|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
1 INFORMAL URBANIZATION: EXAMINING SPATIAL PATTERNS AND PUBLIC POLICY INFLUENCES IN THE METROPOLITAN REGION OF CURITIBA, BRAZIL By JENNIFER JANE CANNON A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORID A IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS IN URBAN AND REGIONAL PLANNING UNIVERSITY OF FLORIDA 2010
2 2010 Jennifer J. Cannon
3 This thesis is dedicated to my husband, Casey His support, friendship, and enc ourag ement gave me strength and inspiration
4 ACKNOWLEDGMENTS I thank my chair Dr. Joseli Macedo for her mentoring, constructive feedback and faith in my ability as a researcher and pl anner. I also thank her for hosting me in her beautiful home t own of Curitiba, a place that became an adored home away from home and a source of inspiration on what planning can accomplish. I thank Andres Blanco, a member of my committee, for his patient coaching, tedious reviews of long drafts, and for giving me confidence in my ideas I thank the participants in my surveys for their honest and open participation. Their relentless strive for a better life l eft me inspired to work hard and effectively towards improved urban living for all The field research was enriched because of the assistance and companionship received from many planners and research partners. I appreciate all the support received from Oscar, at the Institute for Research and Urban Planning of Curitiba (IPPUC ). His frank insights on my research and the region were invaluable. I thank Lorena, Gilberto Celso, and Bernadette from Housing Company of the State of Paran (COHAPAR ) for imparting their knowledge on the complexities of informal settlement and regularization. I am immensely gratefu l for the valuable advice, wealth of information they shared, and accompaniment on field visits. I appreciate the assistance and kindness of Malu and Melissa, at the Coordination Agency of the Metropolitan Region of Curitiba (COMEC ). Those that joined me on the many field visits made my research possible. I thank Lidianne, Anelkis, Lianne, Renata, and Shani for their participation in my field study. All of you made the bumpy rides on bu ses and strolls through the city more enjoyable. I thank Lianne, Re nata, and Lidianne for e nlightening me on the complexities of the Portuguese language. I appreciate the extremely helpful advice that Cerian offered me on remote sensin g
5 I thank my family including my mother, brothers, and sisters for pushing their li ttle sister along and encouraging me to complete what I started even when circumstances were challenging. Even though he is no longer with us, I thank my father, a social worker whom showed me the joy of helping others I thank my in -laws, Errol and Robi n, and my grandparents, Ben and Bobbie, whom offer love and support, and an unrelenting faith in my abilities. Above all, I thank my husband, Casey, for being my rock. I appreciate his belief in my abilities and steadfast support especially when I was fa r away on field visit s I also thank him for nurturing my intellectual curiosity and sharing his adoration of science, research, and passion for making the world a better place I would never have been able to complete this ambitious thesis research with out this su pport and for all these reasons I am filled with gratitude.
6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ...................................................................................................... 4 LIST OF TABLES ................................................................................................................ 8 LIST OF FIGURES .............................................................................................................. 9 LIST OF ABBREVIATIONS .............................................................................................. 11 ABSTRACT ........................................................................................................................ 12 C H APT ER 1 INTRODUCTION ........................................................................................................ 14 Researc h Questions ................................................................................................... 17 Hypothesis Statements ............................................................................................... 17 Research Objectives ................................................................................................... 17 Relevance to the Field ................................................................................................ 17 Methods and Scope of Work ...................................................................................... 18 2 LITERATURE REVIEW .............................................................................................. 2 3 Historical Origins of Curitiba, Brazil ........................................................................... 23 Historical Summ ary of Urban Planning in Curitiba, Brazil ......................................... 25 Informal Settlement: Public Policy and Urbanization Patterns ................................. 28 Defining Informal Settlement ............................................................................... 28 Historical Perspective: Informal Settlement and Government Response ......... 30 Major Factors of Informal Urban Growth ............................................................. 38 Spatial Charact eristics of Informal Settlements .................................................. 41 3 METHODOLOGY ....................................................................................................... 46 Method for Measuring Urban Growth ......................................................................... 47 Land Use Land Cover Classification Categories ................................................ 48 Summary on Land Cover Classification Accuracy .............................................. 49 Method for the Identification of Informal Urban Land Uses ....................................... 51 Steps for Informal Urban Detection: Indicators and Rule -Based Categorization ................................................................................................... 54 Identification of at -risk areas with the use of indicators ............................... 54 Visual interpretation ....................................................................................... 55 Rule-based filtering ....................................................................................... 57 Discussion on Informal Urban Identification Method .......................................... 60 Met hod for Analyzing Spatial Patterns ................................................................ 62 Community and Government Agent Interview Methods ............................................ 63
7 The Case of Curitiba: Public Policy Influences from 1991 to 2007 .......................... 64 Social Interest Initiatives in the Study Area ......................................................... 64 Informa l Settlement Regularization in the Study Area ........................................ 67 The Environmental Protection of Land in the Study Area .................................. 73 4 DATA ANALYSIS AND RESEARCH RESULTS ....................................................... 76 Population Growth Findings ....................................................................................... 76 Urban Growth Data Analysis Findings ....................................................................... 80 Urban Growth Changes from 1991 to 2007 ........................................................ 81 Urban Expansion: Fragmented or Compact? .................................................... 87 Informal Urban Growth Results .................................................................................. 90 Discussion on the Decli ne of Informal Urbanization ........................................... 96 Discussion on Influential Factors for Reduced Informal Urbanization ............... 97 Analysis of Public Policy Influences on Informal Urbanization ................................. 98 Inside of Curitiba: Public Policy Influences on Informal Urbanization ............. 101 Outside of Curitiba: Reduced Informal Urbanization ....................................... 105 D iscussion on Increased Density in Periphery Towns ...................................... 110 Locational Determinants Results ............................................................................. 111 Findings on Hazardous and Ecological Fragile Spatial Trends ........................ 111 Discussion on Hazardous and Ecologically Fragile Trends ............................. 112 Results on Socio-Economic and Bus S ervice Locational Trends .................... 114 Discussion on Socio-Economic and Bus Service Locational Trends ............... 119 Key Findings from Community Surveys ............................................................ 123 Discussion on Community Survey Findings ...................................................... 125 Summary of Research Findings ............................................................................... 125 Discussion on Overall Results ................................................................................. 130 Method Transferability .............................................................................................. 133 5 RECOMMENDATIONS AND CONCLUSION .......................................................... 136 Recommendations for a more Sustainable Future .................................................. 136 Sustainable Urban Form and Growth ................................................................ 137 Social Inclusion .................................................................................................. 138 Environmental Preservation and Restor ation ................................................... 138 Holistic Considerations ...................................................................................... 139 Conclusion ................................................................................................................ 142 APPEN D I X A INTERVIEW AND COMMUNITY SURVEY DOCUMENTS .................................... 148 B SATELLITE IMAGERY DETAILS AND LAND USE LAND COVER CLASSIFICATION .................................................................................................... 157 LIST OF REFERENCES ................................................................................................. 159 BIOGRAPHICAL SKETCH .............................................................................................. 166
8 LIST OF TABLES Table page 2 -1 Rate of informal housing increase from 1974 to 2005, in Curitiba ....................... 33 3 -1 Rules for informal urban categorization ................................................................. 58 3 -2 Ranking results of informal detection for 2007 ...................................................... 59 3 -3 Informal housing growth, excluding homes undergoing regularization, from 1974 to 2005, in Curitiba ........................................................................................ 70 4 -1 A comparison of population growth in Curitiba and the Metropolitan Region of Curitiba ................................................................................................................... 77 4 -2 A comparison of population growth in Curitiba and the periphery towns ............. 78 4 -3 Urban area growth from 1991 to 2007 .................................................................. 80 4 -4 Urban area growth changes from 1991 to 2007 ................................................... 87 4 -5 Informal urban area growth from 1991 to 2007 ..................................................... 93 4 -6 Average annual rate of change for informal urban growth ................................... 93 4 -7 Informal urban percentages ................................................................................... 93 4 -8 Average annual informal area growth, inside and outside of Curitiba ................ 105 4 -9 Urban area growth in social interest zones ......................................................... 105 4 -10 Informal urban areas in periphery towns ............................................................. 109 4 -11 Comparison of periphery towns informal urbanization rates .............................. 109 4 -12 Informal urban areas in environmentally sensitive, hazardous areas ................ 113 4 -13 Combination of spatial variables .......................................................................... 129 4 -14 Summary of spatial trend findings associated with informal urbanization .......... 131 4 -15 Key government program influences on informal urbanization .......................... 132
9 LIST OF FIGURES Figure page 1 -1 Study area map ...................................................................................................... 22 2 -1 Regional view of Curitiba.. ..................................................................................... 23 3 -1 2000 orthophotograph map: variations in informal urban areas ........................... 61 3 -2 Regularization of the Zumbi Informal Settlement: urban form changes from 2000 to 2007, orthophotograph map ..................................................................... 72 4 -1 Portion of urban growth inside and outside of Curitiba, i n the Metropolitan Region of Curitiba ................................................................................................... 77 4 -2 1991 land use land cover classification map ......................................................... 82 4 -3 2002 land use land cover classif ication map ......................................................... 83 4 -4 2007 land use land cover classification map ......................................................... 84 4 -5 Map of urban growth changes from 1991 to 2007 ................................................ 86 4 -6 Urban population density, in the study area .......................................................... 89 4 -7 Comparison of the urban population growth rate with the rate of urban area expansi on ............................................................................................................... 89 4 -8 Map of informal urban growth, 1991 to 2007 ........................................................ 94 4 -9 Map of informal urban growth and regularized areas ........................................... 95 4 -10 Comparison of informal to overall urban area growth rates .................................. 97 4 -11 Map of population densities in informal urban areas and affordable hous ing program areas and social interest zones, in 2000 .............................................. 100 4 -12 Map of social interest zone urban growth from 1991 to 2007 ............................ 104 4 -1 3 Urban population density changes in periphery towns ....................................... 108 4 -14 Conservation and karst areas map ...................................................................... 115 4 -15 Water, floodplains, and steep slopes map .......................................................... 116 4 -16 Map of bus routes, affordable housing initiative areas, and electrical facilities 120
10 4 -17 Urban centers map ............................................................................................... 121 4 -18 Income level concentrations map ........................................................................ 122
11 LIST OF ABBREVIATION S APA reas de Proteo Ambiental or the Environmental Protection Areas CGS Campina Grande do Sul (Municipality in the Study Area) COHAB Companhia de Habitao Popular de Curitiba or the Affordable Housing Company of Curitiba COHAPAR A Companhia de Habitao do Paran or the Housing Company of the State of Paran COMEC A Coordenao da Regio Metropolitana de Curitiba or the Coordination Agency of the Metropolitan Region of Curitiba FRG Fazenda Rio Grande (Municipality in the Study Area) GIS Geographic Information System GPS Global Positioning System IBGE Instituto Brasileiro de Geografia e Estatstica or the Brazilian Institute of Geography and Statistics ILD L oteamentos Irregulars C landestinos or an Irregular Land Division IPPUC Instituto de Pesquisa e Planejamento Ur bano de Curitiba or the Institute for Research and Urban P lanning of Curitiba IS Informal Settlement RMC Regio Metropolitana de Curitiba or the Metrop olitan Region of Curitiba, Brazil SEHIS Setor Especial de Habitaco de Interesse Social or t he Special Sec tor of Social Interest Housing SJP So Jos dos Pinhais (Municipality in the Study Area) SUDERHSA Superintendncia de Desenvolvimento de Recursos Hdricos e Saneamento Ambiental (Superintendant of the Development of Hydrology Resources and Environmental Sanitation) UTP Unidades Territoriais de Planejamento o r the Territory Units of Planning
12 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Arts in Urban and Regional Planning INFORMAL URBANIZ ATION: EXAMINING SPATIAL PATTERNS AND PUBLIC POLICY INFLUENCES IN THE METROPOLITAN REGION OF CURITIBA, BRAZIL By Jennifer Jane Cannon May 2010 Chair: Joseli Macedo Major: Urban and Regional Planning The rapid population influx to urban regions in the global south has led to an unprecedented demand for low -income housing. The inability to meet this demand has resulted in the development of entire neighborhoods referred to as informal settlements that are comprised of self -constructed housing that lack s infrastructure and legal permission. The cumulative effect of informal land occupation has swayed regional urban ization patterns Y et informality is influenced by the limitations posed by the urban system which is a product of various political, economi c, environmental and social forces This study builds on the theory that the s patial trends of informal settlements are a product of various limitations and degrees of tolerance and therefore the patterns are not random Curitibas early execution of a my riad of planning initiatives such as regularization, urbanized lot provision, and sustainable watershed plans serves as an intriguing longitudinal case study given the potential for changes in the customary pattern of informal urbanization. This research is needed since the i mpact of newly implemented public strategies on informal urban growth patterns is poorly understood and public policy has been a powerful influence, whether intended or unintended, on the informal production of space. This study examines informal urban ization patterns before
13 and after the implementation of key policies in the Metropolitan Region of Curitiba from 1991 to 2007. In addition, this research focuses on analyzing the dominant location circumstances of informality in the RMC In many cities, informal settlements typically concentrate along the flanks of metropolitan areas within marginal landsca pes in a segregated pattern. This study found that informal settlement followed this generality given the tendency to oc cupy floodplai ns, conservation lands, and riverbanks ; in a decentralized growth pattern outwards into the urban fringe and periphery near social interest h ousing initiative area s and bus service However, this research found that the high annua l growt h rate of informal settlements fro m 1991 to 2002 diminished as a whole and inside aforementioned ecologically fragile landscapes from 2002 to 2007. The strategy used in Curitiba encouraged consolidated development and lessened the overall rate of informal settlement urban g rowth, most significantly in conservation areas, though it did not diminish the overall consumption o f ecologically fragile landscapes This research found that the government strategy was more sustainable and responsive to the urban poor when both reactiv e (r egularization) and proactive ( increasing the availability of affordable serviced lots ) features were incorporated A balanced reactive, pro active government strategy could help enforce conservation policies Although the approach in greater Curitib a is multi -faceted, it did not succeed in de -segregating low income po pulations This re search provides insights on informal urban ization and concludes with recommendations on ways to support inclusive s ustainable urban development
14 CHAPTER 1 INTRODUCT ION The year 2008 marked the beginning of an era of urban dominance given that urban areas are now the primary place of residence rather than rural hinterlands. In the global south, this urban growth arrived rapidly enough to outpace the provision of basic sanitation such as access to safe drinking water and wastewater facilities In fact, o ne third of all urban dwellers of the global south are left to reside in housing with little to no basic infrastructure (UN -HCA, 2006). Th ese substandard conditions di sproportionately result in tremendous public health detriments that are primarily felt by low income communities (Martinez et. al, 2008). In Brazil, the substandard housing problem is imminent since 84 percent of the population is already urbanbased and 86 percent of all new population growth is expected to take place in urban ar eas (Montgomery et al., 2008). T his rapid urban growth has paralleled an unprecedented demand for low income housing in Brazil that has left an estimated three to eight million in need of housing (FJP, 2001; Fernandes, 2005). Often impoverished urban new -comers resort to their ingenuity to acquire shelter and this has resulted in the development of illegally built neighborhoods comprised of self -buil t housing. These self -construc ted communities, referred to as informal settlements are basically defined by substandard housing conditions and some form of illegality, such as o ccupation of land that the dweller does not possess or development not in compliance with government regulati ons (Gilbert and Gugler, 1994; Ward, 1983 ). The rapid development of informal settlement s and their hidden character makes it challenging to detect their existence, let alone quantify their extent. Despite this challenge estimates indicate that between 30 and 8 0 percent of the urban population of
15 the global south lives in i rregular settlement s (D urandLasserve et al., 2002) and that over half of the new housing stock is built illegally (UN -HABITAT 1996) T h e s e estimate s suggest that informal urbanizat ion can be a major force in the way an urban region expands D istinct patterns have emerged in the types of landscapes left vacant and unpr otected from invasions and where irregular settlement is socially, economically, and politically tolerated in an urb an region. In general, the urban poor settle on the left over land, regularly in the least desirable areas of a city (Gilbert, 1998). Mo st places where land invasions are tolerated have common characteri stics that are rationally grounded based on distinct limitations posed to informal settlers in an urban region. In some instances, informal settlements were planned in a deliberate manner by economic and political interests (Ward, 1983 and Burgess, 1981) and as a result of political circumstances (UN -HABIT AT 1996) Ward (1983) argues that the widely quoted epithet spontaneous is misleading since settlements are premeditated and planned by agents (p. 35) The areas selected for settlement are carefully chosen to ensure tenure longevity and a strong negotiating position for their occupation (UN HABITAT, 1996 ). For example, informal settlements often emerge on land dangerous for construction such as flood plains or steep hillsides (UN -HABITAT 1996) since these areas have little commercial value. Whil e Gilbert and Gugler (1994) draw on the idea that within limits the poors respo nse to poverty is rational, innovative, and nearly always more perceptive than they are given credit for (p. 117) Several studies point out that informal settlement grow th is most likely to occur in the urban fringe and periphery (Gri ffin and Ford, 1980), in vacant marginal areas ( UN-HABITAT 1996) that have access to bus services and employment (Berner, 2001; OHare and Barke, 2003).
16 Household preferences vary from city to city and the poors options for housing are severely l imited by local c onditions (Gilbert and Gugler, 1984). Another influence of in formal settlement is public policy. A study by Dosh and Lerager (2006) found that different public poli cies w ere a d eciding factor of differences in informal settlement construction and choice of whether to occupy public or private land in Lima, Peru and Quito, Ecuador In addition, public policy can sway the spatial patterns of informal urbanization. Areas with less stringent regulations might attract informal development since the construction costs are lower (UN -HABITAT 1996) By exam ining regionwide geographic patterns of informal urbanization in relationship to public interventions the influences that public po licy has on shaping informal urban patterns whether intended or unintended can be recognized T his research examines the set of prevalent locational features that result from limitations posed to informal settlers, in the urban construct of the Metrop ol itan Region of Curitiba, Brazil ( Regio Metropolitana de Curitiba RMC) from 1991 to 2007. A comparison of the spatial patterns before and after the implementation of key public policies will help in det ecting their influence (or lack thereof) on informal urbanization. The government approach in Curitiba combined social interest housing programs regularization, and social interest zoning with sustainable watershed, and conservation p lans The implementation of the multi -faceted government strategy for dealing with informal settlement more sustainably in greater Curitiba may hav e lowered the rate of info rmal urbanization The guiding research questions hypothesis statements, and objectives for this longitudinal study are as follows.
17 Research Questio ns What are the chief locational trends of informal settlement between 199 1 and 2007 in greater Curitiba? H ow has the government strategy influenced (or not) the rate of informal settlement urbanization and prevaili ng spatial patterns ? Hypothesis State ments As indicated in literature, n ew informal settlement growth is most likely to occur in vacant areas located in the urban fringe and periphery that ha ve access to bus service. I n the context of the RMC, inf ormal settlement often is located in floodplains conservation areas, and near rivers and social interest housing programs T he government strategy used in Curitiba helped to reduce the overall informal urbanization rate and encourage consolidated urban growth Research Objectives Measure the inform al urb an growth of greater Curitiba from 1991 to 2007 and compare informal urban areas with locational features to detect major spatial patterns Using the findings, discuss the potential reasons for informal urbanization trends, focusing on th e influ ences of the government approach used in Curitiba and conclude, with recommendations on ways to achieve inclusive sustainable development. Relevance to the Field This r esearch measur es the prevalent types of land circumstances of informal settlement in the RMC to determine how implemented public strategies dynamically influenced informal settlement patterns This type of evaluation is particularly useful for recognizing the combination of features within urban planning approaches t hat facilitate inclus ive sustainable urban development and growth T wo of t he Millennium Development Goal (MDG) targets set out t o improve the lives of 100 million slum
18 dwellers and halve the proportion of the population without basic sanitation and infrastructure (UNDP, 200 0) To support the MDG targets a UNHabitat (2003 ) report stressed the need for sub-city level geographic data showing informal settlement areas to assess the conditions of urban poverty and inform public policy. In addition, Agenda 21 calls for the im provement of the social, economic and environmental quality of human sett lements the monitoring of strategy impacts particularly in areas affecting marginalized groups and the encouragement of sustainable landuse planning practices (UNDSD, 1992) Th e resultant geospatial database, a product of this research, could help in the planning for the improvement of informal settlement conditions and proactively reduc e the tendency for growth to occur in ha zardous environmental circumstances In addition, t he recommendations based on this research, provides insights on how urban regions could promote sustainable urban planning and the right of the urban poor to decent housing conditions The results of this research will be useful on practical and academic le vel s in the fields of sustainable urban development and planning, affordable housing, growth management, and disaster preparedness. Methods and Scope of Work Most academic research within the field of urban planning is distinct in its analysis of complex p roblems and by its inclusion of a space for comprehensive review of how various forces of a problem interface to yield certain outcome s Yin ( 2003, p. 13) points out that a case study is an empirical inquiry that investigates a contemporary phenomenon wi thin its real -life context, especially when the boundaries between the phenomenon and context are not clearly evident. The issue of informal settlement is a complicated phenomenon with unclear boundaries that invo lves economic, social,
19 political, and env ironmental factors in a particular urban context Thus, studying the circumstances of individual urban regions is important for understanding how these contextual forces interact and produce certain outcomes. T he flexibility of a case study meth od provid es such a space needed for researching multi -faceted urbanistic matters such as informal settlement Yin (2003) offers reasons for single -case study research design such as research that tests a unique or extreme event or for a case that is representati ve of typical conditions For this research, Curitiba could be considered a unique case given its early development and implementation of a multi -faceted government strategy for responding to informal settlement that includes social interest, conservation a nd sustainable urban planning elements. In addition, t he premise behind case studies is to draw insights from context -dependent knowledge that arises from experience and direct observations (Flyvbjerg, 2006). Thus, the pluralistic use of multiple sourc es of data that details the dimensions of a given reality validates findings and yields a more complete grounded truth (Yin, 2003, Flyvbjerg, 2006). The scope of work for this case study involves the collection and comparison of geographic and population / housing data analyses, direct field observations and interviews with informal settlers and government agents .1 Essentially this research design allows for a comparison of inf ormal urban spatial patterns before and after government intervention. The firs t type of data analysis consisted of detecting informal settlement growth with the use of geographic data and field visits. This step helped t o measure spatial 1 A mixture of primary and secondary data resources were incorporated in the research. The primary data resources consisted of geographic recordings in the field and survey findings and data reso urces that were secondary were population count data and aerial photographs.
20 patterns and the extent of informal urban land conversions during the study time period. The fi eld visits were a crucial part of this research since it allowed for direct observations to be made on the reality of urban development in the region In a critic on Geographic Information System ( GIS ) and planning, Lejano (2008) warns of the tendency for geographic -based studies to isolate themselves from the most legitimate source of integration, which is experience (p. 273). During the field visits, observations were made on the complexities of urban development and these experiences helped to ground geographic findings in reality. The qualitative data analysis consist ed of interviews and surveys and the review of local government plans The interviews and plan re view helped draw conclusions on the underlying reasons for informal settlement location al trends Overall t his research approac h integrated various data re sources to derive results and conclusions that are useful on multiple levels. T he study area for this research encompasses Curitiba and the significant population concentrations closely co nnected to Curitiba in the RMC T he urbanized portions of ten municipalities that surround Curitiba were included in the study area. Akin to counties in the United States the municipalities in t he study area include all of Pinhais and the urban ized po rtions of Colombo, Fazenda Rio Grande, Almirante Tamandar, Piraquara, So Jos dos Pinhais, Araucria, Quatro Ba rras, Campo Magro, and Campina Grande do Sul (Figure 1 1 ).2 The study area boundary, totaling approximately 344,362 acres was outlined to include land at least 1/2 kilometer from known urban areas extending from Curitiba to ensure displaced informal urbanization 2 The municipalities included in the study area needed to be at least 75 percent urban (IBGE, 2007 ) The urban core of Campo Largo, located west from Curitiba was mostly excluded from the study since this urban area is separated by a significant swath of rural land 10 kilometers in width. The 26 municipalities of the RMC were not all included in the study area since several of these municipalities are separated from Curitiba by large rural areas.
21 was captured. Th e new informal settlement s appearing after 1991 were often located in the municipalities surrounding Curitiba; cons equently, their inclusion was critical for more thoroughly measuring informal urban expansion associated with Curitiba. The choice to include th e towns located in Curitibas urban periphery is a uniqu e aspect of this study and it avoids the arbitrary rest riction that political boundaries pose F ollowing the definition of an urban area as being an expanse of land with a population density that is high relative to the density of the surrounding area ( OSulliv a n, 2009 p. 2 ), the informal urban growth that is an extension of Curitiba wa s captured more thoroughly The urban growth associated with informal settlements and the resultant spatial patterns are products of various political, economic, public policy, and social dynamics. Consequently, the followin g section o ffers a historical perspective and synopsis on Curitibas urban planning. The second part of the literature review provides an informal settlement background that summarizes the historical origins, urban growth influences, and histor y of govern ment responses
22 Figure 11 Study area m ap
23 CHAPTER 2 LITERATURE REVIEW Historical Origins of Curitiba, Brazil Curitiba is the capital of the state of Paran in the southern region of Brazil, strategically located 40 9 kilometers south of So Paulo. R egionally, the city is situated on a plateau at 945 meters of altitude and is bordered to the east by the Serra do Mar mountain range. The nam e, Curitiba means abundant with pine trees (Macedo, 2004) and is in recognition of the Paran Evergreen Pine (Araucaria angustifolia ). Five rivers flow through the C ity of Curitiba including the Passauna, Barigui Atuba Bel m and Iguau Rivers Most of these rivers drain into the Iguau watershed and are surrounded by floodplain s. As shown in Figure 2-1 vario us natural features including mountains (to the east and north), rivers, and floodplains have largely influenced the path of formal urban growth. Figure 21 Regional v iew of Curitiba Source: COMEC, 2008. T he village of Curitiba was founded by Portuguese explorers in 1693. In the beginning, Curitiba served as a gold prospecting supply center and this start helped Curitiba flourish to become an important trading post to cattle ranchers and for those in
24 need of agricultural goods and pack animals ( Schwartz, 2004). In 1842, after Curitiba was chosen as the capital for the state of Paran, the urban center expanded to surpass its urbanization status of village to city This expansion is partially due to the arrival of immigrants, mostly from Germany, Poland, Italy, Ukraine, Japan, Syria, and Lebanon (Rabinovitch, 1992). After gaining railroad access in 1885, Curitiba developed into a lively hub for trade. Over the course of the 20th Century, Curitibas population grew quickly In fact, the population of Curitiba reached 180,000 persons in 1950, then rapidly swelled to 430,000 inhabitants in 1960, and has tripled to approximately 1.8 million in 2007 (IBG E, 2009). The extraordinary growth over the last centur y in Curitiba is reflected by some of the highest urbanization rates in all of Brazil One of the main sources for this rapid growth is rural urban migrations brought on by agricultural modernization and indus trial development (Schwartz, 2004 ). O nce Curitibas population approached million during the 1960 decade informal settlement became more significant (M C et al 2007) .3 T he pace of urbanization in the City of Curitiba seemed to f uel informal settlement growth. The economic and quality of life opportunities offered in Curitiba have been linked to the urban growth rate In 2007, m ost of the formal employment was in the industrial secto r (38 percent), followed by the service (26 percent), c ommercial ( 25 percent), and construction (seven percent) sectors (MTE and IPARDES ).4 C uritiba has become a place with ample employment opportunities and this is evidenced by an unemployment rate that has remained below seven percent for most of 2007 (IPAR DES and IBGE ). Today 3 Translated by author. 4 These figures do not include employment in the informal sector a common source of labor for low income persons.
