Humaninduced Changes And Their Effects On Biodiversity And Ecosystem Function

3.3.1 Climate and atmosphere

Climate warming The anticipated warming from atmospheric loadings of radiatively active gasses is probably the greatest threat to the boreal region worldwide. Although the global circulation models disagree on the details of the extent of warming and drying, all agree that the greatest warming will take place in high-latitude regions, and most agree that mid-continent areas will become drier while maritime areas will become wetter (Schlesinger and Mitchell 1985). Climate warming will affect the boreal regions through a myriad processes: (1) local mortality of boreal species, to be replaced by northern hardwoods or prairies depending on localc and soil type (Pastor and Post 1988); (2) migration of boreal species northward and coastward, also depending on locale and soil type (Davis and Zabinski 1992; Pastor and Johnston 1992); (3) increased probability of fire (Clark 1989); (4) either increased or decreased soil nutrient availability, depending on permafrost, soil water holding capacity and locale (Pastor and Post 1988; Bonan et al. 1990); (5) increased loading of greenhouse gasses, particularly methane, from wetlands (Gorham 1991; Roulet et al. 1992; Updegraff et al. 1995); (6) increasing probability of outbreaks of pests, particularly insects, as trees are drought-stressed and become more susceptible (Mattson and Haack 1987).

Needless to say. all these changes enter into ecosystem and regional feedbacks that affect diversity (Hobbie et al. 1993; Figures 3.1 and 3.2). Because of the strong non-linearities inherent in these systems, no generalizations can be made. While some have suggested that the boreal forest could simply be shifted northward (Emanuel et al. 1985), it appears more likely that warming will be more rapid than the migration rates of the major species and large-scale regional extinctions may result (Davis and Zabinski 1992). Given the patterns of species assembly in the pollen record reviewed above, it is also possible that entirely new species assemblages will arise. Feedbacks from fire, soil nutrient availability and herbivores could accelerate or prevent certain pathways of change, or cause entirely new ones to appear.

Acid deposition There is a huge literature on the effects of acid deposition on the growth and species composition of strands at the southern boundary of the boreal forest, particularly downwind from industrial areas (e.g. Drablos and Tollan 1980; Binkley et al. 1989; Malanchuk and Nilsson 1989; Schulze et al. 1989). The trends are not entirely clear, particularly with regard to diversity, although dieback of conifers, especially at high altitudes continuously bathed in acid fog, appears to be common. Even less clear are the mechanisms (Manion and Lachance 1992). Research has centered on the role of nitric acid in determining forest response. Nitrogen is the most limiting nutrient to productivity in boreal forests (Weetman 1968; Van Cleve and Oliver 1982) and it is possible that dieback could simply be a symptom of nitrogen deficiency induced by the slow decay of spruce litter, leading to spruce being replaced by birch (Pastor et al. 1987). On the other hand, the nitrogen in acid deposition might be a fertilizer, and cause nitrogen saturation, which in turn leads to dieback because of physiological changes in the trees (Aber et al. 1991). In any case, acid deposition remains a serious problem and, although the potential for harm is yet to be proven, there appear to be no beneficial effects on forest growth or biodiversity.

3.3.2 Resource management

Human influence on the boreal forest is as old as the extant, post-Pleistocene forest formations themselves. Some of the pre-modern ways of forest utilization had regional ecological consequences. For instance, swidden agriculture spread into northern European forest many centuries ago and left clear traces in the pollen record. The effect of swidden agriculture and grazing of cattle in forest pastures was a dominant worry in the early evaluations of forest resources all over Fennoscandiavia in the 19th century.

Another human-induced ecological change in boreal forests that started in the pre-modern era is the population decline and extermination of fur-bearing mammals from the western and southwestern Palearctic since the 16th century (Kirikov 1960) and the eastern Nearctic since the 17th century (MacKay 1967), This led to a permanent reduction in the southern range of large mammals such as moose and brown bear (Ursus arctas) in the western Palearctic, and to a decline in predator populations (e.g. the martens; Martes amerieana in North America and M. martes and M. zibellina in Eurasia) with secondary effects on their prey. The decimation of beaver from the last portions of its previous range removed an important factor in landscape dynamics (Naiman et al. 1986). However, most boreal mammal species are not currently endangered except for furbearers, that have been overexploited, predators considered to be pests, or animals requiring large tracts of relatively undisturbed ecosystems. The latter category includes species such as wolves and woodland caribou, particularly at the southern border with northern hardwoods in North America. This is a result of the intense human exploitation of these regions in comparison to the subarctic forests of Russia, Siberia, Alaska, the Yukon, northern Quebec-Labrador and the Northwest Territories of Canada, which by modem standards still contain large expanses of wilderness. However, because of the generally great resilience of boreal species (Holling 1992), numbers of greatly rcduced populations rebound rapidly once direct exploitation stops {Broschart et al. 1989).

