Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Ecological Effects of Forest Clearcutting Management of the harvesting of renewable biological resources re- quires knowledge of the stocks of those resources and of the rates at which they recover after harvesting. Gaining this knowledge is especially difficult when the intervals between cropping are long and rates of recovery are low, as they are in the case of timber harvesting. Not only are trees long- lived, but differences between sites make it difficult to extrapolate results from one location to another. This case study reviews the ways in which these problems have been approached and assesses the current state of prediction of the amounts and significance of nutrient losses from forests after clearcutting. 345
Case Study CARL F. JORDAN, Institute of Ecology, University of Georgia, Athens, Georgia INTRODUCTION A major goal of the forestry profession is to sustain the production of wood in forests. In the early days of the profession, much attention cen- tered on prevention of fire and outbreaks of insect pests as means of sustaining high productivity. These factors still cause extensive losses of trees, but now that many forests are managed as agricultural crops, other problems have increased in importance. One of the common intensive management techniques in modern forestry is clearcut logging. Clearcut logging is often more economical than other forest management tech- niques, but it has the potential to create serious environmental problems out of effects that would be negligible if less intensive techniques were used. Examples include the elimination of habitat of endangered species and increases in soil erosion. This case study reviews research that has evaluated the impact of forest clearcutting on nutrient stocks and site productivity. The first studies of nutrient loss due to clearcutting were concerned primarily with only parts of the nutrient budget of an ecosystem, for example, leaching losses. Improved understanding of ecosystem func- tioning made it apparent that useful interpretation of nutrient losses due to clearcutting required an analysis of both the nutrient stocks in the ecosystem and the nutrient inputs and losses. To evaluate the effect of nutrient loss during clearcutting on future productivity of a site, it is necessary to know the quantities not only of nutrient losses due to leaching and erosion after clearcutting, but also of losses due to tree removal, volatilization, and fixation in the soil. Also important are the rate of nutrient replacement through atmospheric input, rock weathering, and fertilization and the size of nutrient stocks in the ecosystem that can buffer short-term fluctuations in input and output. A nutrient-budget approach to forest management is analogous to budget management in business. Successful business management demands an understanding of the entire operating budget, including revenues, expen- ditures, and financial balance. This need in business management is so obvious as to be taken for granted. Yet, in forest management, it has become clear only recently that successful management for sustained yield requires knowledge of nutrient inputs, nutrient losses, and nutrient stocks. 346
ECOLOGICAL EFFECTS OF FOREST CLEARCUITING 347 These quantities are often difficult to measure, but complete budgets must be constructed, if the impact of clearcutting on nutrient loss is to be evaluated adequately. This case study examines how such understanding evolved. BACKGROUND We do not know when the first agriculturalist observed that ash, litter, and animal carcasses improved crop growth. Undoubtedly, the knowledge that the addition of particular materials to the soil increases productivity arose independently in many regions. However, only in the nineteenth century, after chemical elements had been identified, did it become un- derstood that nutrients constituted the common factor in soil amendments that maintained agricultural productivity (Brady, 19741. The first scientist to make a systematic study of nutrient circulation in forests in relation to growth of trees was Ebermayer. The objective of his study, "Complete Treatise of Forest Litter," was to explain the adverse effect on forest quality of the litter removal then common in middle European forests (Tamm, 19791. The litter was used in cow stables, and some