25 different industries have clustered in the Metrop o litan Region of Curitiba (Regio Metropolitana de Curitiba RMC) and the effects of urbanization economie s have led to economic benefits for job -seekers and employers alike.5 Former Mayor Cassio Tanigu chi (2005) points to quality of life advancements in the city, mostly related to improved education, as attracting growth. Nationally Brazilian urban regions of the south are more affluent in com parison to the northeastern region (Fra nko, 2007) However great i ncome inequality has been an unequivocal characteristic of Brazil for multiple decades and Curitiba is no exception. In 1991 and 2000, all of the municipalities in the study area had a lower Gini coef ficient than Curitiba in other words the periphery municipalities had less drastic in come inequality than Curitiba .6 Although the periphery t owns were less unequal, most had greater rates of poverty than Curitiba (IBGE, 2004 ). Collectively these ratings reflect income divisions between Curitiba and most of the exterior towns in the study area The rapid population growth in Curitiba was partially stimulated by economic growth and quality of life opportunities though it has led to income segregation in greater Curitiba. Along w ith economic considerations, urban planning policies have infl uenced urban growth in Curitiba. Histor ical Summary o f Urban Planning in Curitiba, Brazil Since the mid20th century, several l anduse p lans and infrastructure improvements have been develop ed to manage the urban growth in Curitiba. The first comprehensive master plan for Curitiba was developed in 1943 by the French urban 5 OSullivan ( 2009) describes urbanization economies as the crossover of different industry c lusters or agglomeration economies in one urban area. 6 The respective Gini coefficient ratings for 1990 were 0.55 for Curitiba and an average of 0.458 for the other municipalities (IBGE). The closer the Gini coefficient rating is to one, the higher deg ree of income inequality
26 planner, Alfred Agache (Macedo, 2004). The Agache Plan consisted of elements that recommended a wide spatial distribution of avenues; the layout of districts or specialized centers secondary from the downtown with functions such as civic and university ; several codes and zoning restrictions; and measures to conserve open spaces and guide development to accommodate future gr owth (IPPUC, 2007). The enduring compon ents of the plan are the widened avenues and the designation of specialized centers. Continued street widening and the fear of replicating the sprawling growth of So Paulo inc ited the development of another Master Plan. Subsequently a national competition was held and architect Jorge Wilheim produced the winning master plan for Curitiba (Schwartz, 2004) Although there have been minor modifications to zoning and land use regulations the concepts presented in t he 1965 Master Plan of Curitiba has generally guided urban development till today. The guidelines of the 1965 Master Plan of Curitiba are (IPPUC, 2007): Linear growth of a center served by tangential fast traffic roads; Hierarchy of city streets; Preferre d development of the city along the northeast -southwest axis following historic and spontaneous trends; Multiple centers; Increased density; Expansion and adjustment of green areas; Delimitation of areas of pedestrian domain; and Creation of a unique urban landscape. One of the key elements of the 1965 Master Plan was the establishment of a linear branch road network that is reinforced by land use and bus transport features Basically the linear plan directed highdensity growth and intensive land uses along five a xes and each of these axes are comprised of three parallel str eets .7 The trinary 7 The intense densification surrounding these five axes gradually tapers as one travels away from an axis
27 ar terials (or three parallel streets) have a central road that accommodates two exclusive bus lanes in the interior and one bus lane in each direct ion on either s ide Since m any of the axes connect to centers o f activity, the polycentric radial spatial structure of Curitiba was fortified. In addition to bus transit improvements, t he adoption of the 1965 Master Plan led to the formation of two integral planning or ganizations: the Afforda ble Housing Company (COHAB) and the Urban Research and Planning Institute of Curitiba (IPPUC). IPPUC was instituted to guide the implementation of the plan, specify and lead several projects and manage the c omprehensive development of Curitiba (Macedo, 2004). I n 1971, t he military dictatorship in power at the time, appointed Jaime Lerner, an architect and planner, and former director of IPPUC to the position of mayor of Curitiba (Schwartz 2004 ). During his tenure, Jaime Lerne r challenged conventions by testing ur ban planning innovations in Curitiba. Some of the initiatives that took place during his tenure include the conver sion of a car oriented road to a downtown ped estrian street in 1972, the creation of the Green Exchange recycling and employment pr ogram in 1989; and the conservation of periphery linear parks that function as storage for flood waters in the 1970s to 1980s (IPPUC, 2007). Perhaps, one of the most acclaimed urban improvements a ugmented during Lerner s tenur e is the bus rapid transit system.8 This integrated bus network extends into other metropolitan districts and influence s urbanization patterns, even those associated with informal settlement After 1985, the federal government shift ed from a military di ctatorship to a democracy and this shift a ffected local governments particularly in requirement of 8 The bus rapid transit agency in Curitiba estimates that 2.4 million trips are accommodated daily by the bus service (UR BS, 2007).
28 elections In addition, a new national constitution was adopted in 1988 which lifted censorship, encouraged public participation, and gave Brazilian local g overnments the responsibility to regulate policy areas that were previously managed at the federal or state level (Menegat, 2002). Despite the call for public participation Irazbal pointed out that Curitiba (as of 2005) has la rgely maintain ed a topdown mode of urban governance and this is exemplified by the scarcity of citizen education programs and opportunities for citizen dialogue in the development of plans To summarize, Cervero (1998) classified planning milestones in Curitiba in three distinct p eriods including the formative years (1943 1970) that launched bedrock planning principles that guided development decisions (p.267), then an active period (19721988) of plan execution and lastly the regional transit service r efinement period (1989 1998 ). T he urban planning in Curitiba and the RMC has continued to progress and i n the last several decades attention has been given to implementing planning initiatives to deal with informal settlement and env ironmental degradation. T he following sect ion provides an overview of informal settlement growth and the government approach for handling the issues Infor mal Settlement: Public Policy and Urbanization Patterns Defining Informal Settlement The concept, informal settlement has been defined as clusters of i lle gally developed housing. Gilbert and Gugler (1994) elaborate the i nformal settlement definition to emphasize low -income populations. They suggest that informal settlement communities espouse four characteristics including: self -construct ed housing; lack of planning permission; inadequate infrastructure and services; and low income populations. Informal settlements are commonly referred to as favelas in Brazil, irregular
29 occupations, squatter settlements, shanty towns, slums, loteamentos irregulars clandestinos (irreg ular/clandestine land divisions o r pirate subdivisions ), to name a few I nformal settlements are not always built by the occupiers and instead contracted help is often used. Additionally, not all settlers are homeowners ; in fact many are renters (Gilbert and Gugler, 1994) In addition, both privately and publicly owned lands are occupied by informal settlers Although informal urbanization differ s from city to city and within the same urban region, Fernandes (1997) described two types of informal urban areas in Brazil including favelas and loteamentos irregulars Fernandes (1997) defines favelas as land invasions comprised of occupants without any type of property tenure documentation that invade privately or publicly owned property that often is unsuitable for construction due to ecological or geological conditions. These land invasions are regularly planned and they typically consist of homes or shacks built with varied materials such as red bricks and recycled items (Fe rguson, 2005). Fernandes notes that favelas: all look the same: the usually steep, hilly areas are densely occupied, and the spontaneous pattern of the land division is irregular and inarticulate. Combining improvised streets, alleys and staircases, the road system is confusing and not suitable for access or general traffic. Favelas lack in almost every element of urban infrastructure and collective equipment, and the precarious standard of most dwellings makes for unhealth y and unsafe daily living ( 1997 p. 6 ). In contras t, loteamentos irregulars (also referred to as irregular land divisions or pirate subdivisions) tend to resemble formal development s more than favelas ; h owever they are still considered irregular due to precarious technical conditions, improper infrastructure, and noncompliance with government laws (Fernandes, 1997) E ven though the land may not be invaded, they are perceiv ed as informal (Gilbert, 1981). Fernandes (1997) details two urban development stages of irregular land division s
30 ( loteamentos irregulars ). Initially, lots are delineated by landowners, developers, and in some cases, by occupants Then the lots are commercialized, whereby some of the less desirable lots are first given freely and then, the more desirable lots are sold at higher rates. This process can be profitable since the potential for the municipality to provide infrastructure and services is heightened as this population become s voters (Fernandes, 1997). In addition, Fernandes (1997) describes the emergence of a social process in which a portion of those living in central favelas move and buy lots in peripheral loteamentos irregulars where the security of tenure is greater. Generally, favelas are more precarious and unimproved than irregular land divisions alt hough irregular land divisions tend to be more disconnected from the urban fabric. When applying these concepts to reality, this informal categorization is co mplicated by mixed favelaloteamento conditions imprecise boundaries exceptions to definition generalizations and the changes brought on with land regularization. Following the definitions offered above, t h is research uses the term s informal settlement to indicate favelas and irregular land divisions a nd the two are collectively referred to as inf ormal urban areas or informal urbanization. Historical Perspective: Informal Settlement and Government Response Around the middle of the 20th century, t he emergence and expansion of fave las broadened to become a significant phenomenon that penetrated various urban regions in Brazil Depending on the city, v arious reasons are offered to explain the growth of informal settlements In a study on the growth of favelas in So Paulo, Lloyd -Sherlock (1997) described how cortios ( or tenements ), typically located in central areas of the city were the main types of popular housing during the late 19th century up until the 1940s Then in 1942, w hen rent controls on tenements rendered these units less
31 profitable for developers several tenants were evicted The combination of tenement evictions i ncreased rural -urban migrations and industrialization are thought to have contributed to the initial formation of So Paulo s favelas (Lloyd -Sherlock, 1997) .9 Often there is not one single reason for informal settlement growth and instead a combination of factors that are political, economic, and social related, interact to spur informal settlement growth. Before reviewing the main government approaches for dealing with informal settlements, the r ole of clien telism in interactions between government and informal settlers should be recognized. Throughout different political regimes, the practice of clientelism has been a longstanding characteristic of Brazils political culture and one of the main form s for participating in politics (Macedo, 2000). Gi lbert and Gugler (1994) characterize a patron-client relationship as a reciprocal understanding, made informally through an unequal partnership, where one actor wields more power (the patron) over the other (the c lient) and as an exchange in which the patron typically grants economic and social favors t o the client in exchange for political support. In Brazil, various federal government regimes have used clientelism as a way to respond to housing issues (Macedo, 2000) Even today, the decision of where to upgrade informal settlements outside of Curitiba in the metropolitan region is largely politically -driven .10 9 The 1942 tenancy law prevented rent increases from occurring more than once every two years and led to the eviction of around 45,000 tenants between 1945 and 1947 (Lloyd Sherlock, 1997) In addition, it should be noted that the tenements (or cortios) did not necessarily offer better living conditions than favelas; they often had unsanitary unhygienic living conditions. LloydSherlock (1997) argued that no single factor alone was responsible for the recent growth of favelas and he pointed ou t other factors such as the decline in wages that contributed to the growth of favelas during the 1970 1980 decades. 10 An interviewee said that the decision where to regularize informal settlements in the RMC is politically based.
32 As informal settlements have grown, various government approaches have emerged During the 1950s and 1960s, the federal government opted to focus on the development of public housing. The Affordable Housing Foundation (Fundao da Casa Popular FCP) established in 1946 during the populist regime, was set up to supply low income housing in Brazil (Macedo, 2000). Although the FCPs role was expanded in 1952 to include social services, finance, public works and research, the program suffered from not setting income limits for participation to low -inc ome persons and from clientelism political practices (Mac edo, 2000). The public housing program experienced problems with not being able to meet the mounting housing demand and with discriminatory allocation practices (Sietchiping, 2005).11 In the mid 20th Century, slum eradicat ion interventions emerged as a r esponse to rampant informal settlement growth in many cities These interventions proved to be extremely deficient and actually amplified the number of families in need. In the site and service scheme, centrally located favelas were cleared after evictin g the inhabitants and sometimes these families were relocated to newly serviced lots outside of existing urban areas, often far from jobs In addition, inadequate lots were typically in place and families often became disconnected from their social networ k s. Thus these families often abandoned the subsidized units in favor of new land invasions. In Curitiba, similar favela removal approaches (desfavelamento) commenced in 1967 after i nformal settlement intensified in the city (MC et al 2007). This approach proved ineffective given that informal settlements continued to swell even after 20 percent of the population was relocated and nine neighborhoods were eradicated (MC et al., 11 T he main beneficiaries of these earlier public housing programs were often civil servants and middle income families (Sietchiping, 2005)
33 2007) In addition, the a doption of strict er inspection and building munic ipal codes aggravated the situation. Then d uring the 1970s and 1980s, when the nationwide financial crisis impacted the region, irregular housing growth prolonged Surveys conducted from 1974 to 1979 estimated tha t approximately 4,083 homes in 35 informal settlements expanded to include 6,067 homes in 46 settlements (MC et al. 2007). As shown in Table 2-1, the annual rate of informal housing growth accelerated at a r apid p ace, climbing over 10 percent during the 1980 decade and then to 20 percent during the beginning of the 1990 decade (MC et al., 2007). Then, after 2000 the annual rate of informal housing reduced from 17 percent to 1.7 percent. The factors associated with this decrease, particularly related to the government approach will be discusse d later Table 2 1 Rate of informal housing i ncrease from 1974 to 2005, in Curitiba Year Number of Domiciles Annual Rate of Domicile Increase (%) 1974 4,083 1979 6,067 1974 to 1979 9.72% 1982 7,716 1979 to 1982 9.06% 1987 11,929 1982 to 1987 10. 92% 1996 33,778 1987 to 1996 20.35% 2000 57,333 1996 to 2000 17.43% 2005 62,267 2000 to 2005 1.72% Data Source s : IPPUC, COHAB and domicile c ounts obtained from MC et al., 2007 Table by author. Beginning in the 1960 and 1970 deca des, Turner (1967, 1968 ) an d Mangin (1967) argued for the upgrading of informal settlements as the solution to the affordable housing problem. The advantages of upgrading favelas over public housing relate to increased functionality, flexibility in housing structures, lack of rent, potential for leasing extra rooms (Macedo, 2000) and decreased costs in comparison to public housing (Roy, 2005). In t he 1980 decade, the upgrading strategies were first implemented in Belo Horizonte and Recife Brazil (Fernandes, 2003). The u pgr ading approaches had
34 the advantage of dealing with low -income housing demand more effectively since it facilitates incremen tal develo pment (Peattie and Aldrete Haas, 1981). T he upgrading of informal settlements, referred to as regularizatio n has become one of the main forms of government response s to low -income housing issues in developing country cities (Gilbert, 1999). Ward (2003) p oints out two forms of regularization; the first implements some type of land titling and the second incorporates physica l upgrades service provision, and community development In Brazil, the term, urbanizao de favela indicate s regularization of informal settlements and this intervention can include the regularization or legalization of land occupation the installation of infrastructure, and the inclusion of educational and health facilities and social programs ( Huchzermeyer, 2004).12 In some cases, such as in Lima, Peru informal settlements can be regularized or legalized through the granting of a property titles indi cating security of tenure but will not receive electricity and other infrastructure provisions (Dosh and Lerager, 2006). Thus, regularization and urbanization is administered differently from city to cit y and country to country For example, the Favela-Ba irro program in Rio de Janeiro emphasizes granting property titles and in contrast, the Guarapiranga Program in So Paulo incorporat ed social assistance and public participation and identified new settlement expansion areas (Abiko et al., 2005). Opposite from the development stages of formal development, informal settlements upgrades generally undergo stages of development that begin with land occupation and home construction and ends with infrastructure and tenure security (Abiko 12 The term urbanizao de favela, translated into English, does not necessarily indicate favela urban growth or informal settlement land development Instead, this phrase indicates regularization that includes lot organization and community upgrades such as infrastructure installation.
35 et al., 2005).13 Perhaps, the most difficult stage of regularization is gaining security of tenure. During the 1990s, the security of tenure movement became a campaign supported by internat i onal agencies such as the United Nations Human Settlements Program (UN HABITAT) in respon se to problems with property eviction (Durand-Lasserve et al., 2002). Sietchiping (2005) asserts tha t [t]he security of tenure approach derives from the assumption that when the residents have the sense of appropriation, they also have the confidence to invest, upgrade or improve their environment (p. 284). T he expected benefits associated with land titles grants ha ve been a topic of discussion In a study examining informal housing in Bogot Colombia, Gilbert (1999) concluded that home ownership increased the perception of stability and was used to generate income from rentals; however capital accumulation derived from home sales was low In review of several case studies in different Latin American countries Ward (2003) found that [t] itling took an inordinately long time; it was expensiv e ; and by the time it was concluded, regularization appeared to add little to the perceived l evel of security (p.4) .14 Durand-Lasserve et al. ( 2002 ) found that granting individual property titles is rarely pos sible and as a result, they suggest combining tenure regularization with a range of options that are responsive to the urban and land market such as the provision of service and the increase of affordable shelter options. 13 The upgrading of informal settlements typically begins with a preliminary study that provides an assessme nt of the technical, physical and legal feasibility for regularization and serves to make contact with the inhabitants After the upgrades are deemed feasible, the registration of existing residents commences and the project is designed. The plans for subdividing lots and installing each lot with water, electricity, roads, drainage, telephone and sewer are drafted, normally in a manner that benefits the greatest number of families (Abiko et al., 2005). Lastly, the plans are executed and titles are granted 14 The 12 case studies that underwent review were located in Brazil, Colombia, Costa Rica, Ecuador, El Salvador, Mexico, Peru and Venezuela. Ward (1983) argues that [i]ntervention at the level of the supply of land is likely to have far greater reper cussions (p.34) than regularization.
36 The cost of r egularization has b een a topic of research. A study conducted by Abiko et al. (2005) examining the costs of favela upgrade programs in Rio de Janeiro, Salvador and So Paulo found that upgrades can cost two to three times more than the hypothetical costs of urbanizing a fo rmal residential development. Cost increases can be attributed to the construction in areas unsuitable for development and the need for additional facilities such as the construction of flood control struct ures and drainage storage areas In addition, inc reased infrastructure and relocation costs raised the overall expenses for regularization (Abiko et al., 2005). The choice in where to regularize is influenced by locational factors. Ward (2003) discusses how variations in informal occupants perceived r ights and in their extent of illegality can influence their potential for regularization and longevity of tenure. In summary, r egularization has become one of the main mechanisms for responding to informal settlement yet it has been burdened with irregul ar ity in its implementat ion inefficiency, and high costs In 2001, the Brazilian federal government created a law referred to as the City Statute and formed an agency, called the Ministry of Cities.15 The Ministry of Cities formulat ed nationwide guidelines to address the issues of informal settlements more comprehensively (Fernandes, 2 005). The City Statute proposed the use of various planning tools to make urban land accessible to low -income families, to control speculation in areas designated for social interest, to expedite the process of regularization of tenure, and to concede rights of use without formal property ownership. For example, the City Statute p rovides a provision for settlers to convene as a group to appeal for property 15 The 2001 City Statute is an outcome of the 1988 Constitution of Brazil. Prior to the 1988 Constitution, t he social function of property was first recognized in 1937 (Fernandes, 1997) Then in 1988, the Constituti on legally recognized two local government land provisions: private property rights as the basic principle of the economic order and the social function of property when it conforms to the master plan of the city (p. 20, Fernandes, 19 97).
37 own ership on la nd they have occupied for at least five years ( usucapio coletivo or collective adverse possession) (Macedo, 2008) Another legal tool, the real right to use concession ( C oncesso de Direit o Real de Uso) is a frequently used mechanism to acquire publicly owned land through a 50-year lease contract (Macedo, 2008). In addition, the new policy stipulates that public administrations offer concrete and acceptable conditions for the relocation of residents and that the right to housing continues to prevail ( Fernandes, 2005, p. 6). Overall, the City Statute recognizes instruments for legalizing land tenure for regularized informal settl ements stipulates the production of comprehensive plans and promotes policies that allocate land for social interest purposes. An example of a policy supported by the City Statute is social interest zoning. In cities with rapidly growing economies and populations urban market forces and land speculation tends to inflate property values and intensify shortages in the housing stock and in the amount of serviced property. Fernandes (1997) articulates the consequences of pervasive speculation an increase in the amount of vacant plots and a discontinuous urban fabric. In response to these issues, social interest zoning preven ts land speculation in certain areas through land ownership and subdivision provisions (Maia, 1995). In addition, social interest zoning legally permits regularization procedures and typically has more lenient development code restrictions. Basical ly, soc ial interest zoning allows low income families the opportunity to improve their housing conditions without the pressures of eviction. Social interest zoning has been implemented in some Brazilian cities (examples include Recife Curitiba, and Porto Alegre ).16 An overview of the policies implemented during the study time period, such 16 Typically families residing in social interest zoning contribute to municipal revenues by paying property taxes and public fees (Macedo, 2008).
38 as social interest zoning, in response to the informal settlement issues in the study area is provided in the last section of Chapter 3. Major Factors of Informal Urban Growth Great economic growth in a region can translate into increased housing demand and unmet housing needs Those who are unable to compete in this bidding war for homes and land must fin d other shelter options and most are excluded from purchasing desirable lo ts and housing .17 Generally, without government intervention, properties containing infrastructure that are perceived to be desirable and cost effective for development also tend to not experience informal settlement. In a study examining 52 cities of the global south, Arimah (2000) found that the spatial organization and scale of infrastructure networks can alter the spatial trends of housing demand. In addition, if the provision of infrastructure falls short of demand, the supply of improved land will be come inelastic and result in rapidly escalating land and housing prices (Arimah, 2000). Consequently th e urban market dynamics and the provision of infrastructure could be influencing the spatial trends of informal development. Another factor that impa cts the availability of low income housing and the capacity to self -construct housing is related to the availability of home loans. In fact, Ferguson and Navarrete (2003) suggests that a sustainable housing progr am in the global south should include viable finance options and small subsidies for critical infrastructure in addition to the provision of a broad range of affordable housing solutions including 17 A probit model examining Brazilian housing markets identified statistically significant demographic characteristics as increasing the likelihood for occupancy in informal settlements including: higher unemployment, lower quality of employment, lower level of schooling, higher household sizes, lower age, lower salaries and location in metropolitan regions (Morais et. al, 2005). Although these are common charact eristics of informal settlers population variation exists such as the level of income (extreme poverty to moderate income).
