Chapter IV: Findings
In terms of the critical realism approach described in Chapters I and III above, the results of this research project may be divided conceptually into an intensive phase of direct observation and interviews at the local scale in several cities of Rondônia and an extensive phase of analysis of patterns at the state level. This chapter reports results in these two phases. The first section presents findings based on direct observations and progressive, rolling interviews during a three-month, visit to Rondônia in 1996. These preliminary findings led to the formulation of certain hypotheses, which are presented and tested in the second section of this chapter .

1. Intensive Activities: Observations and Interviews

 

 
 
 
 
 

The observations are divided into a sub-section for each of the municípios in which I spent significant time. In the case of Porto Velho, I have divided the observations into those related to the city of Porto Velho itself and those related to the rural areas of the município of Po rto Velho. The majority of the time I spent in Rondônia was spent in Porto Velho, the capital and largest city of Rondônia. It was in Porto Velho that I gathered data and documents related to Rondônia as a whole, interviewed both ordinary citizens and people knowledgeable about particular aspects of its development, and made contacts that were helpful in organizing my trips to other parts of the state.

I made two separate visits to the nearby município of Candeias do Jamari, spending the first in the town of Candeias itself and the second in the rural zone of Candeias. I spent a week traveling to the interior of Rondônia, dividing my time between Ouro Preto do Oeste, located on BR-364 in the center of the state, and Rolim de Moura, located thirty kilometers south of the main highway in the center-south portion of Rondônia.

1. Porto Velho
1. Urban Zone

 

 
 
 
 
 

According to an interview with a geographer at SEMPLA, the municipal planning agency in Porto Velho and maps provided by the agency, the built-up urban area of Porto Velho grew four-fold between 1980 and 1988, a period during which the population doubled (Field notes, August 6, 1996; SEMPLA, n.d.). The urban area now measures approx imately 100 square kilometers, and encompasses contrasting areas in the central business district, the rapidly growing fringe, and what are now well-established neighborhoods in between. The following sections describe observations and interviews in each of several parts of the city, beginning with the neighborhood in which I lived.

1. Odacir Soares

 

 
 
 
 
 

 
Photograph 2: Home in Odacir Soares

1. Central Business District

 

 
 
 
 
 

 
Photograph 3: Private Computer School in Porto Velho

1. Cidade do Lobo

 

 
 
 
 
 

Cidade do Lobo is located on the extreme southern edge of the city and was settled in 1984. The name, which literally means "City of the Wolf" derives from the name of landowner João Lobo. The settlement was in some ways a spontaneous settlement, but it was also the result of a complicated arrangement in which Lobo offered 1200 lots for sale at a symbolic price of one cruzado to people who had invaded other land holdings which he hoped to develop (Yangopoulos 1989). Much of Lobo’s land had been in agricultural use at the time the settlement was created.

Initially, conditions were rather p oor, with self-built housing and little infrastructure (Yangopoulos 1989). By the time of my fieldwork, the neighborhood was in transition, both because of in situ improvements in housing and the expansion of infrastructure, and because of the influx of relatively affluent migrants. On several occasions, I visited the home of a university professor who owned a house and a double lot in Cidade do Lobo. In many ways, the household resembled my temporary home in Odacir Soares, although the lot was large r and contained more fruit trees (field notes, June 2, 1996).

The property comprised two lots, one of which the professor had purchased in 1991, and the other of which he purchased soon after that. Both lots were purchased from somebody who had invaded it in 1985, when the land was still in rain forest. The professor had bought the land without clear title, and believed that technically the land still belonged to the state. Because it had been continuously inhabited for over a decade, however, he wa s unconcerned about the legal status of the title (field notes, June 23, 1996).

Land invasions have been renewed in nearby Cidade do Lobo II, a short walk from my informant’s home in Cidade do Lobo. The new settlement is unauthorized but well organized. New residents would not have city services or title to the land at first, but by carefully demarcating the area they were invading into well-organized lots, they increased the likelihood that the city government would authorize the settlement and pro vide services after the fact. The process of land invasion in Cidade Lobo II is broadly similar to the process in other Latin American cities, but it is different in some key respects.

First, Cidade do Lobo is an an area on the periphery of Porto Velho adjacent to standing rain forest. For this reason, spontaneous settlement must in some cases be preceded by a substantial clearing effort, perhaps involving the cooperation of many people, and certainly feasible in the space of a day or two. Cidade do Lobo is surrounded by both standing rain forest and land that has already been cleared for agricultural and subsequent ranching use. In some cases, parcels obviously have been cleared specifically for settlement (Photograph 4), but in other cases it is difficult to tell whether the land is being claimed by those who cleared it or whether settlement is opportunistic (field notes, June 23, 1996). The distinction is important because it is an indication of whether expansion of urban settlement is at the expen se of standing rain forest.

Although migration to urban areas began as a process of stepwise migration from rural areas of the state (Yangopoulos 1989), by 1996 the mix of residents in Cidade do Lobo did not reflect this earlier pattern. My informant in Cidade do Lobo indicated that most of the residents now come from other states. Most of them work in the informal sector, although some have regular jobs (field notes, June 23, 1996).

On my final visit to Cidade do Lobo, my informant related an explanation of the origin of the settlement somewhat at odds with the history provided by the British students who had conducted fieldwork in the neighborhood previously (Yanopoulos 1989). He indicated that p art of the land had been given by the government to a group of 200 Japanese families, who attempted to farm it. The farm was unsuccessful, but the land they cleared was invaded by the Cidade do Lobo settlers (field notes, July 14, 1996).

1. Eastern Porto Velho

 

 
 
 
 
 

One such neighborhood is being developed in an abandoned clay mine on the edge of town (Photograph 5). My informant had purchased a plot there as an investment, paying US$ 120 for a corner lot measuring ten by thirty meters. An adjacent lot was selling for only US$ 80. The land is literally a hole in the ground, but it is located next to one of the many private employee associations (country clubs) on the former edge of the city which may make it a valuable investment. The same informant took us to visit a small farm comprising only one hectare on the edge of the city, which he hopes to add to his small portfolio of speculative peri-urban land holdings (field notes, August 7, 1996).

In some neighborhoods the municipal government has become directly involved in housing at a very early stage. At least two neighborhoods of single-family dwellings have been built as public housing in an effort to keep ahead of the demand for housing and infrastructure. These neighborhoods comprise long rows of houses, each on its own lot and each smaller than a typical self-built house (Photograph 6). The neighborhoods are known as Pombal (D ovecote) I & II because of the tiny size of the houses and the minimal provision of infrastructure. Because the houses are simple and the lots are relatively big, the program provides a way for families to obtain basic housing and then make in situ improvements.