of the plant nutrients in it eventually reached arable fields in dung, thus contributing to food production at a time when commercial fertilizer was not available. However, forest growth decreased, particularly in al- ready poor sites, and Ebermayer attributed this to the export of plant nutrients (Tamm, 19791. Interest in the role of nutrients in forest growth continued throughout the early part of the twentieth century, but only within the last several decades have scientists begun to quantify total forest-ecosystem nutrient budgets and changes in budgets due to various logging practices. One reason for interest in nutrient loss resulting from the clearcut method of forest harvesting is the obvious increase in soil erosion, nutrient leaching, and consequent stream eutrophication after such operations (Tamm et al., 19741. Another, which is examined here, is the effect of nutrient loss on the ability of the soil to supply enough nutrients to produce another stand of trees. Observation of soil erosion and increased eutrophication of drainage streams after clearcutting left little doubt that nutrient loss was occurring, but did not indicate whether it was great enough to affect productivity of the site. Because tree growth responds to fertilization in at least some cases, nutrient loss is potentially harmful. But nutrients lost through leach- ing and erosion might be quickly replaced by weathering and other pro- cesses. No long-term studies have been designed specifically to test the effects of nutrient leaching on site productivity. Studies at Hubbard Brook,
348 SELECTED CASE STUDIES New Hampshire (Bormann and Likens, 1979), showed that a northern hardwood forest that had been clearcut and then treated with herbicide experienced substantial amounts of nutrient leaching, but recovery of the forest through natural succession did not appear to be noticeably inhibited by lack of nutrients. Clearcuaing has rarely resulted in nutrient loss so serious that trees could not grow on the denuded site. Occasionally, on very steep slopes in regions of heavy rainfall, such as some areas in western British Co- lumbia, clearcutting has been followed by soil erosion severe enough to expose bedrock, and sites have consequently been permanently deforested (J. P. Kimmins, personal communication). In most cases, however, at least some tree growth has occurred after clearcutting, so the environmental question is not whether trees can grow after clearcutting, but how fast they can grow. The rate is important to industries or agencies concerned with timber as a crop, because trees that grow rapidly yield a greater profit than trees that grow slowly. Rate of forest recovery also is important for nonmarket values of forests, such as the reduction of soil erosion on watersheds, scrubbing of polluted air, support of fish and game, and provision of habitat for some rare species (Farnworth et al., 19811. APPROACHES TO EVALUATING LOSSES Leaching arid Erosion Nutrient leaching losses often increase as a result of clearcutting. After removal of trees, evapotranspiration on the site decreases, so the amount of water percolating through the soil increases; the result is an increase in leaching potential. A decrease in nutrient recycling is also important in increasing leaching. In undisturbed forests, root uptake of nutrients and their incorporation in biomass reduces nutrient loss. Nutrients from de- composing litter and soil organic matter on a clearcut are not taken up by trees, but are exchanged on the clay surfaces of mineral soil, where they are susceptible to loss through leaching and erosion. These ecological ideas are commonly accepted and are discussed in many texts on ecology, agronomy, soil conservation, and forestry. However, emphasis often is on the effect of the nutrients on lake and stream eutrophication and fish productivity, rather than on future productive capacity of the site losing the nutrients. Recent studies have revealed some of the mechanisms by which nutrients are leached. After clearcutting, the activity of nitrifying bacteria in the soil increases (Likens et al., 1969; Vitousek et al., 1979), owing to increased temperatures, decreased competition for ammonium from tree
