Dissertacao Mestrado Grohmann PDF

UNIVERSIDADE DE SO PAULO
INSTITUTO DE GEOCINCIAS
TCNICAS DE GEOPROCESSAMENTO
APLICADAS ANLISE MORFOMTRICA
Carlos Henrique Grohmann
Orientador: Prof. Dr. Mario da Costa Campos Neto
DISSERTAO DE MESTRADO
Programa de Ps-Graduao em Geoqumica e Geotectnica
So Paulo
2004
Todas as cincias exatas so dominadas pela idia da aproximao

Bertrand Russel (1872 - 1970)
ndice
Resumo..................................................................................................................... i
Abstract .................................................................................................................... ii
Agradecimentos........................................................................................................ iii
1 Introduo........................................................................................................... 1
1.0 Da organizao desta dissertao........................................................... 1
1.1 Apresentao ao tema ............................................................................. 1
1.2 A escolha da rea de estudo .................................................................... 2
1.3 Contexto Geolgico ................................................................................. 2
2 Materiais e Mtodos ........................................................................................... 7
2.1 Os programas utilizados.......................................................................... 7
2.2 O Modelo Digital de Terreno .................................................................. 7
3 Anlise morfomtrica em Sistemas de Informao Geogrfica:
aplicaes dos softwares livres GRASS e R..................................................... 9
4 Anlise de superfcies de tendncia de parmetros morfomtricos.................... 10
5 Discusso............................................................................................................ 11
6 Concluses ......................................................................................................... 16
7 Referncias Bibliogrficas.................................................................................. 17
Anexos
ndice de Figuras
Figura 1 - Localizao da rea de estudo ..............................................................................5

Figura 2 - Relevo sombreado da rea de estudo....................................................................5
Figura 3 Mapa geolgico simplificado...............................................................................6
Figura 4 Lineamentos interpretados e diagrama de roscea...............................................11
Figura 5 Perfis em varredura; localizao sobre imagem de relevo sombreado. ...............12
Figura 6 Mapas morfomtricos com estruturas inferidas sobrepostas................................13
Figura 7 Vista tri-dimensional da superfcie de base .........................................................14
Figura 8 Superfcies de tendncia e mapas de resduos .....................................................15
Lista de Anexos
Anexo 1 - Morphometric analysis in Geographic Information Systems: applications of
free softwares GRASS and R
Artigo submetido ao peridico internacional Computers & Geosciences
Anexo 2 - Trend-surfaces analysis of morphometric parameters

Anexo 3 - DXF_Lin
Listagem de programa de computador escrito em linguagem Delphi para clculo de orientao e
comprimento de lineamentos a partir de arquivo vetorial em formato DXF exportado pelo GRASS-GIS
Anexo 4 - Grad_Hidro
Listagem de programa de computador escrito em linguagem Delphi para clculo de gradiente
hidrulico, para ser usado em conjunto com GRASS-GIS
Anexo 5 - Serra do Caraa (Quadriltero Ferrfero, MG): parmetros morfomtricos e

consideraes tectnicas preliminares
Resumo apresentado ao III Workshop Cientfico de Ps-Graduao do IGc-USP (abril de 2003)
Anexo 6 - Serra do Caraa (Quadriltero Ferrfero, MG): morphometric parameters

and preliminary tectonic considerations
Resumo expandido apresentado ao IX Simpsio Nacional de Estudos Tectnicos / III
International Symposium on Tectonics (maio de 2003)
Anexo 7 - Anlise morfomtrica em Sistemas de Informao Geogrfica: aplicao dos

softwares livres GRASS e R
Resumo apresentado ao VIII Simpsio de Geologia do Sudeste (outubro de 2003)
Resumo
Neste trabalho, buscou-se o desenvolvimento e adequao de tcnicas de anlise
morfomtrica em Sistemas de Informaes Geogrficas. Como base de trabalho, foram
utilizados os programas livres GRASS-GIS e a linguagem estatstica R. Os parmetros
morfomtricos estudados foram hipsometria, declividade, orientao de vertentes, perfis em
varredura, densidade de lineamentos e de drenagem, rugosidade de relevo, isobases e gradiente
hidrulico.
A rea de estudo localiza-se na borda leste do Quadriltero Ferrfero (MG), e possui
unidades geomorfolgicas distintas: um relevo montanhoso a oeste (Serra do Caraa), separado
por uma escarpa de centenas de metros de terrenos caracterizados por dissecao fluvial, que
abrigam em sua poro central, um planalto sustentado por conglomerados ferruginosos
cenozicos.
O elemento principal da anlise morfomtrica o Modelo Numrico de Terreno (MDT),
que pode ser interpolado a partir de curvas de nvel ou de pontos cotados. H ainda a
possibilidade de utilizar MDTs produzidos pela agncia espacial americana (NASA),
disponveis gratuitamente via internet.
A metodologia proposta envolve principalmente operaes bsicas de SIG, como
converso entre formatos vetorial e matricial, operaes matemticas simples em mapas raster e
interpolao de valores pontuais em superfcies contnuas.
Anlise de superfcies de tendncia foi realizada para verificar a possibilidade de uso do
mtodo em estudos de morfotectnica. Foram ajustadas superfcies polinomiais de 1 a 6 grau,
e as significncias estatsticas de cada polinmio e do incremento do grau polinomial foram
verificadas com anlise de varincia.
Os mapas morfomtricos produzidos permitiram a identificao de estruturas
provavelmente relacionadas com a configurao atual da paisagem. Os mapas de resduos para
as superfcies de tendncia mais representativas possuem boa correlao com as estruturas
inferidas.
A metodologia proposta pode ser adaptada aos diversos pacotes SIG existentes no
mercado. O uso de programas livres e de cdigo aberto garante o acesso a todos, e sua crescente
popularizao abre novas perspectivas nesse campo.
ii
Abstract
This work aims on developing and adaptation of morphometric analysis in Geographic
Information Systems. Free softwares GRASS-GIS and R statistical language were used as main
applications.
Studied morphometric parameters were hypsometry, slope, aspect, swath profiles,
drainage and lineament density, surface roughness, isobase and hydraulic gradient.
The study area is located in the eastern border of Quadriltero Ferrfero (MG), and has
distinct geomorphological units: a mountainous relief to west (Serra do Caraa), separated by a
scarp with hundreds of meters of fall from fluvial dissected terrains, with a flat plateau in its
central portion, the Chapada de Canga.
The main element for morphometric analysis is the Digital Elevation Model (DEM),
which can be interpolated form contours or scattered points. There is also the possibility of
utilize NASA' s DEMs, freely available via internet.
The proposed methodology deals mainly with basic GIS operations, like conversion
between raster and vector, simple mathematic operations on raster maps and interpolation of
discrete values into continuous surfaces.
Trend-surface analysis was carried out o verify the possibility of use the method for
morphotectonic studies. Polynomial surfaces from 1st to 6th degree were adjusted; the statistical
significance of each surface and the significance of increasing polynomial degree were verified
with analysis of variance.
Produced morphometric maps allow identification of structures probably related with
present-day landscape configuration. Residuals maps for most representative trend-surfaces
keep good correlation with inferred structures.
The proposed methodology can be adapted to the variety of GIS packages available in
the market. The use of free and open-source softwares guarantees access to everyone, and its
increasing popularization opens new perspectives in this field.
iii
Agradecimentos
Se escrever os agradecimentos de um Trabalho de Formatura j no foi fcil, quando
se chega no Mestrado mais difcil ainda. Como sempre, h muitas pessoas que esto ligadas
realizao deste trabalho, direta ou indiretamente.
Em primeiro lugar, agradeo minha famlia, que sempre esteve ao meu lado nestes
anos. Especialmente a minha me, meu irmo e meu pai, obrigado mesmo.
Ao professor Mario da Costa Campos Neto, pela amizade, orientao e constante apoio
em todos os momentos.
Ao CNPq, pela concesso de bolsa de mestrado (proc. 130777/2002-1), e FAPESP,
pelo auxlio para trabalhos de campo (proc. 02/03131-4).
Aos professores Oswaldo Siga Jr, Claudio Riccomini, Jorge Kazuo Yamamoto, Marcelo
Rocha, Teodoro Isnard, Marcos Egydio Silva, Hans Schorscher, Paulo Boggiani, Ivo Karmann
e geloga Mnica Perrota.
A todo os usurios da lista de discusso eletrnica do GRASS-GIS, pela ajuda no
aprendizado dos programas e pelas valiosas discusses, em particular a Hamish Bowman, Glyn
Clements, Roger Bivand e Markus Neteler.
Aos funcionrios do IGc, sempre presentes quando necessrio, Jos Carlos, Wagner,
Tadeu Caggiano, Reynaldo Castellon, Ana Paula Cabanal, Magali Rizzo e Henrique.
A todos os amigos da ps-graduao e da graduao, da USP e de todas as outras
escolas de geologia do pas, pelos momentos vividos durante esses dois anos de trabalho:
Frederico Bambi, Srgio Aracy, Luciana Dolly, Leonardo Pareja, Neivaldo Castro, Anelise
Lima, Eduardo Kago, Janana Mosketo, Thiago Aji, Andr Futinha, Caroline Brulio, Paula
PI, Marcos Zequinha, Arthur Sarita, Gilberto Coma, Carolina Preta, Marcelo Cear, Cristiano
Chiessi, Lucas Falcon, Mrcia Gomes, Cesar Vinasco, Augustin Cardona, Werner Weber,
Hlcio Prazeres, Josiane Aline, Gelson Sapo, Letcia Polegar, Fabrcio Passivo, Pedro Skroto,
Carolina Urtiga, Fernando Lampio, Anderson Feitchera, Fernando Sagui, Ingo Lalas, Camila
Montana, Renato Lesma, Bruno Melado, Samir Samar, Fabrcio Krusty, Adriana Bisteka,
Mrcio Genrico, Fabrizzio Condorito, Charazed, Rafael Cofrinho, Jonas, Jorge Perdido,
Sydnei Schaberle, Daniel Klein, Alexandre Nunsei, Eduardo Deby, Paulo Tibrcio, Murilo Boi,
Sabrina Bifo e Maurcio, Lilian, Roberta, Eva Kaide, Itamar Laffond, Nickie Unonius, Murilo
Valle, Leandro Vida, Fernando 99, Alexandre Lobo, Rogrio Padial, Hermann, Augusto Auler,
Ezio Rubbioli, Leda Zogbi, Carolina Anson, Luciana Alt e demais amigos do GBPE; da UFRJ:
Elisa, Neo e Gizelli; Silas e todos os amigos da UFRRJ; Bam-Bam, Piu, Vagno, Flvia, Ane II,
Alice e demais da UFPR; Barba, Buda e etc da UERJ; Tripa e Popeye da UFMG, Risadinha e
demais da UFOP, Aninha da UFBA, Rodrigo Meireles da UFMT/UFPE, da UNICAMP: Daniel
Fictcio, Alexandre Baleia; de Los Andes 2004: Wagner Lobinho, Edu Liminha, Carlos Isopor,
Seu Creisson, Carmela, Zaka, Vivian, Adriano Rejeito, Vanessa, Joo, Marcio, Paulo, Noele,
Ana Lcia, Alexandre Caveira, Daniel, Ticiano, Giorgio Basilici e Silvio Peralta.
A todos os citados e a todos que eu esqueci, muito obrigado.
1 - Introduo
1.0 Da organizao desta dissertao
A apresentao de dissertaes de mestrado e teses de doutorado na forma de artigos
cientficos prtica recente nas geocincias brasileiras, e implica em algumas adaptaes
quanto forma com relao ao modelo tradicional.
O desenvolvimento principal deste trabalho encontra-se em dois artigos cientficos
submetidos a peridico internacional, e que por serem redigidos em lngua estrangeira no so
includos no corpo do texto, mas anexados ao fim deste.
O captulo inicial traz uma breve apresentao ao tema abordado, seguida dos motivos
que influenciaram na escolha da rea de estudo. O segundo captulo descreve os programas de
computador utilizados e conta com uma pequena reviso de um assunto considerado muito
importante para anlise morfomtrica, a modelagem digital de terreno. O terceiro e quarto
captulos referem-se aos artigos cientficos submetidos, e so apresentados apenas os resumos
em portugus. Os dois captulos seguintes contemplam as discusses sobre os dados produzidos
e as concluses finais do trabalho.
Anexos ao texto encontram-se os dois artigos j mencionados, as listagens de dois
programas de computador escritos durante o desenvolvimento das anlises, e trs trabalhos
resumidos, de aspectos referentes pesquisa, apresentados em eventos cientficos durante o
decorrer do projeto.
1.1 Apresentao ao tema

A confeco e interpretao de mapas morfomtricos uma importante ferramenta em
estudos ligados a neotectnica e geomorfologia, onde a resposta da paisagem natural frente
dinmica interna do planeta muitas vezes mascarada pela rpida ao dos agentes
intempricos. A presena de anomalias na rede de drenagem e de descontinuidades no padro de
relevo podem refletir movimentaes recentes do terreno (Deffontaines 1989,1991; Zuchiewicz
1991; Rodriguez 1993; Salvador & Riccomini 1995; Hiruma & Riccomini 1999; Hiruma et al
1999).
A metodologia empregada na confeco destes mapas depende das ferramentas
disponveis e do grau de conhecimento de cada pesquisador com relao aos programas de
computador utilizados. Assim, comum a utilizao de diversos programas para diferentes
etapas da anlise morfomtrica, tais como vetorizao de drenagens e curvas de nvel,
interpolao de superfcies tri-dimensionais, clculos de rea, anlise estatstica etc. V-se,
portanto, a necessidade de integrao dessas ferramentas, visando, alm de maior simplicidade

na obteno dos dados, tambm a diminuio do custo total envolvido.
Essa integrao possvel com uso de um Sistema de Informao Geogrfica (SIG),
que, resumidamente, uma ferramenta capaz de armazenamento, manipulao e anlise de
dados referenciados geograficamente.
O objetivo deste trabalho foi o desenvolvimento e adequao de tcnicas de anlise
morfomtrica em Sistemas de Informaes Geogrficas. A escolha dos programas utilizados foi
de grande importncia, uma vez que as tcnicas, para serem utilizadas amplamente, devem ser
baseadas em operaes bsicas de um SIG, evitando-se a utilizao de ferramentas ou extenses
especficas de determinados fabricantes.
Foi feita a escolha pela utilizao conjunta do GRASS-GIS (Geographic Resources
Analysis Support System) com a linguagem estatstica R, ambos programas livres e de cdigoaberto, disponveis via internet em http://grass.itc.it e http://cran.r-project.org, respectivamente.
1.2 A escolha da rea de estudo
A rea de estudo escolhida para este trabalho localiza-se na borda leste do Quadriltero
Ferrfero, MG, e compreende a rea da folha topogrfica 1:50.000 Catas Altas (SF-23-X-B-I-1)
(Figura 1).
A regio se mostra propcia ao desenvolvimento de tcnicas de anlise morfomtrica,
pois existem diversos trabalhos que evidenciam a ao de tectnica cenozica (SantAnna &
Schorscher 1995,1997; SantAnna et al. 1997; Lipski et al. 2001; Lipski 2002), e h um forte
contraste entre unidades geomorfolgicas: um relevo montanhoso a oeste de uma rea mais
dissecada, que abriga um planalto sustentado por conglomerados ferruginosos cenozicos em
sua poro central (Figura 2), separados por uma escarpa de centenas de metros.
1.3 Contexto geolgico
O sul do Crton do So Francisco (CSF) dominado por complexos domeados granitognissicos arqueanos (ca. 2,8-2,7 Ga), penecontemporneos ao greenstone belt Rio das Velhas
(Machado et al 1992), que exibem registros isotpicos de retrabalhamento de uma crosta silica
mais antiga (ca. 3,0-2,8 Ga; Teixeira et al 1998). Encontram-se engajados, mais a sul e a leste,
em um cinturo metamrfico paleoproterozico (cinturo Mineiro, Teixeira & Figueiredo
1991), definido enquanto um orgeno acrecionrio h ca.2,1 Ga e caracterizado pela presena
de um arco magmtico plutnico, clcio-alcalino do tipo TTG, associado a granitos
peraluminosos, estes com assinaturas isotpicas de retrabalhamento de crosta arqueana (Noce et
al 1999).
As unidades basais do Super Grupo Minas (Grupos Tamandu, Caraa, Itabira e