39 serviced lots and regularization. In 1964, t he National Housing Bank ( Banco Nacional da Habitao BNH ) of Brazil was set up to finance housing and home ownership. However, in 1967, after a high number of loan defaults, the bank shifted to fund mostly middle income housing (Macedo, 2000). Consequently, the BNH was abolished in 1986. The Federal Savings Bank ( Caixa Econmica Federal) instituted a building materials program for informal settlements; however, there was no national policy that effectively articulated the goals (Fernandes, 2005). U p until 2001 there were few if any available home loan options, particularly for low income populations. Then, in 2002, the Federal Savings Bank funding was reinstated and this organization began to provide housing loans to Brazilians of all income brackets Many researchers point to public policy as a substanti al driver of informal o ccupation. In developing countries, Roy (2005) suggests that affordability accrues in the absence of formal planning and regulation Payne (2001) discusses how c onfusing, complicated regulations, often based on some notional assess ment of what is accepted in Europe or North America favor those with more resources (p. 311). Minimum standards of development and infrastructure intended to protect persons from living in hazardous conditions might push the poor to unplanned hazardous a reas. L engthy administrative barriers for development impede incremental housing improvements and tend to preclude the urban poor from se curing legal housin g (Payne, 2001). The formal zoning requirements for development in central areas tend to foster middle to high income development while the lack of a stringent regulatory environment in the periphery has encouraged low -income settlement in the city outskirts (Fernandes, 1997) A comparative study, contrasting Quito, Ecuador and Lima, Peru found that
40 pu blic policy was the deciding factor of differences in the conditions of building construction and public versus private land occupation (Dosh and Lerager, 2006). In summary, v arious publi c policies could impact where in formal settlements tend to emerge an d their resultant spatial patterns, whether intended or unintended. In Curitiba, urbanization has progressed quickly and the amount of land suitable for development and serviced with infrastructure that is affordable to low -income populations has dimini shed. As discussed in the economic growth section, Curitiba has experienced substantial economic growth and as a result of this, housing and serviced land prices have accrue d and these price increases likely have led to population density increases Curi tiba regulates development with zoning codes and uses several other planning instruments in the municipality though it is uncertain if over regulation has push ed the urban poor to the city outskirts into unregulated areas The newly implemented social interest policies in Curitiba were formulated to cater to low -income populations a nd are considerably less stringent than typical zoning and building regulations (see last section of Chapter 3 for more i nformation on these policies ). Thus, the role that public policy maintains in displacing or influencing informa l settlement patterns merits analysis In addition, after 2002, t he reinstatement of home loan provisions likely supported the purchasin g of homes and imparted capital needed for making home improvem ents or additions. The institution of government fees such as through a municipal tax structure, can affect the availa bility of vacant land in Curitiba During a discussion with the Curitiba Finance Department, the progressive real estate taxation sy stem was explained. Progressively higher tax rate s are charged for property sales above R$30,000 In
41 addition, t he property taxes are greater for vacant land (one to three percent) than they are for residential (0.2 to 1.1 percent) or commercial land use s (0.35 to 0.8 percen t ). Conse quently, Curitiba uses t hese property taxes to incentivize the development of vacant parcels and reduce land speculation activities The urban system of each city involves different political, economic, environmental, and so cial factors and the interac tions between these factors result in different outcomes particularly pertaining to the spatial p atterns of informal urban growth. The urban system also poses limitations in where the urban poor are tolerated, not evicted, and in where their existence is more likely to endure. Based on these limitations the f ollowing section describes the common spatial trends of informal settlements and applies applicable locational theories to the case of Curitiba. Spatial Characteristics of Informal Settlements R esearch mostly presented through case studies offer s some basic guidelines on the typical geographic locations of informal settlements in a n urban region. F ernandes (2005) noted that most Brazilian cities with more than 500,000 residents have informal settlements and 80 percent of cities with a population of 100,000 500,000 have informal settlements This estimate indicates that informal settlements in Brazil are often loca ted in large cities Informal occupations often appear in areas hazardous for development, in vacant areas in the outskirts of cities where regulations are not enforced or nonexistent The UN-HABITAT Program (1996) pointed out that many new informal settlements have emerged on land dangerous for housing cons truction such as floodplains or steep hillsides since the 1960s Sietchiping (2005) described the common types of landscapes at risk of irregular occupation and these include : marginal or lower valued
42 property near indus trial areas and markets, riparian areas or wetlands, steep slopes, dumping grounds, and road right of ways. Many of these locational factors are in areas that ar e difficult, illegal, and expensive to develop Consequently, informal settlement could be tolerat ed by land owners more often on land with low development value Berner (2001) contends that land must meet two prerequisites for informal settlement: first the parcel has to be accessible by public transport, and second, it has to have a source of water. Gilbert (1998) argue s tha t the expansion of self -help housing in Latin America is regularly associated with public transportation improvements. A spatial and temporal analysis of Rio de Janeiro found that informal settlements were primarily perched on the hillsides in a scattered fashion throughout the city with the older favelas more centrally located than the newer ones, and their distribution was strongly influenced by access to public transit and employment (OHare and Barke, 2003). These studies suggest that access to criti cal amenities such as bus service water, and employment could be influential locational factor s In critical response to the notion that Latin American cities are evolving into North American western cities, Griffin and Ford (1980) developed a model ba sed on an examination of over one hundred Latin American cities that depict ed the trends in the socioeconomic organization of Latin American urban areas This model identified the prevalence of informal settlements to develop along the urban periphery. O verall Griffin and Ford (1980) outline a city structure that includes a viable central business district [CBD], a spine/sector, and a series of rings of decreasing residential quality (p. 405). Ford (1996) later up dated this model by adding a market se ction next to the CBD, a shopping mall and industrial sector in the suburban areas
43 The urban region of Curitiba exudes several of Fords spatial patterns, such as in having a distinct CBD T he extent that informal settlements dominate the urban peripher y will be discussed in the data findings section of this research. However, some signs i ndicate that gradual decentralized urban growth is occurring in greater Curitiba. During the 1990 decade, the metropolitan growth rate surpassed the City of Curitiba s rate ( see Table 4 -1 ). Today Curitiba has little vacant land available for development and the only remainin g area for low -income housing is located in the southern urban fringe (IPPUC, 2009). Curitiba has become compact to the point of increasing the am ount of densification and verticalization of development to house newcomers yet t hese high-rise developments are neither simple nor inexpensive to construct. A building structure that is over two stories in height requires special materials and skilled lab or both of which are limited resources for low -income populations (OSullivan, 2009). In addition, the growth of the secondary urban centers and conurbations in greater Curitiba has become more pronounced and contiguous since the 1980s. The term conur bation has been used to describe an urban growth trend observed since the early 20th Century. Park and Burgess (1967) describe the conurbation phenomenon as expanding a citys limits concentrically to merge into a variety of neighboring towns of which conn ect to each other to coalesce into one continuous population agglomeration. For example, adjacent to the eastern edge of Curitiba, the towns of Colombo, Pinhais, and So Jos dos Pinhais have merged into one continuous conurbation. Regionally, the conurba tion development of greater Curitiba reflects tension between the push of polycentric urban growth and pull of concentric urban growth towards the urban core. The role that conurbations maintain in housing low -income populations has been explored little y et is
44 recognized by the local government as substantial residential areas. Serra et al. (2004) found that t he informal housing supply has increased by 205 percent between 1991 and 2000 and nearly half of this stock increase is located in regions 10 to 20 kilometers from central Curitiba (Serra et al., 2004). Yet a substantial portion of the formal housing stock growth (34 percent) is located in the 10 to15 kilometer ring (Serra et al., 2004) This study suggests that both informal and formal housing growt h spurred conurbation development in the RMC.18 Many of the major spatial trends generally apply to the Metropolitan Region of Curitiba ( RMC ) though each city context has slight differences in the major informal urban landscape characteristics given that whatever vacant land is deemed low in value, mar ginal, and undevelopable varies contextually The societal organization and environmental circumstances specific to Curitiba limits and influences informal settlement trends. T he conurbation growth of the R MC could be pushing informal development in the urban fringe and periphery. In addition, the physical terrain of the RMC has a substantial portion of river and floodpl ain areas In fact, around 23 percent of the total stu dy area consists of riparian and lake areas and 9 percent of floodplains. Predominantly located outside of the city limits of Curitiba, around 29 percent of the study area is within conservation area s .19 Consequently, informal urbanization near rivers floodplains and conservation areas could constitute a large proportion of the marginal la ndscapes experiencing informal occupation in greater Curitiba. In addition, a 18 Around 31 percent of the RMC population lives in substandard housing (Serra et al. 2005). 19 Using GIS, the calculation of natural feature areas was based on GIS data from IPPUC, Mineropar and SUDERHSA Conservation areas are defined as UTP and APA conservation regions and parks. Approximately 4 percent of the study area is within karst topography. Generally karst areas were located in terrain, underlain by dissolving rock formations usually comprised of limestone or dolomite.
45 City of Curitiba municipal report suggested that informal settlements tended to locate in areas not of interest to real es tate sectors and clustered along railroads, highways, near high-tension electrical transmission lines, land with low property values ( fundos de vale), flood prone areas, and environmental protection areas during the 1960 to 1980 decades (MC et al. 2007). This research will explore the major spatial trends of informal urban areas likely pre sent in the study area, including location in floodplains and conservation areas and near rivers public housing, and bus service and the tendency for decentralized growth Lastly, the rate of i nformal urbanization could have decreased as the public program s and other features increased the access to viable affordable housing options This literature review offered a theoretical background for this research. The histo rical background of Curitiba summarized the urban pla nning and urbanization of greater Curitiba to provide a context for this research and the subsequent literature review described informal settlement and history of policy approaches. A more detailed des cription of the public policies implemented in the case study area during the study time period and a description of the potential spatial pattern influences is provided in the final section of Chapter 3. A method for exploring the research conjectures is described in the following section.
46 CHAPTER 3 ME THODOLOGY Following a case study method that focuses on the mixed methods research approach, this study used both quant itative and qualitative data The qualitative data included plan review and interview s while the quantitative data analyse s includ ed remote sensed land cover classifications area change measurements, and population/domicile c alculations The spatial trend analysis was guided by interview responses and in return, the interview s helped in explaining the rationale of many of the spatial trend findings. Collectively these linkages served to reinforce each other and enhance results. Basically t he method involves geographic analyses of informal urban growth and identification of spatial trends T o determine the urbanization patterns the extent of the urban area for 1 991, 2002 and 2007 was classified using remote sensing tools. Using the resultant built land covers and following a rule based method that integrated analysis using vari ous geographic datasets and high resolution aerial photographs, the informal urban areas were d emarcated T he informal urban areas were examined for dominant locational circumstances and urban form changes associated with pertinent public programs Interviews w ith members of informal communities and government agents were conducted to determine the perceptions on informal urban spatial patterns and to learn about the potential influences of public policy. The policy influences were measured with the use of geog raphic analyses and the review of several local government plans. Lastly, a set of policy recommendations, emphasizing the attainment of inclusive sustainable urbanization in light of the research results was offered A more detailed description of the m ethodology is described below.
47 Method for Measuring Urban Growth T o identify the locational determinants of informal settlement in greater Curitiba from 1991 t o 2007, geographic analyses was conducted R emote-sensed imagery and aerial photographs datasets were employed to identify the extent of urbanization during the study time period. Remotely sensed data provides an extremely useful instrument for identifying the alterations of land within and proximate to urbanized areas and for measuring the extent of urbanization (Jenks and Burgess, 2000). In addition, remote sensing facilitates regular observations of explicit patterns of land cover changes over a large geographic area (Schneider and Woodcock, 2008). Moller -Jensen et al (2005) used Lands at TM and ETM + imagery to analyze the expansiveness of urban growth in Accra, Ghana and found the approach to be cost effective and simple for monitoring growth and detecting patterns. Studies measuring the extent of residential types of urbanization must overcome challenges related to the complexity of a polluted spectral signature that includes a mixture of vegetation and urban pixels (Jensen, 2005). Consequently, the verification of land cover categories required extensive v isual interpretation of ortho photographs and field visits recorded with a Geographic Positioning System (GPS) This land cover classification required the collection of training samples that are represent ative of individual land cover classes. E ssentially these samples help train t he classification algorithm to create a land cover geographic dataset (Jensen, 2005) .1 Training samples collected durin g 2008 and 2009, were used to classify the 2007 image since these 1 The training classification algorithm is based on means, standard deviations, and covariance matrices fo r each image pixel.
48 conditions most closely reflect 2007 circumstances As a result of th ese field visits, around 85 training samples were collected in various informal urban areas 200 in formal urban areas, and 130 in vegetative areas. A summary of the training sample types is provided in Appendix B. The Landsat TM and ETM+ satellite imagery was chosen as the data source for the land cover classification analysis since th is satellite image type is freely available and it caters to regional studies dating back to 1990. 2 In addition, the Landsat image include s a large swath of land cov ering the entire study area in a single image. Using Erdas Imagine, a supervised classification approach was employed that relied on training samples based on geographic data, field study, and high resolution aerial photographs. As described below, a si mple set of l and cover classes were chosen according to study purposes Land Use Land Cover Classification Categories 1 Water: Water features covering areas such as rivers, lakes, dams, drainage ponds, wetlands with substantial amounts of standing water and canals were categorized. 2 Barren/Bare Soil: Large areas with exposed soil and fields cleared of a great amount of vegetation were classified as bare soil/barren. Often this barren land was cleared in preparation for agricultural, mining, or for formal r esidential/urban development (for the purpose of installing infrastructure ). 3 Vegetation: This category includes areas predominately comprised of fields, vegetated areas and green spaces. Golf courses, agricultural areas with vegetation, parks, yards, a nd community gardens fit within this category. Areas designation in this category may consist of a mixture of vegetation, shrubs and canopy cover. 4 Built/Urban: Built or urbanized areas, such as mixed use, commercial, industrial, and residential developm ents were categorized. Other examples within this category include big -box retail uses, airports, warehouses, industry parking lots, large hotels, large gas stations and railroad features. Mining facilities were included in this definition given the bui lding impact; though, the mining pits were categorized 2 The Landsat TM and ETM+ data sources are INPE, 2008 and UMGLCF, 2008.
49 as Barren/Bare Soil. In addition, this category also could include a housing initiative area supported by a government agency or public -private partnership.3 The decision rules chosen for the supervise d classifications were the maximum likelihood nonparametric rule and the minimum distance parametric rule. After attaining classified images, a majority filter with a three by three pixel moving window was applied to ascertain th e most prevalent land c over types throughout the study area. The resulting classification of the 2007 satellite image was used as a basis for the classification of the 1991 and 2002 satellite images. T o account for errors particularly within the rural urban transition area, e d its were made for all three classifications using available Geographic Information System ( GIS ) datasets that were h igh er in spatial accuracy than the satellite images Jensen (1982) discussed the difficulty with measuring landscape changes in newly deve loped regions that contain a mixture of agriculture, residential, and industrial land uses As expected, the newly developed urban areas required editing to better reflect actual conditions. Summary on Land Cover Classification Accuracy The accuracy was tested for the latest study year since the field visits were conducted closest to the 2007 time period and given that the other classifications followed the log ic produced from the 2007 classification. As shown in A ppendix B the s upervised classification for the 2007 Landsat TM satellite image yielded an overall accuracy of 91 percent. The training sample points used to test the accuracy of each 3 Urban/built land covers often are identified by the presence of impervious surfaces such as roads, buildings and parking lots The urban concept has been defi ned as an area with at least 50 percent impervious surfaces (Schneider and Woodcock, 2008; Arnold and Gibbons, 1996).
50 class were randomly selected with the use of Hawths Tools to eliminate spatial biases .4 The Kappa analysis is a discrete multivariate technique of use in accuracy assessment (Jensen, 2005 p. 506 ) that compares the classified image to the reference data and produces a K statistic. The overall Kappa statistic for the 2007 classification was 81 percent and thi s indicates strong agreement or accuracy between the classification map and the ground reference information (Jensen, 2005, p. 508). The producers accuracy measures the likelihood that the reference pixel is correctly classified and the evaluation detects the error of omission (Jensen, 2005). In other words, this test indicates how well a certain area can be classified. True to its name, the users accuracy is a measure of how useful and reliable the classified image is in its ability to represent the reality in the field. For example, the results suggest that 95 percent of the time the producer was able to consistently identify the built class correctly and 93 percent of the time the user would find built areas while out in the field. The water and bu ilt classes yielded the most accurate results in comparison to the other categories. More training samples were collected for the built and vegetation classes while few bare soil/barren training samples were collected due to accessibility issues. In addi tion, the bare soil/barren category has a similar signature to the built c ategory; thus the results required editing to improve accuracy particularly in the urbanrural transition region. Overall, the identification of built areas proved useful for narrow ing down the scope of review for identif ying informal urban areas and measuring the patterns of regional urban growth. 4 Half of the total training sample points were used to test for accuracy. Hawths tool is a freely available extension that can be use d with ArcView GIS software.
51 Method for the Identification of Informal Urban Land Uses The resultant urban areas derived from the land cover classifications were used to narrow the scope for detecting informal urban land uses. Basically, the classification helped identify land that had been developed from 1991 to 2007 and these developed landscapes were analyzed for informal building patterns and other indicators of informality. For this study, informal urban signifies a cluster of housing that primarily serves low -income persons and possess es inadequate housing facilities and infrastructure that was built in unlawful circumstances. This definition encapsulates informal settlements, favelas, and loteamentos irregulars (irregular land divisions) and follows the definition provided in the literature review Initially, a classification that included an informal urban class was attempted using the Landsat satellite imagery. The shortcoming in the method became quickly apparent when viewing the known locations of informal communities many of these housing clusters were linear in shape and small in size ( around nine acres on average) Thus, it was difficult to separ ate the small patches of informal uses from large swaths of formal urban uses.5 As such, h igh resolution aerial photographs taken during 1990, 2000, 200720 08 and Google Earth aerials were used for deciphering detailed building and lot organization featur es in a separate method (described in the following section) Jensen (2005) mentions that few landuse classification techniques solely rely on satellite 5 In addition, the 30meter spatial resolution offered in the Landsat imagery was too low for detecting small built features. Differentiating between land uses or different human impacts to land, required the use of other bas eline datasets that can decipher human uses (Jensen, 2005). Jensen and Cowen (1999) discuss the importance of high spatial resolution of satellite imagery and suggest a minimum of five meter spatial resolution (0.8 to 16.4 feet) for capturing individual b uildings in the urban/suburban fringe areas. A general rule for classification is that a minimum of four pixels are needed to identify an object this means that the spatial resolution should be at least onehalf the diameter of the smallest object of in terest (Jensen and Cowen, 1999). Several small informal settlement homes are six meters in width, thus the imagery used for their detection should have three meter spatial resolution.
52 imagery; in fact many use a classification hierarchy technique, rulebased modeling, and geographic analyses Thus, a method for capturing informal urban use s was developed that incorporated rulebased m odeling and geographic analyses. These techniques were developed based on the inductive logic drawn from the review of the known locations of informal s ettlements. The Municipal Plan of Social Interest Housing for Curitiba (MC, 2008) pointed to several common locational circumstances of known informal occupations areas The report indicated that 63 percent of the total irregular occupations are located partially or totally in permanent preservation areas 38 percent were in flood areas, and 1 4 percent were under and near electrical transmission lines .6 During the research field visits i t became apparent that informal settlements were often located in floodplains, near rivers, and wetland areas T he fact that one neighborhood was referred to as the Panta nal (wetland) cued in on this spatial linkage. Similar to other geographic studies, the field visits were crucial for linking onthe -ground characteri stic s with indicator features visible from an aerial perspective. In other words, these connections helped develop spatial pattern logic useful for differentiating between informal and formal housing development typologies Secondly, geographic datasets, detailed maps, and plans were collected that were indicative of informal settlements. An informal urban area detection method was customized to the spec ific conditions in th e Metrop o litan Region of Curitiba (Regio Metropolitana de Curitiba RMC ) and wa s formulated based on field visits, interviews with government agents, review of municipal plans in the region, use of census data, and other aerial photograph 6 Translated by author.
53 and geographic datasets.7 In addition, the logic drawn on informal settlement s showed location often in ecologically fragile landscapes (such as floodplains ) and conservation areas ( the Territory Units of Planning or Unidades Territoriais de Planejamento UTP areas and the Environmental Protection Areas or reas de Proteo Ambiental APA areas ) a nd in low er valued land along the urban fringe and periphery. Subseq uently, the informal urban areas were identified ba sed on this logic and a rule-based method that guided the filteri ng of the built areas Within the informal urban areas, land undergoing regularization and that was an irregular land divisions (ILD s ) were categorized. In comparison with the more precariously built informal settlements, t here was difficulty with identifying ILD s through aerial interpretation since these areas might be more regular and formal in appearance Thus the I LD areas recognized in a local government plan produced by the City of Curitiba were preserved and this categorization did not extend outside of the city limits of Curitiba The areas that were categorized a s informal urban undergoing regularizati on signified communities that were in the construction stage s of regularization during the classification year In other words, after emerging from the planning stage of regularization, only the informal urban areas undergoing physical regularization improvements and changes such as road alignment and housing relocation were categorized as regularizing informal urban areas The regularizing areas were identified during field visits with government contacts accordin g to 7 Essentially the best available resources wer e collectively used to filter and guide the identification of informal urban areas.
54 information derived from master plans and when regularization improvements were discernible through aerial photograph analysis .8 In sum, the built classifications (for 1991, 2002, and 2007) were categorized into formal urban, informal urban, and regularization urban use categories throughout the study area. Inside the city limits of Curitiba the ILD areas were categorized Details on t he method for detecting the informal urban categories are provided below. Steps for Informal Urban Detection: Indicators and Rule Based Categorization Using the logic drawn from known informal settlement locations in the City of Curitiba, several steps were taken to delineate informal urban areas. This detection method relied on the best available data outlets, many originating from local government entities in the RMC, and a set of indicators. Th e use of ArcView and other GIS tools facilitated this detection. As discussed above, t he detection proc ess began with the identification of at -risk areas Then t hese areas at -risk for informal occupation were scrutinized for informal development features common in the RMC with the use of high resolution aerial photographs Lastly, these areas were filtered with census data and other indicators, and when possible, veri fication from a person knowledgeable on informal settlement in the RMC Identification of at -risk areas with the use of indicators The areas deemed at -risk for informal housing were identified through the use of a set of indicators that indicates substan dard housing clusters and hazardous living 8 After regularization, an informal community might experience infrastructure and building layout changes. The changes often visible from an aerial perspective were: r oad alignment and widening changes, the installation of drainage structures, removal of extremely precarious homes in environmentally sensitive areas, and the construction of nearby public housing. The extent of these infrastructure improvements varies si nce there are no rigid standards for reaching a minimal or maximum level of regularization. Figure 22 shows an example of the regularization changes before and after regularization.
55 conditions The census data from the Brazilian Institute of Geography and Statistics or Instituto Brasileiro de Geografia e Estatstica (IBGE ) proved to hav e several data features useful for detecting inf ormal se ttlement. T he IBGE domicile dataset has a land tenure referred to as a clustered substandard housing sector type ( or especial de aglomerado subnormal ). The identification of these substandard housing clusters was not reliant on self -reporting from reside nts and instead the census interviewer analyzed the housing situation and classi fied the land tenure based on observations (Serra et al., 2005). However, this dataset was only available for 2000 and for housing areas with over 50 dwelling units Thus many areas would be missed with solely relying on this dataset. As a result several other indicators based on physical characteristics were examined and compared to aerial photograph features including new housing located near existing informal settlements, homes converging in a riparian area (within 50 meters of a river or stream ), and housing located along floodplains, steep slopes, karst areas, drainage valleys, and in an APA or UTP conservation area or park. With the use of detailed orthophotographs, t he s e at -risk areas were analy zed for development features specific to informal settlements and were narrowed down to outline only the informally built urban areas Visual i nterpretation Landuse id entification often involves visual interpretation of detailed aerial photographs since this analysis provides a wealth of information on the human uses of land and land development differences. Based on a review of the aerial features of known locations of informal settlements in Curitiba, common development charac teristics for informal settlements in the RMC were recognized in georectified
56 aerial photographs .9 Generally, the development features discernible from detailed aerial photographs were: Lot s : Irregular lot sizes, often extremely small, arranged in a m anner in which it is difficult to distinguish where the lots begin and end. The lot boundaries might not be visible in denser areas. H omes/Buildings : Scattered placement of buildings that often lack direct access to a road and that fail to include few i f any building setbacks. The homes often cluster around a thin vegetative feature, such as a stream, river, or flood valley or they might be perched in non -quadrangular angles along a hillside. In other words, the arrangement of the clustered homes follow s the shape of the natural feature such as a hill and river. Homes might resemble shacks and be precarious in their construction and the home layout often is diverse and nonuniform. Roads/Blocks : Several unpaved roads that are inconsistent in their wi dth, arranged in non -uniform block sizes and shape s, contrasting in size, width and shape. The roads might be extremely narrow in width or in some areas there may not be any roads near housing clusters The road system generally might be confusing and too narrow for access. Materials: A wide assortment of roofing and housing materials. The formal areas tend to consistently use orange/red terra-cotta roof shingles in residential areas and gray/white flat roof materials in industrial areas while the info rmal areas tended to use a wide range of roofing materials such as tin, wood, and terra -cotta. Extremely detailed aerials are needed to detect the building materials, thus this aerial indicator is less useful. As described above, the detection of th e collective arrangement of the lot, building, and roads was easily recognizable when the orthophotographs were examined at a scale of 1:20,000 meters or less. In addition, the aerial photographs were useful for eliminating other formal types of urban areas s uch as industrial and commercial areas. The use of another land use land cover dataset for the region helped with this filtering process (2000, SUDERHSA). The presence of more informal indicators identified 9 Generally, the pattern and layout of informal settlements showed sev eral physical commonalities such as scattered buildings and lots and an irregular road network. As discussed earlier, the aerial photographs used for identifying informal land uses needed to have around three meter spatial resolution. Basically, the aerial photographs should be detailed enough to decipher the location of individual buildings and roads.
57 through visual identification increases the probability of informality particularly in terms of the lack of standard full infrastructure. As such, a ll of the built areas containing several of the abovementioned informal indicators were filtered based on their location in other indicative regions .10 T he next step requiring filtering based on rules is described below Rule b ased filtering Various indicators were identified and the built areas were filtered with these common locational characteristics in the informal urban identification process. Sev eral of these filters indicate precarious and hazardous development circumstances located in areas which typically are not legal to develop according to land use regulations in municipal plans In addition, other filters used in this identification were b ased on identification from knowledgeable government agents and municipal plans and housing data. For example, the IBGE census GIS data provided information on the location of improvised shelter areas and domiciles without bathing and wastewater faciliti es and areas with concentrations of low income levels .11 These locations often were indicative of inadequate living conditions.12 Consequently, urban areas with clusters of improvised domiciles or homes without bathing and waste water facilities were scr utinized The substandard agglomeration area proved to be the most useful census indicator, thus these areas were examined and narrowed down to outline only the 10 In the RMC and other areas of Brazil, the substandard infrastructure features are often indicative of informal settlements since many informal urban areas do not have infrastructure such as sewage and drainage facilities and fully paved roads. However, other cities with less infrastructure overall might have different commonalities other than substandard infrastructure that are indicative of informal urban areas. 11 Improvised domiciles indicate habitation situations similar to homeless circumstances in the US. 12 Another study used domicile IBGE data resources and housing indicators such as inadequate wastewater facilities and improvised housing to quantify the housing deficit in Brazil (Serra et al., 2005).