ECOLOGICAL EFFECTS OF FOREST CLEARCUlTING 349 roots, decreased allelopathic inhibition of nitrifying bacteria, and other factors (Reiners, 1981; Swift et al., 19791. As a result, ammonium from the mineralization of organic matter is oxidized to nitrate, and nitrate anions and nutrient cations that have been exchanged for hydrogen ions on soil surfaces are rapidly leached. An important effect of forest disturbance, especially in mountainous terrain, is soil erosion. Soil erosion results from exposure of mineral soil to the direct impact of raindrops. The impact breaks soil aggregates, causing pores and channels in the soil to be filled, the soil surface to become less permeable, and the surficial runoff to be greater than when soil surfaces are covered with litter. Although soil erosion often accom- panies clearcut logging, recent studies have shown that it is construction of access roads, not the clearcutting itself, that causes most of the erosion (Douglass and Swift, 19771; removal of logs by cable appears to lead to less erosion. A general approach to determining nutrient loss during and after clear- cutting of forests and during site preparation for a new stand has been to measure leaching and erosional losses from clearcut areas directly and to compare losses with those in undisturbed areas that serve as a control. Measurements have usually been made where drainage from a watershed flows over a weir resting on bedrock, so that subsurface drainage is insignificant. Continuous monitoring of water flow and nutrient content permits measurement of total nutrient loss. Clearcut watersheds are com- pared with control watersheds or watersheds subjected to other treatments, such as conversion to grassland. In some cases where discrete watersheds were not available, losses have been studied by means of lysimeters (soil water collectors). Losses have also been measured by comparing soil and ecosystem nutrient stocks before and after cutting or between cut and control plots. Many studies have shown increases in rate of nutrient loss during forest disturbance. For example, the studies of the watersheds at Coweeta Hydro- logic Laboratory in North Carolina showed higher rates of nitrate and sediment loss in recently cut catchments than in control catchments (Monk 1975; Swank and Douglass, 1975, 1977; Webster and Fatten, 1979~. Studies of nutrient dynamics in conifer forests of the Pacific Northwest also showed an increase in loss rates after clearcut harvesting and slash burning (Feller and Kimmins, 1984; Gessel and Cole, 1965; Miller and Newton, 19834. An important study of nutrient loss after clearcutting was carried out in northern hardwoods at the Hubbard Brook watershed site in New Hamp- shire (Bormann and Likens, 1970; Borrnann et al., 1968; Likens et al., 1970~. Bormann et al. (1968) concluded that
350 clear-cutting tends to deplete the nutrients of a forest ecosystem by (i) reducing transpiration and so increasing the amount of water passing through the system; (ii) simultaneously reducing root surfaces able to remove nutrients from the leaching waters; (iii) removal of nutrients in forest products; (iv) adding to the organic substrate available for immediate mineralization; and (v) in some instances, producing a microclimate more favorable to rapid mineralization. SELECTED CASE STUDIES This conclusion was controversial, because clearcutting at the Hubbard Brook site had been followed by herbicide treatment to prevent regrowth of vegetation. Aubertin and Patric (1974) wrote that "there is a substantial difference between the Hubbard Brook treatment and conventional clear- cutting. Conventional clearcutting also features complete forest cutting; but all saleable wood is harvested and rapid forest regeneration is en- couraged." In an experimental hardwood watershed in West Virginia, Aubertin and Patric measured negligible nutrient losses after clearcutting that was carried out to resemble conventional clearcut logging techniques. Other clearcutting studies in the Hubbard Brook region showed that nu- trient losses are less if herbicides are not used after clearcutting, but can still be important because the soils in the region are shallow (Pierce et al., 19721; harvesting removes a larger proportion of the total nutrient pool there than in areas with deeper soils. Because of the importance of nutrient uptake by vegetation, alternating contour strip cuts with undisturbed forest on a mountainside should de- crease nutrient leaching. Experiments with strip cuts showed that