1. Fire

1. Rural Zone

The first of these visits was to a ranch owned by the sister and brother-in-law of one of my informants in Porto Velho. On the way to the ranch, located about 25 kilometers from BR-364 along one of the unpaved linhas (roads or, literally, lines) that were built as part of the INCRA settlement programs described in Chapter II above, I observed a number of abandoned properties, on which farm buildings and homes were being overtake n by vegetation. The forest had not been cleared to the horizon, but neither did any remain in close proximity to the road, confirming the importance of roads as an agent in forest clearing. The soil supported secondary growth to an unexpected degree, although apparently the soils would not support any economically viable crops or forage (field notes, June 9, 1996).

This ranch was owned by an urban family, who maintain it as a working ranch, a convenient recreational retreat, and a speculative inves tment. Both parents in the family work as accountants in the city of Porto Velho. About forty people attended the weekend festivities, most of them residents of the city of Porto Velho. The owners employ another family as caretakers, who live in a very modest house on the property. The main recent improvement was a satellite dish and television, purchased at considerable expense, apparently for the weekend enjoyment of the owners.

On a tour of the property, I learned a lot about the ambivalence of o wners of ranches like these toward the environmental controversy that surrounds their activities. After driving to the back of the property where a large tract of rain forest had been cleared, the owner of the land confessed that "the ecological crime here is mine," and indicated that he was attempting to reforest some areas of the property. Indeed, he had restored the watershed surrounding a small spring on the property.

A fellow rancher who was visiting did not see ranching activities as environmentally problematic, though. He was from Ariquemes, where he owned a bigger ranch with more cattle and also coffee and other crops. He was more emphatic than the owner in explaining and defending what was happening on the ranches. He cited the destruction of North American forests and pointed out that the owner of the ranch we were visiting was only burning the forest once. He was not destroying it, the visitor said, but rather improving it, because cattle cannot survive on the virgin forest (field notes, June 9, 1996).

Much of the ranch we were visiting, which encompasses an area of 2,000 hectares, had been cleared by the previous owner, but some of the clearings I visited had been burned only a year before. The current owner reported purchasing the entire property from a single seller. Given its large size, that seller could not have been an original INCRA grantee. Rather, it appears that the property had followed a trajectory typical of many ranches in the region, as described in Chapter I I above. The land was consolidated from the holdings of approximately twenty families, all of whom had now abandoned their farms, and many of whom had probably migrated to urban places in Rondônia.

A smaller farm that I visited had a very different history. Located about 40 kilometers to the west of Porto Velho, this ranch is still inhabited by the family who was granted the land by INCRA in 1972. I was taken there by the same informant who had taken me to visit the large ranch. The small farm belongs to the family of one of his coworkers in Porto Velho, a woman who now works in an office of a state agency in the capital. She was born in Sergipe, in the Northeast of Brazil, but was raised on the farm granted by INCRA, the oldest of eight children, and the only one not born in Rondônia.

Unlike the families who had been forced to abandon the farms that were consolidated into the large ranch I visited, this family had continously farmed their land since the original grant of 72 hectar es by INCRA in 1972. Although the household income was supplemented by the adult daughter working in the city, the land remained in productive use after 25 years of occupation. The mixed farm includes a herd of 200 cattle, an orchard of 600 cupuaçu (regional fruit) trees, and several ponds managed for aquaculture. In 1994 the family had spent US$ 3,500 to obtain an electrical connection, and now had television, a refrigerator, and other conveniences. Clearly the household income was being supplemente d by the daughter working in the capital (field notes, June 16, 1996).

The attitude of this long-term resident farmer toward conservation contrasted with the views expressed the week before by the urban ranch owners. A requirement of the INCRA land grants had been that fifty percent of each parcel would remain uncleared. The farm I visited did not have any remaining forest, although a few small stands of trees were present. We visited uncleared forest on an adjacent ranch, however, where the farmer showed me a particular tree that had been used by indigenous people in the area. His tone suggested that the indigenous were ancient, but it was apparent that they had been active in the area shortly before the INCRA program pushed them out of their territory.

Although no standing forest remained on his farm, after visiting the adjacent forest, the farmer commented on the importance of its preservation, so that a balance among forest, cattle, and tree crops could be maintained. His tone was more ear nest and less defensive than those of the ranchers mentioned above. The difference, perhaps, is one of the level of awareness of how those directly involved in the clearing of the rain forest are viewed in Northern media (field notes, June 16, 1996). That is, the men who were living in the cities and running their ranches as businesses were much more aware both of how such activities have been portrayed in the North and of the degree to which the North has been hypocritical in its criticism. This family far mer had some notion that the North was concerned about deforestation, but did not seem to be accutely aware of the discourse.

Both the large ranch and the small farm that I visited were connected in important ways to the city of Porto Velho itself. In the first case, the business decisions of an urban family had important implications for the future of the property. In the second case, the family was able to maintain a mixed farm in part because of the subsidy provided by the urban employment of one of its members. The recent acquisition of satellite television on both farms is symbolic of the ways in which increased connection between the city and the countryside implies increased connection between that countryside and the rest of the world.

1. Candeias do Jamari

 

Candeias is located near the bridge where BR-364 crosses the Rio Candeias. The town is on a bluff over a great bend in the river, and is oriented along the highw ay itself. Buildings in the town are of both wood and other materials, and appear to have been built recently, and most of the roads are newly paved. Young employees in two shops that I visited indicated that they had lived in the town for only a few years, and had migrated directly from cities in other regions of Brazil. Even though the entire município has only 5,000 residents, the town has its own legislative body. It was formed in 1991, its growth from a small fishing village of about ten homes f acilitated mainly by the tax-collection station operated by the state on the highway at the river crossing (field notes, June 26, 1996) to collect tariffs from interstate truck traffic.

Contacts in Porto Velho led me to an artist who lives on the Rio Candeias a few kilometers upstream of the town. He and his wife live on 180 hectares granted to them by INCRA in 1975. The land remains forested, and is located on a 30-meter bluff overlooking a great loop of the Rio Candeais, about 75 minutes by motor boat from town. Although they live far from town, Anká was able to share some insights about Candeias. The economy of the town apparently depends upon state and municipal employment and the provision of related services, such as a new truck stop and bus station. Candeias also appears to serve as bedroom community of Porto Velho, with many residents commuting to the capital for employment on at least an occasional basis (field notes, July 30, 1996).

It was in Candeias that I first began to que stion whether the boom-town model of frontier urbanization in Rondônia is still applicable in the second half of the 1990s. I had no doubt that Candeias had emerged as quickly as the boom towns described almost a decade earlier by Godfrey (1990), but it did not appear to have the ephemeral qualities described for those towns that had rapidly grown from mining camps. Candeias appeared to be a new kind of emerging town, one that was rapidly developing an infrastructure and, more importantly, links to ot her urban places, both within the region and beyond. My thinking about Candeias contributed to the development of the hypothesis on the complexity of Rondônia’s urban network that I describe below.