Piracicaba; Door 1969; Chemale et al 1994) representam depsitos plataformais de uma
margem continental passiva, com fcies distais a SE, dos quais as rochas carbonticas
Gandarela, formao superior do Grupo Itabira, so tidas h 2,4 Ga (Babinski et al 1995).
Rochas turbidticas, diamictticas e vulcanoclsticas do Grupo Sabar, derivadas de uma reafonte a E-SE (Renger et al 1995), repousam em discordncia angular sobre o Grupo Piracicaba.
A idade mxima da deposio Sabar encontra-se h 2,12 Ga (Machado et al 1996).
Uma estruturao em domos e bacias (dome-and-keel, Marshak et al 1992), tida como
extensional, admitida como resultado do colapso do orgeno aps a coliso entre a margem
passiva Minas e a placa/microplaca do arco magmtico Mineiro. Uma estruturao do tipo
foreland fold-thrust belt, vergente para noroeste, aparentemente no metamrfico, precederia o
colapso e controlaria a deposio tipo flysch do Grupo Sabar. O Grupo Itacolomi representaria
molasides sin-extencionais (Alkmim & Marshak 1998).
No entanto, a foliao metamrfica principal do Super Grupo Minas poderia registrar a
chegada, de oeste para leste, do cinturo de cisalhamento dctil por cavalgamento, das nappes
gondwnicas de orgeno Brasiliano (como em parte proposto por Corra Neto & Dayan
1998). Um front oeste para a deformao Brasiliana foi delineado por Alkmin & Marshak
(1998), coerente com o registro K-Ar exposto por Endo (1997).
A aglutinao Brasiliana domina e controla o cenrio tectnico (Fischel et al 1998). De
leste chega por cavalgamento, ou por transpresso lateral oblqua (Ebert & Hasui 1998), uma
lmina metassedimentar (Grupo Dom Silvrio) aparentemente extenso sul da Faixa Araua,
orgeno controlado por subduco e coliso, segundo Cunningham et al. (1996), PedrosaSoares et al (1998). Encontra-se no front dos Gnaisses Mantiqueira, que esto completamente
envolvidos e retrabalhados nesta deformao e metamorfismo (Rettinger et al 1997) e marca o
limite oriental dos terrenos granite-greestone e da margem passiva Minas. Brueckner et al.
(2000) consideram o Grupo Dom Silvrio como uma unidade metassedimentar e meta-gnea
ocenica, em parte derivada de uma fonte Paleoproterozica e metamorfisada durante a
deformao Brasiliana. Os gnaisses tidos como Mantiqueira (Figueiredo & Teixeira 1996),
paleoproterozicos e com assinatura isotpica tanto juvenil, quanto arqueana, ainda esto mal
definidos. Seus limites so incertos, por incerto ainda ser o conhecimento de sua assemblia
litolgica e do contraste com as litologias adjacentes. Seus limites e sua estruturao final
encontram-se, no entanto, intensamente controlados pelas deformaes neoproterozicas.
A Plataforma Brasileira foi afetada em toda sua extenso por deformaes tectnicas
cenozicas que aproveitaram linhas de fraqueza preferenciais herdadas de eras geolgicas
anteriores, resultando em uma compartimentao neotectnica de unidades delimitadas por
descontinuidades crustais resultantes da reativao, em regime transcorrente geral, dos
principais lineamentos precambrianos brasileiros, como o caso da Descontinuidade Crustal

Dois "Brasis" (DCDB), a partir da qual se organizou a hidrografia moderna e consequente
evoluo geomorfolgica cenozica (Hasuy 1990; Saadi 1993).
A poro sul do CSF marcada por uma zona de cisalhamento de direo N50W,
denominada de Descontinuidade Crustal do Alto Rio So Francisco (DCASF: Saadi 1991),
apresentando atividade recorrente desde o Pr-cambriano at o Quaternrio (Saadi 1993). Um
eixo de arqueamento N-S passando por Arcos e Bom Despacho (MG) provoca o reativamento, a
leste e oeste, de empurres associados a planos de descolamento (detachment) que estruturam a
cobertura Proterozica do Grupo Bambui, imprimindo uma morfologia de blocos basculados em
direes divergentes (Saadi 1993).
Segundo Varajo (1991), testemunhos de superfcies de eroso no Quadriltero
Ferrfero possuem estreitas relaes com os domnios litoestruturais. Reativaes, com
movimentao vertical, de falhas antigas seriam as responsveis pelos desnivelamentos
altimtricos atuais (King 1956; Barbosa 1980)
Evidncias de tectnica cenozica no Quadriltero Ferrfero so observadas em diversos
depsitos sedimentares, e indicam a ao de trs eventos deformacionais com campos de
tenses distintos. O primeiro evento extensivo, de orientao NNE-SSW e provavelmente
relacionado a estruturas em horsts e grabens orientados ESE-WNW. Um segundo evento, de
maior expresso, predominantemente compressivo NW-SE. O terceiro evento considerado
como o relaxamento das estruturas do evento anterior (Lipski et al 2001).
Figura 1 Localizao da rea de estudo, com relao ao Quadriltero Ferrfero. Relevo sombreado com
iluminante a N45, inclinao de 20
Figura 2 Relevo sombreado da rea de estudo. Iluminante a N45, inclinao de 30
660000
665000
670000
675000
680000
PPg
Legenda
Quaternrio
(Q)
7785000
Qa
Asb
Asb
Arv
Qa
Arv
Cenozico
Negeno (N)
Arv
Ql
7780000
Qa
Morro da
gua Quente
Asb
Meso
(MP)
PPg
PPm
Arqueano (A)
Pd
7770000
Nc - Fm. Chapada de Canga:

ortoconglomerados oligomticos
cimentados por xido-hidrxidos de ferro
Nr - Fm. Fonseca: Sedimentos arenosos,
argilo-arenosos e argilosos
Nc
PPm
MPei - Grupo Espinhao/Itacolomi

arenitos com estratificao cruzada,
conglomerados polimticos
PPy - Granitos tipo S
PPm - Supergrupo Minas: metapelitos,
metarenitos, itabiritos, carbonatos,
metavulcnicas, metaconglomerados
Arv
Arv - Supergrupo Rio das Velhas:

metavulcanossedimentares, metapelitos
Asb - Complexo Santa Brbara: gnaisses
tonaltico-trondhjemticos
Convenes geolgicas
Nf
Nc
Falha ou zona de cisalhamento

Zona de cavalgamento
Contato, contato aproximado
Fonseca
Asb
K
G
Nc
P
7765000
Paleo
(P)
Proterozico
Asb
7775000
PPg
PPm
PPm
Sta Rita Duro
Ql
Nc
Asb
MPei
PPm
Anticlinal
Sinclinal
Sinclinal invertido
Escala 1:175.000
Ql
NM NG
Arv
Bento Rodrigues
5 km
Projeo Universal Transversa de Mercator

Origem da quilometragem: Equador e Meridiano 45 W Gr.
acrescidas as constantes 10.000km e 500km, respectivamente
22
(2004)
7760000
Ql - Coberturas laterticas
Pd - Diques e corpos bsicos intrusivos
Catas
Altas
Qa
G
Pd
Qa - Coberturas aluvionares
(a declinao magntica cresce -5,5 a cada ano)
(adaptado de Dorr 1969, CPRM 1993, SantAnna 1994, Campos Neto et al 2004)
Figura 3 - Mapa geolgico simplificado da rea de estudo
2 Materiais e Mtodos
2.1 Os programas utilizados
O SIG utilizado foi o GRASS-GIS (Geographic Resources Analysis Support System)
verso 5.0.3, um projeto livre e de cdigo-aberto (U.S. Army CERL 1993; Neteler 1998;
Neteler & Mitasova 2002; GRASS Development Team 2002) gratuitamente distribudo via
internet em http://grass.itc.it, e que oferece um ambiente integrado de anlise matricial (raster)
e vetorial, processamento digital de imagens e criao de mapas e grficos.
A organizao do banco de dados se faz com locations e mapsets; a location
compreende toda a rea de trabalho, enquanto que o mapset a poro ativa e utilizada para
anlise, que pode ser do mesmo tamanho ou menor que a location; vrios mapsets podem ser
definidos para a mesma location (Neteler 1998).
Um conceito importante dentro do GRASS o de region, que define, dentro do
mapset, a rea de interesse e a resoluo espacial dos mapas raster. Tanto a resoluo espacial
quanto as coordenadas do retngulo envolvente da region podem ser facilmente alteradas com o
comando g.region sem a necessidade de reinicializao do sistema ou a criao de novos
projetos.
A anlise estatstica dos parmetros morfomtricos foi realizada com a linguagem
estatstica R (Ihaka & Gentleman 1996; Grunsky 2002; R Development Core Team 2003),
atravs de uma interface com o GRASS (Bivand 2000) que permite que mapas raster e de
pontos sejam tratados como variveis para anlise.
Existem diversos pacotes extras para a linguagem R, que adicionam ao pacote bsico
funcionalidades como mtodos de interpolao por krigagem, ajuste de superfcies polinomiais,
anlise de agrupamento etc. Tanto a base do programa quanto os pacotes extras, alm de extensa
documentao, podem ser obtidos via internet na CRAN (The Comprehensive R Archive
Network, http://cran.r-project.org).
2.2 O Modelo Digital de Terreno

O elemento principal na anlise morfomtrica em SIG o Modelo Digital de Terreno
(MDT), que pode ser construdo a partir da interpolao de pontos cotados ou de curvas de
nvel, estejam elas em formato raster ou vetor.
A representao de superfcies tri-dimensionais complexas em SIGs, pode ser feita de
acordo com duas estruturas principais: redes triangularizadas ou grades regulares.
As Redes Triangularizadas Irregulares (Triangulated Irregular Network TIN) uma

estrutura em que pontos cotados irregularmente distribudos so conectados por uma rede de
arestas que formam tringulos no-sobrepostos; os pontos so conectados geralmente de acordo
com a triangulao de Delaunay, que une os centros de polgonos Thiessen vizinhos (Thiessen
1911, Delaunay 1934 apud Kumler 1994; Yamamoto 1988).
As grades regulares so estruturas mais simples, normalmente representadas por
matrizes de valores de elevao, onde h a localizao (coordenadas) de um nico ponto e o
espaamento da grade e sua orientao determinam a localizao horizontal de todos os outros
pontos (Kumler 1994). Este tipo de estrutura tornou-se conhecida como DEM (Digital Elevation
Model), aps o servio geolgico dos Estados Unidos (USGS) adot-la como padro para
distribuio de dados de elevao.
Como alternativas ao processo de interpolao de dados para gerao de MDTs, hoje
pode-se optar por DEMs produzidos pela agncia espacial americana (NASA) por
interferometria de radar (SRTM - Shuttle Radar Topographic Mission), e disponibilizados
gratuitamente com 30m de resoluo espacial para a rea dos Estados Unidos e 90m de
resoluo para outros pases, no stio http://seamless.usgs.gov.
Outra opo, ainda de uso limitado devido ao alto custo, o da telemetria a laser (Lidar
Light detection and ranging ou ALSM Airborne Laser Swath Mapping), onde pulsos de
laser so emitidos de bases aeroportadas para determinao da distncia ao solo e a posio da
aeronave obtida por DGPS (Differential Global Positioning System) (Carter et al. 2001).
Experimentos com essa tecnologia mostram que pode-se produzir MDTs muito precisos mesmo
em reas densamente florestadas (Haugerud et al. 2003), com resoluo suficiente para
identificao de pequenas escarpas, ligadas atividade tectnica recente, que no haviam sido
reconhecidas em fotografias areas (Sherrod et al. 2004; McKean & Roering 2004).
3 Anlise morfomtrica em Sistemas de Informao Geogrfica: aplicaes dos

softwares livres GRASS e R.
Resumo do artigo Morphometric analysis in Geographic Information Systems:
applications of free softwares GRASS and R submetido ao peridico internacional Computers
& Geosciences (Anexo 1).
O desenvolvimento e a interpretao de mapas morfomtricos so ferramentas
importantes em estudos relacionados a neotectnica e geomorfologia; Sistemas de Informao
Geogrfica (SIG) permitem maior preciso e rapidez neste processo, mas a metodologia
empregada depende das ferramentas disponveis e do grau de conhecimento de cada pesquisador
sobre os programas utilizados.
Uma metodologia para integrao de SIG e estatstica em anlise morfomtrica
apresentada para os parmetros morfomtricos mais comumente utilizados hipsometria,
declividade, orientao de vertentes, perfis em varredura, densidade de lineamentos e de
drenagem, rugosidade de relevo, isobases e gradiente hidrulico.
O SIG utilizado foi o GRASS-GIS 5.0.3, um projeto livre e de cdigo-aberto que prov
um ambiente integrado de anlise raster e vetorial, processamento de imagens e criao de
mapas e grficos. A anlise estatstica foi realizada com a linguagem estatstica R, atravs de
uma interface com o GRASS que permite que mapas raster e de pontos sejam tratados como
variveis para anlise.
O elemento bsico para produo de mapas morfomtricos o modelo digital de terreno
(MDT), que pode ser interpolado a partir de pontos cotados ou curvas de nvel, tanto em
formato raster como vetorial; tambm possvel utilizar os MDTs produzidos pela NASA, com
30m de resoluo espacial para os Estados Unidos e 90m de resoluo para outros pases.
A metodologia proposta pode ser adaptada de acordo com as necessidades e ferramentas
disponveis. O uso de programas livres e de cdigo-aberto garante o acesso a todos, e sua
crescente popularizao abre novas perspectivas de desenvolvimento neste campo.
10
4 Anlise de superfcies de tendncia de parmetros morfomtricos.

Resumo do artigo Trend-surfaces analysis of morphometric parameters submetido ao
peridico internacional Computers & Geosciences (Anexo 2).
Anlise de superfcies de tendncia foi realizada em dados dos parmetros
morfomtricos gradiente hidrulico e isobase. A rea de estudo, localizada na borda leste do
Quadriltero Ferrfero, sudeste do Brasil, possui duas unidades geomorfolgicas principais, uma
caracterizada pela dissecao fluvial e outra de relevo montanhoso, com uma escarpa de
centenas de metros separando-as.
Os mapas morfomtricos foram produzidos com GRASS-GIS e a anlise estatstica foi
feita com o pacote spatial, da linguagem estatstica R. Anlise de Varincia (ANOVA) foi
efetuada para testar a significncia do incremento no grau da superfcie polinomial.
Os melhores resultados foram obtidos com superfcie de 2 ordem para gradiente
hidrulico e de 6 ordem para isobase. A forma e orientao dos contornos dos mapas de
resduos para as superfcies escolhidas foram comparadas com estruturas inferidas a partir de
anlise morfomtrica, e h uma boa correlao entre os dados.
11
5 Discusso
A interpretao de mapas morfomtricos visa a identificao de estruturas que possam
ter uma relao com a configurao atual da paisagem.
Os lineamentos interpretados apresentam orientaes principais segundo NE-SW, e
NNE-SSW em toda a rea, com orientaes secundrias em N-S, NNW-SSE e NW-SE (Figura
4). Na poro norte da rea, ocorre um predomnio de vertentes voltadas para NE, com valores
de declividade entre 0 e 10, enquanto que as escarpas voltadas para SW apresentam
declividade entre 10 e 20, em uma configurao de vales assimtricos alinhados NW-SE, com
vertente abrupta para SW, o que pode ser relacionado com a atitude geral da foliao
metamrfica nessa rea.
Figura 4 Lineamentos interpretados e diagrama de roscea.
Perfis em varredura (ou perfis projetados de Baulig 1926, apud Tricart & Cailleux
1957) permitem uma viso mais ampla do comportamento do relevo (Figura 5) e mostraram a
presena de nivelamentos altimtricos em 2050m, entre 1650-1700m, e um degrau no planalto
da Chapada de Canga entre 850-900m, ligado a uma falha normal de direo E-W com
abatimento do bloco sul. Esta falha pode ser reconhecida como uma anomalia na orientao das
12
curvas do mapa de isobase e sua presena j havia sido apontada anteriormente (Sant' Anna
1994, Sant' Anna et al 1997).
Figura 5 Perfis em varredura; localizao sobre imagem de relevo sombreado.
Com a anlise morfomtrica, diversas estruturas puderam ser inferidas na rea de

estudo, e esto representadas na Figura 5, sobrepostas aos mapas morfomtricos.
Os mapas de densidade de drenagem (Fig. 6A) e de lineamentos (Fig. 6B) mostram a
presena de reas com maior ou menor densidade separadas pelas estruturas inferidas; o
planalto da Chapada de Canga apresenta baixos valores em ambos os mapas.
Na poro sul da Serra do Caraa h uma densidade muito baixa de drenagens em
comparao com altos valores para lineamentos, o que pode ser visto como resultado da
instalao de um sistema de drenagem subterrnea responsvel pela formao de extensas
cavernas. Essas cavernas, como a Gruta do Centenrio, que abriga o maior desnvel conhecido
na Terra em quartzito (Dutra 1997), so condicionadas por falhas verticais ou subverticais, de
direo WNW-SSE (Dutra et al. 2002).
Os mapas de rugosidade de relevo, isobase e gradiente hidrulico apresentam, de modo
geral, comportamentos similares. No mapa de rugosidade de relevo (Fig. 6C) nota-se que a rea
central do mapa tem relevo mais arrasado, com valores mais baixos, enquanto que as Serras do
Caraa (a oeste) e do Pinho (a leste) possuem maior rugosidade. A presena de valores baixos
na poro central da Serra do Caraa deve-se a um nivelamento altimtrico em torno de 2050m.
13
Figura 6 Mapas morfomtricos com estruturas inferidas sobrepostas.