58 developed areas with substandard characteristics, as described in the visual interpretation me thod. T he rule-based method for filtering at -risk features and rating is provided below. Table 3 1. Rules for informal urban c ategorization Rating Rule s 5 o If identified as an informal settlement or invasion by government agencies during field visits or indicated by a local government planning document or data, then the informal urban designation was a ppointed 4 o If the area is in a special substandard agglomeration area ( especial de aglomerado subnormal IBGE ) and deemed substandard through the vis ual interpretation of aerial features, then the informal urban area was recognized. o If the area is located in a n UTP area (particularly if the housing was in a consolidated urban area) and deemed substandard through the visual interpretation of aerial features, then the informal urban area was outlined. o If through review of detailed aerial images, the area shows substantial evidence of regularization at a later date in which there is relocation, construction of public housing, and lot and road alignment then the urban informal or regularization land use was recognized. 3 o If the housing cluster is within 50 meters of the edge of a river or stream feature and the area was deemed substandard by the visual interpretation of aerial features, then the inf ormal urban category was designated. 2 o If the housing cluster (partially or fully) is within a floodplain or karst region and the area was deemed substandard by the visual interpretation of aerial features, then the informal urban class was designated. 1 o If the housing area is within a steep slope or drainage path in a valley, and deemed substandard through the visual interpretation of aerial features, then informal urban was designated. The informal urban areas w ere compared to census data and w ere examined for substandard housing indicators such as homes without bathing and waste water facilities and for income levels below six monthly minimum wages (per 2000, IBGE census data). The areas were ranked according to accuracy and likelihood o f informality. For example, an area identified as an informal settlement by a government agent during a
59 fiel d visit was given a five rating. The areas that espoused several characteristics of different rankings (such as location in a floodplain and ident ification in a master plan) were assigned the highest possible ranking. In general many of the informal urban areas were associated with several indicators. A summary of t he rankings derived from the informal detection method for 2007 are provided below in Table 3-2. Approximately 76 percent of the acreage of informal urban areas was ranked highly as a four or five while the remaining 24 percent was ranked in the lower range from one to three. The rankings show that most of the areas were identified bas ed on guidance from local government agents and plans and crucial census data indicators. Table 3 2. Ranking results of informal d etection for 2007 Ranking Description of Ranking GIS Acres Percent of Total 1 Housing area is located within a steep slop e or drainage path in a valley. 570 9% 2 Located within a floodplain or karst region 246 4% 3 Located 50 meters from the edge of a river or stream feature 697 11% 4 In a substandard agglomeration area (2000, IBGE); or UTP area; or shows substantial phy sical evidence of regularization at a later date 1,047 17% 5 Verified through field visit, by local government agent, or indicated as informal in master plan map or per GIS data 3,697 59% Total 6,257 100% Using the aforementioned method, the informal areas were outlined for 2002 and 2007. Then the geographic locations of built areas within the 1991 land cover classification that intersected with the 2002 informal urban areas were outlined. A set of detailed historical aerial photographs taken in 1991 were available for some areas
60 within Curitiba These 1991 images were reviewed to correct informal urban areas in the 1991 geographic dataset. The informal urban areas that were fully regularized and no longer informal by 2002 would have been missed th ough regularization largely had not been implemented in Curitiba prior to 1991 Given the data limitations for 1991, t he identification of the se informal urban areas generally was less rigorous than later years Overall t he use of filtering helped to impr ove the overall effectiveness of informal urban area detection. Discussion on I nformal U rban Identification Method Many informal urban areas had distinct land development patterns; however the informal urban pattern was not completely homogeneous from neighborhood to neighborhood. There were slight differences depending on the context (see Figure 3-1) In a more established urban neighborhood, the informal building pattern was denser while in a more exterior community, located in an urban-rural transit ion area the bui lding pattern was slightly more dispersed vegetated, and less dense. These varied conditions were observed during field visits and during geographic analyses Generally, the identification of informal urban areas is challenging especiall y when considering the small size of informal settlement clusters (on average informal urban areas were seven acres in size for all three study years) the ir rapid development, and the hidden quality of their existence. Subsequently, the informal urban ar eas that contain more visible physical conditions indicative of substandard housing and development in precarious circumstances might have been identified more often than those informal areas with less apparent physical characteristics. This informal set tlement detection method strives to capture a greater amount of informal urban areas and improve the accuracy particularly in comparison to the
61 Figure 31. 2000 orthophotograph m ap: v ariations in i nformal urban areas
62 method used by IBGE. The extent that the findings were verified varied from town to town. The areas with comprehensive informal settlement location information indicated by municipal plans were available for Curitiba and Fazenda Rio Grande. The areas visited in the field with the acco mpaniment of government agents primarily took place in the eastern conurbations. Thus the rule based rating gives a higher number for these areas to indicate greater identification confidence (see Table 32). Overall, informal detection methods are limit ed by several issues particularly in the inability to account for the full informal settlement population. The lack of available property title information and difficulty with retrieval hindered the detection of the whole informal settlement population in the RMC. In addition, the areas that had a greater proportion of formal land uses in comparison to informal housing were not categorized as informal (needed at least 50% informal housing) and this could have lowered the amount of informal housing detected. In response to these limitations, this method was formulated to utilize the best available data resources and techniques for detecting small pockets of informal settlement. After detecting informal urban areas, the resultant locations were compared for their prevalence with major locational trends using GIS tools and the method for this analysis is described briefly in the following section Method for Analyzing Spatial Patterns The types of spatial patterns measured in the informal urban areas from 199 1, 2002, and 2007 were related to urban growth or major locational features The urban growth trends were measured using urban population density calculations the elasticity of urban sprawl measurement and through visual interpretation In addition, th e informal urban areas were compared with geographic datasets representing common locational determinants including floodplains, rivers, karst features, conservation areas,
63 bus lines, electrical transmission lines, and steep slopes. The use of ArcView Sp atial Analyst and other G IS tools facilitated this analysis and was particularly helpful in deciphering the strength and prevalence of trends (such as the I Morans spatial autocorrelation tool) Often t he informal urban areas converging ins ide locationa l features (such as floodplains, conservation areas, and karst regions) were extracted to measure the portion of informal urban areas that overlap in a particular feature, using GIS tools This simple extraction technique facilitated acreage calculations and helped determine the amount of informal urban areas that are located within a particular geographic feature. In addition, distance measurements (as the crow flies) from a feature of concern were used to determine proximity trends. For example, the am ount of informal urban areas within 500 meter s of a bus line was calculated. Often the rate of growth formula was used to measure whether patterns and trends were accelerating or diminishing from 1991 to 2007 .13 Collectively this geographic analysis helpe d to create a foundation of what constitutes the dominant spatial conditions that informal settlers disproportionately face in the RMC. Community and Government Agent Interview Methods The interviews, a part of this research, were collected mostly durin g the summer of 2009 in addition to a few reconnaissance interviews held earlier in 2008. Informal settlers were interviewed, with the accompaniment of a Housing Company of the State of Paran (COHAPAR ) staff member in the Zumbi informal settlement, a community that is undergoing regularization. The community surveys sought to ascertain the perceptions and opinions of the informal settlers on what factors (if anything) influenced 13 The formula used to calculate the average annual growth rate is: ((Later Yea r Earlier Year)/Earlier Year / # of years) 100.
64 settlement location outcomes.14 These surveys were initially offered as int erviews and if the participant indicated a preference for writing responses, paper surveys were provided. In total, 30 participants provided responses to the community surveys/interviews. In addition, periodic interviews with government agents whom are i nvolved with housing and urban growth management were carried out in 2008 and 2009. These interviews were informal and open ended in nature but were all guided by a drive to comprehend the reasons for informal urbanization trends. The findings derived fr om the community surveys and government agent interviews are interspersed alongside other similar geographic -based results. The purpose of these interviews and surveys was to compare spatial trends that arise from the geographic analysis. Given the nomin al and nonnumerical character of the community survey responses, descriptive statistics su ch as frequencies were calculated to evaluate survey findings. In addition to interviews, a review of the municipal and regional plans was conducted as a part of th e qualitative portion of the research method. The fi ndings from the plan review are provided in the following section. The Case of Curitiba: Public Policy Influences from 1991 to 2007 Social Interest Initiatives in the Study Area In response to informal settlement and affordable housing issues, s everal planning initiatives have been implemented from 1991 to 2007 in Curitiba such as regularization, social interest zoning, and lot and public housing provision programs. Consequently, a review of the major s ocial interest policies and the conceivable ways they could potentially impact the urban growth patterns is provided. 14 Community survey questions and a summary of responses are provided in A ppendix A.
65 T he Special Sector of Social Interest Housing ( Setor Especial de Habitaco de Interesse Social SEHIS ) was adopted by Municipal Decree in the 2000 Master Plan of Curitiba and many areas are designated SEHIS. Prior to its adoption, Decree 901/80 already declared areas of social interest in order for the Affordable Housing Company of Curitiba (COHAB) to implement public housing programs (MC, 2008).15 As articulated in the Municipal Plan (2008), t he SEHIS is viewed as a great advancement especially since different formalization parameters were adopted and additional social interest lots were produced The basis for where and how much land is allocated under SEHIS i s vague other than the fundamental limitation posed by the amount of available land. According to the zoning regulations SEHIS should be designated in areas deemed compatible to the public interes t and where the infrastructure provision a part of regularization might complement other social interest habitation programs (MC, 2008). Besides the recently designated social zones in Parolin and Vila Formosa, most of the SEHIS zones are located in the outer edges of Curitiba. Many of the SEHIS areas had already experienced informal settlement before attaining their social interest designation and several of these areas are in the process of regularization. Consequently, the choice in where to designate SEHIS zoning in Curitiba seems to be linked to locations already experiencing informal settlement. The extent that informal settlement is clustered near SEHIS zones and other social interest habitation programs will be explored along with the overall level of income level segregation. T here are several integral social interest programs allowed in the SEHIS and social interest housing regions Describing all of these programs would take another 15 SEHIS zoning policies were translated by author.
66 chapter, thus a review of a few key social interest program features that influence spatial patterns is emphasized One of th e key programs, referred to as Lot Partnership (or Parceria Lotes ), is a public -private partnership that allows approved low income recipients an opportunity to self -construct a home on a lot serviced with minimum infrast ructure including water, sewage, electricity, and improved roads (COHAB, 2007). The Lot Partnership program requires that the lots be on average, 160 square meters in size (over 20 dwell ing units per acre) and offers financing options for 15 year terms (C OHAB, 2007). Over 26,000 o f these urbanized lots have been provided through this particular lot partnership program. The rapidly growing neighborhood, Bairro Campo de Santana in the southern edge of Curitiba, contains urbanized lots a part of this program Around 68,000 urbanized lots have been offered to low -income recipients since 2007 (MC 2008). Additionally a residential leasing program (referred to as the Programa de Arrendamento Residencial ) offers exclusive leasing of homes for applicable low -in come households with the option to purchase (COHAB, 2007). The housing units a part of the residential leasing program mostly encircle Curitiba in neighborhoods along the south, east, west and north edges.16 The pivotal feature of this serviced lot provi sion program is in its direct impact on increasing the supply of land to low -income populations that is suitable for development. In fact, Berner (2001) contends that housing poverty is largely determined by land supply and allocation (p. 295) and is dim inished even further when there is a scarcity of affordable land suitable for development. Using this idea, t he Municipality of Curitiba 16 The residential leasing program offered approximately 5,535 units by 2007 in the Campo Comprido, Tatuquara, CIC, Bairro Novo, Cajuru, and Santa Cndida Neighborhoods. Depending on the house hold income, the finance interest rate ranges from 0.5 to 0.7% with a repayment term of 15 years (COHAB lecture, 2007).
67 (2008) has allocated the largest share of the total provisions in the form of urbanized lots. Increasing the availabi lity of serviced lots could be seen as a more proactive measure for addres sing informal settlement that is fueled by a dearth of affordable land suitable for development. Another innovative mechanism used in Curitiba by COHAB t hat often replenishes the f inances available in the Municipal Housing Fund for social interest programs is the use of a transfer of development rights (TDR) mechanism ( Solo Criado or created land) (Macedo, 2008). Funds are garnered for affordable housing and social interest program s by charging developers fees for variances and special permissions in approved areas for increasing the built area and intensity beyond what zoning currently allows. In exchange for greater development intensity such as the allowance of additional storie s on a highrise, affordable housing in the form of single -family and multi -family dwelling units, serviced lots, and regularized occupations, has been subsidized. Despite the fact that these funds are subject to fluctuations in the real estate market, t his TDR tool helped a mplify the volume of affordable housing options and serviced land in Curitiba (MC et al., 2007). As a whole, the TDR tool subsidizes infrastructure and sanitation improvements for low -income populations yet it reduces land speculation and enables more intense use of urbanized land (Macedo, 2008). Informal Settlement Regularization in the Study Area Another crucial social interest program in Curitiba and the Metropolitan Region of Curitiba (RMC) is the regularization of infor mal settlements (or regularizao fundiria ). As described earlier, regularization can vary in the types of improvements that are added to an informal settlement. In Curitiba, the implementation of the regularization and urbanization program m ostly began during the 1990 decade (MC, 2008). The
68 objectives of the regularizat ion program are to i nstall basic infrastructure gradually, accelerate the legalization process of tenure, promote educational programs, and strengthen mechanisms to control new invasions (COHAB 2007). The regularization process in Curitiba typically begins with the provision of basic infrastructure (water, electricity, and sewage systems), construction and establishment of social facilities, and registration of an approved lot layout. In the RMC, regularization only recently commenced in 2003 beginning with the Zumbi pilot project in Colombo, as a part of the Right to Housing Program led by the Housing Company of the State of Paran (COHAPAR). The regularization of the Zumbi informal settlement required changes in the neighborhood building pattern such as infill development and a s mall amount of housing relocations from extremely hazardous areas to compactly built new housing, located nearby.17 A new regularization approach, underway in the periphery town of Pinhais in 2009, involves a public private partnership between COHAPAR and a private company called Terra Nova to help carry out part of the regularization efforts ( U P, 2009). Depending on the initial urban form and density of an informal settlement, the act of regularization might increase the rate of infill and consolidation (see Figure 32 ). The organization of lots into more uniform sizes might free up new lots and th e construction of multi -family public housing (often for housing relocations) adds more compactly built 17 The scope of regularization used in Zumbi was exten sive and it included infrastructure provision installation of drainage control s tr uctures, construction of around 280 homes for relocated families and the regularization of around 1,800 lots, construction of a new health center, and the institution of income generation, environmental education, and construction training programs. The Z umbi Program involved COHAPAR, Secretaria de Desenvolvimento Urbano/Paran Cidade/BID, Fundo de Desenvolvimento Urbano, Sanepar, City of Colombo, Copel, and other public organizations Similar to the SEHIS zone in Curitiba, the City of Colombo designated the Zumbi area as zones of Special Inter est and Permanent Preservation (based on an interview in 2008 and translated by author).
69 dwelling units. In addition, the provision of piped sewage infrastructure eliminates the need for septic tank drainage fields and rustic tanks on lots.18 As done in the Zumbi regularization process, some program s might include the construction of a garbage collection warehouse for those residents who are employed as waste pickers. This new waste storage facility might help free up space, formerly used to store waste, and create room needed for additional lots. G enerally, as infrastructure is added and homes become more stable, home owners might invest in the construction of a second story addition. These second stories might serve as rentals to supplement household incomes.19 The collective addition of these new dwelling spaces might increase the housing supply, and as a result, limit the need to invade land. Though some regularized areas might experience price increases because of neighborhood amenities and infrastructure.20 Increases in the supply of rentals t hat are more affordable for low income populations and consolidation from regularization could reduce the informal urbanization rate in Curitiba. Several caveats should be recognized when analyzing the spatial patterns of regularization. In the situation in which an informal urban area is extremely dense 18 A rustic tank is a septic tank without a drainage field. 19 Gilbert and Gugler (1994) found that as informal settlements age and gain infrastructure and services, the proportion of renters increased. In a study examining five legalized informal neighborhoods in Bogot Colombia, Gilbert (1999) found that over half of the informal homes had tenant occupants and collectively the tenant households outnumbered the owners. 20 Varied land pricing have emerged in some regularized communities in Rio de Janeiro and the increase in land prices might spur density increases. In a study analyzing informal land markets in over 500 favelas Abramo (2007) recognized five distinct submarket trends of residential mobility in favelas undergoing consolidation in Rio de Janeiro and pointed out external dete rminants of price based on building freedom and the presence of social reciprocity network s that allow for the bartering of services. In addition, the mere installation of infrastructure and granting of title increases the lot prices by 47 to 49 percent alone (Abramo, 2007). This study contends that price structures in informal settlements hav e formed and that a rational based informal market exists. However, this study does not analyze other towns outside of Rio de Janeiro.
70 without enough roads and suitable space between homes, these communities w ould likely not experience additional consolidation with the implement ation of regularization In addition, n ewly developed info rmal settlements can consolidate naturally as they age and as new housing extensions are built to accommodate extended family members or friends t hat migrate (Gilbert and Gugler, 1994). Largely informal settlements, particularly those newly developed on t he edge of the city tend to consolidate and increase in density as these communities mature. In addition, regularization that includes infrastructure installation and service provisions could attract new informal settlement in nearby locations since there may be opportunities for new -comers to connect to installed infrastructure. Exclud ing regularized housing, Table 33 shows the rate of informal housing increase in Curitiba, Brazil. While the 1987 to 1996 time frame experienced an astronomically incr ease in the rate of informal housing, by 2000 the annual rate of informal housing production declined. The economic crisis in Brazil that culminated in Table 3 3 Informal housing growth, excluding homes undergoing r egularization, from 1974 to 2005, in Curitiba Year Number of Domiciles Annu al Rate of Increase Regularized homes 1974 4,083 1979 6,067 1974 to 1979 9.72% 1982 7,716 1979 to 1982 9.06% 1987 10,187 1982 to 1987 6.40% 1980 decade, 1742 1996 33,078 1987 to 1996 24.97% 1990 to 1996, 700 2000 37,621 1996 to 2000 3.43% 1996 to 2000, 19712 2005 42,814 2000 to 2005 2.76% 2000 to 2005, 19453 Data Source s: IPPUC, COHAB and MC et al., 2007. T able by author the late 1980s likely spurred a greater amount of substandard housing growth and then Curitibas social interest programs gained momentum during the mid 1990s and it could have influenced reduced rates of informal urbanization. This research will explore this
71 finding more thoroughly by evaluating the urban areas surrounding Curitiba to detect whether informal settlement was merely displaced into the outskirts of Curitiba. As a whole Curitiba has initiated several innovative programs undergoing implementation that offer a variety of serviced and regularized lots and public housin g dwelling units. The Municipality of Curitiba (2008) estimates that 57 percent of the total 120,664 social interest and affordable housing provisions (excluding housing undergoing regularization) are urbanized lots and 43 percent are apartments and hous es.21 The concentration of these programs in the so cial interest zones and other affordable housing initiative areas likely sparked rapid development in the urban fringe Moreover, these provisions have accumulated substantially a cumulative estimate s uggest s that almost 40 percent of the total provisions have been allocated since 2000 (MC, 2008 and MC et al., 2007). In addition, the use of public -private partnerships and TDR funding mechanisms has helped to enlarge these effo rts. As such, this tremendous effort in Curitiba could reduce the rate of informal urbanization. A majority of these initiatives take place in Curitiba and not in the surrounding region (COHAB, 2007) and these projects tend to be built on cheaper land located at the edges of the urban region. For example, in 1992 the serviced lot provision program was initiated in Stio Cercado (wes t edge of Curitiba), in Bairro Novo, and Tataquara areas (south edge of Curitiba) (MC et al., 2007).22 As described in the aforementioned section, v arious programs are undergoing implementation in Curitiba and collectively 21 Including housing in regularization processes, around 42 percent of the total provisions are serviced lots. In 2008 the City of Curitiba estimates a rate of production under 3,000 provisions a year (MC, 2008) Translated by author. 22 In addition, the Cajuru social interest program, located on the eastern edge of Curitiba, commenced in 1997 (MC et al., 2007).
72 Figure 32 Regularization of the Zumbi Informal Settlement: urban f orm changes from 2000 to 2007, orthophotograph m ap
73 these programs are r eferred to as the government approa ch or strategy for responding to informal settlement issues. Lar gely, the government response to urban poverty in Curitiba has encouraged informal settlement to concentrate in social interest zones, social initiative areas (or affordable housing program a reas), and near areas undergoing regularization. Curitibas social interest housing program appears to incorporate many sustainable features such as affordable housing options, regularization, and more lenient regulations in social interest areas that all ow for progressive housing construction. Many of these initiatives could spur density increases and consolidated development. The Environmental Protection of Land in the Study Area The 1988 Brazilian Constitution mandated the protection of the natural environment, particularly the forests, fauna, and plant features and the prevention of pollution through the use of zoning tools that limit land uses, occupation intensity, and protect the environment (GEP, 2006). As discussed earlier, Curitiba adopted m ost of the zoning restrictions during the 1960s to 1970s, and other large municipalities a part of the study area including Colombo, Pinhais, and So Jos dos Pinhais have adopted zoning code in 2005 (GEP, 2006). Along with use restrictions, State Decree 1751/1996 authorized all watersheds deemed valuable to the region be declared areas of environmental protection and special interest (GEP, 2006; Macedo, 2000). Referred to as Environmental Protection Areas ( reas de Proteo Ambiental APA areas ) man y of these watersheds are the source of drinking water for the population in the region.23 The criteria for use and 23 Anoth er State Decree (4267/2005), adopted in 2005, added the Baixo Pequeno Watershed to the environmental protection region. State Decrees and zoning policies were translated by author.
74 occupation of APA areas were defined according to local needs and characteristics. Later in 2000, federal law 9985 was instituted, referred t o as the National System of Conservation and Natural Units (O Sistema Nacional de Unidades de Conservao da Natureza SNUC) and this law necessitated the designation of conservation areas into two categories: integral protection and sustainable use (GEP, 2006). The sustainable use category permits some land use as long as biodiversity and natural resources are maintained. In the study area, all of the APA areas were designated for sustainable use and several parks were protected. In addition to APA areas, Territory Units of Planning ( Unidades Territoriais de Planejamento UTPs) were established under state law 1248/98. This law instituted sustainability principles, as a part of the Integrated Sy stem of Watershed Management and Protection in the RM C ( Sistema Integrado de Gesto e Proteo aos Mananciais da RMC SIGPROM/RMC), and articulated which areas could be used for the right of housing and conservation purposes (GEP, 2006).24 The UTPs are divided into four areas including: areas of restricted occupation, areas of advised occupation, areas of consolidated urbanization, and rural areas. These areas were defined according to the local character and in a way to advance the objectives of the state law. The consolidated urban areas mainly consist of informal urban us es (these areas are also referred to as social interest areas) The innovative feature of the law is in the use of development exchanges. The land that is in the publics interest for preservation such as forests and riverbanks are re stricted from development in exchange for occupation in other approved UTP areas 24 Translated by author.
75 All of these conservation areas, mostly located along the edges of the study area, pose limitations to development though the degree that these laws are enforced is uncert ain. 25 T he informal occupation in these periphery areas has merited enough attention to create specialized policies for permitting the informal occupation of land, in response to low -income housing needs. Therefore, the amount of informal settlement an d rate of informal development expansions in UTP, APA, and park conservation areas will be measured.26 Th e above review of the public policies implemented during the study time period described several probable ways that informal urbanization patterns c ould be influenced. The affordable housing approach could have spurred density increases consolidated informal development and concentrated low i ncome development In contrast, the conservat ion policies could have deterred informal urban growth in ec ologically fragile areas. Chapter 4 provides the data analysis and research findings on informal urbanization spatial patterns and in light of these findings, discusses the effects of public policy 25 Figure 4 14 shows the location of conservation areas. The study area includes the Pinhais UTP (established under 808/99 and 4466/2001), Guarituba UTP (established under 809/99 and 6314/2006), Itaqui UTP (established under 1454/99), Quatro Barras UTP (established under 1612/99), and Campo Magro UTP (est ablished under 1611/99). 26 The UTP, APA, and zoning laws were translated by author.
76 CHAPTER 4 DATA ANALYSIS AND RE SEARCH RESULTS Usin g the method explained i n Chapter 3, the data analysis and results are organized to first describe the urban growth trends in the study region and then, the spatial patterns of informal urban areas Initially the m ajor urbanization growth trends including population growth, urban growth, and consolidation are described for the study area from 1991 to 2007 Then the results on the informal urban growth rate s an d spatial pattern s (such as the tendency for decentralization) are offered. The nex t section pro vides a n analysis and discussion on the major public policies potentially influencing informal urban patterns (such as density and consolidation increases) inside and outside of Curitiba. Lastly the major locational features of the informal urban areas (s uch a s location in a floodplain, river buffer, conservation area, near public housing, and bus service) are analyzed fo r their prevalence and degree of influence. At the conclusion of this chapter, a summary of the findi ngs is provided to compare the results and describe the study area holistically Population Growth Findings Measuring the changes in population counts and growth rates is critical for characterizing the patterns of new urban growth such as in detecting population density increases. As shown in Table 4 -1 the populations in Curitiba and the Metrop o litan Region of Curitiba (Regio Metropolitana de Curitiba RMC) b oth have increased since 1975. However, since 1991 the rates have diverged. I n 1991, the RMC annual population growth rate climbed above Curitiba s annual growt h rate and has remained higher till today In addition, the proportion of the RMC population living outside the City
77 of Curitiba has steadily increased since 1991 from 36 to 44 percent. These findings indicate that n ew population growth is decentralizing into periphery municipalities Metropolitan Region of Curitiba 0 10 20 30 40 50 60 70 80 90 100 1970 1980 1991 2000 2007 Year Population Percentage Inside the City of Curitiba Outside the City of Curitiba Figure 4 1 Port ion of urban growth inside and outside of Curitiba in the Metropolitan Region of Curitiba Table 4 1 A comparison of population growth in Curitiba and the Metropolitan R egion of Curitiba Year Curitiba Population Average Annual Growth RMC Population RMC Average Annual Growth 1970 608,417 820,766 Not Available 1975 765,716 5.17% 1,013,279 4.69% 1980 1,025,979 6.80% 1,441,743 8.46% 1985 1,285,027 5.05% 1,772,689 4.59% 1991 1,315,035 0.39% 2,063,654 2.74% 1996 1,476,253 2.45% 2,431,804 3.57% 2000 1,587,315 1.87% 2,725, 629 3.02% 2007 1,797,408 1.90% 3,230,000 2.64% 2009 1,851,215 0.43% Not Available Not Available Data Source: IBGE, various years
78 Table 4 2 A c omparison of p opulation g rowth in Curitiba and the p eriphery t owns Curitiba Almirante Tamandar Araucria C ampina Grande do Sul (CGS) Campo Magro Colombo Fazenda Rio Grande (FRG) Pinhais Piraquara Quatro Barras So Jos dos Pinhais (SJP) 1991 Population 1,315,035 54,030 61,712 19,343 12,129 117,767 24,997 75,516 31,366 10,007 127,455 2000 Population 1,586,848 88,139 94,137 34,558 20,364 183,331 63,031 102,946 72,838 16,149 204,202 Average Annual Growth Rate, 1991 2000 2.30% 7.01% 5.84% 8.74 % 7.54% 6.19% 16.91% 4.04% 14.69% 6.82% 6.69% 2007 Population 1,797,408 93,055 109,943 35,396 22,325 233,916 75,006 112,038 82,006 18,125 263,622 Average Annual Growth Rate, 2000 2007 1.90% 0.80% 2.40% 0.35% 1.38% 3.94% 2.71% 1.26% 1.80% 1.75% 4.16% Av erage Annual Growth Rate, 1991 2007 2.29% 4.51% 4.88% 5.19% 5.25% 6.16% 12.50% 3.02% 10.09% 5.07% 6.68% Data source: IBGE census, various years. The study area municipalities are only provided. The formula used to calculate the average annual growth rat e is: ((Later Year Earlier Year)/Earlier Year 100) / # of years).
79 T able 4 2 dissects the overall regional growth by separating out the study area population counts for each of the ten periphery towns that surround Curitiba. This table shows a high a nnual population growth rate from 1991 to 2007 in several conurbations The highest average annual population growth rates from 1991 to 2007 were in Fazenda Rio Grande (FRG ), Piraquara, So Jos dos Pinhais (SJP) and Colombo. From 1991 to 2007, both of the populations of SJP and Colombo doubled and from 1991 to 2002, the population of Piraquara doubled. Since the beginning of the 21st Century, some of these conurbations have become more significant in their urban magnitude. The se results show th at the conurbations with population concentrations over 100,000 persons and a more urban character are more prone to population increases .1 Generally the increases in urban growth can be attributed to a great amount of rural urban migrations However, in recent years, a portion of the new arrivals are described as inter migrations and urbanurban migrations .2 These particular migrants are moving from one town to another town in the same region instead of originating from the rural hinterlands These tre nds in dicate a decrease in the amount of new -comers accustomed to rural living; thus there might be less of a need for semi rural single -family housing that has space for livestock 1 Urban portion of the population in periphery towns: So Jos dos Pinhais 263,622 persons, 89.75 percent urban; Colombo 233,916 persons, 95.44 percent urban; Pinha is 112,038 persons, 97.81 percent urban; Araucri a 109,943 persons, 91.36 per cent urban; Almirante Tamandar 93,055 persons, 96.01 percent urban; Piraquara 82,006 persons, 80.91 percent urban and Fazenda Rio Grande 75,006 persons, 94.15 percent urban (IBGE, 2007). 2 During the 1995 2000 time period, 50 percent of the migrants and immigrants in the RMC were from the interior portion of the state of Paran; 42 percent originated from other states and countries; and 7 percent were from other urban areas in the RMC (2008, MDC).