1. Ouro Preto d’Oeste

 

Because of contacts made in Porto Velho, I chose to spend several days in the city of Ouro Preto d’Oeste, which translates literally as "Black Gold of the West." It is a city of 25,000 located directly on BR-364 near the center of the state. During my visit I was able to explore many parts of it on foot, interview local residents, and make some basic observations about the timber industry.

During an informal interview in Porto Velho prior to my visit, a state environmental official had indicated that timber consumption within Rondônia comprised a very minor part - no more than ten percent - of the demand for timber harvested in the state. O bservations in Ouro Preto, a major wood-processing center, provided further evidence of this claim. In Ouro Preto, most of the houses are constructed of wood (Photograph 7), but the scale of the timber operations appears to be vastly out of proportion to that which would be required to sustain this construction. A number of saw mills are located on the east side of the highway near the center of the city. The yards of these mills are filled with enormous logs, and trucks are seen leaving the city toward th e south, often able to contain only a few of these logs (July 22, 1996).

Photograph 7: Wood-plank houses, Ouro Preto do Oeste

The recent frontier h istory of Ouro Preto as a city on the frontier is manifested in a number of ways in the cultural landscape of the city. A cathedral built as late as 1995 included a bas-relief depicting pioneer themes, which are also featured on public statuary. Even the name of the local seat of government is telling: Pioneer Palace (field notes, July 22, 1996). The cultural landscape also reflects a pattern of migration influenced by its relative proximity to the South of Brazil. The names of some bars and other businesse s in Ouro Preto reflect ties to Pará and other southern states. Some of the houses are in a German style that suggests the European roots of some of the migrants who have come from the South of Brazil. The population of Porto Velho, composed of migrants from both the Northeast and the South, includes an eclectic racial variety typical of the country as a whole. The majority of the population in Ouro Preto has come from the South of Brazil, however, and therefore many more of its residents are of rela tively recent Western European origin.

The former municipal airport in Ouro Preto is the site of perhaps the most unusual residential settlement I encountered during my fieldwork. Even some of its earliest residents are unaware of its unusual history. I interviewed several long-time residents who could not recall the origin of the community, but they were able to lead me to a local business owner who had lived in the neighborhood since its inception. The municipal airport in Ouro Preto had been loca ted next to a large expanse of rain forest. It ceased operation at the beginning of the dry season in 1985 and was abandoned. In August, at the end of the dry season, the rain forest along the air strip was burned and massive trees were removed. The land was never used for agriculture, but rather was cleared specifically for urban, residential use. The land invasion was spatially well-organized, and now a large neighborhood is oriented on a very regular grid oriented along an extremely long and straight bou levard that was the original landing strip (Photograph 8). The axis of that strip is now one of the major arteries of the city (field notes, July 23, 1996).

Beyond the important observations of its cultural landscape, my experience in Ouro Preto do Oeste, combined with my experience in Rolim de Moura, contributed to the development of the hypothesis regarding timber production discussed later in this chapter. My observations at the edges of the city, combined with my observations at the edges of Por to Velho, led to the hypothesis regarding urban sprawl, also described later in this chapter.

Photograph 8: Airport Garden Neighborhood, Ouro Preto do Oeste

1. Rolim de Moura

 

 
 
 
 
 

When I visited Rolim de Moura, I met first with former colleagues of the English professor who had once worked there. Although a paved road now links Rolim de Moura to Pimenta Bueno, the approach from the north is still rather arduous. It is far easier, however, than it had been in the 1970s and 1980s. The original ferry apparently was not motor-powered, and the crossing was arduous, particularly during the wet season (Photograph 9). By the time of my visit, a motor-powered ferry was in operation, and a new bridge was near completion, implying greater connectivity for Rolim de Moura in the future (Photograph 10; field notes, July 24, 1996).

Photograph 9: Ferry Crossing to Rolim de Moura, 1976

During my first evening in Rolim de Moura, I met three men in the lobby of my small hotel, who gave me some insight into how the elites of the region view its problem s. These elites offered an explanation of the aftermath of rural colonization that differs strongly from that offered in the literature reviewed in Chapter II above. One of the men had migrated as a child in 1972 when his family received one of the original land grants. The family has since prospered, he believes, because they subsequently divorced themselves from government programs. The government programs failed, not because the climate and soils were not suitable for settlement, but because the landless peasants who received the land grants did not have the initiative or sophistication to succeed. He offered as proof the fact that they would frequently sell failed farms, making themselves landless again (field notes, July 24, 1996).

My host from the university introduced me to João Batista, the author whose work had originally piqued my interest in the city of Rolim de Moura. I also interviewed them about the economy of the area. Both of my informants recognized a sharp decline in the popul ation and economy of the município, which they ascribed to political failures. They pointed out that of the 180 saw mills once present in the town, only a handful now operate. The soils were better, too, early in the colonization effort. The professor, who lives on a farm outside of town, reports that the rice grew twice as tall in the early years as it does now (field notes, July 25, 1996). Again, explanations for these declines common in the literature are not accepted by local elites. At the time of my visit, 87 percent of the município had been deforested, so the decline in lumber yards should not have been surprising (EMATER-RO, 1996). The decline in crop yields is inevitable, given the rapid loss of soil fertility following the first few seasons under cultivation following a slash-and-burn clearing.

The improvements in transportation in and out of Rolim de Moura suggest that while it remains fairly isolated and difficult to reach, its connections to other urban centers are growing. Not only have connections to the BR-364 been improved, but Rolim de Moura now also serves as a gateway to the small towns Santa Luzia d’Oeste, Alta Floresta do Oeste, and other towns to the south and west of Rolim de Moura.

Friends in Porto Velho suggested that I seek out Solimar Bergamin, the owner of a music shop in Rolim de Moura, the only person I met there who speaks any English. He had come to Rolim de Moura as an accountant for various timber-related enterprises, but started the music shop a nd a T-shirt printing business when he saw the timber industry declining. In this way, his experience is an exemplar of the economic diversification that I describe in more detail below. I interviewed him about the city and the timber industry while he took me on a driving tour of the area.

