THE NEW JARI: RISKS AND PROSPECTS OF A MAJOR AMAZONIAN DEVELOPMENT PHILIP M. FEARNSIDE and JUDY M. RANKIN
•J ari", the group of indicating that any suggestion to apply in virtually all aspects of the silviculture three enterprises 1 de Jari as a model must be approached with and pulp operations is, and must be, the veloped by shipping extreme caution. The financial crisis at continued supply of wood for pulping magnate D. K. Ludwig, is an estate in Jari leading up to Ludwig's decision to and power generation on a day-to-day Brazil's State of Para and Federal Terri sell (Kinkead, 1981) has now focussed basis. tory of Amapa claiming 1,632,121 ha 2 debate in Brazil in the risks and prospects Native timber, originally of land straddling the Jari River, a north of J ari itself. burned on the site, is now heavily uti ern tributary of the lower Amazon. Sil One viewpoint expressed lized 6 as fuel to fire the boilers supply viculture, pulp manufacture, ric.e grow at recent international meetings (Russell, ing power to the pulp mill, as a com ing, mining and logging are among 1981; Alvim, 1981) is that Jari is itself ponent of the pulp produced by the Kraft Jari's main activites. The Jari project is "technically viable", even though political pulp mill on the site, and as construction currently at a major "crossroads" as final and institutional problems might have materials. Plantation monocultures of details are settled for the sale of this been rendering the Ludwig operation three species supplant the native forest: tremendous operation, with control pass untenable. Such claims must be examined Gmelina arborea has been planted com ing to a consortium of 27 Brazilian closely. The question of Jari's future mercially since 1969, Pinus caribaea var firms led by mining entrepreneur Augus course under new management requires hondurensis since 1973, and Eucalyptus to Trajano de Azevedo Antunes (Veja, examination in light of interdependent deglupta since 1979. The silvicultural 16 de dezembro de 1981). Continuing ecological, economic and social factors plantations continue to be managed with heavy losses s have been a major factor affecting contemporary developments in out artificial fertilizer inputs to balance in Ludwig's decision to sell the project, Amazonia. The present paper is an at nutrient losses due to the harvesting of in which he had invested over US$ 1 tempt to bring new data, garnered from native timber and the plantation species. billion, for the much lower figure of US$ recent site visits 5 and the growing num Continued removal of nutrients from the 280 million (US$ 180 million in debts ber of scientific publications on research system in the form of harvested biomass assumed + US$ 100 million payable over performed at J ari, to bear in the analysis will eventually deplete stocks of available three years) 4. of these factors. nutrients to the point where growth is J ari has often been sug reduced to uneconomic levels. Company gested as a potential model for develop Prospects of Jari's Operations foresters say that a slight decrease in ments throughout the Amazon Basin. Silviculture, Timber and Pulp growth rates has already been noted in While many details of Jari's operations the second cycle of Gmelina. The in have changed since we addressed the Plantation silviculture, evitability of a future added cost for question of Jari's applicability as a devel native timber extraction and pulp mill fertilizer purchase and application, not opment model for the region (Fearnside operation, are the principal activities at currently being considered in silvicultu and Rankin, 1980), the arguments ad Jari in terms of capital invested, land, ral calculations, has yet to be faced by vanced in that discussion remain valid, and employment. The controlling factor project managers.
Dr. Philip M. Fearnside, an ecologist, received his Ph. D. in Biological Sciences from the University of Michigan, Ann Arbor, Michigan U.S.A. in 1978. He is especially interested in the estimation of carrying capacity for human agricultural populations. His work bas been directed at this problem in the Brazilan Amazon since 1973. He is presently a Research Professor at the Instituto Nacional de Pesquisas da Amazonia (INPA), Caixa Postal 478, 69.000 Manaus Amazonas, Brasil. Dr. Judy M. Rankin, an ecologist, received her Ph.D. in Biological Sciences from the Uni versity of Michigan, Ann Arbor, Michigan, U.S. A. in 1978. Her main interests lie in the dynamics of tropical rain forest communities. She is especially concerned with the demography of tropical rain forest trees and natural regeneration, a process holding great potential for sustained yield forestry under Amazonian conditions. She is presently an Associate Research Professor at the Instituto Nacional de Pesquisas da Amazonia (INPA), Caixa Postal 478, 69.000 Manaus, Amazonas, Brasil.
lf\1Ef'IDffiDFI NOV- DEC 1982, VOL. 7 NQ 6 SEPARATA 0378- 1844/82/060329- 11 $ 3.00/0 329 A predictable, sustained of Gmelina at present levels would have kaolin mill ( 4.6%), and the towns of supply of wood is essential to the pulp to be greater than the amounts absorbed Monguba and Monte Dour ado ( 4.6%). mill's functioning and to Jari's existence. by the planted trees, since much of what At present, the boilers consume 1500 kg The cost of down time at the pulp mill is applied is lost to leaching and runoff, wet weight/day of fuelwood chips to is US$ 14,000/hour, meaning that all or, in the case of phosphorus, to fixation supply 80% of the power output, the other operations at the estate must be into unusable compounds in the soil. The other 20% coming from fuel oil (Woess coordinated to guarantee that the mill amounts lost vary greatly, depending on ner, 1980a). 9 Assuming that oil con never has to stop for lack of wood either soil, rainfall, type of fertilizer, and mode sumption cannot be lowered below 20%, for pulp or fuel. This overriding priority and timing of application. The recovery the amount of wood needed for fuel was the motivating force for the testing efficiencies used in Table I are rules of would be a 37.5% addition to the pulp and eventual inclusion of native tree thumb for crops in general. The highly wood need, or a 28.7% addition for only species in Gmelina pulp when early leached soils and high rainfall char the portion of the power output used by Gmelina production was found to be acteristic of the Amazon could mean the mill itself. insufficient to meet pulp mill demands. that recovery would be lower at The costs of fertilization The urgency of feeding the mill also led, Jari. Phosphorus fixation rates (in 6 per hectare must be converted to costs in 1981, to some of the Pinus caribaea hours at 100 ppm P) range from 26.8 per metric ton of pulp produced, since plantations with spacings intended for to 51.6% in representative soils of the this is the unit in which the added ex 16-year saw log rotations being cut for Brazilian Amazon (Fassbender, 1969). pense of fuelwood plantations, among pulpwood, while others were converted These rates are not high by the standards other factors, becomes apparent, and is to faster growing Eucalyptus deglupta of many tropical latosols, but more phos also the unit closest to the market prices (C. E. Russell, personal communication, phorus is lost to fixation at low (and ultimately determining the profitability October 1981). Plantations can be of three types: pulpwood, solid wood, and fuelwood. Costs of producing seedlings and establishing plantations are similar In silvicultural operations generally, many potential management remedies for the three types, but the values of the are not solvable at acceptable costs through traditional three products are very different, solid wood being much more valuable than agricultural remedies such as chemical treatments. pulpwood, which is more valuable than fuel wood. In the U.S., wood in the form of lumber has a value 3. 