A - Densidade de drenagem; B Densidade de lineamentos; C Rugosidade de relevo;
D Isobases; E Gradiente hidrulico
14
As linhas de isobase delimitam superfcies erosionais, portanto as superfcies de base

esto relacionadas a estgios erosionais, ou manifestaes de eventos tectono-erosionais,
principalmente os mais recentes (Golts & Rosenthal 1993). Pode-se pensar em isobases como
uma viso simplificada do relevo atual, do qual foi retirado o rudo do entalhamento
causado pelas drenagens mais recentes, de primeira ordem. Nas figuras 6D e 7, pode-se notar
um relevo suave com drenagem para SE na poro centro-sul da rea, mais alta que a rea a
norte, e separada desta por um desnvel de cerca de 100m.
Figura 7 Vista tri-dimensional da superfcie de base, a partir de ESE.

Exagero vertical de 10 vezes. Iluminante a N135, inclinao de 45.
O mapa de gradiente hidrulico (Figura 6E) ressalta principalmente as escarpas da Serra

do Caraa; reas com baixos valores nas regies sul e sudeste tambm apresentam baixas
densidades de lineamentos. Dentre os modelos de regresso linear aplicados aos parmetros
morfomtricos obtidos, o que apresentou melhores resultados foi entre gradiente hidrulico e
rugosidade de relevo. Apesar de uma considervel disperso dos dados, h uma correlao
positiva entre estes.
A anlise de superfcies de tendncia uma tcnica matemtica usada para separar
dados mapeveis em duas componentes, a de natureza regional das flutuaes locais (Davis
1986, Landim 1998). A componente regional representada por uma superfcie polinomial,
enquanto que as flutuaes locais, ou resduos, so dados pela diferena aritmtica entre o
polinmio e os dados originais. Esta tcnica vem sendo utilizada em exploraes de petrleo no
15
reconhecimento de armadilhas estruturais, feies paleogoegrficas ou quebras no estilo

estrutural de formaes sucessivas (Merrian & Harbaug 1963; Merrian & Lippert 1966;
Sutterlin & Hastings 1986; Davis 1986).
A anlise de superfcies de tendncia dos parmetros morfomtricos gradiente
hidrulico e isobase foi realizada com o intuito de verificar o potencial deste mtodo no estudo
de feies morfotectnicas. Foram calculadas superfcies polinomiais de 1 a 6 grau, e anlise
de varincia (ANOVA) foi utilizada para verificar a significncia estatstica de cada superfcie e
a significncia do incremento no grau do polinmio.
Os melhores resultados obtidos correspondem superfcie de 2 ordem para os dados de
gradiente hidrulico e de 6 ordem para os de isobase.
A correlao entre os mapas de resduos e os mapas morfomtricos boa, e vrias
feies interpretadas como estruturas podem ser vistas em ambos produtos. A escarpa da Serra
do Caraa marcada por valores positivos nos resduos do gradiente hidrulico e valores
negativos nos resduos de isobase.
O planalto da Chapada de Canga, apesar de ser uma importante feio topogrfica, no
bem marcado nos mapas morfomtricos utilizados para anlise de superfcies de tendncia,
provavelmente pelo fato de que esses parmetros se baseiam nas relaes entre a topografia e a
rede de drenagem, e o planalto tem densidade de drenagem muito baixa.
Figura 8 Superfcies de tendncia e mapas de resduos para os

parmetros morfomtricos gradiente hidrulico (A, B) e isobase (C, D).
16
6 Concluses
A anlise morfomtrica uma ferramenta importante nos estudos de neotectnica e
geomorfologia; a integrao dos procedimentos necessrios para obteno dos mapas
morfomtricos em Sistemas de Informaes Geogrficas apresentada com uso do GRASS-GIS
e da linguagem estatstica R, ambos programas livres e de cdigo aberto.
A rea escolhida para o desenvolvimento da pesquisa se mostrou bastante favorvel,
pois existem diversos trabalhos que evidenciam a ao de tectnica cenozica e h um forte
contraste entre unidades geomorfolgicas: o relevo montanhoso da Serra do Caraa, separado
por uma escarpa de centenas de metros de uma rea marcada por dissecao fluvial, que abriga
um planalto sustentado por conglomerados ferruginosos cenozicos (Chapada de Canga) em sua
poro central.
A partir dos mapas produzidos, foi possvel inferir diversas estruturas provavelmente
ligadas configurao atual do relevo. Perfis em varredura mostraram a presena de
nivelamentos altimtricos em 2050m, entre 1650-1700m, e um degrau no planalto da Chapada
de Canga entre 850-900m, devido a uma falha normal de direo E-W com abatimento do bloco
sul. Na poro sul da Serra do Caraa, uma densidade muito baixa de drenagens em comparao
com alta densidade de lineamentos, deve estar relacionada a um sistema de drenagem
subterrnea, responsvel pela formao de extensas cavernas.
A anlise de superfcies de tendncia foi realizada para verificar o potencial deste
mtodo no estudo de feies morfotectnicas. Foram ajustadas superfcies de 1 a 6 grau aos
dados de gradiente hidrulico e isobase. Anlise de varincia foi utilizada para verificar a
significncia estatstica de cada superfcie e a significncia do incremento no grau do polinmio.
Os melhores resultados obtidos correspondem superfcie de 2 ordem para os dados de
gradiente hidrulico e de 6 ordem para os de isobase.
O planalto da Chapada de Canga, apesar de ser uma importante feio topogrfica, no
bem marcado nos mapas de gradiente hidrulico e de isobase, pelo fato de que esses
parmetros se baseiam nas relaes entre a topografia e a rede de drenagem, e o planalto tem
densidade de drenagem muito baixa.
A metodologia proposta pode ser adaptada aos diversos pacotes SIG existentes no
mercado; o uso de programas livres e de cdigo-aberto garante o acesso a todos, e sua crescente
popularizao abre novas perspectivas de desenvolvimento neste campo.
17
7 Referncias Bibliogrficas
Alkmin, F.F.; Marshak, S., 1998. Transamazonian Orogeny in the Southern So Francisco
Craton Region, Minas Gerais, Brazil: evidence for Paleoproterozoic collision and collapse in the
Quadriltero Ferrfero. Precambrian Research, 90, 29-58.
Babinski, M., Chemale Jr., F.; Van Schmus, W.R., 1995. The Pb/Pb age of the Minas
Supergroup carbonate rocks, Quadriltero Ferrfero, Brazil. Precambrian Research, 72, 235-245.
Barbosa, G.V., 1980. Superfcies de eroso no Quadriltero Ferrfero. Revista Brasileira de
Geocincias, 10:89-101.
Baulig, H., 1926. Sur une mthode danalyse altimtrique applique la Bretagne. Bulletin de
lAssociation de Geographie Francaise. 10:7-9
Bivand, R.S., 2000. Using the R statistical data analysis language on GRASS 5.0 GIS database
files. Computers & Geosciences, 26:1043-1052
Brueckner, H.K., Cunningham, D., Alkmim, F.F.; Marshak,S., 2000. Tectonic implications of
Precambrian Sm-Nd dates from the southern So Francisco craton and adjacent Araua and Ribeira belts,
Brazil. Precambrian Research, 99: 255-269.
Campos Neto, M.C.; Basei, M.A.S; Vlach, S.R.F.; Szab, G.A.J.; Vasconcelos, P., 2004.
Migrao de orgenos e superposio de orogneses: um esboo da colagem Brasiliana no sul do Crton
do So Francisco, SE-Brasil. Geologia-USP. no prelo.
Carter, W.E.; Shrestha, R.L.; Tuell, G.; Bloomquist, D.; Sartori, M., 2001. Airborne laser swath
mapping shines new light on Earths topography. EOS (Transactions, American Geophysical Union),
82:549-564.
Chemale, F., Rosire, C.A.; Endo, I., 1994. The tectonic evolution of the Quadriltero Ferrfero,
Minas Gerais, Brazil. Precambrian Research, 65, 25-54.
Companhia de Pesquisas de Recursos Minerais, 1993. Mariana : Folha SF.23-X-B-I- Estado de
Minas Gerais. Programa de Levantamentos Geolgicos Bsicos do Brasil. 183p.
Corra Neto, A.V.; Dayan, H., 1998. Deformation analysis on the central Rio das Velhas
greenstone belt, Quadriltero Ferrfero (MG). In: 14th International Conference on Basement Tectonics,
Ouro Preto, UFOP/EM, Abstracts:128-130.
Cunningham, W.D., Marshak, S.; Alkmim, F.F., 1996. Structural style of basin inversion at midcrustal levels: two transects in the internal zone of the Brasilianao Araua Belt, Minas Gerais, Brazil.
Precambrian Research, 77: 1-15.
Davis, J.C., 1986. Statistics and data analysis in geology (2nd ed.). John Wiley & Sons, New
York, 656p.
Deffontaines, B., 1989. Proposition of a morpho-neotctonic method; application in the Fougres
area, Oriental Britany, France. Bull. INQUA, N.C.12:48-52.
Deffontaines, B., 1991. Contribution of drainage network analysis to neotectonics: method and
application in France. Bull. INQUA, N.C.14:16-17.
18
Delaunay, B., 1934. Sur la sphre vide. Bulletin de lAcademie Des Sciences de lUnion des
Rpubliques Sovitiques Socialistes, Classe des Sciences Mathmatiques et Naturelles (Izvestiia
Akademiia Mauk SSSR, Otdelenie Matematischeskikh i Estestuennykh Nauk). Srie VII / 6:793-800.
Door, J.V.N.II, 1969. Physiographic, stratigraphic and structural developent of the Quadriltero
Ferrfero, Minas Gerais, Brazil. U.S. Geological Survey, Prof. Paper 1119, B1-B19.
Dutra, G., 1997. O maior desnvel do mundo em quartzito. O Carste. Grupo Bambu de
Pesquisas Espeleolgicas. 9:62-66.
Dutra, G.M.; Rubbioli, E.L.; Horta, L.S., 2002. Gruta do Centenrio, Pico do Inficionado (Serra
do Caraa), MG (a maior e mais profunda caverna quartztica do mundo). In: Schobbenhaus, C., Campos,
D.A., Queiroz, E.T., Winge, M., Berbert-Born, M.L. C., 2002. Stios Geolgicos e Paleontolgicos do
Brasil. Braslia: SIGEP. 554pp.
Ebert, H.D.; Hasui, Y., 1998. Transpressional tectonics and strain partitioning during oblique
collision between three plates in the Precambrian of southeast Brazil. In: Holdsworth, R.E., Strachan,
R.A.; Dewey, J.F. (eds.) Continental Transpressional and Transtensional Tectonics. Geological Society,
London, Special Publications, 135, 231-252.
Endo, I., 1997. Regimes tectnicos do Arqueano e Proterozico no interior da placa
sanfranciscana: Quadriltero Ferrfero e reas adjacentes, Minas Gerais. Tese de Doutoramento,
Instituto de Geocincias, Universidade de So Paulo, 243p.
Figueiredo, M.C.H.; Teixeira, W., 1996. The Mantiqueira metamorphic complex, eastern Minas
Gerais State: preliminary geochronological and geochemical results. Anais da Academia Brasileira de
Cincias, 68, 223-246.
Fischel, D.P., Pimentel, M.M., Fuck, R.A., Costa,.A.G; Rosire, C.A., 1998. Geology and SmNd isotopic data for the Mantiqueira and Juiz de Fora complex (Ribeira Belt) in the Abre CampoManhuau region, Minas Gerais, Brazil. In: 14th International Conference on Basement Tectonics, Ouro
Preto, UFOP/EM, Abstracts: 21-23.
Golts, S.; Rosenthal, E., 1993. A morphotectonic map of the northern Arava in Israel, derived
from isobase lines. Amsterdam: Geomorphology, 7:305-315.
GRASS Development Team, 2002. GRASS 5.0 Users Manual. ITC-irst, Trento, Italy.
http://grass.itc.it/gdp/html_grass5/
Grunsky, E.C., 2002. R: a data analysis and statistical programming environment an emerging
tool for the geosciences. Computers & Geosciences, 28:1219-1222.
Hasuy, Y., 1990. Neotectnica e aspectos fundamentais da tectnica ressurgente no Brasil. In:
Workshop Neotecto. Sedim. Cont. Cenoz. SE Brasil, 1, Belo Horizonte, SBG-MG. Boletim 11, Anais:131.
Haugerud, R.A.; Harding, D.J.; Johnson, S.Y.; Harless, J.; Weaver, C.S.; Sherrod, B.L., 2003.
High-resolution lidar topography of the Puget lowland, Washington A bonanza for the earth science.
GSA Today. 13:4-10.
Hiruma, S.T.; Riccomini, C., 1999. Anlise morfomtrica em neotectnica: o exemplo do
Planalto de Campos do Jordo, SP. Revista do Instituto Geolgico. 20:5-19.
19
Hiruma, S.T.; Riccomini, C.; Modenesi-Gauttieri, M.C., 1999. Neotectnica no Planalto de
Campos de Jordo, SP. Revista Brasileira de Geocincias. 31:375-384.
Ihaka, R.; Gentleman, R., 1996. R: A language for data analysis and graphics. Journal of
Computational and Graphical Statistics, 5:299-314.
King, L., 1956. Geomorfologia do Brasil Oriental. Rev. Bras. Geog., 18:147-266.
Kumler, M.P., 1994. An intensive comparison of Triangulated Irregular Networks (TINs) and
Digital Elevation Models (DEMs). Cartographica, 31(2), Monograph 45.
Landim, P.M.B,. 1998. Anlise estatstica de dados geolgicos. So Paulo, Fundao Editora da
UNESP, 226p.
Lipski, M., 2002. Tectonismo Cenozico no Quadriltero Ferrfero, Minas Gerais. Dissertao
de mestrado. Univ. Fed. Ouro Preto, Escola de Minas, Dep. Geol., 171p.
Lipski, M.; Endo, I.; Castro, P.T.A.; Trzazkos-Lipski, B., 2001. Estudo do campo de tenses
Cenozicas no Quadriltero Ferrfero, Minas Gerais. IN: VIII Simpsio Nacional de Estudos Tectnicos,
Recife, UFPE. p.331-333.
Machado, N., Noce, C.M., Ladeira, E.A.; Oliveira, O.A.B., 1992. U-Pb Geochronology of
Archaean magmatism and Proterozoic metamorphism in the Quadriltero Ferrfero, southern So
Francisco craton, Brazil. Geological Society of America Bulletin, 104, 1221-1227.
Machado, N., Schrank, A., Noce, C.M.; Gauthier, G., 1996. Ages of detrital zircon from
Archaean-Paleoproterozoic sequences: implications for Greenstone Belt setting and evolution of a
Transamazonian foreland basin in Quadriltero Ferrfero, southeast Brazil. Earth and Planetary Science
Letters, 141, 259-276.
Marshak, S., Alkmim, F.F.; Jord-Evangelista, H., 1992. Proterozoic crustal extension and the
generation of dome-and-keel structure in a Archaean granite-greenstone terrane. Nature, 357, 491-493.
McKean, J.; Roering, J., 2004. Objective landslide detection and surface morphology mapping
using high-resolution airborne laser altimetry. Geomorphology, 57:331 351
Merriam, D.F.; Harbaug, J.W., 1963. Computer helps map oil structures. Oil and Gas Journal,
61:158-163.
Merriam, D.F.; Lippert, R.H., 1966. Geologic model studies using trend-surfaces analysis.
Journal of Geology, 74:344-357.
Neteler, M., 1998. Introduction to GRASS GIS software. Hannover. Germany. disponvel em
http://www.geog.uni-hannover.de/grass/gdp/neteler/manual98.html
Neteler, M; Mitasova, H., 2002. Open Source GIS: A GRASS GIS Approach. Boston, Kluwer
Academic Publishers, 464 p.
Noce, C.M., Teixeira, W., Qumneur, J.J.G., Martins, V.T.S.; Bolzachini, E., 1999. Isotopic
signatures of Paleoproterozoic granitoids from the southern So Francisco craton and implications for the
evolution of the Transamazonian orogeny. Journal of South American Earth Sciences, 13: 225-239.
Pedrosa-Soares, A.C., Vidal, P. Leonardos, O.H.; Brito Neves, B.B., 1998. Neoproterozoic
oceanic remnants in eastern Brazil: futher evidence and refutation of an exclusively ensialic evolution for
the Araua-west Congo orogen. Geology, 26, 519-522.
20
R Development Core Team. 2003. R: A language and environment for statistical computing. R
Foundation for statistical computing. Vienna, Austria. ISBN: 3-900051-00-3. http://www.R-project.org.
Renger, F.E., Noce, C.M., Romano, A.W., Machado, N., 1995. Evoluo sedimentar so
Supergrupo Minas 500 Ma. De registro geolgico no Quadriltero Ferrfero, Minas Gerais , Brasil.
Geonomos, 2, 1-11.
Rettinger, R.; Oberhnli, R.; Drkop, C., 1998. Thermometamorphic caracterization of the
transition zone between the Archean So Francisco Craton and the Neoproterozoic mobile belt, Minas
Gerais, Brazil. Zbl. Geol.Palont. 1:721-738
Rodriguez, S.K., 1993. Neotectnica e sedimentao quaternria na regio de Volta Grande
do Rio Xingu, Altamira, Par. Instituto de Geocincias, Universidade de So Paulo, So Paulo,
Dissertao de Mestrado, 160p.
Saadi, A., 1991. Ensaio sobre a morfotectnica de Minas Gerais. Belo Horizonte. (Tese para
admisso a cargo de Professor Titular, IGC/UFMG).
Saadi, A., 1993. Neotectnica e tectnica recorrente na poro sul do Crton do So Francisco.
In: Simp. Crton So Francisco, 2, Salvador. Anais...Salvador - SBG/BA. p.230-232.
Salvador, E.D.; Riccomini, C., 1995. Neotectnica de regio do alto estrutural de Queluz (SP-RJ,
Brasil). Revista Brasileira de Geocincias. 25:151-164.
SantAnna, L.G., 1994. Mineralogia das argilas e evoluo geolgica da Bacia de Fonseca,
Minas Gerais. Instituto de Geocincias, Universidade de So Paulo, So Paulo, Dissertao de Mestrado,
151p.
Sant'Anna, L.G.; Schorscher, H.D., 1995. Leques aluviais cenozicos do flanco leste da Serra do
Caraa, MG: a Formao Chapada de Canga. Anais da Academia Brasileira de Cincias. 67:519.
Sant'Anna, L.G.; Schorscher, J.H.D., 1997. Estratigrafia e mineralogia dos depsitos cenozicos
da regio da Bacia de Fonseca, Estado de Minas Gerais, Brasil. Anais da Academia Brasileira de
Cincias. 69:211 - 226.
Sant'Anna, L.G.; Schorscher, J.H.D.; Riccomini, C., 1997. Cenozoic tectonics of the Fonseca
Basin region, eastern Quadriltero Ferrfero, MG, Brazil. Journal of South American Earth Sciences.
10:275-284.
Sherrod, B.L.; Brocher, T.M.; Weaver, C.S.; Bucknan, R.C.; Blakely, R.J.; Kelsey, H.M.;
Nelson, A.R.; Haugerud, R.A., 2004. Holocene fault scarps near Tacoma, Washimgton, USA. Geology.
32:9-12.
Sutterlin, P.G.; Hastings, J.P., 1986. Trend-surface analysis revisited a case history. Computers
& Geosciences, 12:537-562.
Teixeira, W., Cordani, U.G., Nutman, A.P.; Sato, K., 1998. Polyphase Archaean evolution in the
Campo Belo metamorphic complex, southern So Francisco craton, Brazil: SHIRIMP U-Pb zircon
evidence. Journal of South American Earth Sciences, 11:279-289.
Teixeira, W.; Figueiredo. M.C.H., 1991. An outline of Early Proterozoic crustal evolution in the
So Francisco craton, Brazil: a review. Precambrian Research, 53, 1-22.
Thiessen, A.H., 1911. Precipitation averages for large areas. Monthly Weather Review, 39:10821084.
21
Tricart, J.; Cailleux, A., 1957. Cours de Geomorphologie I: Geomorphologie Structurale. C.D.U.
Paris. 252p.
U.S. Army CERL, 1993. GRASS 4.1 Reference Manual. U.S. Army Corps of Engineers,
Construction Engineering Research Laboratories, Champaign, Illinois, 1-425.
Varajo, C.A.C., 1991. A questo da correlao das superfcies de eroso do Quadriltero
Ferrfero, Minas Gerais. Revista Brasileira de Geocincias, 21:138-145.
Yamamoto, J.K., 1988. Representaes grficas espaciais em geocincias auxiliadas por
computador. Revista Brasileira de Geocincias. 18:3-26.
Zuchiewciz, W., 1991. On different approaches to neotectonics: a Polish Carpathians example.
Episodes, 14:116-124.
ANEXO 1
Morphometric analysis in Geographic Information Systems:
applications of free softwares GRASS and R
Morphometric analysis in Geographic Information Systems:

applications of free softwares GRASS and R
Ps-Graduao - Instituto de Geocincias - Universidade de So Paulo, Brasil
Rua do Lago, 562. 05508-800. So Paulo, SP
Phone: +55-11-30914216; fax: +55-11-30914258.
E-mail address: guano@usp.br, carlos_grohmann@yahoo.com.br
Abstract
Development and interpretation of morphometric maps are important tools in studies related to neotectonics and
geomorphology, and Geographic Information Systems (GIS) allows speed and precision to this process, but the applied
methodology will depend on available tools and degree of knowledge of each researcher about engaged softwares. A
methodology to integrate GIS and statistics in morphometric analysis is presented for the most usual mophometric
parameters: hypsometry, slope, aspect, swath profiles, lineaments and drainage density, surface roughness, isobase and
hydraulic gradient. The GIS used was the Geographic Resources Analysis Support System (GRASS 5.0.3), an opensource project which ofers an integrated environment for raster and vector analysis, image processing and maps/graphics criation. Statistical analysis of parameters can be carried out on R, a system for statistical computation and graphics,
through an interface with GRASS that allows raster maps and points files to be treated as variables for analysis. The
basic element for deriving morphometric maps is the digital elevation model (DEM). It can be interpolated from scattered points or contours, either in raster or vector format; it is also possible to use DEMs from NASAs Shuttle Radar
Topographic Mission, with 30m of ground resolution for the USA and 90m for other countries. The proposed methodology can be adapted according to necessities and available tools. The use of free and open-source tools guarantees
access to everyone, and its increasing popularization opens new development perspectives in this field.
Keywords: Morphometric analysis; Neotectonics; Geographic Information Systems; Open source software; GRASSGIS.
1. Introduction
Development and interpretation of morphometric maps are important tools in studies related to
neotectonics and geomorphology, where the answers of natural landscapes to planets interior dynamics is
often masked by fast action of weathering, and the presence of drainage network anomalies and relief
pattern descontiniuties may be related with recent terrain movements (Zuchiewicz, 1991; Rodriguez, 1993;
Salvador and Riccomini, 1995; Hiruma and Riccomini, 1999; Hiruma et al, 1999).
Geographic Information Systems (GIS) allows this process to be fast and precise (Hiruma and
Riccomini, 1999), but the applied methodology depends on available tools and degree of knowledge of
each researcher about engaged softwares (Grohmann, 2003).
This work aims an avaliation of necessary proceedings to development and correlation of
morphometric maps with integrated utilization of GRASS-GIS (U.S. Army CERL, 1993) and R statistical
language (Ihaka and Gentleman, 1996), both distributed as free software under the GNU Public License.
Examples will be presented from the Serra do Caraa region, eastern border of Quadriltero Ferrfero,
southeastern Brazil.
44 00'
43030'
43000'
Nova Era
Caet
Caet dome
Sabar
00
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
sy
nc
lin
Raposo
44o30'
la
Nova Lima
Belo Horizonte
dome
80
Serra
an
da
re
Ibirit
urral
do C
77
d
Moe
Brumadinho
JOO MONLEVADE
BELO HORIZONTE
78
2
0
42 30'
0
20 00'
4
Alvinpolis
a 5
a sy
nclin
60
Rio Casca
77
ITABIRITO
60
Bonfim dome
Bao dome
Acaiaca
MARIANA
10
Ouro Branco syncline
80
PONTE NOVA
40
OURO PRETO
Belo Vale
77
CONGONHAS
00
Ouro Branco
20030'
10
Santa Rita dome
40 km
77
20
6
20
40
60
80
00
20
40
60
Figure 1 Simplified geological map of southeast So Francisco craton (modified from Campos Neto et al, 2004). 1-Quaternary
Chapada de Canga Fm. (itabiritic conglomerate); 2-Miocene Fonseca Fm. (lacustrine deposits); 3-Itacolomi/Espinhao Gr. (coarse and crossstratified sandstone, polymitic conglomerate); 4-Minas Supergroup (basal Moeda and Batatal Fms sandstones and metapelites -, Cau and
Gandarela Fms banded-iron formations and carbonates- and uppers Piracicaba and Sabar Grs metapelites, sandstones, metavolcaniclastic
rocks, conglomerates); 5-Rio das Velhas Supergroup (Archaean -2.7Ga- basalt, komatiite and ryolitic lavas intercalated with sedimentary
units); 6-SE So Francisco craton gneiss (gneiss/migmatite complex, calc-alkaline granitoid plutons and Espinhao-related syenite and alkaline-like granite-gneiss); 7-Dom Silvrio Gr. (metapelites and quartzites with calcsilicated rocks and manganese-rich metasediments); 8Mantiqueira gneisses (amphibole-biotite orthogneisses and migmatites with titanite-rich synorogenic metabasic-intermediate rocks. Acaiaca
enderbite and kinzigite and Santo Antonio do gama amphibolite); 9-Juiz de Fora granulites (orthogranulites from a charnockitic suite, kinzigites
and migmatites); 10-Study area.
2. Geological and geomorphological context

The Quadriltero Ferrfero region, south So Francisco Craton, is characterized by Arquean granite-gneiss domed complexes coeval to Rio das Velhas greenstone belt (Machado et al, 1992), engaged to
south and east in a Paleoproterozoic metamorphic belt (Mineiro belt, Teixeira and Figueiredo, 1991) defined as an acrescionary orogen of ca. 2.1 Ga (Figure 1). A dome-and-kell structure (Marshak et al, 1992),
extensional, is admitted as result of orogen collapse after the collision of Minas passive margin and plate/
microplate of Mineiro magmatic arc.
The study area presents two geomorphological units: the Quadriltero Ferrfero and the Minas
Gerais center-south and east Highlands. The first unit shows average altitudes of 1400-1600m, with maximum in the Serra do Caraa, at 2064m. The second unit is characterized by fluvial dissection. The scarp
that limits this units has hundreds of meters of fall, and leads to believe not only erosional processes, but
also post-cretacic tectonic movements contributed in the morphological evolution.
According to Varajo (1991), remains of planation surfaces in the Quadriltero Ferrfero have close
relations with lithostructural domains. Reactivations with vertical displacement of ancient faults would be
responsible for present day altimetric differences (King, 1956; Barbosa, 1980).
Evidences of Cenozoic tectonics in the Quadriltero Ferrfero are observed in several sedimentary
deposits and indicate three deformational events with distinct tension fields. The first event, extensive,
oriented NNE-SSW and probably related with horsts and grabens oriented ESE-WNW. A second and more
expressive event, mainly compressive NW-SE. The third event is considered to be the relaxing of previous
event structures. (Lipski et al, 2001).
In the study area, the Quaternary Chapada de Canga Formation consists of a flat plateau formed by
a succession of continental itabiritic ironstone pebble conglomerates and directly overlies the Eocene Fonseca
Formation and the Precambrian basement (SantAnna and Schorscher, 1995). Cenozoic deposits are cut by
NE and NW brittle faults and joints, related to reactivation of pre-existing structures in Precambrian basement, process that strongly influenced the development of present day landscape morphology and drainage
network (SantAnna et al, 1997).
3. Methods
The GIS used was the Geographic Resources Analysis Support System (GRASS 5.0.3), an opensource project (U.S. Army CERL, 1993; Neteler, 1998; Neteler and Mitasova, 2002; GRASS Development
Team, 2002), freely available through the internet at http://grass.itc.it, running on a Linux machine. GRASS
ofers an integrated environment for raster and vector analysis, image processing and maps/graphics criation.
The database organization is related to locations and mapsets; the location involves all the
work area, while the mapset is the portion active and used for analysis, it can be smaller or have the
locations same size; several mapsets can be defined for the same location (Neteler, 1998).
The region is an important concept in GRASS and defines, inside the mapset, the area of interest
and the spatial resolution for raster maps. The coordinates of the regions enveloping rectangle can be
easily changed with g.region, a command used in several steps of morphometric analysis. In this paper, we
worked on a default region with 50m of spatial resolution, and created others with resolutions of 1m, 25m,
500m and 1000m.
Statistical analysis of the evaluated parameters can be carried out on R, a system for statistical
computation and graphics (Ihaka and Gentleman, 1996; Grunsky, 2002; R Development Core Team, 2003),
through an interface with GRASS (Bivand, 2000) that allows raster maps and sites (points) files to be
treated as variables for analysis.
There are a number of extensions to R, which adds to the base system alternative interpolation and
analyze methods, such as kriging or akima splines. The R core package and extensions, as well as related
documentation, can be obtained from CRAN (The Comprehensive R Archive Network,
http://cran.r-project.org).
The concept of convex hull of interpolation is still of little use in the geosciences. According to
Eddy (1977), the convex hull of a points (sites) dataset is the minimum area convex polygon which contains all the data, and represents the limits of spatial validity for the interpolating function. The s.hull
command provides the convex hull for a sites file; this polygon can be converted to raster and used as a
mask, thus limiting the interpolation.
Morphometric indices evaluated were: hypsometry, slope, aspect, swath profiles (Baulig, 1296 cited
in Tricart and Cailleux, 1958) lineaments and drainage density (Horton, 1945), surface roughness (Hobson,
1972; Day, 1979), isobase (Filosopov, 1960 cited in Jain, 1980, Golts and Rosenthal, 1993) and hydraulic
gradient (Rodriguez, 1993).
The basic element for deriving morphometric maps is the digital elevation model (DEM). It can be
interpolated from scattered points or contours, either in raster or vector format.
Sites interpolation can be done by inverse distance weighting with s.surf.idw, or by regularized splines with tension (Mitasova and Mitas, 1993; Mitasova and Rofierka, 1993), with s.surf.rst,
command that allows the empirical adjust of mathematical parameters for tension and smoothness of lines.
Vector contours are interpolated in the same way, with v.surf.rst. Interpolation from rasterized contours determines elevations using procedures similar to manual methods, with r.surf.contour.
Alternatively, one might use DEMs from NASAs Shuttle Radar Topographic Mission, with 30m of
ground resolution for the USA and 90m for other countries, freely available at http://seamless.usgs.gov.
Drainage network must be in vector format and two additional layers must be prepared, one with 2nd
and 3rd-order streams for extraction of isobases and the second with 2nd-order streams from its heads, for
calculus of hydraulic gradient. All needed adjusts and classifications are performed on the vectorial edition
module v.digit; v.extract command is then applied to create a new vector layer with the selected
drainage orders only.
3.1. Lineament Analysis

For interpretation of morphostructural lineaments, shaded relief maps are generated with the script
shade.rel.sh and tracing is made in v.digit. This approach has advantages over the use of satellite
imagery, since its possible to set up the position (azimuth and inclination) of scene illumination, thus
emphasize the several existing lineament orientations, due the enhancement of directions perpendicular to
lighting, in spite of parallels ones (Liu, 1987; Riccomini and Crsta, 1988; Grohmann and Campos Neto,
2002).
The properties of orientation and length for each lineament are calculated with DXF_Lin, a software written by the author in Delphi language (Kylix environment) which reads a DXF file exported from
GRASS and writes a text file that can be used as input in structural geology softwares.
In this work four shaded relief maps were used, with lighting 20 above the horizon at N90 N45,
N00and N315. Figure (2) shows interpreted lineaments and rose diagrams for frequency and length, at
10 interval.
Frequency
10%
Length
Figure 2 - Interpreted lineaments over shaded relief image and rose diagram.
10%
3.2. Hypsometry, slope and aspect