Our tour included a sawmill owned by his brother (Photograph 11), where I learned that a single, 2.5 meter section of a large tree can supply all of the wood needed to build a modest house. We also visited a plyw ood manufacturing plant (Photograph 12), which my host had not visited previously. Unlike the sawmills, this business was expanding, and already employed about 50 people at the time of my visit. One reason for the expansion of the plywood industry is that serengeira and mahogany, which are preferred for higher-value processing, have been seriously depleted in the region. Other species of wood, suitable for plywood, are still relatively abundant. The sawmills themselves now rely on wood concessions fr om indigenous groups for the remaining supply of the higher-value species (field notes, July 26, 1996).
Photograph 11: Sawmill in Rolim de Moura

Observations along BR-364

1. Observations at a Hydroelectric Dam

In a ro om adjacent to the control center for power distribution from the Samuel Dam, a miniature museum has been created, describing the efforts to rescue animals that had been threatened by the inundation of rain forest land behind the dam. It is not clear who the intended audience for the museum is, given that access to the dam itself is very closely regulated. In addition to photographs and narrative of the rescue operations, many of the animals themselves have been preserved and are kept in the museum. A room overlooking the control panels is filled with hundreds of stuffed birds, monkeys, sloths, and other animals. This rather grim memorial raises at least some doubt as to the effectiveness of the rescue operations (field notes, June 15, 1996).

During my visit to the hydroelectric plant, it was evident that a significant area of rain forest had been inundated by the reservoir. It was only during subsequent travel many kilometers to the south of the dam that I became aware of the scale of the inundation (field notes, July 21, 1996). The impoundment created by the dam plant is about fifty kilometers long, with an average width of approximately ten kilometers and a maximum width of greater than twenty kilometers. The inundated area, therefore, is approximately 600 square kilometers. Both near the dam and at the margins of the reservoir, standing trees remain in the inundated area, indicating that a significant amount of rain forest was lost as a result of the reservoir’s construction, and that salvage operat ions, if any, were not adequate.

The manager on duty the day of my tour indicated that projected demand is sufficient to justify a plant similar to Uhe Samuel in the center of the state, and that some plans had already been made to construct such a plant. He did not expect, however, that an additional plant would actually be built for at least fifteen more years. Further analysis of trends in energy production is presented later in this chapter.

Informants in Porto Velho provided more insigh ts into the electrical industry. One fairly typical household in which two small air conditioners are used sparingly pays about US$ 100 per month for electricity. Even though the marginal costs of hydroelectric production should be low, the cost to the end user is high for two reasons. First, ELETRONORTE has high overhead attributed to rather lavish expenditures on its employees (for whom an entire subdivision has been built in Porto Velho) and perhaps also to corruption. Two of the seven turbines originall y planned for the Samuel Dam were never installed, apparently because part of the money needed to pay for them had disappeared. Second, many electric customers reverse the direction of their electric meters each evening, so that billable net usage appears to be negligible, increasing fees for those who do not tamper with the meters.

1. Extensive Research Activities: Examining Relationships Between Urban Growth and Deforestation< /LI>

 

 
 
 
 
 

1. Deforestation by Município

 

 
 
 
 
 

 

Name	Total Area (ha)	Area Deforested (ha)	Percent Deforested
Alta Floresta d'Oeste	722,424	76,567	10.6
Alto Alegre dos Parecis	328,378	64, 504	19.6
Alto Paraiso	274,294	51,765	18.9
Alvorada d'Oeste	297,336	71,075	23.9
Ariquemes	455,826	180,639	39.6
Cabixi	164,478	69,616	42.3
Cacaulandia	179,720	62,440	34.7
Cacoal	371,002	186,515	50.3
Campo Novo de Rondônia	550,283	44,223	8.0
Candeias do Jamari	658,896	102,890	15.6
Castanheiras	93,387	43,541	46.6
Cerejeiras	812,180	123,653	15.2
Colorado do Oeste	166,623	90,482	54.3
Corumbiara	436,588	218,084	50.0
Cost a Marques	1,464,131	42,972	2.9
Cujubim	391,083	28,174	7.2
Espigao d'Oeste	462,555	133,527	28.9
Gov Jorge Teixeira	480,235	69,597	14.5
Guajara-Mirim	2,422,570	64,538	2.7
Jamari	407,067	47,487	11.7
Jaru	337,148	180,035	53.4
Ji-Paraná	628,992	179,514	28.5
Machadinho d'Oeste	813,362	61,935	7.6
Ministro Andreazza	67,456	39,356	58.3
Mirante da Serra	120,037	31,522	26.3
Monte Negro	173,635	56,424	32.5
Nova Brasilandia d'Oeste	113,808	44,912	39.5
Nova Mamore	997,696	61,464	6.2
Nova Uniao	58,893	40,358	68.5
Novo Horizonte do Oeste	76,891	39,008	50.7
Ouro P reto do Oeste	220,443	120,730	54.8
Parecis	381,138	23,029	6.0
Pimenta Bueno	808,058	217,395	26.9
Porto Velho	3,522,718	245,207	7.0
Presidente Medici	188,296	103,821	55.1
Primavera de Rondônia	46,369	40,107	86.5
Rio Crespo	181,428	48,685	26.8
Rolim de Moura	138,157	120,426	87.2
Santa Luzia d'Oeste	145,302	71,375	49.1
Sao Felipe d'Oeste	80,389	41,698	51.9
Sao Miguel do Guaporé	765,271	63,432	8.3
Seringueiras	470,169	40,320	8.6
Teixeiropolis	54,759	22,495	41.1
Theobroma	217,863	90,489	41.5
Urupa	82,201	40,360	49.1
Vale do Anari	234,154	23,316	10.0
Vale do Paraiso	100,748	72,971	72.4
Vilhena	1,321,188	113,114	8 .6
State Total			17.1

Source: EMATER-RO 1996
 

1. Land Conversion for Urban Use

 

Clearly, then, land conversion for urban use is a mechanism linking urban growth and deforestation, but the mechanism is not always present, as some neighborhoods eme rge without the need for new clearing. More importantly, the observation that the mechanism exists is useful, but it would be more useful to have an idea of the magnitude of its importance. That is, is urban sprawl a significant cause of deforestation in Rondônia? This may be stated more formally as

H₀: The clearing of land for urban expansion is not significantly related to overall clearing of land in Rondônia.

To test this hypothesis, simple correlation is used to test for a relationship between the amount of urban land and total deforestation at the município level.

I arranged to examine maps and images in the offices of the State Secratariat for Environmental Development (SEDAM), in order to estimate the extent of urban sprawl in each município in Rondônia. Large-scale (1:250,000) color composites of LANDSAT Thematic Mapper Images were available for a variety of years between 1988 and 1995. Highway maps were available to aid in the identification of towns appearing on the images. Images were available for twenty-one of Rondônia’s fifty-two município seats, representing about 85 percent of the total urban population. Because of the high spectral reflectance of roads and buildings in the visible portion of the spectrum, each urban place exhibits both distinct color and distinct texture on the composite or the image, facilitating the calculation of the size of the urban area.