6 times that of wood in the form of fuel (Pimentel et al., 1981: 334). The need to keep the mill more probable) fertilizer application of fertilizing to achieve a sustainable supplied at all times means that oppor rates. In terra roxa (ALFISOL) in Al nutrient balance. The US$ 142,47 /ha/ tunities to produce more valuable pro tamira, Para, the same soil type as Jari's year cost of purchasing and applying ducts must be sacrificed to plant pulp best Gmelina-growing soils, up to 83% fertilizers to pulp plantations would co wood, and in the future firewood, plan of P applied is fixed in 7 days at low rrespond to US$ 9.87 I metric ton dry tations. Losses of potential income from (53 ppm P) application rates (Dynia wood produced at the 38 m3 /ha/year this situation will become more acute et al., 1977). maximum yield level (Woessner, 1980a) when native forest is exhausted. To the cost of purchas and the 0.38 g/ cm3 wood density It is then critical to ing fertilizers one must add the cost of (Woessner, 1980b). The cost would be know how much plantation, at present application. Information is not available US$ 3.79/metric ton of wet wood chips, levels of productivity, is needed to supply on the amount of labor required to since Gmelina,, green wood weight is ap the pulp mill, and the prospects for the fertilize Gmelina plantations, but an idea proximately 260% of dry wood weight, sustainability of such productivity. Plan can be gained from labor requirements according to company technical staff. tation productivity below expectations for cacao plantation fertilization. INCRA The cost of fertilizing the pulpwood has been one of Jari's most persistent (Brazil, Ministerio da Agricultura, IN plantations per metric ton of pulp pro· problems: yields have been, on average, CRA, 1972: 168) uses a figure of 2 man duced would be US$ 21.08, using a con approximately half those anticipated by daysjha/year to fertilize cacao on the version rate of 0.18 metric tons of pulp/ Ludwig at the outset of the project. 7 Transamazon Highway planted at the metric ton of wet wood chips (calculated Finally, information on plantation re same 3 x 3 m density used (after thinn from Woessner, 1980a, indicating that quirements must be translated into pulp ing) for Jari's Gmelina. A portion of present daily mill output of 720 metric production costs and evaluated in terms the cost of supplying this additional labor tons pulp is produced from pulping 4000 of world markets for the pulp produced can be calculated from the workers' pay. 8 metric tons/day of (wet) chips). Adding and for the inputs required. Labor for fertilizing would add US$ 8.48 the 28.7% extra plantation area needed A rough idea of the /hajyear to the costs, raising the total for producing fuelwood on a sustainable costs of supplying soil nutrients removed to US$ 142.47/hajyear. basis for power used directly by the mill, in the plantations can be gained from the In addition, if plan the cost per metric ton of pulp increases composition of the biomass removed in tation wood were to fuel the boilers of to US$ 27.13. the case of Gmelina (Table I) . The the mill's power plant on a sustainable The actual cost would nutrients in the Gmelina bark greatly basis, fertilizer for the fuelwood plan be higher than the figure computed increase the total drain from harvesting, tations would have to be included in the above, since many other costs have not as concentrations are substantially higher calculation. Of the 55 megawatt output been included in the calculation, such in bark than wood for many elements. of the power plant, 76.5% is consumed as transporting the fertilizer, planning, The amount of fertilizer directly by the pulp mill, the balance supervising, administering, and monitor needed to achieve continued production being used for the sawmill (5.5% ), ing its application, and supplying nutri-
330 NOV- DEC 1982, VOL. 7 Nl' 6 II\7ER[)fn[IR ents other than the four considered here. TABLE 1: The Gmelina wood and bark removed REPLACEMENT COSTS OF NUTRIENTS EXPORTED IN GMELINA BIOMASS annually would contain 44.64 kg/ha magnesium, 5.13 kg/ha sodium, 721.1 Item Units N p K Ca Total g!ha iron, 270.7 g/ha manganese, 737.7 g/ha zinc, and 43.3 g/ha copper. Nutrient Content At present price levels, Gmelina wood % dry weight <') 0.13 0.02 0.23 0.15 the US$ 27.13/metric ton cost increase Gmelina bark % dry weight<') 0.78 0.04 0.44 1.35 would represent 6% of the US$ 476.40 Eroded soil <') ppm 1720 140.28 14.50 7.33 average price received for Jari's pulp out Rainwater <') ppm put, but would represent a substantially <0.052 <0.011 <0.10
IIVERDEilOR NOV- DEC 1982, VOL. 7 Nr.> 6 331 would not solve the problem, since the TABLE II: nutrients contained in these plant parts PULP MILL LIQUID EFFLUENT CONTENT(l) are needed to remake the same plant parts in each successive plantation cycle. Added It must be remembered that the plantation Jari River at J ari River at Pulp mill concentration Annual tre.e)s are not capable of extrancing all of Monte Dourado Mondugaim- effluent from pulp Loss from the nutrients present in the soil, many above mills, mediately above in outlet mill in pulp mill nutrient forms being unavailable to the mine and towns pulp mill canal effluent (metric Item (ppm) (ppm) (ppm) (ppm) tons) (2) plants, in addition to the difficulty of capturing all of the stocks of those nu NH+ trients which are in available forms. 4 0.25 0.30 3.35 Data available on soil 2.36 (N> 80.3 (N) nutrient concentration changes in the NO~ <0.005 0.10 0.06 } plantations to date are not sufficient to 3- identify or forecast nutrient depletion 504 <1.00 <1.00 <1.00 0 (S> 0 (S) rates. Through 1980 Jari maintained 22 p 22.1 observation plots, with periodic analyses <0.10 <0.10 0.75 0.65 of soil pH, organic matter, available Cl- 2.50 5.00 312.50 307.50 10,462.5 phosphorus, potassium, calcium, magnesi um, sum of exchangeable bases ( Ca + + Ag <0.05 <0.05 <0.05 0 0 + Mg++ + Na+ + K+) and cation exchange capacity (exchangeable bases AI <0.20 <0.20 8.86 8.66 294.7 + H+ + AI+++). Soil analyses were done before felling the virgin forest, after B 0.30 <0.10 <0.10 0 0 clearing, after burning, and in subsequent Ba 0.29 <0.10 0.14 0.04 1.4 years. Unfortunately, data are only a vailable for the first three years under Ca 9.87 1.92 68.46 66.54 2,264.0 plantations, as the oldest plots are now six years old and the company Cd <0.02 <0.02 <0.02 0 0 switched from annual analyses to a scheme of analyses every three years Cr <0.02 <0.02 <0.02 0 0 prior to deactivation of this and most Cu other research activities as a result of <0.02 <0.02 <0.02 0 0 the financial crisis of 1981. As is usual Fe <0.10 0.17 1.25 1.08 36.7 in other types of agriculture in the region involving burning the forest (Fearnside, K 12.00 5.00 116.25 111.25 3,785,2 1978), the results show sharp increases as a result of burning for pH, phospho Mg 1.26 0.93 5.39 4.46 151.7 rus, potassium, calcium, magnesium, exchangeable bases, and cation exchange Mn <0.01 <0.01 0.10 -0.09 -3.1 capacity, and a decrease in toxic alumi Na 8.12 2.82 205.70 202.88 6,902.9 num. Organic matter also increased some what with burning. In the year following Pb <0.05 <0.05 <0.05 0 0 the burn the levels of these soil quality indicators quickly return towards their Si 29.81 5.80 15.06 9.26 315.1 pre-burn levels, some (especially organic matter) falling below the virgin forest Zn <0.01 <0.01 <0.01 0 0 levels. Most elements, including phospho rus and calcium, stop the abrupt decline Ni <0.02 <0.02 <0:02 0 0 from their post-burning peaks, to remain more or less constant through the second and third years of observation. Not (1) Samples taken by the authors, analyzed by CENA. 3 enough time has elapsed to evaluate how (2) Discharge is 1.1 m'/second, 358 da~jyear. Density assumed of 1 gm/cm for effluent water. these nutrients will change in future years, especially after such events as the end of decomposition of the forest root ditionally, a considerable quantity of present Gmelina harvest rates. Clearly mat, diminished residual effects from solid waste precipitates out in the hold the costs of preparing and transporting burning, and the harvesting phase of inJ ponds before the effluent reaches usable portions of the mill's waste would future rotations. its release point. The nutrient-rich bark be significant, and would have to be One possibility for re removed in the debarkers prior to carefully calculated before implementing ducing the cost of obtaining fertilizers chipping the pulpwood might also be a any full scale nutrient recycling scheme. would be the use of nutrients contained valuable nutrient source: bark accounts As the necessity of supplying nutrients in the pulp mill's effluent (Table II). for 12.3% of the monetary value of the to the plantations makes itself apparent, The value of the N, P, K, and Cain the four nutrients considered in Table I. If re some of these nutrients may prove liquid effluent come to US$ 1.9 million cycled, the bark's share of these nutrients worthwhile returning from the mill site /year, using the prices of Table I. Ad- would be worth US$ 0.6 million/year at to the silvicultural stands.