this technique can reduce nutrient leaching in northern forests (Hornbeck et al., 19751. The effect should be even greater in the humid tropics, where the potential for leaching is extremely high (Jordan, 19821. Once vege- tation becomes re-established in the clearcut strips, the uncut strips can be harvested. ~. Biomass Removal Nutrient losses due to leaching are brief, and often small, compared with those due to tree removal (Cole and Bigger, undated; Hornbeck and Kropelin, 1982; Kimmins, 1977; Sollins and McCorison, 1981; Swank and Waide, 19801. As evidence accumulated that nutrient losses due to biomass removal during clearcut operations were often much greater than losses due to leaching and erosion, scientists shifted their attention to studies of nutrient stocks in forests and soils and to losses from ecosystems due to biomass harvest. Results from various sites and management strat- egies were presented in several major symposia (Ballard and Gessel, 1983; Leaf, 19791. Generalizing about the studies is difficult, because each site had its own combination of soil type, soil nutrient stocks, biomass nutrient
ECOLOGICAL EFFECTS OF FOREST CLEARCU7TING 351 stocks, management technique, rates of nutrient input via rainfall, nitrogen fixation, fertilization, and nutrient losses through leaching and denitrifi- cation. Generalizations valid for sites of a specific type in a given region do not apply to sites of other types. Each nutrient also appears to behave differently. Only by measuring the stocks and dynamics of all the critical nutrients at each site can one estimate accurately the impact of nutrient removal on productivity of the site. Even then, estimation might be in- sufficient, because trees can exhibit uptake beyond immediate needs, and this can be misleading in the prediction of requirements. One general finding is that the extent of nutrient loss depends on which parts of the trees are removed from the site. Leaves have higher concen- trations of nutrients than do stems, so stripping logs of leaves before the logs are removed from the site decreases nutrient losses. For evergreens, branches with leaves are cut off before the tree bole is removed. In the case of deciduous species, an alternative is to harvest during the winter. However, because some nutrients are withdrawn from leaves into stems before the leaves are shed, nutrient losses would be smaller if trees were cut while in full leaf and mechanically defoliated. Harvesting of whole trees, including roots, removes larger proportions of nutrients. Nutrients Remaining in Soil Simply measuring the nutrients lost because of clearcutting and site preparation does not indicate the effect of the nutrient loss in site pro- ductivity. Estimates of the quantities of nutrients remaining in the site and of the rates at which they are replenished are also needed. Scientists attempting to predict the effect of nutrient loss on site productivity must determine nutrient stocks remaining in the soil and the fractions of those stocks available to growing trees (Johnson et al., 19821. Nutrients are held in the soil in various ways (Brady, 19741. Some cations, such as potassium, can be part of the lattice structure of minerals, where they are relatively unavailable to roots; or they can be exchanged on clay surfaces, where they are readily available. Phosphorus can exist in relatively soluble forms, but, in the presence of low pH and high aluminum concentrations, it becomes bound in compounds that are not readily taken up by plants. Soil can contain large amounts of nitrogen; if the nitrogen is bound in organic matter that is resistant to decomposition, however, nitrogen shortages might be critical. Thus, knowing the total stocks of nutrients in the soil is not sufficient. For many agricultural crops, reagent-soluble nutrients in soil have been correlated with crop growth, and solubility of nutrients has been used as an index of the availability of nutrients to crop plants. However, such an