The hypsometric map is quickly obtained from reclassification of DEM with r.reclass; colors
are adjusted with r.colors and contours are extracted in vector format with r.contour. DEM 3-D
visualization (Figure 3) can be done with d.3d or in NVIZ, where one can interactively adjust positioning
of view, lighting and several others parameters.
Slope (Figure 4A) and aspect (Figure 4B) are calculated with r.slope.aspect; slope can be
expressed in degrees or percentage and aspect is counted counterclockwise from east by default, in cartesian
angles. To adjust aspect values to compass angles, one can use a script (Appendix A) written by M. Hamish
Bowman (Department of Marine Sciences, University of Otago, New Zealand).
In Figure (4B), the three highlighted subsets represent regions used for analysis of slope as function
of aspect. First, subregions are created with g.region; then, r.resample is used to create raster layers
with data only inside the subregions. Density plots can then be made to show aspect variation in each
subregion (Figure 5A).
Aspect map can also be reclassified and crossed with slope. In this work aspect was classified into
four categories: NE (0-90), SE (90-180), SW (180-270) and NW (270-360), in separate layers. Multiplying these layers with slope, using r.mapcalc, provides a way to analyse slope variations in each
aspect categorie, as in Figures (5B, C, D).
Figure 3 3D visualization of study area (view from SSE). Note the scarp of Serra do Caraa in the left of image, and the flat plateau
of Chapada de Canga in the center.
50
10
100
20
150
30
200
40
250
50
300
60
Figure 4 A) Slope map.
350
B) Aspect map.
aspect
aspect
NE (0-90)
SE (90-180)
SW (180-270)
NW (270-360)
0.05
NE (0-90)
SE (90-180)
SW (180-270)
NW (270-360)
Density
0.02
100
200
Aspect (degrees)
300
0.01
-5
10
15
Slope (degrees)
20
25
0.00
0.00
0.00
0.000
0.001
0.02
0.01
0.02
Density
0.04
Density
0.003
0.002
aspect
NE (0-90)
SE (90-180)
SW (180-270)
NW (270-360)
0.04
0.06
0.004
western
northern
southern
Density
0.03
subregion
0.03
0.08
30
10
20
30
40
50
60
70
Slope (degrees)
10
20
30
40
Slope (degrees)
Figure 5 A) Density plot of aspect in each subregion. B) Density plot of slope variations with aspect in the northern subregion.
C) Density plot of slope variations with aspect in the western subregion. D) Density plot of slope variations with aspect in the south subregion.
3.3. Swath profiles

Altimetric swath profiles (or projected profiles of Baulig, 1926, cited in Tricart and Cailleux, 1957)
are those were intersections of contours with equally spaced profile lines are marked within a swath, or
band. This kind of profile can provied a broader view of altimetric behaviour, and help to determine inclination of large topographic features in planaltic regions (Meis, 1982). The d.profile command outputs
up to four ASCII files with cumulative length in meters, raster value (elevation), east and north coordinates.
In this work, three swath profiles were made, with 500m of distance between each profile line (Figure 6).
Swath Profile 1
W
altitude (m)
1800
SP.3
1400
1000
600
0
SP.1
10
15
20
25
dist. (km)
SP.1'
Swath Profile 2
2200
WNW
SP.2
ESE
altitude (m)
1800
1400
SP.2'
1000
600
0
SP.3'
10
15
20
25
dist. (km)
Swath Profile 3
1000
altitude (m)
NNW
SSE
800
600
400
0
10
15
20
25
dist. (km)
Figure 6 Swath profiles. Localization of profiles on the shaded relief image.
3.4. Drainage and lineament density

Drainage density maps (Horton, 1945) express the ratio beetwen accumulated length per area
(km/km) and was calculated for 1000x1000m cells (Figure 7A). The same specifications were used in the
lineament density map (Figure 7B).
To determine drainage and lineaments densities, one must convert the vector maps to raster with
high ground resolution (g.region), to ensure details preservation. This map is reclassified, to assign all
the lines a unitary value, and then converted to sites.
The resolution is changed for that choosed to calculate density (e.g. 1000m) and s.cellstats is
used with the sum option to count how many points are in each cell. The result is converted to raster and
multiplied, with r.mapcalc, for a value that represents the ratio of the cell size in high resolution and the
area of low-resolution cell, to achieve the desired relation of length/area, in km/km. For example, if the
vector map was converted to raster with 25m resolution, and one searches density for 1000m cells:
density = n points x (25m/1000) (1)
or, simply
density = n points x 0.025 (2)
This map is finally converted to sites; the resolution is turned back to default value and the sites are
interpolated.
0.5
0.5
1.0
1.0
1.5
1.5
2.0
2.0
2.5
3.0
2.5
3.5
3.0
4.0
3.5
4.5
Figure 7 A) Drainage density map.
B) Lineament density map.
3.5. Surface roughness

The surface roughness map (Hobson, 1972; Day, 1979) was calculated as the ratio beetwen the
surface (real) area and the planar (flat) area of cells with 1000x1000m (Figure 8A). This method is useful
for morphological compartimentation since it considers the shape and not the altitude. Thus, tilted reliefs
have their expression showed, while it could be masked in a hypsometric map, as consequence of altimetric
variations (Grohmann and Campos Neto, 2002, 2003).
The surface area of cells is gotten from the geometric relations between the regular grid and slope
(Figure 8B).
Calculations can be made with r.mapcalc, using the following formula, for a 50m regular grid:
50*(sqrt((exp((tan(A)*50), 2)) + 2500)) (3)
Where A is the slope map.
The map for cells planar area can be achieved reclassifying any preexisting raster layer with a
unique value.
Both maps are then converted to sites; grid resolution is adjusted to the desired value (e.g. 1000m)
and s.cellstats is used with the sum option to get the surface and planar areas; the results are
converted to raster and ratio is calculated in r.mapcalc. This result is converted to sites and interpolated
with default resolution.
A
a = grid cell size

a = slope (degrees)
1.00
1.05
b = tan a a
c = a 2 + ((tan a ) a ) 2
b
a
a = flat area
a c = real area
1.10
1.15
1.20
1.25
1.30
Figure 8 A) Surface roughness map, for cells with 1000x1000m.
B) Geometric relations between regular grid and slope.
3.6. Isobase
The map of isobase (Filosopov, 1960 cited in Jain, 1980; Golts and Rosenthal, 1993) was made
from the intersections of contours with 2nd and 3rd-order stream channels (drainage orders according to
Strahler, 1952a,b).
First, vector contours and streams are converted to raster, to produce maps with values 1 and null
for the streams, and altitude and null for topography. This allows that maps, when multiplied (r.mapcalc),
present result only where there is an intersection of features. The resulting raster is converted to sites and
interpolated (Figure 9A). Additionally, manual interpolation was made (Figure 9B) and maps compared
with linear regression (Figure 9C).
600
600
800
800
1000
1000
1200
1200
1400
1400
1600
1600
1800
1800
1800
Figure 9 A) Automatically generated isobase map.
B) Manually interpolated isobase map.
1600
y=74.3 + 0.919x R=0,96
600
800
Automatic Isobase
1000
1200
1400
n=2397 R=0.92
600
800
1000
1200
1400
Manual Isobase
1600
1800
Figure 9 - C) Linear regression between automatic and manual isobase.
3.7. Hidraulic Gradient

In an attempt of determine areas with similar hydraulic behavior, a map of hydraulic gradient was
elaborated (Figure 10), according to the proposition of Rodriguez (1993), where this parameter is evaluated
for each 2nd-order stream as the ratio of the altimetric difference beetwen head and mouth with the plan
length; the value is attributed to the mid-portion of the stream.
In this work, an automatization of this process was achieved with Grad_Hidro, a software written
in Delphi language (Kylix environment), which relates the spacial position of drainage vector lines with
altitude in DEM.
Initially, the vector layer containing only 2nd-order streams is exported in ASCII format. The program reads this file and returns a sites file with points corresponding to vertices located at the extremities
and middle of each line, which must be imported into the system and converted to raster with 1m resolution. This raster is multiplied with the DEM, to obtain a layer with information only where the drainages
vertices are; this is made to save disk space and processing time in the following steps. The result of the
multiplication is converted to sites and the program is executed again; it assigns DEMs altitude to vertices,
calculate the hydraulic gradient and attribute the value to the mid-point of each stream. The resultant sites
file is imported into GRASS and interpolated with default resolution.
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Figure 10 Hydraulic gradient map.
4. Discussion
Morphostructural lineaments analysis indicate the presence of main trends oriented NE-SW and
NNE-SSW through the whole area, with secondary trends oriented N-S, NNW-SSE and NW-SE (Figure 2).
In the northern subregion, NE-facing slopes prevail, and have values ranging mainly from 0-10; the opposite direction, SW, have more slopes within 10-20, which leads to assimetric ridges oriented NW-SE with
steepest slope in the SW face (Figure 5B). The western subregion, which comprises essentially the Serra do
Caraa, can be divided in north and south; to the north, slopes face mainly NE, with lower dips than those
facing NW and SW; to the south, majority of slopes face SW (Figure 5C). The southern subregion has lined
ridges oriented mostly NE-SW, but the NW-SE trend is also present.
Drainage (Figure 7A) and lineament (Figure 7B) densities agree with general orientation of these
structures. In general, the south portion of the entire area has less lineaments than the north portion; in the
center of the area, over the Chapada de Canga plateau, there is a lack of both drainage and lineaments.
Swath profiles (Figure 6) allowed identification of altimetric levellings at ca. 2050m (profile 1), ca.
1650-1700m (profiles 1 and 2), and a step in the Chapada de Canga plateau between 900 and 850m (profile
3), which should be related to recent terrain movements.
Surface roughness (Figure 8A), and hydraulic gradient (Figure 10) show a similar schema, with a
relatively flat area in the central region, sorrounded by areas of higher roughness and gradient. In both
cases, the Serra do Caraa represents a very fast change of regional trend and has the highest values of all
area. This trend change is also present in the isobase map (Figure 9A), associated with others anomalies,
such as alignment and approach/separation of isolines.
Linear correlation models were applied to all evaluated morphometric parameters. The best results
were achieved among hydraulic gradient and surface roughness (Figure 11). Despite the observed dispersion, the graph shows a positive linear correlation between parameters.
1.30
1.25
y=1.01 + 0.21x R=0,3
1.00
1.05
Surface Roughness
1.10
1.15
1.20
n=2397 R=0.092
0.0
0.2
0.4
Hydraulic Gradient
0.6
0.8
Figure 11 Linear regression between hydraulic gradient and surface roughness.
Conclusions
Within the study of recent evolution of the eastern border of the Quadriltero Ferrfero region,
several morphometric methods were applied in a GIS environment, and proposed methodology is presented.
The integration between GIS and statistics in morphometric analysis allows agility and precision in
determination of necessary parameters. The methods presented in this work can be adapted according to
necessities and available tools. The use of free and open-source tools guarantees access to everyone, and its
increasing popularization opens new development perspectives in this field.
Acknowledgments
The author express sincere gratefulness to those that, directly or indirectly, had collaborated for the
accomplishment of this work. To Mario da Costa Campos Neto, for orientation and incentive; to the Conselho
Nacional de Desenvolvimento Cientfico e Tecnolgico (CNPq), for MSc scolarship cession (proc. 130777/
2002-1); to Jorge Kazuo Yamamoto, for help in Delphi programming; to all members of the GRASS-GIS
mailing list, in special Hamish Bowman, Glyn Clements, Roger Bivand and Markus Neteler.
Appendix A
Shell script for convert GRASS default aspect cartesian angles to azimuth angles
#!/bin/sh
# r.cat2compass.sh
by Hamish Bowman 11 April 2003
if [ -z $GISBASE ] ; then
echo You have to be in GRASS to use this.
exit
fi
if [ -z $1 ]; then
echo USAGE: r.cat2compass.sh AspectRaster
echo
Make a cartesian angled aspect raster into a compass angled
raster
echo
newfile is oldfile_compass.
(ccw from east -> cw from north)
exit
else
INPUT_MAP=$1
r.mapcalc $INPUT_MAP_compass=if(isnull($INPUT_MAP),null(), \
if(($INPUT_MAP < 90), 90-$INPUT_MAP, 360+90-$INPUT_MAP))
##### lighting for SOUTHERN HEMISPHERE (northern sun) #####
r.colors $INPUT_MAP_compass color=rules << EOF
0 white
180 black
360 white
EOF
fi
References
Geocincias, 10(1):89-101.
Baulig, H., 1926, Sur une mthode danalyse altimtrique applique la Bretagne. Bulletin de
lAssociation de Geographie Francaise. 10:7-9
Campos Neto, M.C.; Basei, M.A.S; Vlach, S.R.F.; Szab, G.A.J.; Vasconcelos, P., 2004. Migrao
de orgenos e superposio de orogneses: um esboo da colagem Brasiliana no sul do Crton do So
Francisco, SE-Brasil. Geologia-USP. in press.
Day, M.J., 1979. Surface roughness as a discriminator of tropical karst styles. Zeit. Geomorph.,
Suppl., 32:1-8.
Eddy, W.F., 1977. A new convex hull algorithm for planar sets. ACM Trans. Math. Soft. 3(4):398403.
Filosopov, V.P., 1960. Brief guide to morphometric methods in search of tectonic structures. Saratov,
Saratov Univ. Publ. House. [in Russian]
Golts, S.; Rosenthal, E., 1993. A morphotectonic map of the northern Arava in Israel, derived from
isobase lines. Amsterdam: Geomorphology, 7:305-315.
GRASS Development Team, 2002. GRASS 5.0 Users Manual. ITC-irst, Trento, Italy. Electronic
document: http://grass.itc.it/gdp/html_grass5/
Grohmann, C.H.; Campos Neto, M.C., 2002. Anlises morfotectnicas em Aiuruoca, MG. In: SBG,
Congresso Brasileiro de Geologia, XLI, Joo Pessoa, Anais: 353.
Grohmann, C.H.; Campos Neto, M.C., 2003. Serra do Caraa (Quadriltero Ferrfero,
MG):Morphometric parameters and preliminary tectonic considerations. In: IX Simpsio Nacional de Estudos
Tectnicos / III International Symposium on Tectonics, Armao de Bzios, Boletim de Resumos, 261-262.
Grohmann, C.H., 2003. Anlise morfomtrica em Sistemas de Informaao Geografica: aplicaao
dos softwares livres GRASS e R. In: VIII Simposio de Geologia do Sudeste, Sao Pedro, SP. Boletim de
Resumos, 154.
Hiruma, S.T.; Riccomini, C., 1999. Anlise morfomtrica em neotectnica: o exemplo do Planalto
de Campos do Jordo, SP. Revista do Instituto Geolgico. 20(1/2):5-19.
Hiruma, S.T.; Riccomini, C.; Modenesi-Gauttieri, M.C., 1999. Neotectnica no Planalto de Campos
de Jordo, SP. Revista Brasileira de Geocincias. 31(3):375-384.
Hobson, R.D., 1972. Surface roughness in topography: quantitative approach. In: Chorley, R.J. (ed)
Spatial analysis in geomorphology. Harper and Row, p.225-245.
Horton, R.E., 1945. Erosional development of streams and their drainage basins: hydrophysical
approach to quantitative morphology. Colorado: Geological Society of America Bulletin, 56(3): 275-370.
Ihaka, R.; Gentleman, R., 1996. R: A language for data analysis and graphics. Journal of Computational and Graphical Statistics, 5(3):299-314.
Jain, V.E., 1984. Geotectnica general . Moscou, Mir Ed., v.1, 357p.
King, L., 1956. Geomorfologia do Brasil Oriental. Rev. Bras. Geog., 18(2):147-266.
Liu, C.C., 1987. A geologia estrutural do Rio de Janeiro vista atravs de imagens MSS do Landsat.
Anais do I Simpsio de Geologia Regional RJ-ES, SBG, p.164-188.
Machado, N., Noce, C.M., Ladeira, E.A.; Oliveira, O.A.B., 1992. U-Pb Geochronology of Archaean
magmatism and Proterozoic metamorphism in the Quadriltero Ferrfero, southern So Francisco craton,
Brazil. Geological Society of America Bulletin, 104, 1221-1227.
Marshak, S., Alkmim, F.F.; Jord-Evangelista, H., 1992. Proterozoic crustal extension and the generation of dome-and-keel structure in a Archaean granite-greenstone terrane. Nature, 357, 491-493.
Meis, M.R.M.; Miranda, L.H.G.; Fernandes, N.F., 1982. Desnivelamentos de altitude como
parmetros para a compartimentao de relevo: bacia do mdio-baixo Paraba do Sul. In:Congresso Brasileiro
de Geologia, XXXII, Salvador, SBG/BA, Anais: v.4:1489-1503.
Mitasova, H.; Mitas, L., 1993. Interpolation by regularized spline with tension:I. Theory and implementation. Mathematical Geology. 25(6):641-655.
Mitasova, H.; Hofierta, J., 1993. Interpolation by regularized spline with tension:II. Application to
terrain modeling and surface geometry analysis. Mathematical Geology. 25(6):657-669.
Neteler, M., 1998. Introduction to GRASS GIS software. Hannover. Germany.
available at http://www.geog.uni-hannover.de/grass/gdp/neteler/manual98.html
Neteler, M; Mitasova, H., 2002. Open Source GIS: A GRASS GIS Approach. Boston, Kluwer Academic Publishers, 464 p.
Riccomini, C.; Crsta, A.P. 1988. Anlise preliminar de lineamentos em imagens de sensores remotos
aplicada prospeco mineral na rea dos granitides Mandira, SP. Bol.Inst.Geoc. USP. Srie Cientfica.
19:23-37.
Rodriguez, S.K., 1993. Neotectnica e sedimentao quaternria na regio de Volta Grande do
Rio Xingu, Altamira, Par. Instituto de Geocincias, Universidade de So Paulo, So Paulo, Dissertao de
Mestrado, 160p.
Brasil). Revista Brasileira de Geocincias. 25(3):151-164.
SantAnna, L.G.; Schorscher, H.D., 1995. Leques aluviais cenozicos do flanco leste da Serra do
Caraa, MG: a Formao Chapada de Canga. Anais da Academia Brasileira de Cincias. 67(4):519.
SantAnna, L.G.; Schorscher, H.D.; Riccomini, C., 1997. Cenozoic tectonics of the Fonseca Basin
region, eastern Quadriltero Ferrfero, MG, Brazil. Journal of South American Earth Sciences. 10(3-4):275284.
Strahler, A.N., 1952a. Dynamic basis of geomorphology. Colorado: Bulletin of the Geological Society of America, 63:923-938.
Strahler, A.N., 1952b. Hypsometric (area-altitude) analysis of erosional topography. Colorado:
Bulletin of the Geological Society of America, 63:1117-1142.
Tricart, J.; Cailleux, A., 1957. Cours de Geomorphologie I: Geomorphologie Structurale. C.D.U.
Paris. 252p.
U.S. Army CERL, 1993. GRASS 4.1 Reference Manual. U.S. Army Corps of Engineers, Construction Engineering Research Laboratories, Champaign, Illinois, 1-425.
Varajo, C.A.C., 1991. A questo da correlao das superfcies de eroso do Quadriltero Ferrfero,
Minas Gerais. Rev. Bras. Geoc., 21(2):138-145.
Episodes, 14(2):116-124.
De: "Computers & Geosciences" <compgeo@sympatico.ca>