332 NOV- DEC 1982, VOL. 7 Nl> 6 II\7ER[IEr\[l/l Long term viability of world consumption of white chemical fertilizing Jari's plantations, as in many pulps continues to increase from 43 . 8 of the forms of high input agriculture million metric tons per year in the 1972- around the world, is clouded by the 73 period to a projected 67.1 million diminishing global store of inexpensive metric tons per year in 1990 (United nutrients. Projections of world trends Nations, F.A.O., 1977: 144). Whether indicate both potash and phosphate rock rising prices of inputs will cancel these being exhausted by the year 2061 (United gains is unknown: commodity prices are, States, Council on Environmental Quality unfortunately, more responsive to supply and Department of State, 1980). 11 and demand at any given time than to Utilization of low grade deposits after the cost of investments needed to sustain this date would result in dramatic price production at some future date. increases, as well as requiring presently Political and bureau unachieved technological breakthroughs. cultural remedies such as chemical treat cratic decisions affecting the silvicultural In addition to the cost ments. If a major disease problem arises, operation may have at least as much of suppliyng nutrients to the plantations, for example, the only option usually open impact on its ultimate sustainability as with or without better recycling in the is to switch to a new species unaffected the many biological and agronomic system, plantation area will have to in· by the problem. Preparing the operation threats to continued production. Regu· crease to compensate for wood presently to counter such problems is difficult and lations regarding marketing and pricing derived from native forest. Establishing, costly, but essential. Jari would be wise are particularly important, 14 among maintaining, and harvesting the 37.5% to adopt as a goal the ability to sustain many other factors. additional area for firewood will make the loss of one of its plantation species. supplying energy a major figure in Jari's Finding alternative spe books at some future date, although the cies is not an easy process. The Eucalyp Irrigated Rice cost would be reduced if the contemplat tus deglupta being planted on a com ed hydroelectric dam at Cachoeira de mercial scale since 1979 had performed The irrigated rice pro Santo Antonio is built. well in young trial plots on some sites ject in diked varzea (floodplain) at Sao The 15% of the present prior to initiation of commercial plant Raimundo has recently undergone set. Gmelina-based pulp mix derived from ing, but only a few areas could be tested backs, both agronomic and economic, native forest trees (Woessner, 1980a) prior to adoption. E. deglupta is a species leading to management changes and will also have to be supplied from known for more varied growth than close scrutiny of the project. One major plantations, raising costs accordingly. most, with isolated trees or small plots cause of worry was a sharp fall in rice Prospects for Jari's sil commonly doing much better than large yields in 1979 to about 7 metric tons/ viculture and pulp operation depend, commercial stands (R. G. Lowe, person ha/ year from slighty over 8 metric in addition to establishing an equilibrium al communication, 1980). A nthroce tons/ ha/ year in the previous year. between nutrient inflows and outflows phalus chinensis, a species which has Various contributing causes have been in the system, on low risks of major given excellent results in trials on some identified, and some have been re yield setbacks from pests, diseases, or J ari sites, has not been planted commer medied through management changes. other agronomic problems. Decline m cially because it demands good soil, has One reason for the decline was iron tox soil nutrient levels can be expected to narrower requirements for amount and icity, evinced by accumulation of brown act synergistically with other factors to timing of water than other species, and sludge on the rice roots. Although the further jeopardize silvicultural yields. is susceptible to a disease, not yet arrived area's soils are prone to this problem, Less than optimal soil fertility for tree at Jari, which has attacked plantations iron toxicity can be minimized through growth, such as early Gmelina sites lo in Central America (C. Briscoe, personal careful water management in the paddies. cated on inappropriate soil types, in communication, 1980, see Gibson and A delicate balance exists between ex creases the susceptibility of plantations Nylund, 1976). cessive flooding, leading to iron toxicity, to insect pests (Johnson, 1977). Site Pulpwood plantations, and dryness, leading to aluminum to deterioration through soil compaction and as in many forms of agriculture, are sus xicity. A preflood helps raise pH levels other processes expected to increase with tained through a continued struggle of the and reduce iron toxicity. future rotations will further exacerbate humans managing the system with biolo Increased insect attacks this problem. gical, silvicultural, and technical pro are believed by the technical staff to be The probability of sig blems. The relatively long cycle of the a major factor affecting rice yidds. Army nificant losses to insect and disease out crops, and the large investments in worms (Spodoptera frugiperda) are a break can be expected to increase with volved, make it especially difficult to particularly persistent problem, which can increasing area of monocultures planted,12 steer a successful course where problems quickly inflict severe losses if timely and with the passage of time during must be anticipated and corrective action and effective controls are not taken each which pests and disease organisms not taken several years before yield declines time populations begin to increase. The yet present can become established. 13 become economically significant. To a army worm problem, like the toxicity Defoliation by pests can reduce plan certain extent the sustainability of the problems, is linked to water manage· tation tree growth by varying amounts system depends on high quality research ment: flooding the paddies when the (Ribeiro and Woessner, 1980), adding to and monitoring efforts. worms are young being an important the plantation area that would be needed The relative prices of weapon in the battle to control these to meet the pulp mill's demand for wood. pulp as compared with needed inputs will pests (see Grist, 1975: 320). In silvicultural opera be crucial in determining whether steps Stink bugs (Oebalus tions generally, many potential manage are taken by management to increase poecilus) have also been a problem ment problems are not solvable at ac the likelihood of long term production. believed to contribute to yield declines. ceptable costs through traditional agri- Pulp prices can be expected to rise as Attacks of these insects on rice roots
N\1ERCIEI\[IR NOV- DEC 1982, VOL. 7 N9 6 333 have a synergistic effect with iron to levels throughout the rice area, although high spots. Another engineering change xicity contributing to reduced absorp it has not exploded to inflict the severe began in 1980: construction of an ad tion of nutrients by the rice plant. The losses it is capable of causing. Wild red ditional canal through the middle of each plants respond to roots being punctured rice (Oryza rufipogon) is now present at large field, allowing independent filling by stink bugs by replacing them with Jari's Sao Raimundo rice project in low of each of the paddies into which the additional roots. Foliar symptoms of levels. Wild red rice is a major problem large fields are divided. The new canals nutrient deficiencies can easily lead to in many parts of the world because it is reduce the time required to fill a paddy the conclusion that increased fertilizer difficult to distinguish from domestic from eight days to 36 hours. The new applications are needed, a response the rice (0. sativa), many seeds fall from the canals in presently existing rice produc current technical staff likens to treating panicles before harvest allowing it to re tion fields will be complete by the end a human patient suffering from intestinal seed in the paddies, and any grains of 1983, if the construction schedule is parasites by applying blood transfusions harvested shatter when milled, thus followed. or additional food. Increased attention lowering product quality. The most ef Beginning in 1980, rice to insect problems may reduce the cost fective way to control these and most crops are planted on a rolling basis and increase the effectiveness of future other weeds is to pre-flood the fields to rather than in two distinctly separate responses to these