352 SELECTED CASE STUDIES index usually is not applicable to tree growth. Many tree species are adapted for extracting nutrients from soil when "availability," as mea- sured by solubility, is low. Some of these adaptations are large root biomass, mycorrhizal symbiosis, slow root uptake, and long life, which enable trees to survive during periods of nutrient shortage (Chapin, 19801. The effects of these variables are not easy to measure. Nutrient Replenishment Site productivity depends on relative rates of nutrient replenishment and nutrient loss. The atmosphere contains a stock of nutrients adsorbed on the surface of aerosols, such as dust and pollen, that either settle out of the atmosphere gradually as "dry fall" or are washed out by precipi- tation. Nutrient input from the atmosphere can be substantial. Nitrogen can be contributed to a site by nitrogen-fixing species, which can be present because they occur naturally or through planting of species symbiotic with nitrogen-fixing bacteria. Nutrients can also enter an ecosystem through the weathering of subsoil or parent rock. Nutrients lost through clearcutting can be replaced by fertilizers, but fertilization of forests is often economically infeasible. For most agricul- tural crops, fertilization is profitable, because annual sales are high and the time between investment in fertilizers and return of investment in crop sales is short. The economic benefits of fertilizing forests are more difficult to calculate, because the value of harvestable products that accumulate each year is low, because the period between investment and harvest is long, and because forests have other values besides their use for wood. Budgets arid Sensit~vi~ Analysis Nutrient budgets of ecosystems have proved extremely difficult to mea- sure precisely, because of high variability in both space and time. In addition, such important quantities as the proportions of nutrients available to plants and the rate of mineral weathering are often hard to measure; errors in their measurement could result in large errors in estimates of total ecosystem nutrient budgets. Therefore, these estimates are often viewed skeptically by both scientists and managers. The key question, however, is whether the error is important in relation to the environmental or ecological problem being addressed. As an example of a simple sensitivity analysis to assess the effect of biomass removal on remaining nutrient stocks and the effect of error on the predictions, consider the following hypothetical case. Assume that the standard deviation around the amount of calcium in the soil is relatively
ECOLOGICAL EFFECTS OF FOREST CLEARCU7TING 353 large and that the average value of a set of samples could be different from the true average value by 30%. Is this error important? If calcium were present in the aboveground biomass at 500 kg/ha, all of which would be removed by clearcutting, and present in the soil at 5,000 kg/ha, an error of 30% in the estimate of calcium in the soil would not change the conclusion that clearcutting will not have an important effect on total calcium stock. If calcium were present at 500 kg/ha in the biomass and 100 kg/ha in the soil, an error of 30% would not change the conclusion that clearcutting would have a very important effect on total calcium stock. However, if calcium were present at 500 kg/ha in both the biomass and the soil, it would be important to be able to estimate more accurately the true value of calcium in the soil, because a 30% error could lead to very different management conclusions. This relatively simple type of analysis has yielded predictions of nutrient depletion due to biomass removal in several regions. For example, calcium depletion due to clearcutting of ridge forests in the southern Appalachians might result in reduced growth of the next crop (West and Mann, 1983), and nitrogen is often a limiting nutrient in the conifer forests of the Pacific Northwest (Peterson and Gessel, 1983~. Simulation models of forest growth that incorporate the effects of nu- trients can assist in answering questions about the required degree of accuracy of field measurements and experiments. Only recently have mod- els with increased sophistication been developed. A model like the nitrogen model of Swank and Waide (1980), which predicts yield under various management strategies on the basis of nitrogen dynamics in the treated ecosystems, can be used to assess the effect of an error of a given mag- nitude on the overall conclusion and can aid in a decision as to whether more accurate measurements are necessary. The latter model permits eval- uation of management alternatives and their consequences, such as the effect on forest yield of a change in the length of