Table 8 presents the areal extent of each município seat (urban place) and of each município as a whole. To facilitate comparison, the latest available areal extent of each município is used. Urban places occupy an average of 0.16 percent of each município, ranging from a low of 0.01 percent in Vila Nova Mamoré to a high of 1.08 percent in Rolim de Moura. Vila Nova Mamoré is among the smallest of Rondônia’s cities in population, while Rolim de Moura is very large for its population size, prin cipally because of the unusual design of its street grid described in the section on Rolim de Moura above. The largest city in both population and area is Porto Velho, comprising 99 square kilometers, but it occupies only 0.28 percent of its município, which also has the largest surface area. The total areal extent of urban places was extrapolated on a per-capita basis from the 85 percent of the population in the sample. Because a significant positive correlation (p = 0.0000) was found (b = 0.0269, r² = 0.5364) between the area occupied by urban sprawl and the extent of deforestation in each município, the null hypothesis is rejected. Because the independent variable is the area of each urban place, rather than its population, this finding relates most directly to the question of the impacts urban sprawl itself, rather than to secondary relationships discussed elsewhere in this chapter. The implications of this finding are discussed further in Chapter V below.

 
Table 8: Areal Extent of Selected Urban Places
 

	1995 Land Area (km2)			1991 Population
City	Urban	Município	% urban	Urban	Município	%urban
Alta Floresta do Oeste	6.0	8,598	0.07%	9,379	31,980	29.3%
Alvorado do Oeste	2.2	2,982	0.07%	7,580	21,045	36.0%
Ariquemes	22.0	4,704	0.47%	45,283	83,684	54.1%
Cacoal	19.3	3,809	0.51%	44,091	78,934	55.9%
Candeias do Jamari	2.8	6,868	0.04%	N/A	N/A
Cerejeiras	7.5	10,751	0.07%	14,737	21,608	68.2%
Guajara-Mirim	12.0	25,214	0.05%	27,406	32,583	84.1%
Jar u	9.8	2,910	0.34%	27,675	63,535	43.6%
Ji-Paraná	27.0	6,899	0.39%	75,414	97,799	77.1%
Machadinho do Oeste	3.8	11,716	0.03%	4,891	16,765	29.2%
Mirante da Serra	1.0	1,254	0.08%	N/A	N/A
(Vila) Nova Mamoré	1.2	10,113	0.01%	3,724	7,248	51.4%
Novo Brasilandia d'Oeste	3.6	1,160	0.31%	5,031	15,798	31.8%
Novo Horizonte do Oeste	0.3	834	0.03%	N/A	N/A
Ouro Preto do Oes te	4.0	3,238	0.12%	23,586	83,857	28.1%
Porto Velho	99.0	35,929	0.28%	229,788	287,534	79.9%
Presidente Medici	3.0	1,694	0.18%	10,803	30,064	35.9%
Rolim de Moura	16.0	1,487	1.08%	28,272	59,751	47.3%
Santa Luzia	1.1	1,188	0.09%	4,287	< P ALIGN="RIGHT">12,328	34.8%
Sao Miguel do Guaporé	2.2	7,815	0.03%	2,218	19,458	11.4%
Urupa	0.8	850	0.09%	N/A	N/A
Sample Total	244	150,013	0.16%	564,165	963,971	58.5%
Rondônia Total	287*			659,327	1,132,692	58.2%

*Note: Total urban land area for the state of Rondônia is estimated.

1. Food

 

 
Table 9: Area Planted, Major Crops, 1994

Município	Cotton	Rice	Beans	Manioc	Wheat	Banana	Cacao	Coffee
Alta Floresta d'Oeste	300	3,750	25,289	1,000	7,920	429	0	4,190
Alto Paraiso	0	4,080	280	420	3,984	180	2,003	10,828
Alvorada d'Oes te	3,000	1,500	3,300	350	5,000	350	0	2,530
Ariquemes	0	3,250	1,143	200	1,990	940	6,300	3,000
Cabixi	50	4,000	3,300	540	5,800	260	25	250
Cacaieiros / Novo Horizonte	1,650	3,360	3,094	840	3,528	80	25	3,225
Cacaulandia	1 00	1,500	700	500	4,000	1,083	6,008	3,500
Cacoal	2,496	9,517	11,283	1,410	16,038	1,325	1,772	20,797
Campo Novo de Rondônia	0	550	70	659	637	208	9	42
Candeias do Jamari	0	2,000	125	3,000	900	346	15	300
Castanheiras	70	1,472	1,174	400	2,016	32	0	1,228
Cerejeiras	400	4,400	2,196	594	5,400	240	34	182
Colorado do Oeste	0	8,250	17,800	600	16,200	546	306	163
Corumbiara	638	4,820	9,480	450	9,480	200	472	1,000
Costa Marques	150	2,200	2,400	635	2,200	200	0	700
Espigao d'Oeste	500	900	1,600	475	3,787	175	0	1,550
Gov. Jorge Teixeira	800	3,250	2,760	720	4,720	2,060	2,115	2,530
Guajara-Mirim	0	1,150	600	1,500	1,260	850	0	15
Jamari	0	1,310	25	643	140	70	10	50
Jaru	1,300	11,410	6,558	880	11,690	3,600	6,000	3,625
Ji-Paraná	2,130	7,000	6,000	2,880	7,700	1,010	1,062	2,366
Machadino d'Oeste	80	5,000	1,500	3,800	3,640	980	1,200	6,200
Ministro Andreazza	372	2,264	3,573	470	5,758	375	345	4,990
Mirante da Serra	86	3,480	5,220	525	3,372	1,800	460	3,375
Monte Negro	0	3,900	1,620	800	5,950	1,177	670	6,250
Nov a Brasilandia d'Oeste	600	4,900	9,800	200	9,000	56	20	5,760
Ouro Preto do Oeste	1,122	10,535	9,250	1,960	12,557	6,720	3,906	6,643
Pimenta Bueno	2,748	4,657	6,320	800	7,968	320	22	4,248
Porto Velho	0	1,190	137	3,064	638	750	27	126
Presidente Medici	2,500	6,500	3,500	850	7,500	500	0	3 ,200
Rio Crespo	0	3,949	348	1,240	1,237	726	713	5,699
Rolim de Moura	1,360	5,376	3,200	1,400	7,056	112	0	5,376
Santa Luzia d'Oeste	540	2,700	9,7 98	650	6,462	200	27	3,200
Sao Miguel do Guapore	968	6,650	2,400	200	4,000	40	0	4,000
Seringueiras	60	4,680	720	540	5,346	120	0	3,120
Theobroma	300	2,200	2,000	750	2,200	196	1,933	1,611
Urupa	552	2,200	5,000	657	4,800	500	300	3,500
Vale do Paraiso	70	3,750	1,800	1,000	4,500	2,000	977	5,200
Vila Nova do Mamore	100	2,400	400	463	2,550	141	0	180
Vilhena	0	1,300	300	850	2,600	20	20	80
Total	25,042	157,300	166,063	38,915	211,524	30,917	36,776	134,829