Logging also has a major effect on the diversity of boreal regions. Cutting of timber from boreal forests for household purposes extends back to times immemorial, but extensive cutting that selectively removed the most highly valued tree species (e.g. white pine in US New England and the Lake States; Mladenoff and Pastor 1993) began in the early modern era. Commercial logging currently dominates human-induced changes in southern boreal forest landscapes, and logging may extend to all parts of the boreal forest zone by the end of this century. Systematic forestry originated in central European coniferous forests to forestall regional timber shortage in the 18th century (Radk.au 1983) and spread to ail boreal countries by the early 20th century. However, management methods have undergone changes during process. Modern silviculture, aiming at the establishment of monoculture, even-aged strands growing in uniform conditions created by soil preparation, artificial fertilization and chemical pest control, took over during the decade following World War II. The negative ecological effects of intensive silviculture have become manifest; thus, a timely challenge for ecologists is to give recommendations for forestry practices that would better conform with natural forest dynamics (Pastor and Mladcnoff 1992; Mladenoff and Pastor 1993; Haila 1994). As forests change continuously in several time-scales, human-induced change per se is not a problem; what is at issue is the dynamic relationship between human-induced and natural change (Haila and Levins 1992).

Direct changes in the dominant tire regime, through active fire suppression, have been most important in regions where human land-uses now predominate (Baker 1989, 1993). These changes have altered landscape-scale patch structure and age-class distributions (Baker 1989), as well as composition and within-stand structure (Heinselman 1973, 1981; Baker 1989). In many cases these changes interact with ecosystem processes to further alter future sucees-sional pathways and forest productivity (Fig. 3.2). These alterations occur because of unnatural fuel accumulation, differences in tissue chemistry between conifers and hardwoods, and large regional changes in the mammal and insect fauna (Pastor and Mladenoff" 1992; Mladenolfand Stearns 1993).

Changes in the proportional areas of different types of forest, whether because of logging or tire, had had a major effect on the distribution and abundance of breeding birds, as documented by long-term census data from Finland and North America (Jtirvinen and Vgisanen 1977, 1978; Jarvinene et at. 1977; Helle 1984; Virkkala 1987, 1991; Telfer 1992). Fairly subtle changes in habitat structure can have large ecological consequences if they occur uniformly over large areas (Haartman 1978; Jarvinene et at. 1977; Haila et at. 1980; Hellc & Jarvinen 1986). Logging has different impacts on neotropical migrants in North America than in Eurasia because the largest proportion of neotropical migrants in North America inhabit older-aged stands, while in Eurasia neotropical migrants are more abundant in younger-aged stands (Helle and Niemi 1994). Loss of habitat for neotropical migrants in North America also has important consequences for tropical forests. Neotropical migrants are an important link between boreal and tropical regions, and land-use practices in either one affect the other through habitat for these functionally important species.

"Fragmentation" (that is, a change in spatial configurations of forest patches from a more continuous to a less continuous form) in the boreal forest zone is primarily caused by silviculture or fire. Fragmentation in boreal regions differs from that in agricultural areas because forest patches arc not entirely isolated. Because fragmentation in boreal regions changes the spatial distribution and juxtaposition of age classes and communities, it can have important consequences for the diversity of guilds and ecosystem processes. For example, fragmentation in forested parts of the taiga is directly harmful primarily for winter-resident birds, which depend on some structural characteristics of old growth in critical phases of their life cycle; for instance, the capercaillie (Tetrao urogallus) (Wegge et al. 1992) and Siberian tit (Parus cinctus) (Virkkala 1990) in the western Palearctic.