symptoms. germinate the weeds, followed by drain. and more synchronized cropping periods. Nematodes are an ad ing and multiple herbicide sprayings prior The number of crops obtained from any ditional problem found by the current to flooding and planting. These time given paddy is still limited to two, as team of technical staff, which began consuming and expensive procedures are the remaining time is insufficient for a work at the IRI Research Institute station so far only used for the seed production full third crop. In the future the 1-226 associated with the rice project in the beds. In production fields weed control is variety may allow about 30% of a full latter part of 1980. The problem's se presently limited to one spraying before crop's yield to be harvested from res verity and the possibility of control are planting (either Dalapon of Gramoxone), prouts as a stubble crop during this time still under investigation. Aphids (Aphidi followed by spraying with Starn (Pro slot, but this possibility is still, in the dae) in the rice roots and mites (Acari) , panil) when the rice is growing. One testing phase. Rotation crops other than are two additional problems discovered by the present research team. Rice diseases present include rice blast (Piricularia oryzae), to which the IR-22 variety now being phased Political and bureaucratic decisions affecting the sylvicultural out is very susceptible (but which did operation may have at least as much impact on its ultimate sustainability not inflict economically significant losses during the time that IR-22 served as Jari's as the many biological and agronomic threats to continued production. mainstay). Leaf scald (Rhynchoporium oryzae) has been the most damaging disease for both IR-22 and J-226. J-226 seed is being propagated as a promising replacement for both IR-22 and the J-229 marantaceous weed (Thalia geniculata) rice are also being tested, including sweet progressively supplanting IR-22 since has not been sucessfully controlled with potatoes (Ipomoea batatas), maize (Zea 1979. Brown leaf spot (Helminthospo these procedures, but the problem has mays), sunflowers (Helianthus annus), rium oryzae) is also present, as well as a not grown to economically damaging sorghum (Sorghum vulgare), cow-peas little false smut (Ustilagenoides virens). proportions. Other weeds include an uni (Vigna sinensis), winged beans (Psopho Although diseases represent a constant dentified Phaseolus species, and another carpus tetragonolobus), soya beans (Gly worry to rice growers, J ari's rice has wild rice species (Oryza alta). Two major cine max), peanuts (Arachis hypogaea), not suffered unduly from them so graminaceous weeds in Surinam ( Barett, and an unidentified local legume green far. "Hoja blanca" virus, a major nd. (1975) ) not yet arrived at Jari manure crop. Taro (Colocasia esculenta) problem in Venezuela, has still not (R. Cheaney, personal communication, is being tested as a potential alcohol arrived at J ari. The most effective res 1980) are Leptochloa scabea and Js crop, and a native papyrus (Cyperus ponse to disease problems is generally chaemum rugosum. The latter is con spp.) may be tested as a potential pulp a switch to more resistant varieties, che sidered the "most dangerous grass in source. Rotation crops could help reduce mical treatments representing only a tem Surinam" (Doeven and Peorink, 1955, pest problems, as well as fertilizer costs porary measure. The greatest problem cited by Grist, 197 5: 280) . in the case of green manure crops. is ability to produce sufficient seed fast A number of remedies The search continues enough to effect a variety change when are underway for the various agronomic for rice varieties more tolerant of sub and outbreak occurs (given that a re problems and the decline in rice yields optimal management, and for varieties sistant variety is available in the pool of may not continue if these are effective. with greater tolerance for iron toxicity. strains undergoing testing at the site). A major effort is improvement of the Variety screening continues, as well as Sufficient seed was not available to com physical structure of fields and dikes to research on possible rice management plete the switch from IR-22 to J-229 allow more finely tuned and rapid res changes. Research on annual crops on two years after the change began, and ponses to water management problems. terra firme (unflooded uplands) at Sao J-229 is now already slated for replace Fields are being regraded at considerable Raimundo was discontinued in 1979. ment as soon as enough J-226 seed is expense. Slight depressions and "high Better insect population available. spots" in the present fields have been a monitoring methods have been instituted Weeds are a continual constant headache hampering uniform in the irrigated rice fields (sweep net worry in the Jari rice project, as in irri flooding or drying of fields to control transects rather than 1 m2 quadrats), gated rice everywhere. Barnyard grass insects, weeds or other problems. Army and spraying programs have been adapt (Echinochloa crusgalli) is present at low worms, for example, preferentially attack ed to try and control more insect out-
334 NOV- DEC 1982, VOL. 7 NQ 6 IIVERlJENJR breaks before they become severe. humans involved and the ever-changing menters m various parts of the Amazonas Sprayings have increased from only 2/ biological and other problems threatening and SolimOes Rivers. In the case of the crop to 5-61 crop to achieve better con production. The cost of a single mistake portion of Jari's herd along the Araiolos trol (note: in part to compensate for can often be an entire crop. River below the rice project, it would be aging stocks of pesticides), although In addition to agrono well to monitor the milk and cheese pro army worms are not yet resistant to the mic an human factors, the rice opera duced for pesticide residues. Long term main foliar insecticide used, Folidol tion's future depends heavily on political residents in the Araiolos River area (methyl-parathion). The coordination of and economic decisions beyond the com remark on changes in river water quality applications of various agricultural pany's control. Export restrictions and and fish catch since the implantation of chemicals so that certain chemicals don't price controls are chief among these. the rice operation. Chemicals used in the neutralize the effectiveness of other These restrictions are presently seriously rice project include the pesticides Folidol chemicals is essential. The systemic in affecting the project, probably represent (methyl-parathion), SEVIN (carbaryl), secticide Furadan ( carbofuran), for ex ing a greater source of concern to the Toxaphene (chlorinated camphene), and ample, must be applied before the her company than the dip in rice yields in Furadan ( carbofuran). The principal bicide Starn (propanil), whereas Folidol recent years. Only one rice crop was herbicides used in the fields are STAM (methyl-parathion) must be applied 6-7 ever exported, the first crop of 1978, sold (prop a nil), Dalapon, and Gramoxone; days after Starn (propanil). Human to Italy. Since that time the company lesser quantities of Anthrophene are also errors on these and other details have has been required to sell all production used on leeves. caused production losses. on the Brazilian internal market at The dependence of the controlled prices well below those for mechanized agriculture used in the rice rice of thi~, quality on international com Pasture with Pine operation on fossil fuels continues to be modity exchanges. a potential long term problem. Cost and The performance of the A herd of 6300 zebu availability of fuel will be important rice project in recent years has not been herford hybrid beef cattle 111 is apparently factors in the future. Temporary short encouraging. Ludwig tried unsuccessfully doing well in pastures interplanted with ages of natural gas have required shutt to sell a controlling 50% interest in the Pinus caribaea three years of age, without ing down the rice dryers at least once, project to Atlantic Richfield Corporation doing significant harm to the pine seed although the diesel pumps lifting water (ARCO), but after an 18-month trial lings. Initial use of guinea grass ( coloniiio: in and out of the polder surrounding the period, ARCO exercised its option to Panicum maximum) has been disconti rice fields have not been affected so far. cancel the sale in June 1981. Paralyzing nued in favor of the more covering and Posts have been installed for future expansion plans and great reduction in resistant creeping signal grass ( Quicuiu electrification of the diesel pumps. Elec personnel had failed to reverse the drain da Amazonia: Brachiaria