80 Urban Growth Data Analysis Findings The urban growth from 1991 to 2007 is portr ayed through a series of figures and tab les. Table 4 3 provides the resultant land cover classification acreages Figures 4 2, 4 -3, and 44 show the 1991, 2002, and 2007 classifications geographically with a series of maps Overall the proportion of th e study area with built land covers has increased by ten percent from 1991 to 2007 to result in a total built area that encompasses 34 perc ent of the total stu dy area, in 2007. The total share of built acreages located outside of Curitiba in the study area has steadily increased over the course of the study time period from 37 to 44 percent. Though 56 percent of total built area is located within the city limits of Curitiba and this share still remains highe r than outside the city limits (66,492 acres o f built land, in 2007). However, the margin of difference of urbanized land inside and outside of Curitiba has increasingly narrowed If these rates continue, the urbanization outside of Curitiba will eventually become larger in size than the area of d evelopment in Curitiba and these urban areas might continue to grow together to create larg er conurbations Table 4 3. Urban area growth from 1991 to 2007 1991 2002 Land Cover Acres Portion of Total Area Land Cover Acres Portion of Total Area Built 8 5,824.71 24.92% Built 110,918.19 32.21% Vegetation 232,752.37 67.59% Vegetation 206,351.77 59.92% Water 6,083.38 1.77% Water 9,409.47 2.73% Bare Soil 19,701.18 5.72% Bare Soil 17,682.20 5.13% Total 344,361.64 100.00% Total 344,361.64 100.00%
81 Table 4 3. Continued 2007 Land Cover Acres Portion of Total Area Built 118,397.55 34.38% Vegetation 200,005.90 58.08% Water 7,603.32 2.21% Bare Soil 18,354.87 5.33% Total 344,361.64 100.00% *Acreages were derived using GI S tools and are approximate. Urban Growth Changes from 1991 to 2007 As shown in Table 4 4 and Figure 4 5 the proportion of total built lands cape has increased t o cover 34 percent of the total study area by 2007. The average annual change in urban land conversions has reduced from 2,281 acres to 1,496 acres per year; indicating a significant decline in the rapidness of urban growth. Other than the municipalities of Campo Magro and Araucria (on the west edge), most towns experienced a reduction in the r ate of urbanization during the 20022007 time period in comparison to the rapid urban growth from 1991 to 2002. Se veral major construction changes occurred during the 1991 to 2002 time period. The development of a dam referred to as the Represa do Rio I rai in the eastern edge of the study area transformed the region and likely resulted in hydrological modifications, particularly in downstream areas These impacts likely prevented the natural flow of water to floodplains in much of the eastern and south ern portions of the study area and stabilized peak flow dispersed from flood events. In addition, after 1991, the construction of a major arterial, referred to as Roan Contorno Leste, connected Quatro Barras to So Jos dos Pinhais (SJP). Both the road addition and the dam installation likely facilitated development in the region.
82 F igure 42 1991 land use land cover c lassification m ap
83 Figure 43 2002 land use land cover c lassification m ap
84 Figure 44 2007 land use land cover c lassification m ap
85 Holistically, the areas urbanizing the most rapidly are located at the edges of th e main urban fabric, often in periphery municipalities A great amount of new urban growth emerged in Piraquara during the 1991-2002 decade. Much of this new urban growt h in Piraquara is associated with informal development The municipalities of Pinhais, Fazenda Rio Grande ( FRG ), and Almirante Tamandar also experienced substantial growth, associated with a mixture of formal and informal uses during the 19912002 decade that continued into the 2002-2007 time period. An interviewee recalled the rapid intense growth in the periphery towns of Colombo and FRG The responden t referred to FRG as merely a bus stop in 1980 and today considers this same area a town with over 75 ,000 residents SJP too experienced steady urban land conversions over the course of the entire study period. The newly urbanized areas in the northern and western edges of SJP are associated with informal housing while much of the urban growth in the int erior, south, and east are associated with industrial urban growth sometimes associated with the regional airport Except for the southern fringe of Curitiba, most of th e new large pockets of urban growth that were built from 2002 to 2007 are more formal in nature. The urbanized areas of Campina Grande do Sul and Quatro Barras mutually experienced a great deal of growth mostly formal in nature. In addition, a Volkswagen/Audi plant was built next to BR 376, in the southern extremity of the study area, f ar from the main urban fabric. Similarly, Araucria experienced new urban growth associated with various formal uses and informal settlement along the edges proximate to environmentally sensitive areas. Collectively this urban growth has accumulated to f orm a contiguous urban extension along most of the perimeter of Curitiba In addition, four secondary centers
86 Figure 45. Map of urban growth c hanges from 1991 to 2007
87 have consolidated substantially around Curitiba and these include: SJP located t o the southeast; Almirante Tamandar to the north; FRG to the south; and Araucria to the southwest of Curitiba. These centers fortify the polycentric organization of the urban system. However, the urban growth pressure for these urban centers to connect can be observed with the emergence of new conurbation development (part of which is informal) that has appeared at the fringe over the last several decades T he next section examines the pattern of urban expansion from 1991 to 2007 in the study area. Table 4 4 U rban area growth c hanges from 1991 to 2007 Land Cover Average Annual Change 1991 2002 Average Annual Change 2002 2007 Built 2,281.23 Acres 1,495.87 Acres Vegetation 2,400.05 Acres 1,269.17 Acres Water 302.37 Acres 361.23 Acres Bare So il 183.54 Acres 134.53 Acres *Acreages were derived using GIS tools and they are approximate. Urban Expansion: Fragmented or Compact? Tendencies towards consolidated or fragmented urban growth were measured over the course of the study time period. Us ing a basic formula referred to as the elast icity of urbanized land formula, urban sprawl and the rate of consolidation can be detected.3 A number over one indicates urban sprawl whereas a number below one indicates more compact growth (OSullivan, 2009). 3 The concept, urban sprawl, typically associated with forms of urban growth in the US, is defined as scattered and expansive low density development with buildings separated by swaths of land, with a lack of mixed use neighb orhoods, inaccessible road network, and noncontiguous development (referred to as dispersed, leap frog, or strip) without a centralized activity center in the outskirts of an urban area (Ewing et. al, 2002). The elasticity of urban growth is the ratio of change in new urban land conversions divided by the percent change in urban population growth.
88 0 5 10 15 20 25 30 1991 2002 2007 Urban Population Density (Persons per Acre) Study Area without Curitiba Study Area with Curitiba Curitiba Figure 46 Urban p opulation density, in the study a rea 1.3% 1.8% 2.3% 2.8% 3.3% 3.8% 1991-2000 2000-2007 Annual Growth Rate Study Area Urban Growth Rate (Area) Study Area Urban Population Growth Rate Figure 47. Comparison of the urban population growth rate with the rate of urban area
89 0 5 10 15 20 25 30 1991 2002 2007 Urban Population Density (Persons per Acre) Study Area without Curitiba Study Area with Curitiba Curitiba Figure 46 Urban p opulation density, in the study a rea 1.3% 1.8% 2.3% 2.8% 3.3% 3.8% 1991-2000 2000-2007 Annual Growth Rate Study Area Urban Growth Rate (Area) Study Area Urban Population Growth Rate Figure 47 Comparison of the urban popu lation growth rate with the rate of urban area expansion
90 From 19912002, the elasticity of sprawl value was 0.92 while from 2002-2007, the value, 0.4, was much lower These findings suggest that the urban gro wth pattern from 2002 to 2007 was more consolid ated. As portrayed in the urban growth change figures, the rate of urban land conversions decreased since 2002 as the rate of consolidation and infill development escalated. Figure 47 reiterates these findings by comparing the urban population growth r ate with the rate of urb an land conversions in the entire region. The urban land conversion rate dipped below the urban population growth rate from 2002 to 2007 indicating that the rate of urban expansion was lower than the annual urban population growth rate. In other words, there were fewer new urban land conversions yet more p opulation additions In addition, Figure 4-6 shows that the urban population density in the study area (whether inside or outside of Curitiba) increased from 1991 to 2007. The f indings on the growth of informal urbanization and a description on how informal urbanization played a role in increasing the rate of consolidation is provided in the following section. Informal Urban Growth Results Figure 4 8 visually maps out the in formal urban growth changes in the study area and Tables 4 -5 to 47 offer various acreages and rates of growth for the informal urban areas from 1991 to 2007. In several tables, the acreages of properties undergoing regularization were separated since reg ularized properties transition from informal to formal urban land uses. As discussed in the method, t he properties were identified as undergoing regularization if they had experienced some sort of substantial community upgrade in the informal area beyond the planning stages .4 Table 4 7 shows that for all 4 There are several informal communities in periphery towns, such as Campo Magro and Guarituba that by 2009 were emerging from the planning stages and entering i nto construction stages of regularization.
91 three study years, over 80 percent of the informal urban properties in Curitiba were informal settlements rather than irregular land divisions By 2007 around 30 percent of the total informal urban areas inside of Curitiba underwent regularization. I n contrast only 4.5 percent of the total informal areas outside of Curitiba were undergoing regularization. More regularization has tak en place in Curitiba; in fact a round 85 percent of the total portion o f regularizing informal urban areas is located inside Curitiba This is not surprising since t h e City of Curitiba has been i mplementing reg ulariz ation longer than the other jurisdi ctions in the study area. Analogous to the reduced rate in urban expansion, t he results reveal a substantial decline in the annual rate of new informal urbanization from 2002 to 2007. Whether the values include or exclude areas undergoing regularization the rate of additional informal urban acreage added to the study area from 1991 to 2002 was substantially higher than the rate of additional informal urban acreage from 2002 to 2007. In fa ct, Table 4 6 sho ws that the rate of i nfor mal growth decreased by over 150 acres per year and this trend continued even when including t he properties undergoing regularization Correspondingly Figure 4 -8 sho ws a great amount of new informal growth that appeared by 2002 and few new informal areas in 2007. Generally, most of the informal areas grew adjacent to existing informal areas Th e urban form for much of the informally built areas is often linear and elongated in shap e. These shapes often conform to the s ilhouette of riparian features. In some areas, the informal areas were linear in shape since they insulated a particular intens ive land use such as railroads, electrical transmission lines, or an industrial complex. Often an individual informal However, these were not considered as undergoing regularization given that the construction activities had not taken place by 2007.
92 urban area would enlarge over time near a natural feature, particularly if no regularization program was underway. T he dominant ecologic al and hazardous features of a local region might have a powerful influence on shaping informal settlement s and this was visually obvious in the RMC. Comparing the 1991 informal urban areas to the 2002 sites shows a general outward shift whereby more informal areas are located farther out in the periphery. The decentralizing pattern of informal urban growth is not unusual and is verified by the more interior loc ation of informal settlements that were in existence since 1978 (IPPUC map). Generally less informal settlement takes place in the dense urban core and older serviced areas. The urban fringe portion of Curitiba, encircling the urban core within a distance of eight kilometers, had the highest portion of informal settlement for all three study years (over 70 percent). However, the percentage in the periphery (beyond eight kilometers) has increase by five percentage points since 2002 (21 to 27 percent). These findings suggest that informal urban development gradually is decentralizing outwards farther from the urban core. The informal settlement identification found few new informal settlements in areas isolated from the urban fabric since 2002. Though, earlier, during the 1991 -2002 decade, more expansive informal growth occurred than during 2002 2007, particularly in the Municipality of Piraquara. Generally, most of the new in formal urban areas, larger in size, were more often contiguous to t he existing urbanization or settlement rather than occupation in areas isolated from the urban fabric Regionally, the informal urban spatial patterns seem to be guided by a tendency to locate on land without infrastructure but with a potential for obtaining infrastructure given its close location to serviced areas.
93 Table 4 5. Informal u rban area g rowt h from 1991 to 200 7 Inside Curitiba IS, no Reg. IS, in Reg. ILD, no Reg. ILD, in Reg. Total, inside Curitiba 1991 1,519.2 *N/A 104.2 *N/A 1,623.4 2002 1,623.8 748.1 257.2 35.6 2,664.7 2007 1,654.5 717.3 339.3 141.9 2,853.1 Outside of Curitiba, in th e study area IS, no Reg. IS, in Reg. Total, outside Curitiba Overall Total 1991 1,423.7 *N/A 1,423.7 1991 3,047.1 2002 3,227.6 *N/A 3,227.6 2002 5,892.3 2007 3,249.1 155.20 3,404.3 2007 6,257.4 Total Share Outside of Curitiba Inside of Curitiba Acr onyms: 1991 46.7% 53.3% IS: Informal Settlement 2002 54.8% 45.2% ILD: Irregular Land Division 2007 54.4% 45.6% Reg. Regularization Process *Notes: growth is presented in approximate acreages derived using GIS tools Table 4 6. Average annual r ate of change for informal urban g rowth Description During 19912002: During 20022007: During Entire Study Time Period : Excluding Regularization areas 187.4 3.7% 26.8 0.53% 137.2 4.5% Including Regularization areas 258.7 8.5% 73.0 1.2% 200.7 6.6% *No tes: results are presented in acres per year (used GIS) and annual rate of increase. Table 4 7. Informal urban p ercentages Inside Curitiba Outside Curitiba, in the study area ILD to Total IS to Total Total in Reg. Total no Reg. Total in Reg. Total no Reg. 1991 6.4% 93.6% *N/A *N/A *N/A *N/A 2002 11.0% 89.0% 29.4% 70.6% *N/A *N/A 2007 16.9% 83.1% 30.1% 69.9% 4.6% 95.4%
94 Figure 4 8 Map of i n formal urban growth, 1991 to 2007
95 Figure 4 9 Map of in formal urban growth and regularized areas
96 Informal settlers attempt to connect to serviced areas regardless of how poor the land conditions are for development. In addition, the amount of informal urban areas in comparison to the total built areas in the study area increased from two percent i n 1991 to 5.3 percent in 2002 and 2007. Overall, the informal urban areas comprise a small percentage of the total built area. Figure 49 shows where informal settlements and i rregular land divisions are und ergoing regularization. A local governm ent interviewee mentioned that irregular land divisions typically are easier to regularize given the propensity for more lot organization, preservation of minimum road widths needed for infrastructure, possession of some valid property titles, and location outside of conservation areas. Despite this point, there was a higher portion of informal settlements undergoing regularization than irregular land divisions perhaps this difference is a reflection of the greater amount of informal settlements in the r egion. Discussion on the Decline of Informal Urbanization The coinciding reductions in overall urban growth and informal urban growth show that the informal urbanization rate does in fact influence the overall regional growth though the rate of influence is uncertain. Figure 4 -10 compares th e urban growth rate to the informal urban growth rate for the total study area. As the informal urban growth rate declin ed, the urban growth reduced; h owever, their reduction rates are not fully in accordance. This r ate comparison suggests that there might have been an increase in other types of formal urban growth such as in dustrial and formal residential from 2002 to 2007 that leveled out the reduction in urban growth The reasons behind the reduced informal urban growth rate after 2002 are multiple. As such, a discussion on key changes and public policy influences is offered.
97 1991 to 2002 2002 to 2007 Annual Urban Growth Rate Annual Informal Urban Growth 0% 1% 2% 3% 4% 5% 6% 7% 8% 9% Annual Area Growth Rate (%) Annual Urban Growth Rate Annual Informal Urban Growth Figure 4 10 C omparison of informal to overall urban area growth rates Discussion on Influential Factors for Reduced Informal Urbanization During the 1990 decade Brazil was still recovering from a national economic crisis and a substantial portion of n ew -comers from 1991 to 2002 were rural -urban migrants that were in search of economic opportunity However, change s in the type of migrants have emerged and not all new -comers originate from rural areas. An urban migrant could have different housi ng preferences and this could affect the expansiveness of the development pat tern. The survey responses, a part of this research, showed that 23. 3 percent of the respondents originated from the interior part of the state of Paran, 20 percent from the Metropolitan Region of Curitiba ( RMC ) but outside of Curitiba, 17 percent from outside of the state (some from So Paulo), and 13
98 percent from the Ci ty of Curitiba.1 T hese findings suggest that a portion of RMC new comers are originating from u rban areas These new urbanites might be accustomed to living in apartments and other types of shared residential spaces Thus, the consolidation of development patterns might partially reflect changing housing preferences of urban n ew -comers An interviewee pointed out key infrastructure changes that might be influencing informal settlement growth. Many streets were paved in the urban portions of the RMC as a part of a region wide campaign referred to as Universal Streets ( or Plano Mil ). In fact 1,000 kilometers of roads were paved in the city and along transit routes This street paving program began with a c ampaign in 1997, then plan development in 1998 and the spreading of asphalt on the roads from 2000 to 200 2 T he respondent also pointed out that the installation of water, electricity and sewage infrastructure in every home is a state goal. In certain areas, these new infrastructur e improvements migh t have affected growth patterns. V arious influential public policies and social interest programs were implemented during this time period. The influence of these social interest programs was analyzed using geographic measurements, observations, and avai lable population and domicile counts. Analysis of Public Policy Influences on Informal Urbanization Several public policy changes emerged during the study time period that could have impacted the rate of informal growth.2 Inside the city limits of Curi tiba, various 1 Full survey responses are provided in Appendix A. Another survey conducted by COHAPAR (2008) found that 25 percent of the Zumbi in formal settlement residents migrated from the exterior and t he rest originated from the RMC (based on an interview in 2008). 2 A complete summary of the public policies administered during the study t ime period is provided at the end of Chapter 3.
99 social interest programs were implemented b eginning in the early 1990s Outside of Curitiba, the Territory Units of Planning (or UTP) conservation/social i nterest program commenced in 2000 and regularization of informal settlements was implemented only in a select few neighborhoods after 2002. Many of the public policies could encourage population density increases. To explore this potential effect, a map showing the overall population densities for 2000 in comparison to i nformal urban areas and social interest programs in Curitiba is provided (Figure 411) .3 The map shows high population densities along the edges of the urbanized fabric and in many of the informal urban areas and social interest program areas. In fact, the population densities in the southeastern and west edges are comparable to the population densities in the downtown area. In comparison to 1991 population densities, several of the social interest initiative areas and informal settlements undergoing regularization expe rienced overall population density increases. To analyze the urbanization influence of the social interest programs further the patterns of informal urbanization and rate of development was analyzed inside program boundaries. Inside of Curitiba, the so cial interest housing zones (SEHIS) and the Affordable Housing Company of Curitiba ( COHAB ) social interest initiative program boundaries were available while outside of Curitiba, the data was less reliable; thus 3 Figure 411 shows the population densities provided in the 2000 census dataset (IBGE). This data only shows the population density for areas that IBGE defined as urban consequently the white areas, in t he map indicate no data. The census dataset was not available for 2007. The informal urban areas include areas undergoing regularization and areas not undergoing regularization.
100 observations on the built pattern in and near UTP conservation areas were made in the subsequent section.4 Figure 411 Map of population densities in informal urban areas and affordable housing program areas and social interest zones in 2000 4 After 200 2 COHAB only had jurisdiction over implementing social interest housing programs inside of Curitiba and other agencies such as COHAPAR are leading social interest housing programs outside of the Municipality of Curitiba.
101 Inside of Curitiba: Public Policy Influences on Informal Urbanization Inside of Curitibas city limits, s everal areas were designated as the special sector of social interest h ousing zone (SEHIS) in 2000 and various regularization programs and serviced lot and housing provision programs took place in these zones. As shown in Figure 4 12 the SEHIS areas experienced rapid growth during the study time period. Initially in 1991, only 42 percent of the 6,075 acres of SEHIS zoned property was developed and then by 2007, 88 percent of the total area was built (see Table 4 9 ). This means that around 2,800 acres of new urbanization, partly produced informally was built over the course of 16 years. Figure 4 1 2 portrays the rate of rapid consolidation visually and illustrates the initial stages of developm ent in the SEHIS zones by showing the transition of cleared lots from bare soil/barren to built land cover in later years Another fact or influencing the consolidation of urban growth is related to regularization (se e last section of Chapter 3). R egular ization in Curitiba mostly began during the 1990 d ecade while in areas outside of Curitiba in the study area, regularization commenced later in 2002, in a few areas Regularization could have increased the supply of housing and rentals though there was no data available to measure rental sup ply changes and limited data to compare d omicile supply change s. As shown by the COHAB social interest initiative bo undaries in Figure 41 2 o ther programs were implemented in and nearby the SEHIS areas.5 During a go ver nment agent interview, a respondent offered reasons for the high urban growth in the social interest zoning region. T h e respondent described the B airro Novo region, situated in a 5 The social housing interest initiative boundaries (labeled as COHAB initiative areas) are show n in the 2007 SEHIS zoning map, located on the bottom of the page, in Figure 4 9.
102 SEHIS zone in the south as having higher housing densities than downtown Curitiba In addition, t he respondent described a public -private partnership program in which a serviced plot is given to a beneficiary whom is respons ible for the construction of a home. The respondent stated that t he housing programs in Bairro Novo foc us on lot provision and lately public housing is rarely constructed .6 The interviewee also described the rapid urbanization in and near various social interest program areas in the Tat u quara and Campo de Santana neighborhoods located in the south edge o f the city Overall, in Curitiba there are many innovative social interest programs undergoing implementation that offer serviced and regularized lots and dwelling units. The concentration of these programs in the SEHIS zones and social initiative areas o f Curitiba likely influenced the rapid development in the urban fringe. Moreover, these social interest provisions have accumulated substantially In addition, a little under half of the overall provisions (including housing regularization) are serviced lot provisions (MC, 2008; MC et al., 2007; and COHAB, 2007). Examining the new urban growth from 1991 to 2007 fulfilling a social interest function in Curitiba showed that 27 percent of all the new urbanization emerging during the study time period co ntained informal urban areas and/or social interest programs or zoning in Curitiba The share of th is total area (estimated at 4,233 acres) dedicated to social interest initiative f unctions (part of COHAB) is 50 percent, to SEHIS zones (not overlapping) i s 21 percent, and to informal settlement is 29 percent. The share of the 6 As described in the literature review, the lot partnership program has provided more lots than the other programs. In this program, t he city is responsible for improving the roads and the state is responsible for providing water, energy and sewage infrastructure Although there is a standard building plan for the homes, enough flexibility is offered for variation in home design and in the allowance of home expansion. Dif fe rent types of financing for apartment s and homes are offered (based on interview held in 2009 and COHAB, 2007)
103 new informal settlement s in Curitiba emerging after 1991 that are undergoing regularization is 33.5 percent. These numbers suggest that approximately 81 percent of the new urban growth with a social interest function in Curitiba is either undergoing regularization or is a part of some other social interest initiative and/or zoning. This finding indicates that a substanti al portion of new urban land co nversions, one in every three a cres, has been dedicated to a social interest function within the municipality of Curitiba The average housing density in the social interest areas with lot provision programs is around 20 homes per acre in 2005 In addition, the hou sing density in area s involved with regularization inside of Curitiba is around 24 dwelling units per acre while the non -regularized areas have on average, four less homes per acre than the regularizing areas. 7 Consequently the se new social interest housing areas and regu larized neighborhoods support a high housing density. Overall this analysis show ed that almost one in every three urban acres newly conve rted to urban uses between 1991 and 2007 was dedicated to informal settlement and for soci al interest initiative purp oses and 81 percent of this total is either undergoing regularization or is a part of a social interest program or zone. T hese findings show that the social interest strategy in Curitiba has contributed to density increases and the rate of consolidation i n a substantial portion of newly emerging urban growth inside the city of Curitiba Though a majority of these programs take place in Curitiba and not in the surrounding region. Consequently the urban growth impacts (consolidation and density increases ) brought on by the social interest programs described above were mostly felt in side of Curitiba but not outside of Curitibas city limits in the study area 7 Domicile counts inside of each informal urban neighborhood were only available for 2000 and 2005 and not for earlier years.
104 Figure 412 M ap of social interest zone urban growth from 1991 to 2007
105 However, as shown in a comparison of the informal urbanization rates inside and outside of Curitiba in Table 48, the reduced rate of new informal growth from 2002 to 2007 was felt both inside and outside of Curitiba. This reduced informal growth rate outside of Curitiba ind icates that the informal growth associated with Curitiba had not been displaced into the periphery. Table 4 8. Average annual informal area growth, inside and outside of C uritiba Inside Curitiba Outside of Curitiba, in the study area Acres Percent Ex cluding Reg. Areas Acres Percent Excluding Reg. Areas 1991 2002 94.7 5.8% 1.4% 258.7 8.5% 8.5% 2002 2007 37.7 1.4% 1.2% 73 1.2% 0.1% *Notes: The annual rate of growth is shown in acres added per year and percent change. Acreages are approximate and wer e calculated using GIS tools Reg. indicates Regularization areas. Ta ble 4 9. Urban area growth in social interest zones 1991 Percent 2002 Percent 2007 Percent Area Built 2,521 Acres 41.5% 4,839 Acres 79.7% 5,326 Acres 87.7% Annual Rate of Developm ent 1991 2002: 8.4% 2002 2007: 2.0% *Total area of Special Sector of Social Interest Housing zones (SEHIS) is approximately 6,075 acres. Acreages are approximate and were calculated using GIS tools. Outside of Curitiba: Reduced Informal Urbanizati on Outside of Curitiba, the decreased rate of urban growth from 2002-2007 is influenced by other factors since the implementation of regul arization only began in a select f ew areas by 2007. The decreased urban expansion rate might be attributed to develop ment consolidation that typically emerges after initial settlement In fact, new informal settlements immersed in the rural -urban t ransition area seem to settl e at lower housing densities during the initial stages of occupation and a s the se informal
106 settl ements in rural -urban transition areas age, they consolidate. As the land becomes more secure for informal settlement, others join the community such as relatives by settling on nearby vacant land unprotected from their occupation. Given that some of th e periphery towns have a greater portion of informal urban area s than others, the informal urban growth in each of the surrounding periphery towns are presented in Tables 411 and 4-12. The urban popul ation density graph, Figure 413, shows density changes from 1991 to 2007 and is based on urban population counts (IBGE, various years ) and built area calculations.8 FRG and Campo Magro experienced steady density increases while Almirante Tamandar, Pinhais, and Campina Grande do Sol experienced slight decreases in density. The addition of new industrial and formal ly built residential areas (with lower densities) in Pinhais and Campina Grande do Sol possibly triggered a decline in urban population density while the tremendously hazardous building issues in t he karst areas of Almirante Tamandar might have reduced the rate of development. As illustrat ed in the bar graph (Figure 413), t he population densit ies for most of the larger periphery towns wit h populations over 100,000 increased substantial ly since 1 991. In fact, t he periphery towns of Colombo, SJP, and Piraquara increased in density tremendously and t hese towns also have the highest proportion of infor mal urban areas and UTP social interest/ conserv ation areas within their limits Examining the info rmal urban growth from 1991 to 2007 visually shows areas designated for social interest in the UTP program areas did experience infill development. 8 The density, pre sented as persons per acre, might appear higher than other estimates for the region given that most other estimates include rural, non developed land in the calculation. Only the urban land acreages derived from the study classifications were used to calc ulate the population density.