: carlos_grohmann@yahoo.com.br
Objet: Computers & Geosciences Submission received: 03-197
Date: Sat, 17 Jan 2004 19:14:59 -0500
I apologise for the delay in acknowledging the receipt of your paper. There was a large
backlog of papers to clear as a result of our adding electronic submission to the journal reviewing
process.
I received your paper on December 31, 2003. I have assigned it manuscript number 03-197,
please refer to this number in any future correspondence. I am now seeking, from our database of
specialists, two independent reviewers for the evaluation of your work. You can expect a decision
from us regarding publication in approximately fourteen weeks from todays date.
Sometimes the evaluation of your paper can be delayed due to difficulties in securing
reviewers, please be patient. In the meantime, if you have any questions or concerns, do not hesitate
to contact me.
Sincerely,
Kathryn MacKinnon
NOTE: Many of our reviewers are now requesting we send electronic copies of the paper to speed up the
reviewing time if you have not already included an electronic copy of your paper please send one
attached to a reply to this message.
Managing Editor
Computers & Geosciences
P.O. Box 26128
72 Robertson Road
Nepean, ON K2H 9R0
Phone +1 613 828-8500
Fax +1 613 828-0696
E-Mail: candg@iamg.org
ANEXO 2
Trend-surfaces analysis of morphometric parameters
Trend-surfaces analysis of morphometric parameters

Ps-Graduao - Instituto de Geocincias - Universidade de So Paulo, Brasil
Rua do Lago, 562. 05508-800. So Paulo, SP
Phone: +55-11-30914216; fax: +55-11-30914258.
E-mail address: guano@usp.br, carlos_grohmann@yahoo.com.br
Abstract
Trend-surface analysis was carried out on data from morphometric parameters hydraulic gradient and isobase.
The study area, located in the eastern border of Quadriltero Ferrfero, southeastern Brazil, presents two main
geomorphological units, one characterized by fluvial dissection and the other of mountainous relief, with a scarp of
hundreds of meters of fall between them. Morphometric maps were evaluated in GRASS-GIS and statistics were made
on R statistical language, using the spatial package. Analysis of Variance (ANOVA) was made to test the significance
of each surface and the significance of increasing polynomial degree. The best results were achieved with 2nd-order
surface for hydraulic gradient and 6th-order surface for isobase. Shape and orientation of residuals maps contours for
choosed trends were compared with structures inferred from several morphometric maps, and a good correlation is
present.
Keywords: Trend-surface analysis; Morphometric analysis; Geographic Information Systems; GRASS-GIS; R Statistical Language, Analysis of Variance.
1. Introduction
Trend-surface analysis is the mathematical method used separate map data into components of
regional nature from local fluctuations (Davis, 1986). Regional trends are computed as polynomial surfaces
of successive powers; residual values, corresponding to local fluctuations, are the arithmetic difference
between original data and trend surface. Residuals maps play an important role in trend-surface analysis
given that they can favour on identify or accentuate features of interest. This technique has been widely
used by petroleum geologists, to predict structural behaviour of stratigraphic units in search for traps,
paleogeographic features, or to recognize structural breaks between successive units (Merrian and Harbaug,
1963; Merrian and Lippert, 1966; Sutterlin and Hastings, 1986; Davis, 1986).
Morphometric maps are important tools in studies related to neotectonics and geomorphology, where
the answers of natural landscapes to planets interior dynamics is often masked by fast action of weathering,
and the presence of drainage network anomalies and relief pattern discontinuities may be related with
recent terrain movements (Zuchiewicz, 1991; Rodriguez, 1993; Salvador and Riccomini, 1995; Hiruma
and Riccomini, 1999; Hiruma et al, 1999).
This work presents trend-surface analysis of the morphometric parameters hydraulic gradient
(Rodriguez, 1993) and isobase (Filosopov, 1960 cited in Jain, 1980; Golts and Rosenthal, 1993), with
integrated utilization of free softwares GRASS-GIS (U.S. Army CERL, 1993) and R statistical language
(Ihaka and Gentleman, 1996), in the Serra do Caraa region, eastern border of Quadriltero Ferrfero,
southeastern Brazil.
44000'
43030'
43000'
Nova Era
Serra
Ibirit
urral
do C
Moe
Brumadinho
lin
ync
da s
60
2
42030'
20000'
a 5
6
Rio Casca
ITABIRITO
60
Bonfim dome
Bao dome
Acaiaca
MARIANA
80
10
PONTE NOVA
40
OURO PRETO
Belo Vale
77
3
4
Alvinpolis
77
sy
Nova Lima
G
an
da
re
la
80
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
aaaaaaaa
nc
lin
e
Raposo
44 30'
77
Caet
Caet dome
Sabar
00
Belo Horizonte
dome
JOO MONLEVADE
BELO HORIZONTE
78
CONGONHAS
00
Ouro Branco
20 30'
10
Santa Rita dome
40 km
77
20
6
20
40
60
80
00
20
40
60
Figure 1 Simplified geological map of southeast So Francisco craton (modified from Campos Neto et al, 2004). 1-Quaternary
Chapada de Canga Fm. (itabiritic conglomerate); 2-Miocene Fonseca Fm. (lacustrine deposits); 3-Itacolomi/Espinhao Gr. (coarse and crossstratified sandstone, polymitic conglomerate); 4-Minas Supergroup (basal Moeda and Batatal Fms sandstones and metapelites -, Cau and
Gandarela Fms banded-iron formations and carbonates- and uppers Piracicaba and Sabar Grs metapelites, sandstones, metavolcaniclastic
rocks, conglomerates); 5-Rio das Velhas Supergroup (Archaean -2.7Ga- basalt, komatiite and ryolitic lavas intercalated with sedimentary
units); 6-SE So Francisco craton gneiss (gneiss/migmatite complex, calc-alkaline granitoid plutons and Espinhao-related syenite and alkaline-like granite-gneiss); 7-Dom Silvrio Gr. (metapelites and quartzites with calcsilicated rocks and manganese-rich metasediments); 8Mantiqueira gneisses (amphibole-biotite orthogneisses and migmatites with titanite-rich synorogenic metabasic-intermediate rocks. Acaiaca
enderbite and kinzigite and Santo Antonio do gama amphibolite); 9-Juiz de Fora granulites (orthogranulites from a charnockitic suite, kinzigites
and migmatites); 10-Study area.
2. Geological and geomorphological context

The Quadriltero Ferrfero region, south So Francisco Craton, is characterized by Arquean granitegneiss domed complexes coeval to Rio das Velhas greenstone belt (Machado et al, 1992), engaged to south
and east in a Paleoproterozoic metamorphic belt (Mineiro belt, Teixeira and Figueiredo, 1991) defined as
an acrescionary orogen of ca. 2.1 Ga (Figure 1).
The study area presents two geomorphological units: the Quadriltero Ferrfero and the Minas
Gerais center-south and east Highlands. The first unit shows average altitudes of 1400-1600m, with maximum
in the Serra do Caraa, at 2064m; the second unit is characterized by fluvial dissection. The scarp that
limits these units has hundreds of meters of fall, and leads to believe not only erosional processes, but also
post-cretacic tectonic movements contributed in the morphological evolution.
According to Varajo (1991), remains of planation surfaces in the Quadriltero Ferrfero have close
relations with lithostructural domains. Reactivations with vertical displacement of ancient faults would be
responsible for present day altimetric differences (King, 1956; Barbosa, 1980).
Evidences of Cenozoic tectonics in the Quadriltero Ferrfero are observed in several sedimentary
deposits and indicate three deformational events with distinct tension fields. The first event, extensive,
oriented NNE-SSW and probably related with horsts and grabens oriented ESE-WNW. A second and more
expressive event, mainly compressive NW-SE. The third event is considered to be the relaxing of previous
event structures. (Lipski et al, 2001).
In the study area, the Quaternary Chapada de Canga Formation consists of a flat plateau (Figure 2)
formed by a succession of continental itabiritic ironstone pebble conglomerates and directly overlies the
Eocene Fonseca Formation and the Precambrian basement (SantAnna and Schorscher, 1995). Cenozoic
deposits are cut by NE and NW brittle faults and joints, related to reactivation of pre-existing structures in
Precambrian basement, process that strongly influenced the development of present day landscape
morphology and drainage network (SantAnna et al, 1997).
Figure 2 Shade relief map of the study area. Lighting at N45, 20 above the horizon.
3. Methods
The GIS used was the Geographic Resources Analysis Support System - GRASS 5.0.3 (U.S. Army
CERL, 1993; Neteler, 1998; Neteler and Mitasova, 2002; GRASS Development Team, 2002), which offers
an integrated environment for raster and vector analysis, image processing and maps/graphics creation.
Statistical analysis were carried out on R, a system for statistical computation and graphics (Ihaka
and Gentleman, 1996; Grunsky, 2002; R Development Core Team, 2003), through an interface with GRASS
(Bivand, 2000) that allows raster maps and points files to be treated as variables for analysis.
Morphometric parameters were developed in GRASS according the propositions of Grohmann
(2004). The map of hydraulic gradient was elaborated to determine areas with similar hydraulic behaviour
(Figure 3A). This parameter is calculated for each 2nd-order stream channel as the ratio of the altimetric
difference between head and mouth with the plan length; the value is attributed to the mid-portion of the
stream and points values are interpolated.
Source of Variation
Sums of
Squares
Degrees
of
Freedom
Mean
Square
Regression of Degree p
SSRP
MSRP
Deviation from Degree p
SSDP
nk1
MSDP
SSRP+1
MSRP+1
F Test
MSRP/MSDP
Regression of Degree
(p+1)
Deviation from Degree
(p+1)
Regression due to Increase
from p to (p+1) Degree
SSDP+1
nm1
MSDP+1
SSRI =
SSRP+1 - SSRP
m-k
MSRI
Total Variation
SST
n1
(a)
MSRP+1/MSDP+1
MSRI/MSDP+1
(b)
(c)
Table 1 General ANOVA for the significance of increasing the degree of a polynomial trend from p to (p+1) degree; polynomial
equationof degree p has k coefficients, not counting the b0 term; equation of degree (p+1) has m coefficients, not counting the b0 term; number
of observations is n. R2: percent goodness-of-fit. (a) test of significance of the p-degree trend surface. (b) test of significance of the (p+1)-degree
trend surface. (c) test of significance of increase in fit of the (p+1) degree over p degree.
Trend-surface
analysis area
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
-0.2
0.050
-0.1
0.075
0.0
0.100
0.125
0.1
0.150
0.2
0.175
0.3
0.200
0.4
0.225
0.5
0.250
Figure 3 A: Original hydraulic gradient surface, with area used for trend-surface analysis in dotted line; B: Data points, n=453; C: 2 ndorder trend surface; D: Residuals of 2nd -order.
degree
1st
2nd
Source of
Variation
Regression
Sums of
Squares
0.257
Degrees of
Freedom
2
Mean
Square
0.129
0.032
Deviation
6.899
450
0.015
0.968
Regression
1.007
0.201
0.131
Deviation
6.149
447
0.014
0.869
0.75
0.25
Regression
1.190
0.132
0.149
Deviation
5.967
443
0.013
0.851
0.182
0.045
Regression
1.620
14
0.115
0.202
Deviation
5.536
438
0.013
0.798
0.431
0.086
Regression
1.793
20
0.090
0.216
Deviation
5.364
432
0.012
0.784
0.172
0.029
Regression
2.073
27
0.077
0.244
Deviation
5.084
425
0.012
0.756
0.280
0.04

3rd

4th

5th

6th
R2
F Test
F(0.05)
8.394
3.00
14.645
2.21
18.173
2.60
9.815
1.88
3.386
2.37
9.157
1.75
6.814
2.21
7.219
1.57
2.313
2.10
6.419
1.52
3.347
2.01
Table 2 - General ANOVA for the significance of increasing the degree of hydraulic gradient trend surfaces from 1st to 6th degree.
A
Trend-surface
analysis area
600
800
1000
1200
1400
1600
1800
600
-200
700
800
-100
900
1000
1100
100
1200
1300
200
1400
300
1500
1600
Figure 4 A: Original isobase surface, with area used for trend-surface analysis in dotted line; B: Data points, n=4512; C: 6th-order
trend surface; D: Residuals of 6th-order.
degree
1st
2nd
Source of
Variation
Regression
4509
52448.0
0.709
30.6 x 106
0.457
4506
40170.0
0.543
18.5 x 106
192.9 x 10
21.4 x 10
141.1 x 10
4502
31345.5
9.97 x 106
152.9 x 106
Deviation
Regression
Deviation
Regression
Deviation
Regression
Deviation