humidicola), a tricity would be supplied from a genera on cash resources. The rice project's grass species currently being promoted tor burning crude oil, unless Jari's hydro 1980 cash flow was estimated at approxi for pastures throughout the Amazon electric dam is authorized and built. mately US$ 10 million (Kinkead, 1981). terra firme (uplands) by the Brazilian The future prospects of government based on EMBRAP A experi the rice project depend on many factors. ments 16• Biological problems can be expected to Water Buffalo become more severe as additional weeds, pests, and diseases arrive in this new rice Jari's herd of 5800 Kaolin Mine producing area, and as pest populations water buffalo in unaltered wirzea (flood develop resistance to chemical treatments plain) grasslands helps reduce the pro Jari's open-pit kaolin being applied ( Fearnside and Rankin, ject's costs for dairy products and meat. (china clay) mine presently represents a 1980). At the same time management J ari personnel are pleased with this small area of forest destruction when practices can be expected to improve, as animal's adaptation to native Amazon compared with other land uses, a contrast more is learned about how to grow rice floodplain, as has been the experience of which is even greater if the difference in this new area, and as new varieties, private ranchers and government experi- in value of the products is considered. equipment and other improvements in physical parts of the agroecosystem be come available. Costs related to the project's isolated setting and minimal infraestructure (no road link yet exists) can be expected to decrease, as can costs from the large staff required for engi Foliar symptons of nutrient deficiencies can easily lead neering works in present fields and, even tually, for completion of the project's to the conclusion that increased fertilizer applications now stalled expansion phase. Economies of scale would also be expected to reduce are needed, a response the current technical staff likens per hectare costs more as the project expands from its present 3062 ha size to to treating a human patient suffereng from its projected 14,165 ha. intestinal parasites by applying blood transfusions To a great extent the prospects for sustainability and profita or additional food. bility of the rice operation depend on the skill and dedication of the people who run it. Agriculture of this type is a con tinual balance between the ability of the
IIUER[}Ef\[IR NOV- DEC 1982, VOL. 7 N9 6 335 Jari officials say the size of the deposit rapid incr·eases in cleared areas since that consequent increase in the risk of pollu could supply the mining operation aJ the time (Fearnside, 1982). Beyond Jari tion effects. Although the 2000 m11 / 1980 extraction rate for 250 years, itself, probable limitations on the spread second average discharge of the Jari although the extraction rate is expected of large silvicultural operations in Ama River (Lang, 1979: 63) would appear to be increased in 1982 according to the zonia would include how much of the to dwarf the effluent inputs, the head company 17• The small disturbance of Amazon is suitable for this kind of of the Brazilian Navy station at the site the current operation could therefore be plantation, and how much plantation says at least one fish die-off has occurred expected to increase tremendously as ex area would be needed ·to supply projected downstream of the mill. The location of traction proceeds through the coming market demands (not to imply that all the mill near the mouth of the Amazon two centuries. The problem of reclaiming market demand should be supplied or River is fortunate in limiting potential these much larger areas will therefore that all suitable areas should be used). damage were water pollution to become eventually have to be faced. It is well to remember that macroecolo a severe problem (and is also one of the gical effects are generally cumulative in many features of the Jari site not dupli nature, being the combined result of cable were the project to be copied Environmental Risks many smaller strains on the system, none elsewhere in Amazonia). Silviculture of which is sufficient in itself to provoke Jari maintains a pollu the change. tion monitoring laboratory at Monduga, The size at which re which submits reports of daily measure placement of natural forest with planta ments of biological oxygen demand tions would provoke significant wider Pulp Mill (BOD), disolved oxygen (DO) and other environmental effects is unknown, but indicators to SEMA every week. SEMA probably large in comparison with Jari Water pollution from also occasionally sends its own personnel itself. Reduced rainfall is the macroeco pulp mill is treated by filtering solid to collect samples. The Jari River is logioal effect most closely tied to defore sludge, followed by holding the effluent classified by SEMA as "class 2"; appli station in general (Salati et al., 1978; for 16 days in a series of 1 m deep cable water quality standards for this Fearnside, 1979). Increased severity of the dry season is a greater reason for preoccupation about deforestation than are decreases in annual rainfall totals. J ari's silvicultural plantations could be Most of J ari's other operations, such as timber extraction for solid expected to have less impact than the wood, kaolin mining, and irrigated rice, involve the Amazon region's fast-growing areas of cattle pasture, but would have more than destruction of natural habitats. land uses which maintain intact rain forest canopy cover. Transpiration is largely a function of leaf area, implying that the transpiration potential of the plantations is less than that of rain forest. ponds where bacteria act to degrade category call for a BOD/ 5 days at 20°C Evapotranspiration, mostly transpiration, organic materials. The 159 ha of ponds of 5 mg/ 1 or less, and DO of at least has been found to account for approxi include capacity for emergency drainage 5 mg/1 in any sample (Brazil, Ministe mately half of the water vapor generating of digestors, not included in original rio do Interior, SEMA, 1976: 21). rainfall between Bell!m and Manaus (Vi plans. The relatively inexpensive ponds, lla Nova et al., 1976; Salati et al., 1978). constructed with simple earth dikes, allow Large-scale modification in the evapo much cheaper pollution treatment than Other Operations transpiration potential could therefore be would be possible in temperate zones expected to cause similar changes in rain where covered concrete tanks are needed Most of Jari's other fall for the region. Precipitation records to maintain water at the required 300C operations, such as timber extraction for for rain gauges maintained by Jari at temperature. In late 1980, Brazil's solid wood, kaolin mining, and irrigated Monte Dourado since 1968, and at other Special Secretariat of the Environment rice, involve the destruction of natural locations in the silvicultural area since (SEMA) took the additional precaution habitats. Some, such as rice, involve the 1970, 1971, 1974, and 1975, do not of requiring a further enlargement of the use of chemicals which can affect eco reveal any significant trends in total rain pond area, since the ponds were nearly systems beyond those altered directly in fall, or in such dry season severity indi full. The enlarged ponds are under implanting the production system. cators as number of days of rain in each construction (P. Nogueira Neto, perso Jari's buyers are seeking of the dry season months, minimum nal communication, Feb. 1981). government approval for several projects number of days of rain in any month, Effluent, a brown liquid attempted by Ludwig but blocked by and minimum rainfall in any month. The with a distinctive odor (Table II), is re regulatory impasses. These include cons time sequence is, however, too short to leased into the Jari River at the river tructing a hydroelectric dam and import allow conclusions. bottom (about 18 m deep). An 80 em ing a second pulp mill ( Veja, 13 de ja Risk of any eventual tide at the mill site in Monduga causes neio de 1982). Other projects contemplat climatic effect is minimized by the large water to flow upstream each day as the ed by Ludwig, such as expanding the areas which must be converted before tide comes in, restricting effluent release industial sector to produce paper on the any measureable change is provoked. times to the hours when water is moving site, might eventually also be realized if Jari's area is small in comparison with at its maximum rate downstream. The 1.1 the various projects already in operation the 3.5 million ha of cattle pastures es ms/second average discharge is therefore prove profitable. timated to exist in Brazil's Legal Amazon concentrated into much larger effluent Environmental risks of by 1978 (Serrao et al., 1979) and the pulses when released into the river, with the various projects depend very much