Equally important are the indirect, secondary effects of forest fragmentation which change the "matrix" and bring about new juxtapositions and interactions between species, and thus also changes in ecosystem functions (Saunders et al. 1991; Haila et al. 1993; Mladenoff et al. 1993; Haila 1994). In forests surrounded by farmland, nest predation rate is elevated (Andren and Angelstam 1988). Increased vole densities in clearcuts also increases the population densities of their predators (Rolstad and Wegge 1989). Indirect effects of forest fragmentation on ecosystem functions in boreal forests and are probably related to edge effects and the distribution of deciduous trees and saplings in the forest landscape (Angelstam 1992).

General assessments of the effect of forestry on invertebratas are given by Niemela et al. (1993, 1994a), Heliovaara and Vaisanen (1984). Mikkola (1991) and Esseen et at. (1992). As the habitat requirements of invertebrates are usually related to the quality of microsites within forest stands, their population trends are primarily influenced by changes stand structure. As we have already indicated above, two mutually interdependent aspects need to distinguished here, namely, change in average strand structure that has a regional effect on species abundance, and loss of particular, specific micro-sites locally (Siitonen and Martikainen 1994).

3.4 RESEARCH NEEDS 3.4.1 Long-term surveys

There is an urgent need for systematic survey research on the biological diversity of different ecological complexes (Haila and Kouki 1994). As already pointed out in this chapter, the diversity of boreal regions has both a spatial and a temporal dimension thus surveying existing patterns of biodiversity and monitoring changes in biodiversity are two sides of the same task. Research on different scales is needed simultaneously (Caughley 1994).

The main requirement in survey research is that it be systematic, both concerning the coverage of the area surveyed and the methodology. Taxonomic knowledge of species in the boreal forest zone is more complete than that for most parts of the world, but data on quantitative distribution patterns of different taxa across environmental gradients and natural patterns of variation in populations and communities are urgently needed. This would allow assessment of the role of deterministic factors such as habitat, soil and climate, and stochastic factors related to local population dynamics and vagaries in the movements and fates of individual organisms. Such data are a necessary baseline for predicting changes that follow from changes in forest structure, and for giving recommendations concerning forest management practices.

All aspects of forest biodiversity cannot be captured at the same time. Surveys need to be focused on particular groups of organisms, such as functional groups. Thus, it is doubtful whether a single comprehensive biodiversity survey program is feasible. When data accumulate, comparisons across taxa become possible, and the generality of patterns inferred from particular groups can be tested.

Because of the large-scale spatial dynamics that characterize the boreal forest, remote sensing and geographic information systems (GIS) can be useful tools for guiding long-term surveys in several ways. Satellite imagery is ideal because of its synoptic coverage spatially and temporally. These data can be useful in addressing improved forest habitat mapping and classification. Research in this area needs to include application of new techniques to classify forests at the species level (Wolter et al. 1995), and detection of habitat structural characteristics. Also addressable at these scales are detection of changes in landscapes, measuring ecosystem properties and processes, and providing input into large-scale spatial modelling of processes and landscape change (Hall et al. 1991; Ranson and Williams 1992; Miadenoff and Host 1994).

3.4.2 Experiments

With respect to biogeography, the Holarctic boreal forest is not homogeneous. It contains two distinct latitudinal zones that for historical and climatic reasons are biogeographically distinct, and within cach of these zones there are six longitudinal zones that can be distinguished biogeographically (Figure 3.3). This biogeographical complexity suggests that the experimental studies of human-induced perturbations of boreal forest ecosystems use a nested experimental design, with the six longitudinal regions as the first level of nesting, and the two latitudinal zones within these regions as the second level of nesting.

We suggest that the most likely large-scale disturbances to the Holarctic boreal forest resulting from human activity will be (1) increasing wildfire resulting from warming, (2) changes in herbivore densities, including increases in insect outbreaks in evergreen conifer forests caused by stresses to trees resulting from a combination of warming and increased atmospheric pollution, and (3) the effects of increased logging. These three perturbations

Longitudinal Gradient

E. Siberia Alaska Central

North America

Eastern North America

Fenno- Trans-Urals Scandia

North (permafrost)

South (non-permafrost)

Latitudinal Gradient

Burn or Log

No-Burn, Harvesting, No-Log Fire

Gradient

Exclosure

Herbivores

Mammalian Herbivore Gradient

Further Treatments as Needed:

1. C02 enhancement

2. Soil warming

3. Acid precipitation

4. Kill conifers (insect outbreak)

Experimental Responses:

1. Litter decay and nutrient availabi'ity

2. Tissue chemistry

5. Phenology

Figure 3.3 Nested experimental design to test the interactions of ecosystem processes and biodiversity across boreal regions suggest as primary experimental treatments (I) fire, (2) mammalian herbivore exclosures or selective killing of evergreen conifers to stimulate insect outbreak (Crawley 1989), and (3) experimental logging be conducted in each of the six longitudinal and two latitudinal zones in a nested experimental design.

We also suggest the effect of these three perturbations on ecosystems will be to initiate secondary successions whose rate and outcome will be strongly controlled by interactions between mammalian herbivory, plant growth and establishment, and decomposition processes of plant litter, and secondarily affected by interactive effects of warming, C02, and atmospheric pollutants (NO,, SO*) on plant growth, plant chemistry and decomposition processes. This suggests further nesting within the above design, such as additions of CO2. atmospheric pollutants and soil warming.

This design has several attractive features: (1) It will answer questions about human-induced disturbances to ecosystems of the Holarctic boreal forests that are already of international concern, and also whether the same disturbances are likely to be induced by future human activity. (2) In answering these questions it will take into account gradients in biodiversity and ecosystem function that characterize the boreal forest. Furthermore, by taking into account longitudinal variations, this experimental design has the further advantage of encouraging collaborative multinational experimentation. (3) In addressing these questions, it recognizes the importance of interactions between ecosystem processes, natural disturbances and human perturbation of the environment. (4) The experimental design is realistic. Similar experiments now being conducted by the Alaskan taiga long-term ecological research project demonstrate that the experiment is technically tractable.

3.4.3 Models of boreal forests

The patterns and dynamics that characterize the diversity and ecosystem processes of boreal regions cannot be fully explained by comparative studies or manipulative experiments. As previously noted several times in this chapter, the oscillations of populations, patterns of diversity, and ecosystem processes in boreal regions occur over decades and even centuries. This is far longer than most experiments. Furthermore, the analysis of variance traditionally used to evaluate experimental data cannot fully elucidate the non-linearities underlying such dynamics because it is based on a linear model. Finally, the expected magnitude and rate of global warming is such that no comprehensive experiment is possible, and long-term observations in the absence of experiments and models will simply establish the development of patterns, not necessarily their mechanisms.

One approach would be simply to develop regressions of various diversity indices against environmental variables. However, such an approach is static and does not capture the interactions between plants, resources and herbivores that control the changes in diversity noted in the experiments cited above. Such regression models are descriptive, not explanatory.

To obtain a full view of the functioning of boreal ecosystems and the relationships between plant and animal species life characteristics, diversity and rates of element flows, these experimental and observational data need to be assembled into a simulation model. Such a model should encapsulate the sequence of events and important flows of information and nutrients that comprise feedbacks between organisms and their environment.

There are several candidate models that do precisely this at a number of scales. At the levels of interactions among individuals within communities, there are the boreal versions of the JABOWA/FORET models which have been used to simulate the effects of global warming and timber harvesting on diversity and element cycling (Pastor and Post 1986, 1988, 1993; Bonan and Shugart 1989; Bonan et a!. 1990; Cohen and Pastor 1991, 1995; Pastor and Mladenoff 1993). At larger landscapes scales, a variety of approaches are being developed for different purposes. These include simple Markov models (Baker 1989; Hall el al. 1991; Pastor et al. 1993b) and more spatially dynamic GIS models that include stochastic processes such as disturbance and seed dispersal (MladenofT and Host 1994), These models are still in their infancy, and need further forma! mathematical analysis before their behavior can be fully understood (Cohen and Pastor 1991, 1995). However, they are a framework for incorporating experimental and observational data into the theoretical view of organismal-ecosyxtem feedbacks discussed throughout this paper.

Was this article helpful?

0 0
Waste Management And Control

Waste Management And Control

Get All The Support And Guidance You Need To Be A Success At Understanding Waste Management. This Book Is One Of The Most Valuable Resources In The World When It Comes To The Truth about Environment, Waste and Landfills.

Get My Free Ebook


Post a comment