and mammals found in grasslands are not found in pinyon–juniper woodlands. Therefore, increasing acreages in pinyon–juniper woodland will favor these species. Likewise, shifts from ponderosa pine to mixed conifer also affect wildlife composition.

The rapid change of aspen to mixed conifer is one of the most significant concerns to wildlife management. Although both communities have many species in common, there is a distinct suite of birds and mammals that occur only in aspen. In addition, aspen provides a broadleaf component that is important to certain invertebrates, such as the western tiger swallowtail (Papilio rutulus) and red-spotted purple (Basilarchia astyanax) butterflies.

The size and abundance of meadows is declining in the Southwest as conifers encroach. Forty-six percent of the species found in meadows are not found in mixed conifer forests. The fragmentation of meadows limits the habitat's ability to support wildlife species that require large meadows.

AIR QUALITY

In Arizona the number of areas that do not meet National Ambient Air Quality Standards has increased over time, while by comparison air quality in New Mexico remains more pristine. However, efforts to maintain good air quality in New Mexico have increased, as demonstrated by mandatory no-burn policies during air temperature inversions that affect Albuquerque. Most Forest Service resource inventory and monitoring for air pollution effects has focused on establishing current conditions. Few long-trend analyses have been conducted. However, the National Acid Precipitation Assessment Program and the Interagency Monitoring of Protected Visual Environments (IMPROVE) network have summarized long-term data for western states.

Terrestrial and aquatic ecosystems and visibility have been affected by atmospheric pollutants. Sensitivity to acidic deposition is a function of geologic characteristics of the watershed, soil and vegetation type, hydrologic characteristics, chemical and biological characteristics of the ecosystem, and precipitation volume. Mercury, which could be from the atmosphere, and other trace metals have bio-accumulated in fish tissue, and health advisories have been issued in some areas. A loss of acid-neutralizing capacity due to high sulfur or nitrogen deposition has been documented in the Southwest. Episodic acidification has been detected at some high elevation lakes or streams. Paleobotanical reconstructions infer changes in lake species that may relate to changes in lake chemistry caused by atmospheric pollutants. These changes indicate a shift to more acid-tolerant species may be occurring in some high elevation lakes in New Mexico. Ozone, which is toxic to plants, is also a concern, particularly near large urban areas where combustion from various sources generates ozone precursors.

Sulfate aerosols from all sources (including power plants and copper smelters) and other fine particles and gases in the atmosphere impair scenic vistas in the Southwest. For example, data from the Interagency Monitoring of Protected Visual Environments network show that in autumn sulfur concentrations have steadily increased since 1980 at Grand Canyon National Park. The National Academy of Sciences (1993) state that "the average visual range in most of the western United States...is about one-half to two-thirds of the natural visual range that would exist in the absence of air pollution."