336 NOV- DEC 1982, VOL. 7 NQ 6 IIVERDffiDR on how large they become. Jari's size NOTES the company, anticipated a mean yield of ( 305 such estates would occupy Brazil's 25 m'/ha/year for Gmelina, and 15 m1/ha; 1. Jari Florestal e Agropecubia, Ltda. and year for Pinus. (Rollet, 1980: 8), and re entire Legal Amazon) means that po ported that 35 m1/ha;year were "expected" tential impacts at the local level merit Caulim da Amazonia S/A are now being integrated into the firm Companhia do from Eucalyptus (Rollet 1980: 14). Al careful monitoring. Jari. The third Ludwig firm, Sao Raimundo though J ari staff declined to estimate Agropecuiiria Ltda., is not included in the average yields for us as well, they stated sale, although negotiations are underway that yields obtained on good soils were (Veja 27 de janeiro de 1982: 92). 35 m• /ha/year for Gmelina, 25 m' /ha/year General Conclusion for P. caribaea, and very approximately 2. This figure is J ari's last claim under 40 m' /ha/year for E. deglupta. It is worth Jari's management policy Ludwig. Jari's new owners also claim 1.6 noting that these E. deglupta yields are million ha (A Critica, 13 de janeiro de much higher than those reported for plan must strive for indefinite sustainability, 1982: 2), but have apparently so far only even though this implies higher current tations in other tropical countries. In the received assurances from the Brazilian gov Philippines, for example, a maximum of outlays. Among the preconditions re ernment that 900,000 ha are theirs, with only 25 m' /ha;year has been obtained (Ta quired for sustainable management of the the "possibility of expansion" to 2 million gudar, 1974 cited by Johnson, 1976: 208). silviculture operation are: additional ha (Veja, 13 de janeiro de 1982). Woessner (1980a) gives maximum yields plantations to supply the fuel and pulp 3. Jari's reported cash deficit in 1979 was at rotation age at Jari of 42 m' ;ha/year US$ 150 million, the US$ 70 million in for B. deglupta 38 m' /ha/year for Gmelina, currently derived from native forest, an revenues (pulp + rice) being offset by and 25 m' ;ha/year for P. caribaea, but extensive research and monitoring effort, operating costs of US$ 110 million, con observes that "on less optimum sites the and the replacement of nutrients removed struction costs of US$ 50 million, and debt yields may be reduced by 'h". Actually, in harvested biomass. These changes do service costs of US$ 60 million. In 1980 the on the least optimal sites the economic deficit was US$ 14 million, revenues of not assure Jari's future, although long yield can be considered to be zero, since US$ 9(} million being absorbed by operating 20.7% of the plantation is reportedly "not term prospects are clearly bleak without costs in the pulp project of US$ 10 million, managed" mostly due to poor soils (T.H. them. operating costs in the rice project of US$ Svendsen and S. Lassen, personal commu Silviculture, irrigated 10 million, construction costs of US$ 20 nication July 1981). Jari's problem is that rice, and other sectors of Jari face bio million, and debt service of US$ 64 million relatively little of the area is optimal, and (Kinkead, 1981). Other reports indicate logical and technical difficulties, as well it can ill afford to base calculations on the project lost US$ 80 million in 1980 the best possible yields. Average yields are as environmental risks. Hopefully inform on an operating budget of US$ 250 million much lower than the approximately 40 ed debate on these issues will take place (Brooke, 1981). m'/ha/year originally anticipated by Lud with an eye to minimizing the probability 4. Proceeds from the sale will go to the wig for the Gmelina he expected to use for and potential impact of possible pro Fundacao Daniel Ludwig para a Pesquisa the entire plantation: the areas now planted do C~ncer (Daniel Ludwig Cancer Research blems, and with the objective of contri with P. caribaea instead of Gmelina can Foundation) in Sao Paulo, plus a percen yield a maximum of 63% of this figure, buting to wiser choices of development tage of dividends distributed to stockholders probably yielding less over most of the patterns offering a sustainable basis for through the year 2021: 5% for 10 years area, while Gmelina and E. deglupta, al· serving human needs in the region 1s. beginning in 1987,4% for the next 10 years, though capable of achieving the expected and 3% for the last 15 years (Veja, 13 de yields on the best sites, fall far short on janeiro de 1982). Note: the total debts of more typical solis. At more representative GLOSSARY the project have been reported to be US$ site indices for Gmelina on the order of 350 million (Veja, 16 de dezembro de 21, yields are around 20 m'/ha/year (see ARCO Atlantic Richfield Corporation. 1981: 77), or US$ 170 million in addition note 10). One report gives Gmelina yields to the remaining balance of the pulp mill for sites of index 21 or over as 28 m' /ha/ CENA Centro de Energia Nuclear na loan. 3 Agricultura (Center for Nuclear year, expected yields of 20 m /ha/year for 5. From 5-12 November 1980 we observed 3 Energy in Agriculture), Piracica P. caribaea, and 20 m /ha/year (based on commercial and experimental plantations experimental plots) for E. deglupta (T.B. ba, Sao Paulo. of Gmelina arborea, Pinus caribea var. Svendsen and S. Lassen, personal com CNPq Conselho Nacional de Desenvol hondurensis, and Eucalyptus deglupta, and munication July 1981). One earlier estimate, vimento Cienti/ico e Tecnol6gico experimental plantations of various other also made without benefit of production (National Council for Scientific species. Also visited were the nursey, pulp figures from the company, anticipated a and Technological Development), mill and power plant, mill effluent treat mean yield of 25 m3/ha/year for Gmelina Brasfiia. ment ponds, wood transport and storage and 15 m1/ha;year for Pinus (Rollet, 1980: BMBRAPA Empresa Brasileira de Pesqulsas facilities, laboratories for soil, pulping 8), and reported that 35 m'/ha/year were Agropecudrias (Brazilian Enter research, and pollution control, herbaria, "expected" (Rollet, 1980: 14). Very ap prise for Agriculture and Cattle entomological and wood collections and proximately, present yields on the order of Ranching Research), Brasilia. other laboratories and technical offices. half the original expectation seem probable Other sectors observed were the vdrz;ea over the estate as a whole. FAO Food and Agriculture Organiza rice project, kaolin mine, water buffalo tion of the United Nations, Rome. 8. According to Jari technical personnel, herd, beef cattle herd (in pastures in young company laborers earn (Nov. 1980) Cr$114 INCRA Instituto Nacional de Coloniza,iio pine plantations), and native forest extrac ;day when calculated at 30 days/month e Reforma Agraria (National Ins tion operations, as well as the native forest paid for 13 months/year to include the tile for Colonization and Agrarian and the Jari and Araiolos Rivers. We also extra month's salary standard in Brazil. Reform), Brasilia. visited the project in August 1978. They also receive production incentives 6. 60-90% standing crop ~ 20 em DBH. In INPA Instituto Nacional de Pesquisas averaging 30% of their total pay (but over 400 species present, 230 are of same ranging up to 300%). This gives a total da Amaz;bnia (National Institute use. Of usable species, 82-86% can be of Cr$57,798/year. Additionally, the com for Research in the Amazon), utilized with Gmelina for pulp (Woessner, pany probably pays the standard benefits Manaus, Amazonas. 1980a). in Brazil: 22.33% in health insurance and IRI IRI Research Institute, New York.. 7. Various growth rates have been reported, social security, 8.33% in government medi but an average figure has apparently never Sociedade Brasileira para o Pro cal assistance, and 8.33% for disability SBPC been disclosed by the company. Kalish gresso da Ciencia (Brazilian So leave allowance. These would increase the (1979) reports that "the exact growth ciety for the Progress of Science), cost to the company to Cr$75.132/worker/ rate was not disclosed by J ari foresters, year, or Cr$262.70/day worked using a Slo Paulo. but it is believed to be in the range of 286-day year (discounting Sundays and a SBMA Secretaria Especial do Meio-Am 25-35 m' /ha/year for Gmelina and slightly one month paid vacation), or US$4.24/ biente (Special Secretariat of the lower for pine." One estimate, also made worker/day worked at Cr$62/US$ (Nov. Environment), Brasflia. without benefit of production figures from 1980).