107 To examine the rate of consolidation in periphery towns with a substantial portion of informal urbanization Tables 4 -10 and 4-11 compares the amount of informally built land in each town compared to the total amount of informal area in all of the periphery towns The urban growth in Piraquara, the town with one of the largest informal settle ment in the RMC c ontained around 19.4 percent of total share of informal urban area thus this area likely was influenced by informal development patterns .9 Comparing density with rates of informal urbanization shows a great deal of densification in Piraquara from 2002 to 2007. Viewed jointly, the urban areas of Colombo and So Jos dos Pinhais have over 1,450 acres of informal uses in 2007 collectively, 42.7 percent of the total amount of informal urban areas in the periphery towns of the study area. These two urban centers also experienced substantial increases in population density and reduced rates of informal urban land conversions from 2002 to 2007 in informal urban areas In addition, Colombo and FRG had some neighborhoods unde rgoing regularization by 2007 and th ese efforts could have influenced the substantial reduction in the informal urban land conversion rate from 20022007 .10 Lastly, the periphery towns, Almirante Tamandar (13 percent) and Pinhais (10.5 percent) also have a substantial amount of informal ur ban areas but these areas experienced slight decreases in urban population density from 2002 to 2007. Thus it is uncertain whether the informal urban areas in these towns consolidated. 9 The Guarituba i nformal settlement had entered into the plan development stages as a part of a regul arization effort by 2007 but physical improvements began around 2009 10 The overall built land conversion rate for FRG had dropped, though not as drastically as the informal land conversion rate. Thus formal urban growth has continued in FRG.
108 Urban Population Density in Periphery Towns 0 5 10 15 20 25 30 Almirante Tamandar Araucria Campina Grande do Sul (CGS) Campo Magro Colombo Fazenda Rio Grande (FRG) Pinhais Piraquara Quatro Barras So Jos dos Pinhais Persons per Acre 2007 2002 1991 Figure 4 13 Urban population density changes in periphery tow ns
109 Table 4 1 0 In formal urban area s in periphery towns Municipality Informal Areas, 1991 Acres Portion of Total Informal Areas, 2002 Acres Portion of Total Informal Areas, 2007 Acres Portion of Total Almirante Tamandar 245.19 17.22% 406.34 12.59% 4 40.58 12.96% Araucria 114.43 8.04% 166.71 5.17% 188.69 5.55% Campina Grande do Sul 13.53 0.95% 24.83 0.77% 29.32 0.86% Campo Largo 17.75 1.25% 19.99 0.62% 20.87 0.61% Campo Magro 41.40 2.91% 50.52 1.57% 61.69 1.81% Colombo 412.32 28.96% 741.33 22.97% 783.65 23.06% Fazenda Rio Grande (FRG) 75.03 5.27% 200.37 6.21% 173.06 5.09% Pinhais 210.92 14.82% 344.19 10.66% 359.34 10.57% Piraquara 60.14 4.22% 641.04 19.86% 660.20 19.42% Quatro Barras 5.70 0.40% 14.25 0.44% 14.88 0.44% So Jos dos Pinhais 227 .27 15.96% 618.04 19.15% 666.71 19.61% Total 1,423.67 100.00% 3,227.62 100.00% 3,398.98 100.00% Notes: A small portion of the eastern edge of the Municipality of Campo Largo, outside of the main downtown urban area of Campo Largo and on the western edg e of Curitiba was included The total acreage of informal urban area s includes areas outside of the City of Curitiba. Table 4 1 1 Comparison of periphery towns informal urbanization rates Informal Urban Land Total Built Areas Municipality 1991 2002 Annual Rate 2002 2007 Annual Rate 1991 2002 Annual Rate 2002 2007 Annual Rate Almirante Tamandar 5.97% 1.69% 3.51% 1.81% Araucria 4.15% 2.64% 1.51% 1.76% Campina Grande do Sul (CGS) 7.59% 3.62% 6.61% 1.56% Campo Magro 2.00% 4.42% 1.61% 2.15% Colombo 7.25% 1.14% 3.61% 1.44% Fazenda Rio Grande (FRG) 15.19% 2.73% 6.78% 2.88% Pinhais 5.74% 0.88% 2.93% 2.83% Piraquara 87.82% 0.60% 14.62% 2.26% Quatro Barras 13.64% 0.88% 8.16% 3.11% So Jos dos Pinhais 15.63% 1.57% 4.70% 2.48% Notes: CGS and Qua tro Barras have less than one percent of the total amount of informal areas. Campo Largo was excluded.
110 Discussion on Increased Density in Periphery Towns These findings show that the holistic urban growth patterns in several of the larger periphery to wns (Colombo, SJP, and Piraquara) may have experienced densification in informally built areas The density increases and consolidation in the towns of FRG and Colombo could be connected to regularization but Piraquara and SJP likely experienced density i ncreases for other reasons such as consolidated development in the social interest regions of the UTP program and continued densification processes after initial settlem ent Thus, t he urban growth pattern changes underway in informal urban areas, sometime s a result of community upgrade and social interest programs ha ve influenced the density changes in the urban fabric only partially outside of Curitiba Moreover, there is too little of information on the specific housing increases to accurately measure t he influence that the UTP social interest program has on consolidating informal development and restricting development in co nservation areas. Holistically, t he regularization in Colombo, FRG, and Curitiba could sup port density increases. I n the other areas such as Piraquara, the densification process could have been effected by UTP program implementation and i nitial informal settlement consolidation processes Collectively the urban population density in the entire stud y area has increased overall from 1990 to 2007. In addition to informal urban pattern chan ges, this trend might generally reflect regionwide infrastruct ure improvements overall reduced supply of urban land ideal for development, and decreases in the rate of population growth. Overall these findings indicate that the social interest strategy inside of Curitiba and partially outside of Curitiba likely has played a role in the consolidation and densification changes in the region These results show a regional shift towards more compact growth but this growth may not be fully sustainable given
111 the tendencies for infringement in environmentally sensitive areas. As such, results are provided below that compares the quantity of informal urban areas in hazardous and ecological fragile areas and then, the results of the spatial trend analysis are offered. Locational Determinants Results Findings on Hazardous and Ecological Fragile Spatial Trends The amount of informal urban land within the karst, conservation lands, floodplains, and the 50 m eter buffer of rivers in 1991, 2002, and 2007 was calculated and t hese acreages are provided in Table 4 12.15 Maps showing informal urban growth in conservation lands, karst regions, floodplains and riverbanks are i n Figures 4 14 and 4 -1 5 Table 4 1 2 demonstrates the strength of these locational determinants by describing the share of informal settlement within each of them Viewing the properties collectively that are either in a floodplain, co nservation area, 50 meter river buffer, or karst region, showed that 95 percent of th e informal urban areas were in one or more of these features in 2002 and 2007 and 92 percent in 1991. In other words, on average, over 90 percent of the informal urban areas were loca ted within a hazardous and/or environmentall y sensitive landscape feature that i s common in the Metrop o litan Region of Curitiba (Regio Metropolitana de Curitiba RMC) Along with a reduced informal urbanization rate since 2002, the annual rate of informal urban growth taking place in conservation areas, karst regions, riparian areas, and floodplains has declined The rate of informal urban occupation in conservation lands reduced dras tically from 23 percent to 1.3 percent during 2002 2007 in 15 The steep slope parameter could not be assessed given data restrictions. Although a small percentage of informal settlement has occu rred near high voltage electrical transmission lines and railroads, these indicators were not as prevalent outside of the Municipality of Curitiba.
112 comparison to the earlier decade.16 Likewise, the annual rates of informal urban growth within floodplains and karst regions both dropped by over five percent after 2002. The least substantial decline in the rate of informal urbanization occurred in riverbank areas Even though the annual rate of growth has largely declined, the total development footprint of informal areas located in conservation lands, karst features, riverbanks, and floodplains increased over the entire study time period. Additionally, over 30 percent of the informal urban communities were situated in a r iverbank and around 20 percent of the informal urban area occupied 100 year floodplains during the entire study time period.17 In other words, as time lapsed, the sheer acreage of informal urban within floodplains continued to gradual ly increase. The survey responses complement these findings given that 77 percent of respondents mentioned that settling outside o f a floodplain was not an important locational factor for them. When droughts emerge the danger of settling within a floodpl ain might not be visibly apparent. Discussion on Hazardous and Ecologically Fragile Trends Largely this decline in the rate of urbanization in environmentally sensitive lands show s progress towards achieving a more sustainable form of development As dis cussed in the literature review, the UTP conservation areas in the study area contain areas designated for consolidated informal settlement development. Perhaps these areas have consolidated and this density transfer reduced the impact to other protected 16 Conservation areas are defined as UTP or APA conservation areas or parks. 17 The floodplain areas wer e geographi cally located by combining floodplain GIS data from IPPUC that represents 100 year floodplains and areas inundated with flooding during a 1983 flood event and SUDE RHSA land use land cover data that provided a flooded areas category ( rea alagada) The land cover data produced by SUDERHSA was derived by visually interpreting 2000 orthophotographs at the 1:20,000 meter scale.
113 sections in the conservation areas. Ov erall the rate of decline in conservation areas was drastic it went from 23 percent to 1.3 percent. Still the gradual expansion in the amount of informal urbanization in these ecologically fragile and hazardous l andscapes should serve as a warning for looming water quality impacts and increased destruction in the case of a natural disaster scenario. As informal urban growth continues in these landscapes the severity of damages and greater loss of life from a significant natural disaster or flood event w ould increase for the most vulnerable sectors of the population. The overall expansion of informal housing, precariously constructed within ecologically fragile lands, indicates a need for sustained efforts on low -income housing initiatives and merits continuous monitoring. If the rate of informal settlement growth accelerates in ecologically fragile terrain that is hazardous for development, the affordable housing, rental and land supplies and accessibility issue s should be critically examined. Lastly, the accuracy of assessing informal urban ization in at -risk landscapes is heavily dependent on the accuracy of the source geographic datasets that represent the natural features. Several of the geographic datasets required cross reference and adjustment with more accurate and reliable resources. Table 4 1 2 In formal urban areas in environmentally sensitive, hazardous areas Informal Urban Areas in Conservation Areas 1991 2002 2007 570 2,009 2,139 A cres 18 .7 % 34.1 % 34.2 % (share of total Informal Urban) Annual Rate Increase 1991 2002 2002 2007 23 % 1.3 %
114 Table 4 1 2 Continued Informal Urban Areas Located inside Karst Areas 1991 2002 2007 58.1 116 124 A cres 1.9 % 2 % 2 % (share o f total Informal Urban) Annual Rate Increase 1991 2002 2002 2007 9.1 % 1.4 % Informal Urban Areas Located inside the 50 meter River/Stream Buffer 1991 2002 2007 1,095 1,874 2,052.5 A cres 35.9 % 31.8 % 32.8% (share of total Informal Urb an) Annual Rate Increase 1991 2002 2002 2007 6.5 % 1.9 % Informal Urban Areas Located inside Floodplains 1991 2002 2007 661.3 1,127.3 1,175.6 A cres 21.7 % 19.1 % 18.8 % (share of total Informal Urban) Annual Rate Increase 1991 2002 2 002 2007 6.41% 0.86% Data sources for natural feature GIS datasets : IPPUC, COMEC, SUDERHSA, and Mineropar. The conservation area calculation includes UTP and APA conservation areas and parks The acreages we re calculated using GIS tools and are approximate. Results on Socio Economic and Bus Service Locational Trends In conjunction with recognizing physical land trends various other infrastructure and so cio -spatial characteristics prevalent in the informally developed fabric were analyzed In general, informal settlements were near regularized areas though it was i
115 Figure 414 Conservation and karst areas map
116 Figure 41 5 W ater, floodplains, and steep slopes map
117 i mpossible to separate their independent influence since they are inter twi ned with other indic ators such as location near other informal settlements and public housing. I nformal urban areas often converge near social interest initiative areas .18 Figure 4 -16 provides a visual comparison of the location of informal urban areas to social initiative areas built in Curitiba by 2007 Generally the social initiative areas and informal urban areas congr egate beside each other along the edges of Curitiba. In regularizing area s, part of t his concentration might have been intentional s ince public housing built for household relocations typically is located as close as possible to the informal areas so as to not disrupt social networ ks A separate analysis was performed to account for the social interest initiative a re as that appear t o be connected with an informal area undergoing regularization and this analysis showed that around 45 percent of the informal urban areas inside of Curitiba were within 500 meters of social interest initiative areas from 1991 to 2007.19 Another calculati on including all the social interest areas showed that around 7 5 percent of th e informal urban areas in Curitiba are within 500 meters of social interest initiative areas. Largely, the juxtaposition of informal settlement with social interest initiative areas is a trend and this concentration has increased socio economic segregation. Another sign of a tendency towards socio economic segregation is the spatial clustering of income brackets. A spatial autocorrelation tool, referred to as Moran I was 18 As discussed in the literature review, the social interest areas contain a variety of programs that offer public housi ng and urbanized lots 19 The geographic dataset showing social interest initiative areas did not differentiate between public housing, serviced lot, and other types of social interest program areas. Thus the individual influence of informal settlement loc ated near public housing could not be determined.
118 use d to measure the rate of income clustering and dispersal in the RMC for 2000. The calculation yielded an index score of 0.86 and this score signified great clustering that has less than one percent chance of random occurrence .20 Figure 4 -18 depicts the income concentrations spatially and shows the clustering of informal urban areas outside of middle and high income neighborhoods. An examination of 87 percent of the informal urban areas inside the census defined urban limits showed that 97 percent of the informal urban areas (for 2002) were located in low and moderate income regions.21 Together the results yielded from the social interest proximity analysis and income level clustering both imply that socioeconomic segregation is a pattern in the region. Location near bus access proved to be a crucial factor for informal settlement A large majority of the informal urban properties in 2007 were located within 500 meters of bus service (96 percent) The extensive bus rapid transit network in great er Curitiba reaches various urban extremities, thus residential access to bus service is not rare. However, it should be determined whether the bus service was installed befor e or after informal occupation. The only informal urban clusters with out any nearby bus route are located in the southwestern edge of Araucria. In addition, a large majority of the survey respondents felt that bus service was an important factor in settlement location (93 percent of respondents). A large majority of the informal s ettlements tend to concentrate in the urban fringe and periphery portions of the city structure (see Figure 417 and previous section 20 The Morans I tool was used to measure the clustering of income for 2000, in the areas defined as urban by IBGE (census) 21 Around 63 percent of the 2002 informal urban areas were located in low income c ensus blocks (on average, earn less than three monthly wages a month) and 34 percent were located in moderate income blocks (on average, earn over three but less than six monthly wages a month). Income values were not available for 1991 and 2007.
119 describing decentralized growth). These concentrations are not perfectly concentric and instead, they emerge outside of t he cores of surrounding towns and Curitiba Generally the older an urban center is, the more likely the urban cores of periphery towns are to be formally urbanized. These older urban hubs are more likely to possess full infrastructure, better transportat ion access, and basic sanitation services; thus the property values in these serviced areas are higher. Consequently the more mature, i ntensely urbanized centers secondary to Curitiba, have experienced their own form of concentric urban growth that contains informal settlement along the urban fringe and in marginal land s During a field visit, a government agent pointed out the increased amount of formal residential and commercial development in the more intensely urbanized center of Pinhais. The cente r of Pinhais is the area closest to downtown Curitiba and therefore, one of the wealthiest in the region The respondent said that those with less income tend to live closer to riparian areas on the edge of the urban core. The government agent s aid the c ost diffe rs drastically in Curitiba and Pinhais ; for example, they suggested that t he cost of a square meter parcel of land in Curitiba was twice the amount of land in Pinhais Consequently, these low land and housing costs in the peripheral areas might a ttract informal and formal housing development alike and this could be explored further by examining data on property and building value appraisals Discussion on Socio -Economic and Bus Service Locational Trends The findings suggest that informal urban areas in the study area are often located near bus service and social interest initiat ive areas, in the urban fringe and periphery. The trend for most informal urban areas to be located near bus service would be a critical feature to improve their li velihood given that most informal residents do not
120 Figure 4 1 6 Map of bus routes, affordable housing initiative areas, and electrical facilities
121 Figure 417 Urba n centers map
122 Figure 418 Income level concentrations map
123 possess a car Althoug h Curitiba excels in providing bus access to most of the citizens in the region the convenience of this service diminishes the farther a resident is located from an express bus line and their intended destination.22 The tendency for informal urban growt h to occur in the urban fringe and periphery coincides with the location of low property values, land without infrastr ucture, and several rivers and floodplains. The property value of land in floodplains and riverbanks particularly in the city outskirts i s extremely low ; consequently property owners are more likely to tolerate and sometimes encourage info rmal settlement in these locations .23 Several of these trends could be the outcome of social processes particularly related to increasing income segregat ion that is also organized by the development quality of the land and utility levels. Key Findings from Community Survey s The surv ey respondents pointed out their preferences for where to live and these responses cue in on the rational considerations for informal occupation of land For instance, 83 percent thought it was important to settle near existing informal settlements and 93 percent thought it was crucial to be near infrast ructure and regularized areas. In addition, 73 percent thought it was imp ortant t o live close to family and 71 percent found a housing location near em ployment centers and opportunities ideal. Another ideal circumstance, mentioned by 97 percent of respondents, is to location near social services such as school s, daycares, and h ealth centers The community survey respondents offered their viewpoint on the main reasons for the accelerated growth of informal settlements since 1990. Over 30 percent said that 22 Accor ding to a survey, 50 percent of the employed residents living in the Zumbi Neighborhood spent more than one hour on buses to reach their place of employment (based on an interview, 2008). 23 P roperty owners may benefit from informal settlement given the potential for regularization procedures that incorporate infrastructure provision and government compensation for land acquisitions
1 24 the search for employment spurred informal urbanization while 19 percent pointed to the need for less expensive housing and land. Other significant reasons for informal growth were a better quality of life and so cio economic benefits (16 percent) and the lack of money, resources, and financing (13 percent) .24 Du ring a field vis it, an interviewee said that everyone wanted to live in Curitiba during the 1980 and 1990 decades because the good reputation of Curitiba had spread. A myriad of reasons generally associated with economic gain and quality of life improvements brought new -comers to the city and spurred informal urban growth. Other intriguing findings from the survey revealed that many informal residents embraced living outside of the busy urban core. Many respondents did not feel that it was desirable to live near central Curitiba a nd one respondent found the crime there intolerable. Many respondents expressed happiness with not living in a more urban environment The respondents discussed advantages and disadvantages to living in the community and the range of th e se responses is provided i n A ppendix A. As a whole, t hese responses illustrate the settlers needs for infrastructure, bus access, and services, a better quality of life and closer proximit y to employment opportunities, schools, and health centers. Many of the disadvantages to living in an informal settlement are related to the negative externalities commonly associated with concentrated poverty such as crime, prejudice of the favela, and lack of security. Lastly, the survey asked respondents what were the main obstacles for living in the central urban areas. While 63 percent did not want to live in the central urban areas, 17 24 Another survey of the Zumbi Informal Settlement found settlement reasons associated with the pursuit of employment a better qual ity of life, and availability of inexpensive housing opportunities not available in the City of Curitiba (based on an interview in 2008).
125 percent of the respondents found land prices to be too expensive and eight percent felt that entrance into city living was limited t o those whom are trained professionals. Discussion on Community Survey Findings The informal settlers largely expressed preference to settle near areas with a potential for a higher quality of life. These areas often are located on land near infrastruct ure and regularized areas that offer the potential for infrastructure connection. T he location of informal settlements away from desirable community features or services such as sc hools and full infrastructure c ould be recognized as a potential trend and a product of the urban system The preference of settlers to live near other informal settlements might reflect the social acceptance these areas offer and in contrast, the exclusiveness of formal areas Overall these responses touch on the importance o f having access to economic services and other quality of life opportunities where they live and point to high land and housing prices as a barrier to living in urban centers Summary of Research Findings The measurement of urban growth in greater Curit iba derived from remotesensed land cover classifications, showed substantial urban expansions and decentralized urbanization outwards into the urban fringe and periphery from 1991 to 2007 at the same time as population density increased. The urban growt h extended outwards towards secondary urban c enters in a polycentric pattern and in several areas the urbanization from these secondary centers connected together to form conurbations that concentrically has expanded Curitiba. The overall urban growth bec ame more compact and consolidated from 2002-2007 in comparison to the 1991-2002 decade. There are multiple forces associated with this shift, many of which delve into economic, social, political, and environmental matters such as regionwide infrastructur e
126 improvements However, a critical force in density increases and reduced urban expansion is related to informal settlement urbanization. The rate of new informal urban growth declined drastically f rom 2002 to 2007 in comparison to the rapid urbanization that unfolded from 1991 to 2002 In addition, the rate of informal urbanization reduced more drastically than the overall rate of urbanization in the study area. Overall, informal urban ization has decentraliz ed and the regional patterns show that inform al urban areas generally are located o utside of Curitibas downtown area and around the core of older more established urban centers in the periphery The frantic informal settlement growth during the 1990 decade was reflective of hard economic times. Whi le the consolidated growth beginning around 2000 signaled economic recovery and it was partially impacted by the implementation of a robust social interest sustainable planning program. In fact, after 2002, population densities increased in many of the p eriphery towns with a larg e amount of informal urbanization and social interest initiative s Isolating the influence of government initiatives in specific areas revealed insights on the urbanization trends of this study time period ; however this analysis was more successful inside of Curitiba given data restrictions Inside of Curitiba, t he rapid pace of compact development in the social interest zones and initiative areas triggered c onsolidated informal development and facilitated density increases In fact, approximately 27 percent of all types of new urban growth consisted of informal settlemen t and social interest initiative areas in Curitiba from 1991 to 2007 and 81 percent of these newly urbaniz ed areas are either undergoing regularization or are a part of another s ocial interest program or zone. Inside of Curitiba, most of the informal
127 urban areas undergoing regularization increased in housing density from 2000 to 20 05 at an average increase of 4 domiciles per acre T he myriad of social interest pr ograms provided a substantial amount of service d lots, public housing, and regularized informal settlements. Curitibas focus on the provision of serviced lots is pro active and was a driving force in capturing a substantial portion of potential ly expansi ve informal urbanization. Instead of land invasions, these new -comers were directed to serviced land that is more suitable for development The allowance of progressive housing improvements and gradual construction was a feature that appealed to lot reci pie nts In addition, t he act of regularization could increase densities and the supply of rentals .25 In Curitiba, many of these programs were well underway during the 1990 decade; consequently the urban pattern effects such as consolidation and increased density of urban growth were felt mostly inside of Curitiba Outside of Curitiba less regularization was underway by 2007 and as a consequence, the resulting implications were less significant. In 2000, the UTP conservation policies restricted develop ment in certain areas and encouraged the consolidation of informal development in social interest region s already experiencing settlement. However, the extent that the UTP program influenced reduced informal urbanization is uncertain. Holistically, vario us social interest initiatives could have increased density and the rate o f consolidation inside and outside of Curitiba however t he extent that specific public polic y has influenced urban growth patterns should be examined at a later date through a more d etailed study that examines the housing densities in individual neighborhoods and key program locations 25 As discussed in the literature review, regularization might eliminate th e need for septic tank facilitie s, necessitate t he development of compact housing units and lots and spur the investment of second stor y home expansions that often serve as low income rentals.