6th
0.291
Regression

5th
Mean
Square
48.7 x 106
326.48 x 10

4th
Degrees of
Freedom
2
Deviation

3rd
Sums of
Squares
97.5 x 106
Regression
Deviation
181.0 x 10
55.5 x 106
39.9 x 106
226.5 x 10
14
16.2 x 10
107.4 x 10
4497
23894.9
6.74 x 106
256.6 x 10
77.4 x 10
30.1 x 106
274.6 x 10
59.3 x 10
18.0 x 106
0.577
0.423
33.7 x 106
R2
0.677
0.323
20
12.8 x 10
4491
17227.6
5.01 x 106
0.767
0.233
27
10.2 x 10
4484
12237.4
2.57 x 106
0.821
0.179
F Test
F(0.05)
929.50
3.00
761.68
2.21
460.39
2.60
683.67
1.88
318.13
2.37
677.17
1.75
281.75
2.21
744.79
1.57
291.07
2.10
768.40
1.52
194.39
2.01
Table 2 - General ANOVA for the significance of increasing the degree of isobase trend surfaces from 1st to 6th degree.
Isobase lines delineate an erosional surface, so isobase surfaces relate to erosional stages, or
manifestations of erosional-tectonic events, especially younger ones (Golts and Rosenthal, 1993). The map
of isobase (Figure 4A) was made from interpolation of the intersections of contours with 2nd and 3rd-order
stream channels (drainage orders according to Strahler, 1952a,b).
Points files were imported into R as data frames; trend surfaces were computed using the R package
spatial, which allows the fit of a polynomial surface up to 6th degree, and inserted back into GRASS as
raster maps; residual surfaces were obtained by subtracting trends from the original surface.
A problem that can affect not only trend surfaces, but also contour maps, moving-average and other
forms of fitted surfaces is the presence of edge effects, which occur when there are few (or no) control
points on the map boundary, so there are almost no constraints on the form of the surface (Davis, 1986;
Landim, 1998). In order to avoid these effects, the area used for analysis is smaller than the area of the
morphometric maps, assuring the presence of a buffer, in which edge effects are concentrated (Figures
3A, 3B, 4A, 4B).
The goodness-of-fit of trend surfaces can be statistically tested by analysis of variance (ANOVA),
comparing the variance of regression to the variance of deviations (residuals). When a series of equations
of successively higher degree are fitted to the data, this analysis can be expanded to analyze the contribution
of additional partial regression coefficients and measure the appropriateness of increasing the order of
equations (Davis, 1986). The general ANOVA table for test the significance of increasing degree of
polynomial surface is presented in Table (1). ANOVA data for evaluated morphometric parameters is
presented in Table (2) for hydraulic gradient and Table (3) for isobase.
The best F-values for hydraulic gradient were achieved with 2nd-order polynomial (Figure 3C),
both for the significance of regression and for its contribution due to degree increase. All F-values for
isobase are very high, and the best correlation is of 6th-order surface (Figure 4C), with a R2 of 0.821.
DEPRESSED
AREA
UPLIFTED
AREA
UPLIFTED
AREA
Figure 5 Morphological scenario and inferred stuctures for the study area, according to Grohmann and Campos Neto (2003).
4. Discussion
Morphometric analysis of the study area were carried out by Grohmann and Campos Neto (2003),
where several parameters were employed to infer structures that can be related with landscape configuration;
a morphological scenario was proposed, with a depressed central area enclosed by uplifted hills (Figure 5).
The presence of these structures can be seen on residual maps for the analyzed parameters (Figures 3D,
4D), in the change of shape and orientation of contours.
The 2nd-order residuals for hydraulic gradient show a fast transition to higher values in the eastern
region, with a strong alignment of contours in NW-SE and NE-SW directions, which marks the scarp of
Serra do Caraa. The 6th-order residuals for isobase also have good agreement with inferred structures; in
this map, a region of very negative deviations marks the scarp.
There is a nearly E-W structure in the central portion of the study area, admitted as a normal fault
(SantAnna et al, 1997) associated with a morphological step of the Chapada de Canga plateau, which is
well marked in swath profiles made by Grohmann (2004, Figure 6).
Despite the fact that the Chapada de Canga plateau is a very distinctive topographic feature, it does
not show much expression on morphometric maps, what can be explained by the fact that it is an area of
very low drainage density, and that evaluated morphometric parameters are based on the relations of
topography with drainage network.
NNW
altitude (m)
1000
SSE
800
600
400
10
15
20
25
dist. (km)
Figure 6 Swath profile oriented NNW-SSE through the center of the study area (Grohmann, 2004). Note the step of the flat plateau in
its mid-portion.
5. Conclusions
Trend-surface analysis was applied to morphometric data, and analysis of variance was used to
determine the most representative polynomial surface. The residuals maps were compared with a proposed
morphological scenario, and there is a good correlation between inferred structures with shape and orientation
of contours. In the central portion of the area, there is a flat plateau, which has little expression in
morphometric maps, although it represents a distinct topographic feature. This can be seen as a result of
low drainage density in the flat area, since most morphometric methods are based on relations of topography
with drainage network.
Acknowledgements
The author expresses sincere gratefulness to those that, directly or indirectly, had collaborated for
the accomplishment of this work. To Mario da Costa Campos Neto, for orientation and incentive; to the
Conselho Nacional de Desenvolvimento Cientfico e Tecnolgico (CNPq), for MSc scolarship cession
(proc. 130777/2002-1); to Jorge Kazuo Yamamoto, for help with spatial statistics.
References
Geocincias, 10(1):89-101.
Campos Neto, M.C.; Basei, M.A.S; Vlach, S.R.F.; Szab, G.A.J.; Vasconcelos, P., 2004. Migrao
de orgenos e superposio de orogneses: um esboo da colagem Brasiliana no sul do Crton do So
Francisco, SE-Brasil. Geologia-USP. in press.
Davis, J.C., 1986. Statistics and data analysis in geology (2nd ed.). John Wiley & Sons, New York,
656p.
Filosopov, V.P., 1960. Brief guide to morphometric methods in search of tectonic structures. Saratov,
Saratov Univ. Publ. House. [in Russian]
Golts, S.; Rosenthal, E., 1993. A morphotectonic map of the northern Arava in Israel, derived from
isobase lines. Amsterdam: Geomorphology, 7:305-315.
GRASS Development Team, 2002. GRASS 5.0 Users Manual. ITC-irst, Trento, Italy. Electronic
document: http://grass.itc.it/gdp/html_grass5/
Grohmann, C.H.; Campos Neto, M.C., 2003. Serra do Caraa (Quadriltero Ferrfero,
MG):Morphometric parameters and preliminary tectonic considerations. In: IX Simpsio Nacional de Estudos
Tectnicos / III International Symposium on Tectonics, Armao de Bzios, Boletim de Resumos, 261-262.
Grohmann, C.H., 2004. Morphometric analysis in Geographic Information Systems: applications
of free softwares GRASS and R. Computers & Geosciences. (submitted)
Hiruma, S.T.; Riccomini, C., 1999. Anlise morfomtrica em neotectnica: o exemplo do Planalto
de Campos do Jordo, SP. Revista do Instituto Geolgico. 20(1/2):5-19.
Hiruma, S.T.; Riccomini, C.; Modenesi-Gauttieri, M.C., 1999. Neotectnica no Planalto de Campos
de Jordo, SP. Revista Brasileira de Geocincias. 31(3):375-384.
Ihaka, R.; Gentleman, R., 1996. R: A language for data analysis and graphics. Journal of
Computational and Graphical Statistics, 5(3):299-314.
Jain, V.E., 1984. Geotectnica general . Moscou, Mir Ed., v.1, 357p.
King, L., 1956. Geomorfologia do Brasil Oriental. Revista Brasileira de Geografia, 18(2):147-266.
Landim, P.M.B,. 1998. Anlise estatstica de dados geolgicos. So Paulo, Fundao Editora da
UNESP, 226p.
Machado, N., Noce, C.M., Ladeira, E.A.; Oliveira, O.A.B., 1992. U-Pb Geochronology of Archaean
magmatism and Proterozoic metamorphism in the Quadriltero Ferrfero, southern So Francisco craton,
Brazil. Geological Society of America Bulletin, 104, 1221-1227.
Merriam, D.F.; Harbaug, J.W., 1963. Computer helps map oil structures. Oil and Gas Journal,
61(47):158-163.
Merriam, D.F.; Lippert, R.H., 1966. Geologic model studies using trend-surfaces analysis. Journal
of Geology, 74(3):344-357.
Neteler, M., 1998. Introduction to GRASS GIS software. Hannover. Germany.
available at http://www.geog.uni-hannover.de/grass/gdp/neteler/manual98.html
Neteler, M; Mitasova, H., 2002. Open Source GIS: A GRASS GIS Approach. Boston, Kluwer
Academic Publishers, 464 p.
Rodriguez, S.K., 1993. Neotectnica e sedimentao quaternria na regio de Volta Grande do
Rio Xingu, Altamira, Par. Instituto de Geocincias, Universidade de So Paulo, So Paulo, Dissertao de
Mestrado, 160p.
Brasil). Revista Brasileira de Geocincias. 25(3):151-164.
SantAnna, L.G.; Schorscher, H.D., 1995. Leques aluviais cenozicos do flanco leste da Serra do
Caraa, MG: a Formao Chapada de Canga. Anais da Academia Brasileira de Cincias. 67(4):519.
SantAnna, L.G.; Schorscher, H.D.; Riccomini, C., 1997. Cenozoic tectonics of the Fonseca Basin
region, eastern Quadriltero Ferrfero, MG, Brazil. Journal of South American Earth Sciences. 10(3-4):275284.
Strahler, A.N., 1952a. Dynamic basis of geomorphology. Colorado: Bulletin of the Geological Society
of America, 63:923-938.
Strahler, A.N., 1952b. Hypsometric (area-altitude) analysis of erosional topography. Colorado:
Bulletin of the Geological Society of America, 63:1117-1142.
Sutterlin, P.G.; Hastings, J.P., 1986. Trend-surface analysis revisited a case history. Computers &
Geosciences, 12(4B):537-562.
U.S. Army CERL, 1993. GRASS 4.1 Reference Manual. U.S. Army Corps of Engineers, Construction
Engineering Research Laboratories, Champaign, Illinois, 1-425.
Varajo, C.A.C., 1991. A questo da correlao das superfcies de eroso do Quadriltero Ferrfero,
Episodes, 14(2):116-124.
Data: Sun, 14 Mar 2004 19:43:49 GMT

De: gbonhamc@NRCan.gc.ca
Para: guano@usp.br
Assunto: Confirmation new paper No. guano.AT.usp.br/20040314/1 submitted to CAGEO
This is an automated email

14-Mar-2004
Dear Mr Grohmann,
Thank you for submitting the paper entitled:

'Trend-surfaces analysis of morphometric parameters' to Computers & Geosciences
The e-submission number for your paper is CAGEO_guano.AT.usp.br_20040314/1
Your paper will be evaluated as quickly as possible; and we will be in touch again in due
course.
Yours sincerely,
Dr. Graeme Bonham-Carter
Editor-in-Chief
Computers & Geosciences
For queries about the E-submission website please contact authorsupport@elsevier.com

For queries specific to this submission please contact: gbonhamc@NRCan.gc.ca
ANEXO 3
DXF_Lin
Listagem de programa de computador escrito em linguagem Delphi

para clculo de orientao e comprimento de lineamentos a partir de
arquivo vetorial em formato DXF exportado pelo GRASS-GIS
unit Unit1;
interface
uses
SysUtils, Types, Classes, Variants, QTypes, QGraphics, QControls, QForms,
QDialogs, QStdCtrls, QMenus;
type
TForm1 = class(TForm)
Button1: TButton;
OpenDialog1: TOpenDialog;
MainMenu1: TMainMenu;
Abrir1: TMenuItem;
Fechar1: TMenuItem;
Memo1: TMemo;
SaveDialog1: TSaveDialog;
Label1: TLabel;
Label2: TLabel;
Salvar1: TMenuItem;
procedure Button1Click(Sender: TObject);
procedure Abrir1Click(Sender: TObject);
procedure Fechar1Click(Sender: TObject);
procedure Salvar1Click(Sender: TObject);
private
{ Private declarations }
public
{ Public declarations }
end;
var
Form1: TForm1;
entrada:textFile;
nomeArquivo:string;
implementation
{$R *.xfm}
procedure xyz(var x,y,z:real);

var
cod0,cod10,cod20,cod30:integer;
begin
readln(entrada,cod10);
readln(entrada,x);
readln(entrada,y);
readln(entrada,z);
end;
procedure TForm1.Button1Click(Sender: TObject);

var
polyline:string[8];
titulo,vertex,seqend,rumo,comprimento:string;
cod0,cod1,cod8,cod66:integer;
x0,y0,z0,x1,y1,z1,x2,y2,z2,x11,x22,y11,y22:real;
azim,arco,hipoten,cat_op,cat_adj,cat_op2,cat_adj2,tangente:real;
label 10;
begin
assignfile(entrada,nomeArquivo);
reset(entrada);
memo1.lines.clear;
//pula o header.
while not(eof(entrada)) do
begin
readln(entrada,polyline);
if polyline=POLYLINE then goto 10;
end;
//comea a ler os parametros
10:readln(entrada,cod8);
readln(entrada,titulo);
xyz(x0,y0,z0);
readln(entrada,vertex);
xyz(x1,y1,z1);
readln(entrada,vertex);
xyz(x2,y2,z2);
//define os ns da linha
if x2>x1 then
begin
x11:=x1;
y11:=y1;
x22:=x2;
y22:=y2;
end
else
begin
x11:=x2;
y11:=y2;
x22:=x1;
y22:=y1;
end;
// linha vertical N-S

if x1=x2 then
begin
azim:=0;
hipoten:=y2-y1;
end
else
begin
cat_op:=x2-x1;
end;
// linha horizontal E-W
if y1=y2 then
begin
azim:=90;
hipoten:=x2-x1;
end
else
begin
cat_adj:=y2-y1;
end;
// calcula comprimento e azimute

tangente:=(cat_op/cat_adj);
arco:=ArcTan(tangente);
azim:=(arco*180)/Pi;
if azim<0 then azim:=azim+180;
cat_op2:=Sqr(cat_op);
cat_adj2:=Sqr(cat_adj);
hipoten:=Sqrt(cat_op2+cat_adj2);
// salva o arquivo de sada
rumo:=FloatToStr(azim);
comprimento:=FloatToStr(hipoten);
memo1.Lines.Add(rumo+
+comprimento);
//le mais parametros

readln(entrada,seqend);
readln(entrada,polyline);
if polyline=POLYLINE then goto 10;
closeFile(entrada);
end;
//procedure para abrir arquivo

procedure TForm1.Abrir1Click(Sender: TObject);
begin
if openDialog1.Execute then
nomeArquivo:=extractFileName(openDialog1.FileName);
if not(fileExists(nomeArquivo)) then showmessage(Arquivo no existe.);
end;
// sai do programa
procedure TForm1.Fechar1Click(Sender: TObject);
begin
close;
end;
procedure TForm1.Salvar1Click(Sender: TObject);

begin
if saveDialog1.Execute then
nomeArquivo:=extractFileName(saveDialog1.FileName);
if (fileExists(nomeArquivo)) then showmessage(Arquivo j existe.)else
Memo1.Lines.SaveToFile(nomeArquivo);
end;
end.
ANEXO 4
Grad_Hidro
Listagem de programa de computador escrito em linguagem Delphi para clculo

de gradiente hidrulico, para ser usado em conjunto com GRASS-GIS
unit hidraulic_gradient;
interface
uses
SysUtils, Types, Classes, Variants, QTypes, QGraphics, QControls, QForms,
QDialogs, QStdCtrls, QMenus, Math;
type
TForm1 = class(TForm)
Memo1: TMemo;
Memo2: TMemo;
Memo3: TMemo;
MainMenu1: TMainMenu;
OpenDialog1: TOpenDialog;
SaveDialog1: TSaveDialog;
Arquivolinhas1: TMenuItem;
Arquivopontos1: TMenuItem;
Salvar1: TMenuItem;
fechar1: TMenuItem;
Memo4: TMemo;
Button1: TButton;
SalvarVrtices1: TMenuItem;
Button2: TButton;
Label1: TLabel;
Label2: TLabel;
Label3: TLabel;
Label4: TLabel;
procedure fechar1Click(Sender: TObject);
procedure Salvar1Click(Sender: TObject);
procedure Arquivolinhas1Click(Sender: TObject);
procedure Arquivopontos1Click(Sender: TObject);
procedure SalvarVrtices1Click(Sender: TObject);
private
{ Private declarations }
public
{ Public declarations }
end;
var
Form1: TForm1;
entrada1:textFile;
nomeArquivo1:string;
entrada2:textFile;
nomeArquivo2:string;
nomeArquivo:string;
implementation
{$R *.xfm}
procedure TForm1.fechar1Click(Sender: TObject);

begin
close;
end;
procedure TForm1.Salvar1Click(Sender: TObject);

begin
end;
procedure TForm1.SalvarVrtices1Click(Sender: TObject);
begin
end;
procedure TForm1.Arquivolinhas1Click(Sender: TObject);
begin
nomeArquivo1:=extractFileName(openDialog1.FileName);
if not(fileExists(nomeArquivo1)) then showmessage(Arquivo no existe.);
end;
procedure TForm1.Arquivopontos1Click(Sender: TObject);
begin
nomeArquivo2:=extractFileName(openDialog1.FileName);
if not(fileExists(nomeArquivo2)) then showmessage(Arquivo no existe.);
end;