ln7EI?DEni:IR NOV- DEC 1982, VOL. 7 N9 6 337 9. According to mill officials, 20% repre 14. December 1980 pulp prices of US$ 322/ REFERENCES sented a goal for 1981, and 30% of the metric ton on the Brazilian internal market, power generated was coming from fuel oil as compared with US$ 515/metric ton at A! vim, P. de T. (1981): Os riscos de desen as of 1980. Some oil generated power must European ports, mean that the 20% of volvimento na Amazonia. Paper presented be used as a supplement to maintain even output required to be sold within Brazil, at Comissao Economica para a America power output in the face of inevitable at the 750 metric ton/day nominal capacity, Latina (CEPAL), Seminru-io Regional so variation in the energy supply from burn represents a loss of US$ 10.4 million/ ing wood chips. bre Expansao da Fronteira Agropecuaria year, less a relatively small amount for e Meio Ambiente na America Latina, Bra difference in shipping costs. 10. The site index represents an estimate of the silia, 10-13 de novembro de 1981. average top height in meters of the 100 15. The cattle herd is reportedly being slaught Arkcoll, D. B. (1979): Uma avaliacao das tallest trees in a stand at age 10 years. J ari ered without replacement, to feed J ari's opcoes agrosilviculturais para a Amazonia. has gradually increased the site index at work force during the current economic pp. 101-111 In: Academia de Ciencias do which Gmelina stands are converted to crisis (C.F. Russell, personal communica Estado de Sao Paulo (ACINSP), Anais other species, either immediately or at the tion, Oct. 1981). do Simp6sio sobre Ciencias Bdsicas e Apli end of a cycle. Most recently, sites of index cadas. Publicacao ACINSP N9 19. (Sao 19 or over are allowed to resprout for 16. The good results of the pasture/pine in Paulo, ACINSP) 219 pp. further Gmelina crops, but new plantings terplanting raises the important possibility Barett, S. C. H. (nd. (1975) ) : Report on the of Gmelina are only made on sites of index of pine plantations as a potential use for weeds of cultivated rice at J ari, Lower 21 of above. Sites of indices of approxi the extensive areas of fast-degrading pas Amazon Manuscript, Department of Bota mately 15 to 18 are kept in Gmelina until tures throughout the Amazonian terra fir ny, University of California, Berkely, Ca harvested, and then converted to P. caribaea me. It is worth remembering two points of lifornia. 16 pp. or E. deglupta. E. deglupta is planted on caution: 1) if an enterprise wished to Brazil Ministerio da Agricultura, Instituto Na sites with index 13-17 and P. caribaea on establish a pulp operation following the cional de Colonizacao e Reforma Agraria sites with indices below 13. Jari finds Jari pattern, the native forest wou!a re (INCRA). (1972: Projeto lntegrado de Gmelina yields relate to site indices as present a valuable source of fuel and sup Colonizafiio Altamira-1. (Brasflia, INCRA) follows (R. G. Lowe, personal communi plementary pulpwood, making it more at 217 pp. tractive to entrepreneurs to establish such cation, 1981), assuming a 6 year rotation Brazil, Ministerio do Interior, Secreta ria Espe plantations on currently forested areas site index: metric/tons wet inside bark cial do Meio Ambiente (SEMA). (1976): rather than cattle pastures, and 2) many yield/ha;year): 13:7. 7; 15:8 .1; 17:11. 5; LegislafiiO Bdsica. (Brasilia, SEMA) 24 of the negative features of cattle pastures 19:14.7; 21:19.3; 23:22.8; 25:26.5; 27: pp. remain (effects on rainfall, income distri 30.5. Note: Jari staff use appoximate Brooke, J. (1981): Billionaire's dream founders Gmelina wet density of 1 metic tonjm'. bution, employment, etc.), and 3) pine plantations also have deleterious effects on in Amazon Jungle. The Washington Post, May 31, 1981. pp. 1 & AlB. 11 . The long term availibility of phosphate the soil and undemonstrated long term Cox, G. W. and Atkins M. D. (1979): Agri rock is a matter of debate. One study sustainability. Although pastures supplanted cultural Ecology: an Analysis of World (Institute of Ecology, 1972) indicates that by Pinus caribaea are more economically Food Production Systems. (San Francisco, phosphate rock of a least 8% P (currently and socially desirable than pastures sup W. H. Freeman & Co.) 721 pp. mined deposits contain 12% or more) planted by second growth, care must be taken not to foster illusions either that A Critica (Manaus) 13 de janeiro de 1982. would last until 2062, based on continued a way has been found to make pastures "Venda do Jari tem rubrica de Ludwig' . exponential increase (5.25%/year) in sustainable or that the various negative Cad. 1 p. 2. phosphorus fertilizer use. The assumption aspects of 'pasture as a development option of continued trends at this rate has been Doevan, J. G. P. and Poernik, H. G. (1955): have ceased to be preoccupations. Above Weeds in rice and their control in Surinam. criticized as overly pressimistic (Wells, all, such very praiseworth efforts to im 1976). Population and cultivated area can Tropical Agriculture (Trinidad) 32: 115- prove pasture management must ~ot be 123. not expand unchecked. Although obvious used as a rationalization for encouragmg the Dynia J. F., Moreira, G .N.C. and Bloise, -~· that the historical exponential increase in clearing of more of the Amazon's rain phosphate use cannot continue for long, M. (1977): Fertilidade de solos da regiao forests for pasture. With these warnings in da Rodovia Transamazonica II. Fixacao the eventual lowering of use rates will mind the pasture/pine experience at Jari do f6sforo em podz6lico vermelho-amarelo probably be tied to unprecedented price shouid be a valuable demonstration of one increases. e terra roxa estruturada latoss6lica. Pes way to make use of those parts of the quisa Agropecuaria Brasileira 12: 75-80. Amazon terra firme which have already Falesi, I. C. (1972): Solos da Rodovia Trans 12. The relation of plantation area to the num been cleared for cattle pasture (see Rankin, amaz6nica. Instituto de Pesquisa Agrope ber of pests follows logically from island 1979). biogeographic theory (MacArthur and Wil curu-ia do Norte (IPEAN) Boletim Tecnico son, 1967). Examples include that of sugar 17. In late 1981 the Kaolin operation was N9 55. (IPEAN, Belem) 196 pp. cane native to New Guinea: an isolated shut down temporarily due to a drop in Fassbender, H. W. (1969): Retenci6n Y tras one hectare patch in Iran 800 years old world prices, but is expected to resume formaci6n de fosfatos en 8 latosoles de Ia has only pest, whereas Brazil's substantial normal operation soon (C.E. Russell, per Amazonia del Brasil Fitotecnia Latiname cane areas have 50-60 different pests (W. sonal communication, Oct. 1981). ricana, 6 (1): 1-10. W. Benson, Statement at SBPC meeting, Fearnside, P. M. (1978): Estimation of Carry Fortaleza 1979). Eucalyptus plantations in 18. We are grateful to the National Council for ing Capacity for Human Populations in ? Brazil's state of Minas Gerais have experi Scientific and Technological Development. part of the Transamazon Highwa_y Colo_m enced increasing insect pest attacks as local (CNPq) for providing travel funds th.rough zation Area of Brazil. Ph.D. Dissertation insects have adapted to the new species the National Institute for Research Ill the in biological sciences, University of Michi (Moraes et al., 1982). Municipios with Amazon (INPA). We thank the Center for gan, Ann Arbor. (Ann Arbor,_ Michigan, older plantations have more severe pest Nuclear Energy in Agriculture (CENA) University Microfilm International) 264 problems, as do municfpios with over 5% for analyzing our soil, water, and vegeta pp. tion samples. This paper in no way re of the land area planted to Eucalyptus Fearnside P. M. (1979): 0 processo de deser presents the views of CNPq, INPA, or t~e (G.W.G. Moraes, presentation at SBPC tifica~iio e os riscos de sua ocorrencia no many colleagues with whom we have dis meeting, Campinas, 1982). Brasil. Acta Amazonica, 9 (2): 393-400. cussed the developments at Jari. Any errors Fearnside, P. M (1980): The prediction of soil are our own responsibility. 13 . Many people do not realize the rapidity erosion losses under various land uses in with which the cumulative probability of an the Transamazon Highway Colonization event increases, even when the annual Area of Brazil. pp. 1287-1295. In: J. I. probalitiy is quite small. An event with Furtado (ed.) Tropical Ecology and De a 1% probability of occurrence per year, velopment: Proceedings of the 5th Interna for example, has a more than 50% chance tional Symposium of Tropical Ecology, of occurring within 69 years. A 2%/year 16-21 April 1979, Kuala Lumpur, Malay risk would reach this level in less than 35 sia. (Kuala Lumpur, International Society years. of Tropical Ecology (ISTE) ) 1383 pp.