128 Along with recognizing urban growth patterns, the chief locational determinants of informal settlement were identified. Location near bus access prov ed to be crucial for info rmal occupation. In addition, around 45 percent of the informal urban areas in Curitiba were within 500 meters of social interest initiative areas from 1991 to 2007. T he strongest geographic indicators for informal urban settlem ent are location in floodplains, conservation lands and i n riv erbanks. Viewing the properties collectively showed that over 90 percent of the informal urban areas were located either in a floodplain, conservation area, 50 meter river buffer, or karst regi on from 1991 to 2007. However, t he results found that the rate at which i nformal development encroaches i n the aforementioned ecologically fragile and hazardous lands has dropped overall since 2002. Although the a mount of informal urban invasions in ecol ogically fragile areas has not declined overall the rate of new informal urban land conversions in these areas has substantially dropped (by at least five percent in all categories) This decline in the rate of urban land conversions in conservation areas might be attributed to an overall decline in informal urbanization and new conservation laws that encourage consolidated informal development in certain social interest areas These research findings imply that location in a floodplain, conservation ar ea, or karst region, in the urban fringe and periphery ; and close proximity to a river and bus route could be useful for predicting and monitoring the growth of informal settlement In fact, if several of these indicators are combined, over 97 percent of the informal urban areas were selected for all three study years ( see Table 4 -13 ). However, the strength of some of these indicators has declined (such as location in conservation area, riverbank,
129 and floodplain); thus the set of indicators useful for predicting informal settlement location might change. Table 4 13 Combination of spatial variables Year Informal Urban Areas within Combination of Indicators Total Informal Urban Acreage Share 1991 2,983.4 3,047.1 97.9% 2002 5,787.9 5,892.4 98.2 % 20 07 6,134.9 6,257.4 98.1 % The combination of features includes areas within 500 meters of bus line; and/or in a karst region, conservation area or floodplain; and/or within 50 meters of a river or stream. Largely, informal settlement accumulates near other informal settlements, public housing, and social i nitiative and regularization areas In addition, informal urban areas mostly were located in low -to moderate income neighborhoods. Other less prevalent indicators are location in karst regions ste ep slopes and settlement near electrical transmission lines Most respondents felt that location near infrastructure was crucial. In addition, they indicated that close proximity to bus service access, family relations employment centers and opportunit ies were all important. In addition, t he choice in where to settle might be limited to areas outside of desirable community features such as schools and in areas without infrastructure given the strong potential for increased land prices An outline of the findings on spatial trend findings is pro vided in Table 4 14 and Table 4 15 provides a comparison of key government program influences on informal urbanization tr ends. Both Tables 414 and 4-1 5 summarize major findings derived from this research. The findings stress the need for continual evaluation of the government strategy and public policy influences o n informal urbanization outcomes. This
130 evaluation should be iteratively developed and adapted to ensure that accurate results are yielded. This re search was designed to measure the collective influence of the government approach in greater Curitiba to discover ways how they might influence informa l urbanization outcomes and encourage inclusive sustainable development throughout the region. As a wh ole, the findings suggest that new informal settlement growth is most likely to occur in vacant areas located in the urban fringe and periphery that have access to bus service in the RMC and that location of informal settlement in floodplains, conservation areas, and near rivers and social interest housing programs is a common locational trend however growth in conservation areas, rivers and floodplains has diminished Consequently the strength of these trends has weakened. Lastly, the results indicate th at t he government strategy used in Curitiba did help to reduce the overall informal urbanization rate and encourage consolidated urban growth. Discussion on Overall Results Collectively these findings indicate that in the case of greater Curitiba, informal settlement location is not random and the prevalent patterns are an outcome of limitations posed on informal settlers M uch of the settlement s ten d to concentrate on marginal hazardous lands without infr astructure in areas where their occupation is tole rated and in some cases encouraged because of the potential for land owners to gain infrastructure and government compensation. The settlement location often is near urban infrastructure, perhaps because this land is perceived by residents as having the p otential for acquiring infrastructure. In addition, informal settlements typically concentrate around areas where they are socially accepted such as near other informal
131 settlements low income neighborhoods, and social interest initiative areas The exis tence of common spatial patterns supports the conjecture that there are prevalent Table 4 14 Summary of spatial trend findings associated with informal urbanization Spatial Pattern Description Decentralization of informal urban areas and location in t he urban fringe and periphery Steady decentralization from 1991 to 2007. From 1991 to 2007, over 70% of all informal urban areas were located in the urban fringe. From 2002 to 2007, the portion of informal urban areas in the urban periphery increased from 21 to 27%. Informal urbanization in conservation areas The share increased from 18% in 1991 to 34% in 2002 and 2007. Though the annual growth rate in these areas drastically decreased from 23% (19912002) to 1.3% (2002 2007). Informal urbanization in fl oodplains The share was similar for all three study years: 21% in 1991, 19% in 2002, and 18.8% in 2007. Although the annual growth rate in floodplains has decreased from 6.4% (19912002) to 0.9% (20022007). In addition, 77% of survey respondents did not find location outside of a floodplain an important factor in where they live. Informal urbanization within 50 meters of a stream or river The share was similar for all three study years: 36% in 1991, 32% in 2002, and 33% in 2007. However, the growth rat e has decreased from 7% (19912002) to 2% (20022007). The rate of decrease is not as significant as the decline in conservation areas and floodplains. Informal urban areas location near bus service Most informal urban areas were located within 500 meters of bus service (96%) and most respondents thought location near bus service was important (93%). The bus service in greater Curitiba is often extended to populations in need. rationalities for the informal occupation of certain types of landscapes T hese residents are not allotted the opportunity to live in a des irable neighborhood with i nfrastructure
132 and services and instead their choice in where to settle is determined by where they are tolerated and permitted and how the urban system limits settlem ent options Table 4 15 Key government program influences on informal urbanization Spatial Patt erns & Public Policy Influence Description Reduced informal development in floodplains & riverbanks from 20022007. Drastic rate reduction in conservation a reas: from 23% (19912002) to 2.2% (2002 2007). A combination of social interest programs, development exchanges, and conservation plan (UTP and APA) restrictions. Table 4 15. Continued Inside of Curitiba: Informal settlement location near social in terest housing programs and SEHIS zones 75 percent of the informal urban areas were within 500 m eters of social interest initiative/zoning areas. Regularization interventions often include public housing development nearby an informal settlement for housin g relocations. Inside of Curitiba: Consolidation and density increases within informal urban areas From 1991 to 2007, rapid consolidated development in S EHIS zones and social interest initiative areas. Social interest zones and initiatives Fr om 1991 to 2007, 27% of all new urban growth was dedicated to a social interest function (81% of these areas are part of a social interest or regularization program) in Curitiba. A round 42% of the total social interest provisions are given in the form of s erviced lots in Curitiba. The social interest programs support high housing densi ties. Regularization Around 30% undergoing regularization in 2002 and 2007 and these areas supported slightly higher housing densities than nonregularized areas (d om icile counts 20002005). Couldn't measure rental increase. Thus, regularization efforts could have partially contributed to density increases though need to measure rental stock changes. Outside of Cur itiba: Partial consolidation and density increases in informal urban areas UTP social interest areas & regularizing areas could've experienced increased consolidation in a few areas. Visually it appears many informal areas have consolidated and informal urbanization in conservation areas has reduc ed drastically though more data and further study is needed
133 The spatial pattern findings exemplify the ways in which informal settlement communities disproportionately live in hazardous deplorable living environments such as in a floodplain The strength of these trends illustrates the types of divisions and contrasting circumstances between formal and informal and the urban poor and rich in greater Curitiba. This divide is not always black and white in Curitiba and there are shades of gray in which a mi xture of informal and formal residents live side by side particularly in areas transitioning between the se divisions .1 A reduction in the strength of the informal -formal divisions and trends reflective of social inequality might indicate progress in term s of supporting inclusive sustainable development Thus, these trends and patterns should be monitored and researched further. The overall reduction in new land invasions shows signs that other housing options are capturing low -income new -comers. One o f the goals of Curitibas social interest plan was to use mechanisms to reduce the rate of new land invasions. This study shows a reduction in the amount of informal settlement even in areas beyond Curitibas political boundaries These findings are moderated by the uncertainties in whether all informal urban areas were captured through the identification method and by data limitations Method Transferability Th e basic steps involved with this informal urban identification approach could be appli ed e lsewh ere i f it is customized to local ci rcumstances. The premise behind the method is to focus informal urban identification b y first recognizing the common areas at -risk of informal urban development in a particular region. In the RMC, t hese at -risk 1 Based on field visit observations, several areas had a mixture of informal and formal housin g.
134 areas often included dominant natural features of an urban area that is deemed formally undevelopable (examples include steep mount ains prone to landslides, riparian areas or floodplains ). However, the set of informal settlement and irregular land division i ndicators vary according to the urban context Accordingly, the development of logic based on local knowledge and the commonalities of known informal settlement locations (such as finding common building patterns ) would be critical for detection methods In addition, local knowledge would be critical for narrowing down the scope of review and for verifying results In the RMC, t he use of aerial photographs with high spatial resolution was crucial for deciphering the informal urban areas more precisely. Often the informal settlements in Curitiba were small in size and linear in shape; therefore the use of aerial with high enough spatial re solution to see these building features was crucia l. Depending on the urban region, informal areas might vary and wh en urban regions have larger informal settlements the detection process might be easier and require less detailed orthophotographs Though, the aerials likely will need to provide high enoug h spatial resolution (at least ten meters) to see building patterns Another unique difference of Curitiba in comparison to other urban regions is in t he overall amount of informal urban uses a larger amount of irregular housing neighborhoods is more typical for urban regions of the global south. Nonetheless, both the context of the region and the spatial character of at -risk areas influence the method of identification. The use of a GPS unit was crucial for collecting training sample s and the GIS and Spatial Analyst tools facilitated geographic and spatial examin ations. In fact, the overlain comparison of geographic features such as rivers with urban areas and
135 detailed aerial photographs would have been im possible without GIS tools and data. The use of the best available data resources including geo -rectified ae rial photographs various GPS-recorded field visits in a range of irregularly built areas population and domicile counts, interviews, and government plans served to complement one another in the discovery and measurement of informal urbanization condition s Using freely available tools, such as Google E arth satellite imagery would enhance efforts Although narrowing down the scope of examination to ur ban areas at risk of settlement is useful, a method should account for atypical informal urban areas through other ways such as through verification procedures. I n many highly urbanized areas of the global south, i dentifying the location of all the existing informal settlements is nearly impossible and time consuming given their rapid developm ent and elusiv e circumstances. T hus a balance should be made between attaining highly accurate results a nd consuming time Overall, urban planners and local government in any urban region experiencing rapid increases of informal urbanization should make it a priority to detect informal urban growth and work hard to iteratively advance these methods After this process begins, subsequent updates and modifications should be considerably less demanding.
136 CHAPTER 5 RECOMMENDATIONS AND CONCLUSION Recommendatio ns for a more Sustainable Future Surprising few studies set in the global south use concepts of sustainable development as a means of evaluation even though rapid urban growth merit s this examination today more than ever. T he long held definition of sustainabili ty is that actions today should not compromise the ability of future generations to meet t heir needs (UN ESA, 2007). This chapter narrows down the sustainable development discussion by focusing on the results of this study. Jenks and Burgess (2000) applie d the sustainability concept to evaluate development by examining the density and compactness of urban form ; the amount of mixed uses ; types of building layout and diversity; use of passive solar design ; support of sustainable transport; and inclusion of g reen spaces (Jenks an d Burgess, 2000). In most Latin American countries, mixed use development and diverse building layout is common and the passive solar design and inclusion of green spaces concepts are beyond the study scope Thus of these sustainable urban development fa ctors, informal urban growth at risk of d eveloping expansively ; disproportionately on environmentally sensitive land; and segregated by income and hazardous substandard living circumstances were used to formulate recommendations Thi s research, set in the bustling metropolis of Curitiba, revealed sever al advancements, useful principles, along with challenges related to informal urban growth. Campbells (1996) theory of sustainable planning suggests that these interests be recognized and then calls for the formulation of winwin solutions that complement s these i nterests Using only the applicable results of this study, the informal urbanization in the study area was evaluated.
137 In terms of achieving sustainable urban form, encouraging social inclusion and reduc ing the rate of environmental degradation, Curitiba has advanced and digressed. Since entering the 21st Century, Curitiba has promoted a more sustainable urban form with the reduced rate of new informal urbanization, increased r ate of consolidation and compact development, and decreased rate of informal land consumption in ecologically fragile areas at -risk of informal settlement. However, the extent that in formal urban development consume s ecologically fragile lands that are ha zardous for de velopment has not decreased overall In addition, the region of Curitiba appears to have become more segregated by income. Many of the affordable housing program areas are located next to informal settlements along the urban fringe and peri phery of Curitiba and are clustered in areas with cheaper land. Consequently, Curitibas informal urban areas are concentrated in certain landscapes As such, recommendations were developed based on the study results and theories of sustainable planning and urban development Sustainable Urban Form and Growth New locations for soci al initiative programs should be considered in areas in need of infill development. The redevelopment and adaptive reuse of vacant industrial facilities should be considered for housi ng and social interest purposes and incentives should be given to d evelopers whom build moderate and low income housing. When deciding where to establish so cial interest zones and regularize informal settlements, opportunities to promote soc ial integration and support low income housing expansion nearby all types of development s should be considered. This vacant land allotted for mixed income housing development should not be hazardous and ecologically fragile. The regularization program in t he study area helped increase the density ; this effect was more possible when infrastructure was installed (such as piped sewage facilities) and when additional housing units were added Early intervention in newly settled areas, that helps guide the hous ing and lot layout location outside of ecologically fragile areas will help actively prevent difficult relocations and environmental degradation. The development of moderate to high housing density in urban areas that are affordable to low income populations is seen as another basic factor for promoting social inclusion and sustai nable urban growth. However, the construction of extremely high housing
138 densities that are precariously built should not be encouraged (particularly those at risk of collapsi ng following a natural disaster or storm event ) Public policy should promote and incentivize the attainment of sustainable housing densities that can endure inclement we ather and natural disasters and have enough space for basic infrastructure. A minimu m lot size should be established (as is done in many of Curitibas housing programs). At least medium housing density should be encouraged in areas not encroaching into environmentally sensitive areas Perhaps, regular quality of life surveys could be co nducted in regularized areas to ensure that housing consolidation and increased density has not d egraded the quality of life or impaired the functionality of infrastructure. Social Inclusion The social implications of the prevailing urban pattern shou ld be evaluated for the degree of inclusion that low to moderateincome populations are offered as valid citizens deserving of basic sanitation, infrastructure, services ( such as medical and education) decent housing conditions, security, schools, and par ticipation in public decisions. The fact that Brazil has mandated the inclusion of the urban poor as integral citizen s deserving of infrastructure and social interest support is a step in a more inclusive sustainable direction. Ignoring the existence of informal settlement and failing to build a comprehensiv e social interest strategy will spur expansive unsustainable development. I nformal communities prefer to be located within walking distance from public transit routes; thus regularization and social interest zones should be established in areas with sufficient bus access. Curitibas dedication to providing affordable bus service that is conveniently located promotes social inclusion however perpetuating long commute times might detract from this pro gress The social interest zoning proved to be an effective tool at proactively guiding new development. However, the concentration of the urban poor in social interest zoning pockets might be creating dense islands of poverty. As such, areas undergoi ng consideration for social interest zoning designation and regularization should look for opportunities to promote social integration and mixed income neighborhoods In addition, other initiatives that pro actively increases the supply of affordable low income housing and rentals and reduces segregation tendencies should be considered. A housing voucher program could be instated that provides monetary assistance for low income residents to live in formal housing areas that are in good condition. Inclus ionary zoning tools that require a small percentage of new housing developments to include dwelling units that are affordable to moderate income levels could also be explored. Environmental Preservation and Restoration Curitibas focus on the provisi on of urbanized lots in various social interest programs helped proactively guide development that might have encroached environmentally sensitive areas in a cost effective way. However, different types of new comers might have different preferences. I n other cities, low income new comers might prefer renting an apartment over self constructing a home on a serviced lot. Identifying the preferences of new comers and using these findings could be used as a guide for programs.
139 A regional program that ed ucates informal settlement residents on the local ecology and hazards of living within certain types of landscapes in conjunction with programs that foster and incentivize environmental stewardship might be a us eful approach. Providing simple education material on what is a floodplain and the dangers with living in these areas might be useful.2 Adding tax credit value or granting monetary incentives for property owners to preserve the most ecologically fragile lands should be considered. Choosing to regularize in areas that are located near vacant land without hazardous or ecologically fragile conditions could be used as a tool to guide new growth. Though regularization alone will not signif icantly reduce informal settlement particularly in ecological h azardous lands The degree that all types of urban growth encroaches ecologically fragile and hazardous features such as hillsides, floodplains, riverbanks, wetlands and c ritical habitat areas should be comprehensively examined especially since informal u rbanization might emerge near this type of formal development The extent that regularization community upgrades (such as the restoration of riverbanks) improves t he water quality of watersheds, particularl y those supplying potable water, should be measured to identify opportunities to improve water quality. In addition, the use of innovative low impact development improvement s (such as inexpensive pervious pavers and bioswales ) could be integrated as a part of regularization Holistic Considerations Par tn erships between agencies working on informal settlement issues throughout the RMC appear to be limited. In all of the facets of sustainability, a need remains to engage in civil dialogue at regular intervals to more effectively deal with informal urbani zation and develop a strategy for promoting inclusive sustainable developm ent that is proven to work These interactions should be facil itated and should include a broad range of stakeholders such as urban planners, different agency representatives, socia l workers, community leaders, informal settlers, elected officials, and sc holars These periodic multi stakeholder events involving multiple jurisdictions could help in the development of a learning community, united in furthering sustainable urban development. Region wide efforts have begun with the development of a plan with sustainability objectives, referred to as the Integrated Development Plan for the RMC (GEP, 2006) and with Curitibas release of several reports on social interest, housing, and con servation area plans (MC, 2008 and MC et al., 2007) As mentioned above, the integration of these plans, the sharing of lessons learned, and evaluation of useful me chanisms encourages praxis. A regional perspective for such a plan that captures the urban rural transition area in its p urview could increase plan effectiveness given the tendency for decentralized and displaced informal settlement. 2 An example of this type of environmental stewardship program took place in the Zumbi Informal Settlement. T o ensure that the crucial riparian ecosystems were maintained and preserved, treeplanting community events were launched and the participati on of the residents was strongly promoted. Estimates indicate that approximately 4,0 00 trees were planted (Interview, 2008).
140 Public participation should be an integral part of sustainable development. The process of regularizing the Zum bi informal settlement included several community meetings and allowed for a couple of community driven decisions However, the extent that the public is a part of the decision making on how to resolve conflicts and develop more sustainably is limited in Curitiba More opportunities for public input and civil dialogue should be afforded. Developing a resource center ( akin to a little city hall) in informal settlement hubs might improve transparency and nurture civil dialogue between citizens and government. As observed during research field visits, several informal settlements in the Metropolitan Region of Curitiba ( RMC ) undergoing the beginning stages of regularization, gained a computer lab in the center of the community that also served as a government offic e The act of establishing a computer lab that is staffed with government agents whom are available to share new information and listen, track, and consider community input facilitates this needed dialogue. Other resources and interactions with s takeholders could help promote regular civil dialogue on ways to resolve issues and become more sustainable. Gaining funds for social interest programs and sustainable projects is challenging. Curitiba used public private partnerships and transfer of development right (TDR) tools to help gain private sponsorship for social interest programs. Other tools to incentivize the private development of low to moderate income housing (built sustainably) should be considered. In addition, some community improvem ents and training a part of the Zumbi community regularization was funded by corporate sp onsors. Providing a venue for private sponsorship in addition to TDR tools might increase funding revenues Using the planners triangle for sustainable development (Campbell, 1996) a policy balance should be made in lessening environmental degradation throu gh conservation of land and development restrictions along with the promotion of social interest programs that create viable affordable housing and rental options for low income populations. Quantifying the housing and land need of low income populations is crucial for keeping pace with social interest needs The needs for affordable housing could be measured through fluctuations of informal settlement urban gr owth, housing deficit calculations bas ed on census data (similar to the method used by Serra et al. in 2005), and by quantifying the number of families on the waiting list for public housing and serviced lots Another signal of dis equilibrium between hous ing and land supply and demand takes the form of rapid inflations in land and home prices The monitoring
141 of these signals could be integrated in a sustainable strategy and help with balanc ing public policy programs Overall this discussion is reflective of the many functions and purposes of land: social, environmental, and economic. The environmental function is crucial for many reasons such as the need for clean water and air; the so cial function of land is fundamental particularly for providing a pl ace to live; and the econo mic functi on of land fulfills the need for resources and economic development. The recent policy recognition of the social function of property in Brazil encourages planning to improve the wellbeing of the urban poor T hus polic ies should recognize these various land functions and formulate plans and initiatives to meet the diverse needs in a balanced, sustainable manner. Curitiba seems to have begun forging this balance as evidenced by the reduced rate of informal urbanization. Similar to what was done for this research, the implementation of programs should be monitored for reduced rates of informal urbanization and their rate of convergence in ecologically fragile and hazardous la ndscapes, and the extent that living condition s and quality of life improves in different neighborhoods The monitoring would benefit from a regional focus given the tendency for informal settlement displacement. With the use of facilitated stakeholder input the goals and program suggestions can be culled into categories such as short term and long term; urgent and not urgent; and by degree of importance. This culling might help focus the strategy and manage the implementation of a vision. Achieving inclusive sustainable development is as much a process as it is a plan. This process should embrace needed adaptations based on experiential learning and active reflection, particularly as conflicts and challenges ari se
142 Overall, these recommendations emphasize ways to formulate an inclusive sustainable urban development strategy with short -term and long-term objectives Such a strategy would more likely yield sustainable outcomes when it promotes cost effective solutions; prevents impacts to hazardous environmentally fragile landscapes; and improves water quality In addition, this strategy should support the social integration of low income populations as citizens deserving of equal access to affordable housing options with basic sanitation and decent living conditions needed social servi ces and sustainable transport options Conclusion This longitudinal case stud y of the Metropolitan Region of Curitiba (RMC) showed distinct spatial patterns in informal urbanization that are based on rational considerations and limitations posed by the urban system In the study area, informal settlement tended to locate in floodplains, conservation lands, and riv erbanks from 1990 to 2007. In addition, t he regional pattern of informal settlement growth decentralized as new settlements most commonly emerged in the urban fringe and perip hery of the main c ity and outside of the older more established secondary urban cente rs Largely, informa l settlement s ac cumulate d near other informal settlements, public housing and social initiative areas, and bus servi ce in the study area thus several socioeconomic indicators were recogn ized along with environmental based indicators The characteristics of the urban region influenced these patterns on many levels For example, t he urban region of Curitiba has many river s and floodplains and only a small portion of karst landscapes thus there was m ore informal settlement constructed alongside rivers and in floodplains Each city context differs in the types of land left vacant and deemed low in value and undevelopa ble In addition, informal settlers mad e
143 adjustments in where they occupy land according to the degr ee of tolerance that others bestow. Another crucial finding was in the reduction in the amount of new land invasions from 2002 to 2007. The imple mentati on of regularization and the government approach to affordable housing in greater Curitiba could have contributed to the reduced rate of informal urban expansion and population density increases T hese public policy influence s would have been felt more di rectly inside of Curitiba than in the surrounding urban areas given the early implementation of regularization and other social interest programs in the main urban area. Another change in the prevailing spatial trends from 2002 to 2007 is in the decline in the rate at which informal development occurred in ecologically fragile and hazardous lands This decline could be partially attributed to the combination of new ly implemented sustainable planning laws and social interest initiatives The sustainable planning areas (Unidades Territoriais de Planejamento or the Territory Units of Planning ) located outside of Curitiba in the urban periphery calls for informal settlement consolidation in designated social interest areas in exchange for the conservation of ecologically valuable land The allowance of informal urbanization in designated areas mig ht have helped to reduce informal urbanization taking place in conservation lands from 2002 to 2007. This study showed that the prevailing pattern of urban growth in informal settlements has changed since the implementation of a mor e comprehensive government strategy and t he se policy influence s should be given greater attention in future research.
144 Implemented jointly and effectively the complementary interface of social interest and conservation public policies can encourage sustainable urban development. Likewise, the interface of affordable housing program provisions and social interest zoning with regularization can serve to pro actively and reactively respond to the needs of low income populations. The inclusion of proactive responses to the low income housing needs, particularly with serviced lot provisions, supports urbanization outside of ecologically fragile and hazardous land. As long as informal sett lement prolongs in the RMC, regularization typically lessens the environmental impact and improves the housing conditions but the response is reactive in nature and fails to address the affordable housing and land supply issues adequately. In some cases regularize d housing might be surrounded by environmental hazards and this type of development could r equire drainage structures and modifications to the natural system. Another consequence of the government approach was the tendency for public housing locations to be built near informal settlement or for settlers to locate near public housing. Holistically, t his concentration could contribute to greater socio economic segregation. As such, the outcomes of Curitibas public policies should be explored fu rther to learn how to better promote sustainable urban development and develop ways to reduce income segregation without fracturing critical social networks In addition, further research on the policy implications outside of Curitiba, particularly related to density increase, would be useful. These findings are moderated by the uncertainties in whether all informal urban areas were captured through the identification method and by data limitations.
145 The implications of this res earch point to several research needs A need remains to develop an approach to independently measure the strength of public policy influence s of urban growth in a way that accounts for intervening variables contributing to urbanization and consolidation and density changes Some of t he intervening variables pointed out in this study that should b e examined further include: the contribution of urban market dynamics pertaining to the supply of serviced land, homes, and rentals; the role of population and migration shifts in sh ifting housing preferences ; the role of housing loans; and the influence of naturally evolving consolidation processes involved with informal settlement The se variables could be combined in a model to better understand the causes and effects of informal urban development patterns Additionally, to differentiate between whether informal urban growth is attr acted to a particular feature or whether the feature was built as a result of informal settlement, the year that the fe ature was developed and the date of initial informal settlement should be compared. Another research need is to explore ways to use the informal settlement data to assemble an effective disaster preparedness plan. For example, the data created through this research could be used to dev elop a strategy for mobilizing at -risk populations out of harms way and it could guide educational strategies that strive to promote home construction improvements to prevent the development of hazardous, unsanitary and unsustainable living conditions La rgely the research results showed several issues and socioenvironmental conflicts related to informal settlement in the Metropolitan Region of Curitiba and then explored the influences of key public policies Though any comparison to Curitiba should r ecognize its unique characteristics such as the regions lower amount of
146 informal housing compared to other Latin American urban regions and its activ e implementation of various planning initiatives. As shown in this research, t he sheer amount of informa l settlement is astounding and reflective of a world wide problem that should not be ignored. Urban regions that fail to address informal settlement and the low income housing dilemma, as described herein, could confront even greater more aggravating problems with poor water quality, infectious diseases, concentrated crime, income segregation, and un -sustainable development In the aftermath of a natural disaster in informal urban areas the devastation would disproportionately impact the most vulnerable sectors of the population if informal occupation in hazar dous areas prolongs The urban planning and growth management in an urban region experiencing informal settlement cannot be effectively implemented to achieve sustainable outcomes without programs t hat address low income housing and informal settlement issues alongside other economic transportation, and environmental as pects of planning Holistically, t he prevailing pattern of informal settlement largely translates into grave environmental impacts and segregated hazardous living conditions. A ll residents of a city should be afforded opportunities to secure decent living conditions, regardless of whe ther one is poor or rich. Yet in the absence of an effective government approach and viable housing o ptions low income new -comers tend to occupy low valued environmentally sensitive land If this land -use pattern perpetuates natural resources, par ticularly related to water, runs the risk of becoming polluted beyond repair. This bleak scenario shoul d be recognized in order to gain the momentum needed to support needed strategies, exchange lessons learned from city to city, and truly realize a more sustainable future.
147 In sum mary t his research showed positive advancements on the path towards susta inable development in greater Curitiba given the inc reased compact urban growth, reduced informal settlement in enviro nmentall y sensitive areas and the improvement in the lives of the urban poor who no longer live i n hazardous living conditions The gove rnment approach was more effective for managing urban growth when bot h reactive (regularization) and proactiv e (provision of urbanized lots and availability of home loans ) features were incorporated and in return, the implementation of such a balanced approach could help enforce conservation restrictions Although the strategy in greater Curitiba is multi -faceted, it failed to de -segregate low income populations and it may encounter problems with meeting future demands for low income housing Curitiba is a city proclaimed as a commendable example of urban planning Consequently, these findings and other research should continue to evaluate promising strategies and adaptively formulate guidelines and improvements on how to develop in inclusive sustainable way even in urban regions experiencing informal settlement.