Although production data are readily available, detailed information about the food market was difficult to obtain. Although ample data exist about the production of food in Rondônia, no data could be found to indicate the e xtent to which the expansion of agriculture in the region is driven by the demand for food in local urban centers. This question has occurred to government researchers in Rondônia, and a plausible methodology has been suggested, using data generated from road-side tax-collection stations along BR-364 (field notes, August 1, 1996). In theory, these data would provide a great deal of information about the food market in Rondônia, and indeed several informants indicated that such a study was alread y underway at EMBRAPA, a research center located between Porto Velho and Candeias. I telephoned and visited contacts at EMBRAPA, however, who informed me that the effort to quantify interstate food shipments had been abandoned as too difficult.

Food prices and dietary choice, however, do provide some clues. Beef is extremely inexpensive and popular, comprising a much more important part of the diet in the north than elsewhere in Brazil. Although a lot of the beef is consumed locally, it would not be accurate to say it is produced for local consumption. It is consumed locally precisely because so much beef is produced for other markets that the local price is especially attractive. Manioc is consumed extensively, and is strictly a regional food, so its production can be viewed as locally induced (field notes, June 21, 1996). Most packaged foods are expensive, and originate in Paraná or elsewhere in the South (field notes, July 8 and August 6, 1996). Even many local crops, such as rice, are trans ported to the South for packaging before they are sold in grocery stores in Rondônia’s cities, apparently because much of the commercial food business is dominated by businesses that have packaging operations located in the South. Almost everybody in Rondônia’s cities grows at least some food at home: virtually every house I visited or observed had at least one fruit tree. Farmers’ markets are an important source of food for some people in urban centers, however, indicating that at least some of the area devoted to fruits and vegetables in Rondônia is used to stock urban markets.

This led to the following hypothesis:

H₀: Agricultural land in Rondônia is used to supply urban markets in the state, rather than export markets.

To test this hypothesis, the production of food crops is compared to population size. To reduce the effects of spatial scale on the analysis, the comparison is made both at the município scale (n = 40) and at the microregi on scale (n = 8). In each case, simple correlation is calculated between the area devoted to six basic food crops in 1994 and population size in 1995. At the município scale (n = 40), b = -0.7461, but r² = 0.0439 (p = 0.1942). At the microregion scale, (n = 8), b = 0.0278, and r² = 0.0169 (p = 0.7575). Thus, the null hypothesis is rejected, as no significant correlation between local markets and clearing for agric ulture is found. The implications of this finding are discussed in Chapter V below.

1. Construction and Wood Consumption

 

Table 10: Yield of Forest Products, Rondônia 1965 to 1990

 

Year	Brazil Nuts (T)	Rubber (T)	Cut wood (M3)	Charcoal (M3)	Fire Wood (M3)
1965	824	5,997
1966	1,025	3,295
1967	1,587	4,000
1968	3,313	4,965
1969	2,412	4,824
1970	3,230	4,357
1971	3,357	4,336
1972	4,480	5,706	3,822
1973	2,050	5,548	10,292
1974	1,725	4,227	18,451
1975	3,127	1,118	3 0,962
1976	3,247	2,415	44,897
1977	2,103	3,713	66,935
1978	1,503	1,903	101,676
1979	1,826	2,103	318,109
1980	1,201	2,016	307,001
1981	784	3,119	491,914
1982	833	4,177	581,372
1983	1,466	3,779	768,712
1984	1,392	4,018	1,256,307
1985	563	4,621	1,320,213
1986
1987	316
1988
1989	907	4,629	2,255,352	963	967,667
19 90	1,472	6,105	1,903,646	867	809,340

De Bortollo’s (1996) account o f the development of Rondônia’s timber industry is instructive. Numerous, small sawmills (pica-pau) emerged around 1970 to serve the housing needs of new settlers in Rondônia (De Bortollo 1996). There was no tradition of export-oriented timber, and only a very small portion of the wood that was removed in the clearing of the land made its way to the timber mills. As the development projects grew during the middle 1970s, the fledgling timber industry drew some investment from the Center-So uth (e.g., São Paulo). These projects, intended to supply the local market, were not sufficiently successful to maintain a thriving timber industry. It was not until the late 1970s, when investors arrived who had a strong interest in harvesting wood for export to a national market, that the industry reached a significant size.

As noted in the field observations presented above, houses of recent construction in urban areas of Rondônia are typically constructed of wood (Photograph 7) or o f brick (Photograph 6). According to an interview with Solimar Bergamin, my contact in Rolim de Moura with timber-industry experience, a typical wood house measures four meters by five meters and requires about 70 planks, each measuring 2.5 meters by 30 centimeters by 2.5 centimeters, for a total volume of 1.3 cubic meters per house. A 2.5-meter section of a single large tree (Photograph 13) provides enough lumber for one such house. The construction of brick houses also results in the consumption of some w ood, because fire wood is used to fuel the brick ovens, which are common throughout the state.

Photograph 13: Timber for processing, Rolim de Moura
 

Photograph 14: Felled Trees Along BR-364, Jamari, Rondônia

< IMG SRC="Image15.gif" WIDTH=539 HEIGHT=315>

The question of whether local demand for timber is a significant factor in overall demand is posed in the form of the following hypothesis:

H₀: The demand for timber in Rondônia’s urban places does not explain increases in timber harvests.

Table 11: Timber for Housing, By Município
 

 

Município	Average Size of Wooden Houses (m2)	Estimated Wood Needed Per House (m3)	Prevalence of Wood as Housing Material	Number of Urban Houses	Total Wood Demand for Urban Houses	1994 Deforestation to Date (ha)
Ariquemes	58.1	3.8	56%	10,445	39,437	485,204
Cacoal	53.7	3.5	64%	10,364	36,174	225,871
Cerejeiras	60.1	3.9	16%	3, 416	13,334	188,157
Colorado do Oeste	43.2	2.8	96%	3,500	9,830	160,098
Costa Marques	42.9	2.8	38%	1,381	3,852	223,291
Espigao dOeste	54.7	3.6	53%	2,585	9,188	133,527
Guajara Mirim	42.3	2.7	25%	6,073	16 ,689	126,002
Jaru	42.1	2.7	3%	6,250	17,115	340,121
Ji Parana	52.6	3.4	40%	17,574	60,062	179,514
Ouro Preto do Oeste	79.8	5.2	38%	5,357	27,778	328,436
Pimenta Bueno	51.7	3.4	61%	5,970	20,076	322,229
Porto Velho	38.2	2.5	20%	53,838	133,773	467,981
Pres. Medici	55.3	3.6	50%	2,521	9,060	219,808
Rolim de Moura	51.1	3.3	85%	6,645	22,086	274,350
Vilhena	54.0	3.5	50%	8,263	28,977	331,198
Rondônia		3.1		144,182	447,431	4,005,787

1. Analysis of Electrical Power Trends

 
Throughout the 1980s, Rondônia was self-reliant in electrical power generation. Both consumption and production grew, with gross production generally exceeding consumption by approximately 25 percent (da Costa 1992). By 1995, however, consumption at least equaled production (Figure 7).
 