338 NOV- DEC 1982, VOL. 7 N9 6 Jf\1Efl0ffiDR Fearnside, P. M. (1982): Deforestation in the Paper Industry, Brussels, Belgium, May 1978. Centro Internacional de Agricultura Brazilian Amazon: how fast is it occuring? 1979. (San Francisco, Miller Freeman Tropical (CIAT) Series EG-05. (Cali, Co Interciencia, 7 (2) : 82-88. Publications) 160 pp. lombia, CIAT) 488 pp. Fearnside, P. M. and Rankin, J. M. (1980): MacArthur, R. H. and Wilson, E. 0. (1967): Sioli, H. (1969): Okologie im brasilianischen Jari and development in the Brazilian The Theory of Island Biogeography. Mo Amazonasgebiet. N aturwissenschaffen, 65: Amawn. Interciencia, 5 (3): 146-156. nographs in Population Biology 1. (Prin 248-255. Gibson, I. A. S. and Nylund, J. (1976): Sudden centon, New Jersey, Princeton University Strong, D. R. Jr., McCoy, E. D., and Rey, J. R. death disease of cadam (Anthocephalus ca Press) 203 pp. (1977): Time and the number of herbivore damba Roxb. Miq.). Commonwealth For Moraes, G. W. G., Brun, P. G. and Soares, species: the pests of surgarcane. Ecology, estry Review, 55 (3): 219-227. L. A. (1982): Ocorrencia de lepid6pteros 58: 167-175. Grist, D. H. (1975): Rice, 5th ed. (New York, desfolhadores de euca!ipto em Minas Ge Tagudar, E. T. (1974): Development of indus Longman) 601 pp. rais e as perspectiva para seu controle. trial plantations inside Paper Industries Guimeriies, G. de A., Bastos, J. B. and Lopes, Ciencia e Cultura, 34 (7): 575 (abstract). Corporation of the Philippines. In: Pro E. de C (1970): Metodos de analise fisica, Pimentel, D., Chick, S. and Vergara, W. (1981): ceedings of the Forest Research Symposium quimica e instrumental de solos. Instituto Energy from forests: environmental and of Industrial Forest Plantations, Manila, de Pesquisas e ExperimentafiiO do Norte wildlife implications. Interciencia, 6 (5): June 26, 1974. (IPEAN) Series Quimica de Solos, 1 (1): 329-335. United Nations, Food and Agriculture Orga 1-108. nization (F. A. 0.). ( 1977) : World Pulp Rankin, J. M. (1979): Manejo florestal ecol6- Herrera, R., Jordan, C. F., Klinge, H. and Me and Paper Demand, Supply and Trade- 2. gico. Acta Amazonica 9 (4) suplemento: dina, E. (1978): Amazon ecosystems: their F.A.O. Forestry Paper 4/2. (Rome, UN 115-129. structure and functioning with particular FAO) 243 pp. United States, Council on emphasis on nutrients. Interciencia 3 ( 4): Ribeiro, G. T. and Woessner, R. A. (1980): Environmental Quality and Department of 223-232. Efeito de diferentes niveis de desfolha ar State. (1980): The Global 2000 Report to Institute of Ecology. (1972): Cycles of elements tificial, para avalia~iio de danos causados the President. New York, (Pergamon Press) pp. 41-89 In: Man and the Living En por sauvas (Atta, spp.), em arvores de 3 vis. vironment. (Madison, Wisconsin, University Gmelina arborea Linne e de Pinus caribaea Veja (Siio Paulo). 16 de dezembro de 1981. of Wisconsin Press). var hondurensis Barr & Golf. Anais da "0 adeus de Mr. Ludwig.'. p. 112. Johnson, N. E. (1976): Biological opportuni Sociedade Entomol6gica do Brasil, 9 (2): Veja (Siio Paulo). 13 de janeiro de 1982. "0 ties and risks associated with fast-growing 261-272. pais afinal comprou o sonho do Jari.' pp. plantations in the tropics. Journal of For Rollet, B. (1980): Jari: succes ou echec? un 68-73. es try, April 1976: 206-211. exemple de developpement agro-sylvo-pas Veja (Siio Paulo). 27 de janeiro de 1982. "Jari Johnson, N. E. (1977): How susceptible are toral et industriel en Amazonie Bresilienne. com a cortin a aberta." pp. 90-92. tropical tree plantations to insect depre Bois et Forets Tropiques, N9 192: 3-34. Villa Nova, N. A., Salati, E., and Matusi, E. dations?" pp. 209-214 In: BIOTROP Pro Russell, C. E. (1981): Jari: a lesson in tropical ( 1976): Estimativa da evapotranspira~iio na ceedings Symposium on Forest Pests and ecosystem management? Paper presented Bacia Amazonica. Acta Amazonica 6 (2): Disease in Southeast Asia, April 20-23 at International Society for Tropical Eco 215-228. 1976, Bogor, Indonesia. BIOTROP Special logy and Resource Management in the Wells, F. J. (1976): The long-run Availability Pub. N9 2. (Bogor, Indonesia, South East Tropics". Bhopal, India, October 5-10, of Phosphorus: a Case Study in Mineral Asian Ministers of Education Organization 1981. Resource Analysis. (Baltimore, Johns Hop (SEAMEO) Regional Center for Tropical Salati, E., Marques, J. and Molion, L. C. M. kins University Press). Biology (BIOTROP) ) 225 pp. (1978): Origem e distribui~iio das chuvas Woessner, R. A. (1980a): Forestry operations Kalish, J. (1979): Jari. Pulp and Paper Inter na Amazonia. Interciencia, 3 (4): 200-206. and wood utilization at Jari. Manuscript. national January 1979: 37-52. Serrao, E. A. S., Falesi, I. C., de Viega, J. B. Paper presented at the World Woods Con Kinkead, G. (1981): Trouble in D. K. Lud and Teixeira Neto, J. F. (1979): Producti ference, Santiago, Chile, December 1980. wig's jungle. Fortune April 20, 1981: 102- vity of cultivated pastures on low fertility 14 pp. 118. soils in the Amazon of Brazil. pp. 19.5-225 Woessner, R. A. (1980b): Gmelina arborea Lang, S. (1979): Jari Florestal e Agropecuaria In: P. A. Sanchez and L. E. Tergas (eds.) Roxb, variation in wood density, height and Ltd a. pp. 63-69 In: L. E. Haas ( ed. ) New Pasture Production in Acid Soils of the diameter of international provenance trial Pulps for the Paper Industry: Proceedings Tropics: Proceedings of a Seminar held at Jari. Silvicu/tura, julho-agosto de 1980: of the Symposium on New Pulps for the at CIAT Cali, Colombia, 17-21 April 50 (abstract).
lf\1ER0Ef\0R NOV- DEC 1982, VOL. 7 N9 6 339