rotation. The importance of sensitivity analyses is often not appreciated by eco- logical and environmental scientists. Perhaps the reason is that sensitivity analyses often contradict what many scientists have been taught, which is that the greatest possible accuracy should always be sought. Sensitivity analyses show us where a large amount of less accurate data might be more valuable than a smaller amount of more accurate data. CONCLUSION Studies to evaluate the effect of clearcutting on nutrient depletion and site productivity have progressed a long way. Early studies were concerned with nutrients removed by leaching and erosion after clearcutting. Then
354 SELECTED CASE STUDIES it was recognized that nutrient stocks remaining in the ecosystem after logging were important, and studies of those stocks were carried out. Nutrient dynamics and their effects on stocks and productivity were in- cluded in the analyses. Recent work has used systems analysis to keep a better account of the multitude of continuously changing factors in the ecosystem. The proof of the effectiveness of the approach will lie in tests of the predictions generated by the models. From the perspective of applications, ~ , the most important prediction is that of wood yield. Because of the time required for such validation, it Is too early to appraise the approach. Regardless of how accurate the predictions prove to be, they will be useful to the scientists who formulated the models on which they were based. Discrepancies between predictions and results will point up weaknesses in the models and indicate what studies are necessary to improve the accuracy of the predictions. REFERENCES Aubertin, G. M., and J. H. Patric. 1974. Water quality after clearcutting a small watershed in West Virginia. J. Environ. Qual. 3:243-249. Ballard, R., and S. P. (vessel, eds. 1YbJ. lUPKU ~ympv~luIll all rulers am alla ~V11~111~O Productivity, Seattle, Washington, August 22-28, 1982. PNW-163. U.S. Department of Agriculture Forest Service, Portland, Greg. Bormann, F. H., and G. E. Likens. 1970. The nutrient cycles of an ecosystem. Sci. Am. 223(4):92- 101. Bormann, F. H., and G. E. Likens. 1979. Pattern and Process in a Forested Ecosystem. Springer, New York. Bormann, F. H., G. E. Likens, D. W. Fisher, and R. S. Pierce. 1968. Nutrient loss accelerated by clear-cutting of a forest ecosystem. Science 159:882-884. Brady, N. C. 1974. The Nature and Properties of Soils. Macmillan, New York. Chapin, F. S. 1980. The mineral nutrition of wild plants. Annul Rev. Ecol. Syst. 11:233- 260. Cole, D. W., and C. M. Bigger. Undated. Effect of Harvesting and Residue Removal on Nutrient Losses and Productivity. Fifth Annual Report. University of Washington College of Forest Resources, Seattle. Douglass, J. E., and L. W. Swift. 1977. Forest service studies of soil and nutrient losses c.q.~sed bv roads. mechanical site preparation, and prescribed burning in the Southeast. IN 489 5()] in D. L. Correll, ed. Watershed Research in Eastern North America. A _ _ _ , , ~ ~ 1_ _ ~ ~_. ~_~ ~ =~ I;-tal ~til~iP~ Workshop to (compare Results. ~nesapeaxe nay w~llr~l lU1 l~illVllVlil~l~llt~ ~ A, Edgewater, Maryland. Smithsonian Institution, Washington, D.C. Farnworth, E. G., T. T. Tidrick, C. F. Jordan, and W. M. Smathers. 1981. The value of natural ecosystems: An economic and ecological framework. Environ. Conserv. 8:275- 282. Feller, M. C., and J. P. Kimmins. 1984. Effects of clearcutting and slash burning on streamwater chemistry and watershed nutrient budgets in southwestern British Columbia. Water Resour. Res. 20:29-40.
ECOLOGICAL EFFECTS OF FOREST CLEARCU7TING 355 Gessel, S. P., and D. W. Cole. 1965. Influence of removal of forest cover on movement of water and associated elements through soil. J. Am. Water Works Assoc. 57:1301- 1310. Hornbeck, J. W., and W. Kropelin. 1982. Nutrient removal and leaching from a whole- tree harvest of northern hardwoods. J. Environ. Qual. 11:309-316. Hornbeck, J. W., G. E. Likens, R. S. Pierce, and F. H., Bormann. 