Documents made available by the state planning agency (SEPLAN, 1996) after the completion of my fieldwork confirms that small hydroelectricity plants are under a ctive consideration. The documents, intended to promote economic investment in Rondônia, were prepared at the micro-region level -- equivalent to the first seven municípios in the state -- but have been compiled into one document that outlines the prospects for investment throughout the state. As my contact in Rolim de Moura indicated (field notes, July 24, 1996), the unreliability of electricity is viewed as a major impediment to industrial development. For this reason, the discussion of each of the seven microregions includes a discussion of the availability of power and of future plans.

Although the documents are promotional in nature, they are circumspect in their treatment of the electric-power situation in the state. They cite interupted service in all of the microregions except Porto Velho, which is the location of the Samuel hydroelectric plant. In the other six microregions, private diesel-powered energy sources are used in about fifty percent of existing businesses. The discussi on of Ariquemes mentions its high-tension connection to the Samuel plant as a mitigating factor, but nonetheless describes the same service limitations described for most of the other microregions. The discussion of Ji-Paraná mentions the future power plant described by the engineer I interviewed at Samuel, and indicates that the new plant will have a capacity of 512 megawatts. Interestingly, the document does not provide a time horizon for that project, perhaps because it is as uncertain as my conta ct suggested it is.

The most interesting aspect of the discussion of electricity in these documents was the specific identification of sites for the small-scale hydroelectric projects that were mentioned by my contact at SEDAM (field notes, July 21, 1996). At the time of publication in 1996, only two such projects had been completed, but 43 more were planned (Table 12). Both of the completed projects are located in Rolim de Moura and together represent less than one percent of total planned capaci ty. Note that the aggregate potential of all of these plants would be less than half of the proposed plant in Ji-Paraná, and about equal to the capacity of the Samuel plant. The projects would be located throughout the state, in every microregion except Porto Velho, which has an adequate supply of electricity from the Samuel plant (Table 13). The result of these developments, if fully realized, would be the geographic dispersion throughout the state of the land-use impacts of further increases in hyd roelectric power generation.

 
Table 12: Micro-hydroelectric Projects, by Status
 

Status	Number	Aggregate Capacity (KW)
Identified	16	69,250
Under Study	9	33,750
Under Construction	18	109,920
Completed	2	1,750
Total	45	214,670

Source: SEPLAN, 1996

Table 13: Potential Micro-hydroelectric Plants, by Microregion
 

Microregion	Number of Potential Projects	Potential Aggregate Capacity (KW)
Ariquemes	7	25,850
Cacoal	8	78,150
Guajará-Mirim	3	17,250
Ji-Paraná	6	20,610
Rolim de Moura	12	31,410
Vilhena	9	41,400
Rondônia	45	214,670

Source: SEPLAN, 1996

1. Inter-Urban Changes

A relatively recent m aster’s thesis completed at UNIR addresses the question of economic diversification directly. Feitosa (1993) found a significant increase in the diversity of Rondônia’s economic activities during the period of rapid urban population growth. Between 1980 and 1989, for example, the number of industrial establishments in Rondônia increased by over six hundred percent (Feitosa 1993, 11). The major categories of industries operating in 1989 are shown in Table 14, which indicates the number of enterpr ises in each category and the number of municípios in which these enterprises could be found. Industries closely tied to resource extraction, such as timber and mineral processing, were most abundant in 1989, but a number of higher-order industries were also represented, such as clothing, food processing, and publishing.

Table 14: Industries Operating in Rondônia, 1989
 

Type of Industry	Enterprises	Number of Municípios Represented
Timber	1,018	18
Food products	455	18
Repair, Maintenance, and Conservation Services	192	18
Construction	182	9
Furniture	132	18
Non-metallic mineral production	99	14
Metallurgy	88	11
Publishing and Graphics	51	13
Mineral extraction and treatment	40	4
Clothing	29	9
Beverages	6	5
TOTAL	2,292	18

Source: After Feitosa, 1993.

More specifically, diversification of Rondônia’s timber industry was evident during the 1980s (Feitsosa 1993). In addition to an increase in the number of sawmills between 1980 and 1989, a number of other processing facilities, such as plywood makers, emerged during the decade (Table 15). As of 1989, one-third of timber-processing businesses performed a value-added function beyond simple milling.

Table 15: Wood Processing Facilities, Rondônia, 1980 and 1989
 

	Sawmills		Value-Added Wood Processing		Total Wood Processing		Proportion Value Added
Município	1980	1989	1980	1989	1980	1989		1980	1989
Porto Velho	12	17	3	38	15	55		20%	69%
Guajara-Mirim	2	4	1	6	3	10		33%	60%
Costa Marques	12	38		19	12	57			33%
Ariquemes	49	159	4	27	53	186		8%	15%
Jaru	11	35		32	11	67			48%
Ouro Preto do Oeste	18	49	2	21	20	70		10%	30%
Ji-Paraná	20	52	5	46	25	98		20%	47%
Presidente Medici	8	46	3	30	11	76		27%	39%
Cacoal	64	46	2	26	66	72		3%	36%
Pimenta Bueno	18	53		34	18	87			39%
Espigao do Oeste	8	34	4	21	12	55		33%	38%
Vilhena	22	43		21	22	64			33%
Colo rado		39		17		56			30%
Cerejeiras		45		10		55			18%
Rolim de Moura		66		19		85			22%
Alta Floresta		18		1		19			5%
Alvorada		32		2		34			6%
Santa Luzia		5				5
Total	244	781	292	1,521	268	1,151		9%	32%

Source: After Feitosa 1 993.

Building on Feitosa’s work, I have endeavored to test the following hypothesis:

The 1996 classified telephone listings ("Yellow Pages") for Rondônia (Teleron 1996), were obtained in the field. The listings are identical in form to classified listings found in classified telephone directories in the United States. They have limitations as a data source (some businesses do not have telephones, for example, while others may be listed in multiple ca tegories), but the listings provide far more detail than any available commercial or industrial census. They include listings for several thousand businesses in 1,040 categories. Listings are statewide, including 27 of Rondônia’s 52 município seats. Of these, 380 could be identified as non-government, service-sector categories, and are used in the analysis.