148 APPENDIX A INTERVIEW AND COMMUNITY SURVEY DOCUMENTS
150 Research Survey on Informal Settlement Urbanization Patterns and their Drivers in the Metropolitan Region of Curitiba from 1991 to 2007 Thank you for participating in this survey. Your responses may enhance the understanding of the potentially powerful role that different public policies and other factors have on shaping the urbanization patterns of informal settlements (informal settlements are also referred to as favelas ) in the greater Metropolitan region of Curitiba, Brazil. Please answer the questions honestly and to the best of your ability. 1) How long have you lived here? 2) Where did you live before moving here? 3) What are t he advantages o f living here? What are the disadvantages? 4) Indicate your opinion on whether the following factors influence (or do not influence) the choice in a loc ation of an informal settlement or r espond with yes if the factor is important and no if the factor is not important .1 Factors extremely important very important fairly important somewhat important not so important makes no difference a. Close proximity to family relations b. Close proximity to existing employment c. Clo se proximity to employment centers & opportunities d. Close proximity to the central city e. Close proximity to a bus route f. Area that is less noticeable to Governmental Entity g. Area that is more noticeable to a Governmental Entity h. Area near Existing Squatters i. Area near social services (schools) j. Area near infrastructure (electricity, potable water or sewage services) k. Area outside of floodplains l. Government offers titles to prop erty m. Area near regularization n. Other: 1 Some respondents found the ranking of importance confusing. Thus the survey was simplified and respondents were asked whether these factors were important or were not important.
151 5) What are some barriers for living near central Curitiba or Colombo ? 6) In your opinion, what are the main reasons for the accelerated growth of informal settlements since 1990? 7) Do you any other comments about living here that you would like to share with me? Your participation is greatly appreciated. Another short survey is anticipated for this research. You may be contacted within a year If interested in results of any of these surveys, please contact Jennifer Canno n at firstname.lastname@example.org. Thank you! Appendix A. Community Survey Responses2 Question 1: How long have you lived here? (Please note that the Zumbi Settlement was thought to have experienced settlement beginning in the early 1990s.) o 0 to 5 years: 4 responses, 13.3 percent o 5 to 10 years: 6 responses, 20 percent o 10 to 15 years: 9 responses, 30 percent o Above 15 years: 11 or 37 percent o Average of 13.1 years Question 2: Where did you live before moving here? o From the interior o f the state of Paran: 7 responses, 23.3 percent o From the RMC outside of Curitiba and Zumbi: 6 responses or 20 percent o From Outside of the State: 5 responses, 16.67percent o From Curitiba: 4 responses 13.33 percent o L ived in Zumbi most of their lives: 2 responses, 6.67 percent o Unknown, left blank: 6 responses, 20 percent Question 3: Advantages and d isad vantages of living in an informal s ettlement Advantages Disadvantages Employment/work opp ortunities 18.18% No disadvantages 20.00% Good place overall 15.15% Violence/danger/crime 30.00% 2 In addition to the author, Lianne Guerra Jepson verified and contributed to the translation of these surveys.
152 Near big city/downtown and everything is close 12.12% Far from family 3.33% Sewage system, water, basic sanitation, good infrastructure 9.09% Many thi ngs left to do in Neighborhood (to improve the neighborhood) 3.33% People, social action, companionship, unity of residents 9.09% Need better infrastructure 3.33% Easy access to bus 9.09% Health 3.33% Peace, tranquility 6.06% School is far from ho using 3.33% Education/school 6.06% Prejudice of the favela 3.33% Far from floods 3.03% Lack of security 3.33% Good security 3.03% (27% had no responses) Many legal persons 3.03% Health center 3.03% Quality of life 3.03% Quest ion 4: Indicate your opinion on whether the following factors influence (or do not influence) the choice in a location of an informal settlement. Respond with yes if the factor is important and no if the factor is not important. 4A: Close proximity to family relations, yes = 1, no = 2 Frequency Percent Valid Percent Cumulative Percent Valid 1 22 73.3 73.3 73.3 2 8 26.7 26.7 100 Total 30 100 100 4B: Close proximity to existing employment, yes = 1, no = 2 Frequency Percent Valid Percent Cumulative Percent Valid 1 20 66.7 71.4 71.4 2 8 26.7 28.6 100 Total 28 93.3 100 Missing System 2 6.7 Total 30 100 4C: Close p roximity to employment opportunities, yes = 1, no = 2 Frequency Percent Valid Percent Cumul ative Percent Valid 1 20 66.7 69 69 2 9 30 31 100
153 Total 29 96.7 100 Missing System 1 3.3 Total 30 100 4D: Close proximity to the central city, yes = 1, no = 2 Frequency Percent Valid Percent Cumulative Percent Valid 1 13 43. 3 44.8 44.8 2 16 53.3 55.2 100 Total 29 96.7 100 Missing System 1 3.3 Total 30 100 4E: Close proximity to a bus route, yes = 1, no = 2 Frequency Percent Valid Percent Cumulative Percent Valid 1 27 90 93.1 93.1 2 2 6.7 6.9 100 Total 29 96.7 100 Missing System 1 3.3 Total 30 100 4F: Area that is less noticeable to governmental entity, yes = 1, no = 2 Frequency Percent Valid Percent Cumulative Percent Valid 1 7 23.3 26.9 26.9 2 19 63. 3 73.1 100 Total 26 86.7 100 Missing System 4 13.3 Total 30 100 4G: Area that is more noticeable to a governmental entity, yes = 1, no = 2 Frequency Percent Valid Percent Cumulative Percent Valid 1 22 73.3 78.6 78.6 2 6 20 21. 4 100 Total 28 93.3 100 Missing System 2 6.7 Total 30 100 4H: Area near existing irregular occupations, yes = 1, no = 2
154 Frequency Percent Valid Percent Cumulative Percent Valid 1 24 80 82.8 82.8 2 5 16.7 17.2 100 Total 29 96.7 100 Missing System 1 3.3 Total 30 100 4I: Area near social services (i.e. schools, daycares, health centers), yes = 1, no = 2 Frequency Percent Valid Percent Cumulative Percent Valid 1 29 96.7 96.7 96.7 2 1 3.3 3.3 100 T otal 30 100 100 4J: Area near infrastructure (such as electricity, potable water or sewage services), yes = 1, no = 2 Frequency Percent Valid Percent Cumulative Percent Valid 1 28 93.3 93.3 93.3 2 2 6.7 6.7 100 Total 30 100 100 4K: Area outside of floodplains, yes = 1, no = 2 Frequency Percent Valid Percent Cumulative Percent Valid 1 7 23.3 23.3 23.3 2 23 76.7 76.7 100 Total 30 100 100 4L: Government offers titles to property, yes = 1, no = 2 Frequency Percent Valid Percent Cumulative Percent Valid 1 25 83.3 89.3 89.3 2 3 10 10.7 100 Total 28 93.3 100 Missing System 2 6.7 Total 30 100 4M: Within or near area with regularization, yes = 1, no = 2 Frequency Percent Valid Perce nt Cumulative Percent
155 Valid 1 27 90 93.1 93.1 2 2 6.7 6.9 100 Total 29 96.7 100 Missing System 1 3.3 Total 30 100 Question 5. What are the obstacles/barriers for living near central Curitiba or Colombo? Response Category Number Percent of total No obstacles because they don't want to live there 15 62.50% Land is expensive 4 16.67% Must be a trained professional 2 8.33% More stores, shopping and recreation spaces for children that take up the space 1 4.17% Neighborhoods in the center are more dangerous 1 4.17% Bus delays 1 4.17% Total Responses 24 Question 6: In your opinion, what are the main reasons for the accelerated growth of informal settlements since 1990? Response Category Number of r esponses Percent of total Lack of work in the interior 4 12.50% To find work 6 18.75% Lack of opportunities 2 6.25% Better Quality of Life 2 6.25% Socio economic factors 1 3.13% Need for less expensive land 3 9.38% Lack of Affordable Housing opportunities 3 9.38% Lack of money, resources and financing 4 12.50% Violence 1 3.13% Conflict with land occupation 1 3.13% Location Near family 1 3.13% Government contributions 1 3.13% Near bus 1 3.13% Difficult to work in Curitiba 1 3.13% Live near Colombo 1 3.13%
156 Total (There are over 30 responses since some respondents offered more than one reason.) 32 100.00%
157 APPENDIX B SATELLITE IMAGERY DETAILS AND LAND USE LAND COVER CLASSIFICATION Landsat TM and ETM+ satellite images used for land use land cover c lassification Type Date Re solution Source Geometric Correction TM 9/12/1991 Spring TM: 30 X 30 Meter for all bands except the thermal band Radiometrically and geometrically corrected (INPE) 20 usable Ground Control Points, total RMS error of 13.3470 ETM+ 9/2/2002 Spri ng ETM+ has 15 X 15 Meter resolution for panchromatic Radiometrically and g eometrically corrected to 50 RMS error (UMGLCF) Used this image for the geometric correction TM 7/6/2007 Winter TM: 30 X 30 Meter for all bands except the thermal band Radiomet rica lly and geometrically corrected (INPE) 29 usable Ground Control Points, total RMS error of 9.889 A ccuracy results for the 2007 land use land cover classification Class Name Reference Total Classified Total Number Correct Pro ducers Accuracy U sers Accuracy Built/Urban 305 313 291 291/305 = 95.41% 291/313 = 92.97% Water 18 10 10 10/18 = 55.56% 10/10 = 100% Vegetation 118 125 109 109/118 = 92.37% 109/125 = 87.2% Bare Soil/Barren 22 15 10 10/22 = 45.45% 10/15 = 66.67% Totals 464 464 421 Overall classification a ccuracy = 90.73% Kappa (K^) s tatistics Overall Kappa Statistics = 0.8092 81% Class Kappa Percentage Built 0.7949 80% Water 1.0 100% Vegetation 0.8283 83% Bare Soil/Barren 0.6501 65% Training samples collected i n 2008 and 2009
158 Category Based on Geographic Data and Field Study Based on Geographic Data Total for Each C ategory Bare Soil 11 52 63 Vegetation 130 188 318 Irregular Land Division 1 16 17 Informal Urban 84 131 215 Regularization Areas 9 14 23 Formal Urban 218 240 458 Water 18 39 57 Total Training Samples 471 680 1151 All of the training samples were verified based on available high quality geographic data and geo rectified aerial photographs. T o ensure that pure training samples were retr ieved for each of thes e categories, the points were adjusted in ArcGIS and moved into the center of a patch of land cover, at least 90 square meters in size, in areas that were representative of individual categories Areas with a mixture of land covers were designated according to the more dominate land cover. Training sample collection focused on retrieving a broad representation of built landscapes especially in the informal urban areas Several areas were inaccessible out in the field (such as bare so il areas) and as a result aerial photographs (w ith a resolution greater than ten meters in accuracy) and other geospatial datasets were used as a basis for creating additional training sampl es.
159 LIST OF REFERENCES Abiko, Alex; Cardoso, Luiz Reynaldo de Azevedo; Rinaldelli, Ricardo; Haga, Heitor Cesar Riogi. (2005). Paper: Informal settlement, slums, and upgrading. A part of the Proceedings of the Third International Urban Research Symposium held in Brasilia. Land and Urban Policies for Poverty Re duction. The World Bank and IPEA Instituto de Pesquisa Econmica Aplicada. Abramo, Pedro. (2007). Paper: Informal land markets in favelas and residential mobility of the poor, a contribution to a research agenda. IPPUR -UFRJ. Rio de Janeiro, Brazil Arimah, Ben C. (2000). Housing-sector Performance in Global Perspective: A Cross City Investigation. Urban Studies. 37; 2551. Arnold, Jr., Chester and Gibbons, C. James. (1996). Impervious surface coverage. Journal of the American Planning Ass ociation. Spring 1996, Vol. 62, Issue 2, pp. 243 Berner, Erhard. (2001). Learning from informal markets: innovative approaches to land and housing provision. Development in Practice. Volume 11, Numbers 2 &3. pp. 292-307. Burgess, Rod. (1981). Ideol ogy and Urban Residential Theory in Latin America. In Geography and the Urban Environment edited by D. T. Herbert and R. J. Johnson. Chichester: John Wiley and Sons. Campbell, Scott. (1996). Green cities, growing cities, just cities?: Urban planning and the contradictions of Sustainable Development. Journal of the American Planning Association. Volume 62, Number 3. pp. 296-312. Cervero, Robert. (1998). The Transit Metropolis: A Global Inquiry. Island Press: Washington D.C. Co mpanhia de Habitao Popular de Curitiba (COHAB) (2007). Lecture: 42 Years of COHAB. A Coordenao da Regio Metropolitana de Curitiba (COMEC). (2008). http://www.comec.pr.gov.br/ Retrieved map image from website in 2008. Dosh, Paul and Lerager, Jam es. (2006). Surprising trends in land invasions in Metropolitan Lima and Quito. Latin Ameri can Perspectives. 33; 29; pp. 29-54. Durand -Lasserve, Alain; Fernandes, Edesio; Payne, Geoffrey; and Smolka, Martim. (2002). Secure tenure for the urban poor. CIVIS, Cities Alliance Issue 3. pp. 1 -8.
160 Ewing, R eid, Pendall, Rolf and Chen, D on (2002). Measuring sprawl and its impact: the character and consequences of metropolitan expansion. Smart Growth America, Washington, DC. http://www.smartgrowthamerica.com/sprawlindex/sprawlindex.html Accessed on February 27, 2009. Ferguson, Bruce. (2005). Paper: Informal settlement, slums, and upgrading. A part of the Proceedings of t he Third International Urban Research Symposium held in Brasilia. Land and Urban Policies for Poverty Reduction. The World Bank and IPEA Instituto de Pesquisa Econmica Aplicada. Ferguson, Bruce and Navarrete, Jesus. (2003). New approaches to progre ssive housing in Latin America: a key to habitat programs and policy. Habitat International. Volume 27, pp. 309-323. Fernandes, Edesio. (1997). Access to urban land and housing in Brazil : three degrees of illegality Lincoln Institute of Land Poli cy, Working Paper. Fernandes, Edesio. (2003). Paper: t he Legal Regularization of Favelas in Brazil. Lincoln Institute of Land Policy Conference Fernandes, Edesio. (2005). Paper: Principles, bases and challenges of the national programme to support sustainable urban land regularization in Brazil. A part of the Proceedings of the Third International Urban Research Symposium held in Brasilia. Land and Urban Policies for Poverty Reduction. The World Bank and IPEA Instituto de Pesquisa Econmic a Aplicada. Franko, Patrice. (2007). The Puzzle of Latin American Economic Development. Third Edition. Row m an and Littlefield Publishers, Inc. : Boulder. Flyvbjerg, Bent. (2006). Five Misunderstandings About Case-Study Research. Qualitative Inquiry. Volume 12, Number 2, April. Ford, Larry. (1996). A new and improved model of Latin American city structure. The Geographic Review. Volume 86 Number 3 Funda o Jo Pinheiro (FJP). (2001). Deficit habitacional no Brasil: Municipious selecionados e m icrorregies geogrficas 2nd edio. Centro de Estatistica e Informaes : Belo Horizonte. Gilbert, Alan. (1981). Pirates and invaders: l and acquisition in urban Colombia and Venezuela. World Development. Volume 9, Number 7, pp. 657 678. Gilbert, A lan. (1998). The Latin American City. The Latin American Bureau: UK. Gilbert, Alan. (1999). Home is for ever? Environment and Planning A. Volume 31, pp. 1073-1091.
161 Gilbert, Alan and Gugler, Josef. (1994). Cities, poverty, and development: Ur banization in the third world. Second Edition. Oxford University Press : Oxford Governo Do Estado do Paran (GEP). (2006). Plano de Desenvolvimento Integrado da Regio Metropolitana de Curitiba (PDI). ( v arious published maps and reports). Griffin, E rnst and Ford, Larry. (1980). A model of Latin American city structure. The Geographic Review. Volume 70, Number 4. Huchzermeyer, Marie. (2004). Unlawful occupation: Informal settlement and urban policy in South Africa and Brazil. Africa World Press, I nc.: Trenton, NJ. Instituto Brasileiro de Geografia e Estatstica (IBGE). (Reports on Municipality and Metropolitan regional growth from 1950, 1970, 1975, 1980 1985, 1991, 1996, 2000, 2002, 2007, and 2009). http://www.ibge.gov.br Accessed on multiple dates from July 2008 and March 2009. Instituto Nacional de Pesquisas Espaciais (INPE). (2008). Landsat TM Imagery originating from USGS and NASA. Instituto Paranaense de Desenvolvimento Econmico e Social (IPAR DES). (2007). (Reports on unemployment for 2007). Instituto de Pesquisa e Planejamento Urbano de Curitiba (IPPUC). (2007). Forty Year Review of Planning. (Report on planning history ). Instituto de Pesquisa e Planejamento Urbano de Curitiba (IPPUC). (2009). Various reports and maps r etrieved from website August, 2009. http://www.ippuc.org.br/informando/index_informando.htm Irazbal Clara. (2005). City Making and Urban Governance in the Americas: Curitiba and Portland. Ashgate: Burlington. Jenks, Mike and Burgess, Rod. (2000). Compact cities: s ustainable urban forms for developing countries. London: Spoon Press. Jensen, John R. (1982). Detecting Residential Land-Use D evelopment at the Urban Fringe Photogrammetric Engineering and Remote Sensing. Vol. 48, No. 4, April 1982, pp. 629643. Jensen, John R. (2005). Introductory digital image processing, a remote sensing perspective. Pearson, Prentice Hall, Upper Saddle River, NJ.
162 Jensen, John R. and Cowen, Dave C. (1999). Remote Sensing of Urban/Suburban Infrastructure and Socio-Economic Attributes. Photogrammetric Engineering & Remote Sensing Vol. 65, No. 5, May 1999, pp. 611 -622. Lejano, Raul P. (2008). Technology and institutions: a critical appraisal of GIS in the planning domain. Science Technology & Human Values. 33, pp. 653678. Lloyd-Sherlock, Peter. (1997). The recent appearance of favelas in So Paulo City: An old problem in a new setting. Bulletin of Latin American Research Vol. 16, No. (3), pp. 289305. Macedo, Joseli. (2000). Land Use Policies and Urbanization of Informal Settlements: Planning Initiatives for Environmental Protection Areas in Curitiba, Brazil. Dissertation, University of Florida. Macedo, Joseli. (2004). Curitiba profile. Cities. Volume 21, No. (6), pp. 537 -549. Macedo, Joseli (2008). Urban land policy and new land tenure paradigms: l egitimacy vs. legality in Brazilian cities. Science Direct, Land Use Policy Vol. 25, pp. 259-270. Maia, Maria Leonor. (1995). Land use regulations and rights to the city: squatter settlements in Recife, Brazil. Land Use Policy. Elsevier Science Ltd. Vol. 12, No. 2, pp. 177-180. Mangin, William. (1967). Latin American squ atter settlements: A problem and a solution. Latin American Research Review. Vol. 2, No. (3), pp. 65 98. Martinez, Javier, Mboup, Gora, Richard Sliuzas, and Stein, Alfred. (2008). Trends in urban and slum indicators across developing world cities, 1990 2003 Habitat International. Menegat, Rualdo. (2002). Participatory democracy and sustainable development: integrated urban environmental management in Porto Alegre, Brasil. Environment and Urbanization. 14; 181., pp 181 -206. Moller -Jensen, Lasse; Kofie, Richard Y.; Yankson, Paul W. K. (2005). Large area urban growth observations a hierarchical kernel approach based on image texture. Danish Journal of Geography. 105 (2); pp. 39-47. Montgomery, Mark R., et al. (2008). The urban transform ation of the developing world. Science. Vol. 319, pp. 761-764. M orais, Maria da Piedade; Cruz, Bruno de Oliveira; Oliveira, Carlos Wagner de Albuquerque. (2005). Paper: Residential segregation and social exclusion in Brazilian housing markets. A part o f the Proceedings of the Third International Urban Research Symposium held in Brasilia. Land and Urban Policies for Poverty Reduction. The World Bank and IPEA Instituto de Pesquisa Econmica Aplicada.
163 Municpio de Curitiba (MC). (2008). Plano Municipa l de Habitao e Habitao de Interesse Social: Sintese do Diagnostico. (Housing and Social Interest Housing Plan). March. Municpio de Curitiba, COHAB -Curitiba, and IPPUC (MC et al.). (2007). Plano Municipal de Regulariz ao Fundria em reas de Pres ervao Permanente. M inistrio do Trabalho e Emprego (MTE) and Instituto Paranaense de Desenvolvimento Econmico e Social ( IPARDES) (2007 ). Saldo do emprego com carteira assinada no Paran -2002January 2010. (Report from the IPARDES website. www.ipardes.gov.br Retrieved March 2010. OHare, Greg and Barke, Michael. (2003). The favelas of Rio de Janeiro: A temporal and spatial analysis. GeoJournal. Vol. 56, pp. 225240. O'Sullivan, A rthur (2009). Urban economics. Seventh Edition. New York: McGraw Hill. Park, Robert E. and Burgess, Ernest W. (1967). The City, Suggestions for Investigation of Human Behavior in the Urban Environment. The University of Chicago Press. Payne, Geoffrey. (2001). Lowerin g the ladder: regulatory frameworks for sustainable development. Development in Practice. pp. 308318. Peattie, Lisa and Aldrete Haas, Jose A. (1981). Marginal settlements in developing countries: research, advocacy of policy, and evolution of programs. Annual Review Sociology 7. pp. 157-175. Rabinovitch, Jonas. (1992). Curitiba: towards sustainable urban development. Environment and Urbanization. Volume 4, No. 2., pp. 62-73 Roy, Ananya. (2005) Urban informality, toward an epistemolog y of planning. Journal of the American Planning Association; Spring, Volume 71, Number 2, pp. 147 -156. Schwartz, Hugh. (2004). Urban Renewal, Municipal Revitalization: the Case of Curitiba, Brazil. Hugh Schwartz: Alexandria, VA. Serra, M.V.; Dowall, D avid E.; Motta, Diana; Donovan, Michael. (2004). Paper: Urban land markets and urban land development: An examination of three Brazilian Cities: Brasilia, Curitiba and Recife. Informal settlement, slums, and upgrading. A part of the Proceedings of th e Third International Urban Research Symposium held in Brasilia. Land and Urban Policies for Poverty Reduction. The World Bank and IPEA Instituto de Pesquisa Econmica Aplicada.
164 Serra, M. Vitor, William J. ODell, Joseli Macedo, Marc T. Smith, Maria P. Morais, Santiago Varella and Diep Nguyen. (2005). Affordable Housing Needs Assessment Methodology: The Adaptation of the Florida Model to Brazil. Texto para Discusso No.1083. IPEA Instituto de Pesquisa Econmica Aplicada. Schneider, Annemarie and Wood cock, Curtis E. (2008). Compact, dispersed, fragmented, extensive? A comparison of urban growth in twenty -five global cities using remotely sensed data, pattern metrics and census information. Urban Studies. Vol. 45, Number 659 pp. 659692. Sietchiping, Remy. (2005). Paper: Slum growth model: theory, method, implementation and evaluation. A part of the Proceedings of the Third International Urban Research Symposium held in Brasi lia. Land and Urban Policies for Poverty Reduction. The World Bank an d IPEA Instituto de Pesquisa Econmica Aplicada. Taniguchi, Cassio (excerpt in collection of case studies edited by Charlesworth, Esther). (2005). City Edge, Case Studies in Contemporary Urbanism. Architectural Press: Amsterdam. Turner, John C. (196 7). Barriers and channels for h ousing development in modernizing c ountries. Journal of the American Institute of Planners. Volume XXXIII, Number 3. pp. 167181. Turner, John C. (1968 ). Housing priorities, settlement patterns, and urban development i n modernizing countries. Journal of the American Institute of Planners. Volume XXXI V, Number 6 pp. 353363. Universidade Positivo (UP). (2009). Lecture: Regularizao Fundiria: velho problema, nova leitura. Multiple present ers. University of Ma rylands Global Land Cover Facility (UMGLCF) (2008). Landsat satellite imagery originating from USGS and NASA. United Nations Development Programme (UNDP). (2000). End poverty 2015, Millennium Development Goals. Accessed on January 2010 at http://www.un.org/millenniumgoals/environ.shtml United Nations D ivision of Sustainable Development (UNDSD). (1992). A D ivision of the Department of Economic and Social Affairs. Agenda 21. Accessed o n January 2010 at http://www.un.org/esa/dsd/agenda21/ United Nations Economic and Social Affairs (UNESA) ( 2007) Indicators of Sustainable Development: Guidelines and Methodologies October 2007, Third Edition. United Nations: New York.
165 United Nations Center for Human Settlements (UN HABITAT ) (1996) An Urbanizing World: Global Report on Human Settlements 1996, Oxford University Press. United Nat ions Human Settlements Program (UN HABITAT), ( 2003). Slums of the World: the face of urban poverty in the new millennium? Mon itoring the Millennium Development Goal, Target 11. Working Paper United Nations Habitat and Cities Alliance (UN HCA), Human Settlements Programme. (2006). Analytical Perspective of Pro poor Slum Upgrading Frameworks. http://www.unhabitat.org/ Nairobi, Kenya. Accessed February 2008. Urbanizao Curitba S/A (URBS). (2007). Lecture: Public Transportation System. Ward, Peter (1983). Land for housing the poor: how can planners contribute? In Land for Housing the Poor edited by S. Angel, R. W. Archer, S. Tanphiphat and E. A. Wegelin. Singapore: Select Books. Ward, Peter. (2003). Paper: Land Regularization in Latin America: Lessons in the Social Construction of Public Policy. Lincoln Institute of Land Policy Conference Paper. Yin, Robert K. (2003). Case Study Research Design and Methods. Third Edition. Sage Publications: Thousand Oaks, California.
166 BIOGRAPHICAL SKETCH Jennifer Cannon was born in Spokane, Washington and graduated from Mead High School in 1995 She moved to Bellingham, Washington to earn a Bachelor of Arts in Environmental Policy and Planning at the Huxley College of the Environment, at Western Washington Univ ersity, in 2000. Jennifer Cannon is an urban, environmental, and community planner that has expanded her scope of work from the Pacific Northwest to the Southeastern United St ates, to international planning topics She worked as a community planner in S eattle, Washington and in th is position, planned and implement ed the public outreach for a range of transportation improvement and environmental cleanup projects. She then worked as a Geographic Information System an alyst, environmental planner, and proje ct manager in Gainesville, Florida before returning to academia to c omplete a Master of Arts in urban and regional planning and pursue certification in Latin American s tudies at the University of Florida. As a part of an international housing research i ni tiative, Jennifer researched methods for quantifying housing deficits and informal housing in Brazil in a graduate student assistant position. Jennifer was a scholarship recipient for a study abroad course in Brazil and i n 2008 she was awarded a field r esearch grant from the Latin American Studies Department.