1975. Strip cutting as a means of protecting site and streamflow quality when clearcutting northern hard- woods. Pp. 209-225 in B. Bernier and C. H. Winget, eds. Forest Soils and Forest Land Management. Proceedings of the Fourth North American Forest Soils Conference. Les Presses de l'Universite Laval, Que. Johnson, D. W., D. C. West, D. E. Todd, and L. K. Mann. 1982. Effects of sawlog vs. whole-tree harvesting on the nitrogen, phosphorus, potassium, and calcium budgets of an upland mixed oak forest. Soil Sci. Soc. Am. J. 46:1304-1309. Jordan, C. F. 1982. Amazon rain forests. Am. Sci. 70:394-401. Kimmins, J. P. 1977. Evaluation of the consequences for future tree productivity of the loss of nutrients in whole-tree harvesting. For. Ecol. Manage. 1:169-183. Leaf, A. L., ed. 1979. Impact of Intensive Harvesting on Forest Nutrient Cycling. Pro- ceedings of a Symposium at Syracuse, New York, August 13-16, 1979. Northeast Forest Experiment Station, Broomall, Pa. Likens, G. E., F. H. Bormann, and N. M. Johnson. 1969. Nitrification: Importance to nutrient losses from a cutover forested ecosystem. Science 163:1205-1206. Likens, G. E., F. H. Bormann, N. M. Johnson, D. W. Fisher, and R. S. Pierce. 1970. Effects of forest cutting and herbicide treatment on nutrient budgets in the Hubbard Brook watershed-ecosystem. Ecol. Monogr. 4Q:23-47. Miller, J. H., and M. Newton. 1983. Nutrient loss from disturbed forest watersheds in Oregon's coast range. Agro-Ecosystems 8:158-167. Monk, C. D. 1975. Nutrient losses in particulate form as weir pond sediments from four unit watersheds in the southern Appalachians. Pp. 862-867 in F. G. Howell, J. B. Gentry, and M. H. Smith, eds. Mineral Cycling in Southeastern Ecosystems. ERDA Symposium Series. CONE 740-513. U.S. Energy Research and Development Administration, Wash- ington, D.C. Peterson, C. E., and S. P. Gessel. 1983. Forest fertilization in the Pacific Northwest: Results of the regional forest nutrition research project. Pp. 365-369 in R. Ballard and S. P. Gessel, eds. IUFRO Symposium on Forest Site and Continuous Productivity, Seattle, Washington, August 22-28, 1982. PNW-163. U.S. Department of Agriculture Forest Service, Portland, Oreg. Pierce, R. S., C. W. Martin, C. C. Reeves, G. E. Likens, and F. H. Bormann. 1972. Nutrient loss from clearcuttings in New Hampshire. Pp. 285-295 in Watersheds in Transition. American Water Resources Association Symposium. Colorado State Uni- versity, Ft. Collins, Colo. Reiners, W. A. 1981. Nitrogen cycling in relation to ecosystem succession. Pp. 507-528 in F. E. Clark and T. Rosswall, eds. Terrestrial Nitrogen Cycles. Proceedings of a Workshop. Ecological Bulletins (Stockholm) 33. Swedish Natural Science Research Council (NFR), Stockholm. Sollins, P., and F. M. McCorison. 1981. Nitrogen and carbon solution chemistry of an old growth coniferous forest watershed before and after cutting. Water Resour. Res. 17: 1409-1418. Swank, W. T., and J. E. Douglass. 1975. Nutrient flux in undisturbed and manipulated forest ecosystems in the southern Appalachian mountains. Pp. 445-456 in Symposium
356 SELECTED CASE STUDIES de Tokyo. Publication 117 de ['Association Internationale des Sciences Hydrologiques, Reading, Eng. Swank, W. T., and J. E. Douglass. 1977. Nutnent budgets for undisturbed and manipulated hardwood forest ecosystems in the mountains of North Carolina. Pp. 343-364 in D. L. Correll, ed. Watershed Research in Eastern North America. A Workshop to Compare Results, Chesapeake Bay Center for Environmental Studies, Edgewater, Maryland. Smithsonian Institution, Washington, D.C. Swank, W. T., and J. B. Waide. 1980. Interpretation of nutrient cycling research in a management context: Evaluating potential effects of alternative management strategies on site productivity. Pp. 137- 158 in Forests: Fresh Perspectives from Ecosystem Analysis. Proc. 40th Annul Biol. Colloq. Oregon State University Press, Corvallis. Swift, M. J., O. W. Heal, and J. M. Anderson. 1979. Decomposition in Terrestrial Ecosystems. Studies in Ecology. Vol. 5. University of California Press, Berkeley. Tamm, C. O. 1979. Nutnent cycling and productivity of forest ecosystems. Pp. 2-21 in A. L. Leaf, program chairman. Impact of Intensive Harvesting on Forest Nutrient Cy- cling. Proceedings of a Symposium at Syracuse, New York, August 13-15, 1979. North- east Forest Experiment Station, Broomall, Pa. Tamm, C. O., H. Holmen, B. Popovic, and G. Wiklander. 1974. Leaching of plant nutrients from soil as a consequence of forestry operations. Ambio 3:211-221. Vitousek, P. M., J. R. Gosz, C. C. Gner, J. M. Melillo, W. A. Reiners, and R. L. Todd. 1979. Nitrate losses from undisturbed ecosystems. Science 204:469-474. Webster, J. R., and B. C. Fatten. 