A matrix of urban places and business categories was created, with a binary code representing the presence or absence of a given bu siness type in each place. The number of service-sector categories present in each settlement is counted, and each city is assigned to classes based on the number of enterprise types found in it. The classes are tested against four indicators of a central-place heirarcy identified by Hochberg and Miller (1992, after Marshall 1989). Centrality, as used here, refers to the number of kinds of services represented in a given city. The indicators are:

1. Discrete stratification of centra lity. The differences in the centrality of towns within each level should be small, and the differences between each level clearly marked. (That is to say, the clustering should be meaningful, with greater variation between categories than among them.)
2. Numerical pyramid in the frequencies of successive orders. There should be fewer central places in each level than in the one immediately below it.
3. Incremental basket of goods. It should be possible to specify a group of goods and services that are available at all levels. At each level there should be goods that are generally not available at the next lower level. (In fact, the ratio of goods available at a given level to those available at the next lower level should be 2:1 or more.)
4. Interstitial placement of orders. Lower-level centers should appear on the map in the interstices between the higher-level centers.

Table 16: Munic&iacut e;pios by Service-sector Class
 

City	Count	Class
Porto Velho	323	I
Ji-Parana	158	II
Vilhena	112	II
Cacoal	100	II
Ariquemes	91	II
Pimenta Bueno	69	II
Rolim de Moura	55	III
Ouro Preto do Oeste	47	III
Guajara-Mirim	42	III
Jaru	42	III
Colorado do Oeste	27	III
Espigao do Oeste	26	III
Cerejeiras	19	IV
Presidente Medici	19	IV
Alta Floresta do Oeste	13	IV
Urupa	11	IV
Alvorada do Oeste	8	IV
Machadinho	5	IV
Nova Brasilandia do Oeste	5	IV
Costa Marques	3	IV
Jamari	3	IV
Nova Mamore	3	IV
Santa Luzia do Oeste	3	IV
Mirante da Serra	2	IV
Monte Negro	2	IV
Cabixi	1	IV
Ministro Andreazza	1	IV

Table 17: Characteristics of Service Classes
 

Class	I	II	III	IV
Intra-class range	0	89	29	18
Inter-class difference (with next higher class)	&nb sp;	165	14	7
Number of cities in the class	1	5	6	15
Number of services found in this class	323	223	107	40
Ratio of class to next higher		1.45	2.08	2.68
Number of services exclusive to this class and above	138	91	51
Ratio of exclusive services to total services in class	43%	41%	48%

Figure 8: Service Center Classes

Note: o = urban place not includ ed in database

1. Future Economic Activity

Figure 9: Municípios of Rondônia

The government of Rondônia has more specific economic plans related to the shift in emphasis from rural to urban populations and economies. Two documents obtained during fieldwork describe government plans: a four-year plan of economic goals for the period 1995 to 1998 (SEPLAN 1995) and a document encouraging private investment in Rondônia (SEPLAN 1996). The four-year plan, recognizing the presence of a labor surplus in Rondônia’s cities -- and the limitations of an economy based on primary-sector activities -- proposes a new generation of "development poles" in existing urban centers. The purpose of financial incentives and infrastructure investment in these poles will be to build upon the existing extractive industries to create employment in secondary and tertiary activities (SEPLAN 1995, 114). Most of the site-specific plans (Table 18) are in secondary-sector activities based on local raw materials. Notable exceptions include plans for the expansion of gr anite mining in Jaru, a computer-hardware industry in the border zone at Guajará-Mirim, and eco-tourism based in the border zone at Costa Marques.

Table 18: Urban Growth Poles - Planned Activities
 

Município	Proposed Activity(ies)
Vilhena	Soy processing, including animal feed
Cacoal	Cotton processing, including animal feed from residue
Pimenta Bueno	(Finished) clothing industry Ceramic and limestone industries
Rolim de Moura	Citrus fruit and juice processing
Ji-Paraná	Fish packing (based on aquaculture in Ariquemes and Cacoal)
Ouro Preto do Oeste	Furniture industries Coffee processing, including instant coffee
Jaru	Granite mining and processing of granite for façades and paving stones Leather processing and shoe production
Ariquemes	Tin smelting and artisan industries using semi-precious stones
Porto Velho	Processing of manioc and fruit pulps Furniture industries Fish packing (based on aquaculture in Guajará-Mirim, Nova Mamoré, Porto Velho, Candeias do Jamari, and Jamari)
Guajará-Mirim	Computer hardware and peripherals
Colorado do Oeste	Fish packing (based on aquaculture in Vilhena, Cerejeiras, and Pimenteiras)
Costa Marques	Eco-tourism Processing of vegetable products (Brazil nuts, rubber, medicinal extracts, and cosmetics)

Source: SEPLAN 1995, 122.

Consistent with the four-year plan, the state planning agency is actively seeking private investment for a variety of economic activities located in specific microregions of Rondônia, as shown on Table 19. For some activities, it is suggested that outside support would be needed, but in many cases state an d municipal governments are providing infrastructure and even buildings for projects of this kind. In Ariquemes, for example, several million dollars (U.S. equivalent) are proposed to fund a business incubation center for the wood-processing sector, particularly furniture making. According to Feitosa (1993, cited above), Ariquemes had more sawmills than any other city as of 1989. Although it also had more value-added wood-processing facitilies than most cities, the potentially redundant labor force (following a decline in forestry) is large, and both state and municipal planners are eager to encourage businesses that will absorb some of this labor.

Table 19: Potential Activities by Microregion, 1996
 

Activity	Vilhena	Cacoal	Ariquemes	Guajara Mirim	Porto Velho	Ji- Parana	Rolim de Moura
Dairy processing	X	X	X	X	X	X	X
Fruit juice and pulp processing	X	X	X	X		X	X
Furniture	X	X	X	X	X		X
Cattle raising	X	X	X	X
Coffee		X	X			X	X
Cold storage	X	X	X		X
Food processing		X		X		X	X
Tanning		X	X		X	X
Wood			X		X	X	X
Non-me talic mineral products		X			X		X
Agroforestry management	X		X
Cacao		X	X
Cotton processing						X	X
Fruit cultivation	X						X
Manioc products					X		X
Metallic minerals			X		X
Animal feed		X
Coffee - instant						X
Cold storage - fish						X
Cold storage - meat							X
Ecotourism				X
Fish				X
Fish culture		X
Rubber processing						X
Tropical fruit processing					X
Wood lamination						X

Source: SEPLAN 1996.