1979. Effects of watershed perturbation on stream potassium and calcium dynamics. Ecol. Monogr. 49:51-72. West, D. C., end K. L. Mann. 1983. Whole-Tree Harvesting: Fourth Year Progress Report for 1982. Nutnent Depletion Estimates, Postharvest Impacts on Nutnent Dynamics and Regeneration. Environmental Sciences Division Publication No. 2184. Oak Ridge Na- tional Laboratory, Oak Ridge, Tenn. Committee Comment As illustrated by this case study, methods of analyzing nutrient budgets of ecosystems have developed slowly as scientists gradually have rec- ognized the need to perform a complete accounting of nutrient stocks and fluxes if the significance of changes in flux rates induced by clearcutting is to be understood. This represents discovery of methods that were already well known in the business world, and it might be asked whether ecologists would have progressed more rapidly if they had been better versed in budget analyses as practiced in various disciplines. The answer is probably yes, but it is also clear that ecosystems are so different from businesses that those techniques have to be modified for use in studying ecosystem nutrient dynamics. Such features of ecosystems as differences in availa- bility of nutrients due to variation in soil types and in the types of plants growing on them have no close parallels in the business world. Whereas much progress has been made in measuring total stocks and fluxes of nutrients, site variation is great enough that very few general
- ECOLOGICAL EFFECTS OF FOREST CLEARCU7TING 357 predictions are possible. There is no reason to believe, in principle, that developing a broad predictive ability is impossible, but it seems evident that the necessary data base is large and that its components might not yet all be identified. That is true even if we simply wish to predict growth rates of trees. If we also wish to predict the dynamics of other constituents of an ecosystem, such as herbivores, carnivores, and detritivores as well as interactions among plant species even more extensive data are re- quired. Results of such analyses could be surprising. For example, lower availability of nutrients, in combination with physical stress, such as drought, might render trees much more susceptible to attacks by defoliating insects or fungi (Fearnside and Rankin, 19851. These attacks might de- crease yields and change growth forms of the trees, leaving yields of usable wood products much lower than would be predicted simply on the basis of the availability of nutrients to support tree growth. An important conceptual advance illustrated by this case study is sen- sitivity analysis. Sensitivity analysis is especially important when systems consist of many interacting factors that are of uncertain influence or that are difficult to measure accurately. Time and budgetary constraints prevent accurate measurement of all factors of interest, and choices need to be made as to which factors should receive the most attention. Predictions generated by models are unequally sensitive to variation in different pa- rameters. Sensitivity analysis is a powerful method for deciding which factors need to be measured most accurately and which ones require only crude estimation. Sensitivity analysis helps to avoid consequential errors errors that could lead to inappropriate management decisions. The natural inclination of most scientists to measure all quantities as accurately as possible might actually lead to poorer predictive abilities, given the in- vestment of comparable resources, than an approach designed to obtain accurate measurements only for the quantities identified as critical. The study of ecosystem nutrient dynamics is made difficult not only by the number of interacting factors, but also by the long duration of the most important processes. Development and testing of models might re- quire decades of work that taxes the patience of investigators, funding agencies, and managers who must make decisions, whether or not suf- ficient data are available for them to predict the consequences of their decisions. This emphasizes the need for cooperation among all who are involved in developing plans for managing and using forest ecosystems. Reference Fearnside, P. M., and J. R. Rankin. 1985. Jari revisited: Changes and the outlook for sustainability in Amazonia's largest silvicultural estate. Interciencia 10:121-129.