var
name,desc,input,output,label_grass,forma1,forma2,n_jota:string;
org,d_date,d_name,m_name,m_date,m_scale,other,zone:string;
west,east,south,north,thresh,verti,x1_j,x2_j,y1_j,y2_j:string;
tralha:string[1];
utm_x:string[6];
utm_y:string[7];
barra:string[3];
cota:string[4];
espaco:string[2];
x_float,y_float,cota_float:array [1..10000] of real;
L:array [1..600] of char;
x1,y1,z1,x2,y2,z2:array [1..1000] of string;
utm_e1,utm_e2,utm_n1,utm_n2,utm_e_mid,utm_n_mid:array [1..1000] of real;
utm_n,utm_e:array [1..2500] of real;
i,j,k,n_vertex,n_j,n_i,dv,dx,dy:integer;
x1_int,x2_int,y1_int,y2_int,z1_int,z2_int,mid2:integer;
dh,dx2,dy2,dz,grad,meiox,meioy,mid1:real;
label valores1,valores2;
begin
//limpa os memos
memo1.lines.clear;
memo2.lines.clear;
memo3.lines.clear;
memo4.lines.clear;
//inicia leitura do arquivo dig-ascii
assignfile(entrada1,nomeArquivo1);
reset(entrada1);
// leitura do header
readln(entrada1,org);
readln(entrada1,d_date);
readln(entrada1,d_name);
readln(entrada1,m_name);
readln(entrada1,m_date);
readln(entrada1,m_scale);
readln(entrada1,other);
readln(entrada1,zone);
readln(entrada1,west);
readln(entrada1,east);
readln(entrada1,south);
readln(entrada1,north);
readln(entrada1,thresh);
readln(entrada1,verti);
//inicia j,k. leitura das coordenadas do arquivo dig_ascii

j:=1;
k:=1;
while not (eof(entrada1)) do
begin
read(entrada1,L[j]);read(entrada1,espaco);readln(entrada1,n_vertex);
mid1:=(n_vertex/2);
mid2:=Round(mid1);
for k:=1 to n_vertex do
begin
read(entrada1,utm_n[k]);readln(entrada1,utm_e[k]);
utm_n[k]:=Round(utm_n[k]);
utm_e[k]:=Round(utm_e[k]);
end;
utm_e1[j]:=utm_e[1];
utm_e2[j]:=utm_e[n_vertex];
utm_n1[j]:=utm_n[1];
utm_n2[j]:=utm_n[n_vertex];
utm_e_mid[j]:=utm_e[mid2];
utm_n_mid[j]:=utm_n[mid2];
// imprime vrtices no memo1

x1_j:=FloatToStr(utm_e1[j]);
y1_j:=FloatToStr(utm_n1[j]);
memo1.Lines.Add(x1_j+,+y1_j);
memo1.Lines.Add(x2_j+,+y2_j);
j:=succ(j);
end;
n_j:=j-1;
n_jota:=IntToStr(n_j);
memo1.Lines.Add(n_jota+ linhas);
//inicia leitura do arquivo de sites-topografia
reset(entrada2);
readln(entrada2,name);
read(entrada2,desc);read(entrada2,input);read(entrada2,output);
readln(entrada2,label_grass);
readln(entrada2,forma1);
readln(entrada2,forma2);
// inicia i e leitura das coordenadas do arquivo sites-topografia
i:=1;
while not(eof(entrada2)) do
begin
read(entrada2,utm_x);read(entrada2,barra);read(entrada2,utm_y);
read(entrada2,barra);read(entrada2,tralha);readln(entrada2,cota);
//imprime x,y,z no memo2
memo2.Lines.Add(utm_x+ +utm_y+ +cota);
// converte string para float e coloca no vetor
x_float[i]:=StrToFloat(utm_x);
y_float[i]:=StrToFloat(utm_y);
cota_float[i]:=StrToFloat(cota);
i:=succ(i);
end;
n_i:=i-1;
//inicia comparao das variaveis

//compara valores para vertice 1
for j:=1 to n_j do
begin
for i:=1 to n_i do
begin
if abs(utm_e1[j]-x_float[i]) <= 1.5 then
begin
if abs(utm_n1[j]-y_float[i]) <= 1.5 then goto valores1
end;
end;
valores1:
x1[j]:=FloatToStr(utm_e1[j]);
y1[j]:=FloatToStr(utm_n1[j]);
z1[j]:=FloatToStr(cota_float[i]);
end;
//compara valores para vertice 2
for j:=1 to n_j do
begin
for i:=1 to n_i do
begin
if abs(utm_e2[j]-x_float[i]) <= 1.5 then
begin
if abs(utm_n2[j]-y_float[i]) <= 1.5 then goto valores2
end;
end;
valores2:
x2[j]:=FloatToStr(utm_e2[j]);
y2[j]:=FloatToStr(utm_n2[j]);
z2[j]:=FloatToStr(cota_float[i]);
end;
//imprime no memo3
for j:=1 to n_j do
begin
memo3.Lines.Add(x1[j]+ +y1[j]+ +z1[j]);
memo3.Lines.Add(x2[j]+ +y2[j]+ +z2[j]);
end;
// comea a calculeira
for j:=1 to n_j do
begin
x1_int:=StrToInt(x1[j]);
x2_int:=StrToInt(x2[j]);
y1_int:=StrToInt(y1[j]);
y2_int:=StrToInt(y2[j]);
z1_int:=StrToInt(z1[j]);
z2_int:=StrToInt(z2[j]);
// linha N-S
if x1_int = x2_int then
dx:=0
else
dx:=abs(x1_int-x2_int);
// linha E-W
if y1_int = y2_int then
dy:=0
else
dy:=abs(y1_int-y2_int);
dx2:=dx*dx;
dy2:=dy*dy;
dh:=sqrt(dx2 + dy2);
dz:=abs(z1_int-z2_int);
// linha horizontal
if z1_int = z2_int then
grad:=0 else
grad:=dz/dh;
meiox:=utm_e_mid[j];
meioy:=utm_n_mid[j];
memo4.Lines.Add(FloatToStr(meiox)+ +FloatToStr(meioy)+
+FloatToStr(grad));
end;
end;

var
org,d_date,d_name,m_name,m_date,m_scale,other,zone:string;
west,east,south,north,thresh,verti,x1_j,x2_j,y1_j,y2_j:string;
L:array [1..1500] of char;
utm_e1,utm_e2,utm_n1,utm_n2:array [1..1500] of real;
utm_n,utm_e:array [1..1500] of real;
j,k,n_vertex:integer;
espaco:string[2];
begin
//limpa o memo 1
memo1.lines.clear;
//inicia leitura do arquivo dig-ascii
reset(entrada1);
readln(entrada1,org);
readln(entrada1,d_date);
readln(entrada1,d_name);
readln(entrada1,m_name);
readln(entrada1,m_date);
readln(entrada1,m_scale);
readln(entrada1,other);
readln(entrada1,zone);
readln(entrada1,west);
readln(entrada1,east);
readln(entrada1,south);
readln(entrada1,north);
readln(entrada1,thresh);
readln(entrada1,verti);
//inicia j,k. leitura das coordenadas do arquivo dig_ascii

j:=1;
k:=1;
while not (eof(entrada1)) do
begin
read(entrada1,L[j]);read(entrada1,espaco);readln(entrada1,n_vertex);
for k:=1 to n_vertex do
begin
read(entrada1,utm_n[k]);readln(entrada1,utm_e[k]);
utm_n[k]:=Round(utm_n[k]);
utm_e[k]:=Round(utm_e[k]);
end;
utm_e1[j]:=utm_e[1];
utm_e2[j]:=utm_e[n_vertex];
utm_n1[j]:=utm_n[1];
utm_n2[j]:=utm_n[n_vertex];
// imprime vrtices no memo1

memo1.Lines.Add(x1_j+ +y1_j+ 1');
memo1.Lines.Add(x2_j+ +y2_j+ 1');
j:=succ(j);
end;
end;
end.
ANEXO 5
Serra do Caraa (Quadriltero Ferrfero, MG):

parmetros morfomtricos e consideraes tectnicas preliminares
Resumo apresentado ao III Workshop Cientfico de Ps-Graduao do IGc-USP

(abril 2003)
SERRA DO CARAA (QUADRILTERO FERRFERO, MG):

PARMETROS MORFOMTRICOS E CONSIDERAES TECTNICAS PRELIMINARES
Carlos Henrique Grohmann de Carvalho 1 guano@usp.br
O sul do crton do So Francisco dominado por complexos domeados granito-gnissicos arqueanos

penecontemporneos ao greenstone belt Rio das Velhas. A aglutinao Brasiliana controla o cenrio tectnico.
O Grupo D. Silvrio, extenso sul da Faixa Araua, encontra-se no front dos Gnaisses Mantiqueira, e
marca o limite oriental dos terrenos granito-greenstone e da margem passiva Minas.
Deformaes tectnicas cenozicas afetaram toda a Plataforma Brasileira, reativando, em regime
transcorrente geral, os principais lineamentos precambrianos brasileiros.
Para o estudo das movimentaes cenozicas atuantes na configurao atual do relevo, prope-se a
correlao entre dados geolgicos, imagens de satlite e modelagem digital de terreno, em ambiente integrado
de Sistema de Informaes Geogrficas SIG.
A anlise morfomtrica foi realizada em uma rea de aproximadamente 1400 km2, correspondente
folha topogrfica 1:50.000 Catas Altas (SF-23-X-B-I-1), borda leste do Quadriltero Ferrfero. O ambiente
SIG utilizado foi o sistema GRASS 5.0.2 (Geographical Resources Analysis Support System), um projeto
livre e de cdigo aberto, disponvel gratuitamente via internet em http://grass.itc.it.
A rede de drenagem e as curvas de nvel foram digitalizadas e importadas para o sistema como
entidades vetoriais. As curvas de nvel foram convertidas para formato raster (matricial) e interpoladas
pelo mtodo de spline regularizado com tenso para gerar um modelo numrico de terreno, de grade regular,
com 50m de resoluo espacial. A partir do modelo numrico de terreno foram calculados valores de
declividade e de orientao de vertentes para clulas de 250x250m.
Para o traado de lineamentos morfoestruturais, foram utilizados quatro mapas de relevo sombreado,
com iluminante artificial posicionado 20 acima do horizonte, em N90, N45E, N00 e N45W.
O mapa de densidade de lineamentos expressa a relao de comprimento por rea (km/km2) de clulas
com 500x500m. Os mesmos parmetros foram usados no mapa de densidade de drenagem. O Mapa de
isobases foi feito manualmente, a partir da interseco das curvas de nvel com drenagens de 2a e 3a ordem.
O mapa de rugosidade de relevo foi calculado a partir da relao entre a rea real e a era planar de
clulas com 1000x1000m.
A anlise conjunta dos dados morfomtricos permitiu a determinao de estruturas orientadas NESW, NW-SE e NNE-SSW.
As estruturas NE-SW concordam, de maneira geral, com a foliao metamrfica regional, e sua
presena em mapas morfomtricos pode ser vista como resultado de reativaes, com abatimento prograssivo
para norte.
Orientaes segundo NW-SE e NNE-SSW, concordam com zonas de cavalgamento de baixo ngulo,
e podem tambm estar reativadas como falhas rpteis normais, responsveis pelo atual cenrio morfolgico,
com uma rea central deprimida em relao s suas bordas.
Lipski, M.; Endo, I.; Castro, P.T.A.; Trzazkos-Lipski, B., 2001. Estudo do campo de tenses Cenozicas no Quadriltero Ferrfero, Minas
Gerais. VIII Simp. Nac. Est. Tec, UFPE
1 - Bolsista CNPq Programa de Ps-graduao em Geoqumica e Geotectnica

Orientador: Prof. Dr. Mario da Costa Campos Neto
ANEXO 6
Serra do Caraa (Quadriltero Ferrfero, MG):

morphometric parameters and preliminary tectonic considerations
Resumo expandido apresentado ao

IX Simpsio Nacional de Estudos Tectnicos / III International Symposium on Tectonics
(maio 2003)
Serra do Caraa (Quadriltero Ferrfero, MG): morphometric parameters and preliminary

tectonic considerations
Carlos Henrique Grohmann de Carvalho (guano@usp.br) - IGc-USP - CNPq scholarship holder
Mario da Costa Campos Neto IGc-USP
Introduction
The Brazilian Platform was affected in all its extension by cenozoic tectonic deformations which
reactivated the main precambrian brazilian lineaments, resulting in a subdivision of blocks bounded by
crustal descontinuities. (3, 10).
The southern portion of the So Francisco Craton is marked by a Precambrian N50W-trending shear
zone, renewed in the Quaternary (10). Erosional surfaces in the Quadriltero Ferrfero have close relations
with litostructural domains (11). Reactivations with vertical displacements, of ancient faults, would be
responsible for the present-day altimetric level differences (6, 1).
Evidences of cenozoic tectonics are observed in several sedimentary deposits, and indicate three
deformational events with distinct stress fields. The first one, extensive, with NNE-SSW orientation and
probably related with grabens and horsts oriented ESE-WNW. A second event, more expressive, with NWSE compression. The third event is considered the relaxing of the previous one (7).
This paper intents to present morphometric analysis made with Geographical Information System
(GIS) for the eastern portion of the Quadriltero Ferrfero (fig.1), as well as their possible relations with
recent tectonic events.
Methods
The morphometric analysis were carried out for an area of about 1400 km2, correspondent to the
1:50.000 topographic sheet Catas Altas (SF-23-X-B-I-1). The main GIS used was the GRASS open source
GIS (8).
The topographic contour lines and drainage network were digitized from the topo sheet and imported
into the system as vector entities. Contour lines were converted to raster format and interpolated to produce
a Digital Elevation Model (DEM) with 50m of spatial resolution.
Four shaded relief maps, with lighting at N90, N45E, N00, N45W, and with inclination of 20
above the horizon, were used to trace lineaments (9).
A lineament density map was produced as the relation of total length per area (km/km2) of cells with
500x500m. The same parameters were used in the drainage density map (5). The map of isobase (2) was
elaborated manually.
The surface roughness map (4) was calculated as the ratio beetwen the real (surface) area and the flat
(plan) area of cells with 1x1km. This method is useful for morphological compartimentation since it considers
the shape and not the altitude. Thus, tilted reliefs have their expression shown, while it could be masked in
a hypsometric map, as consequense of altimetric variations.
Results
The surface roughness map (fig.2) shows a relatively flat area in the central region, sorrounded by
areas with higher roughness. A similar schema can be seen in the map of isobase (fig.3). Inferred structures
are represented in both maps, and were used to delimit five unformal morphological domains (north, south,
east, west and central). Lineament and drainage densities agree with orientation of these structures.
Morphostructural lineament analysis indicate the presence of main trends oriented NE-SW and NNESSW through the whloe area. Secondary trends are oriented N-S, NNW-SSE and NW-SE (fig.4).
The northern morphological domain comprises an essentially plan area, with lineament frequency
peak at the NE direction, and length peaks at NE, NNW and NW. The eastern and southern domains have
greater roughness values and lineaments oriented mainly NE-SW and NW-SE; ENE-WSW trend occurs
only in the southwest of the area.
The western domain comprises almost only the quartzite massif of the Serra do Caraa. It has the
biggest roughness values, and is a strong isobase anomaly. Lineament orientation differs from all the other
domains, with high counting of N-S, NW-SE and E-W trends.
The central domain encloses the Chapada de canga, a 900m-high plateaus supported by laterite.
Surface is nearly flat, and lineaments show all directions mencioned above.
Slope and aspect (fig.5) maps show, in the south and north domains, NE-SW lined ridges, with
steepest slope in the NW face. This orientation agrees with the regional metamorphic foliation and should
be related to it. N-S and NW-SE lined ridges are less commom, best viewed in the east and west domains,
and the related lineaments are probably younger.
Conclusions
Within the study of recent geologic evolution of the eastern border of the Quadriltero Ferrfero,
several morphometric methods were applied in a GIS environment. The system used was the GRASS opensource GIS, availabe at no cost for the user.
In the study area, main structures could be inferred (fig.6). NE-SW structures agree with the
metamorphic foliation, and its presence in morphometric maps can be seen as reactivations, with progressive
abatement to north.
Structures oriented NW-SE and NNE-SSW, agree with low-angle thrusts, and may also be reactivated
as brittle faults, responsibles for the present morphological scenario, with a depressed central areal,
sorrounded by uplifted hills.
References
1 - Barbosa, G.V. 1980. Superfcies de eroso no Quadriltero Ferrfero. Rev.Bras. Geoc., 10(1):89101.
2 - Golts, S. & Rosenthal, E. 1993. A morphotectonic map of the northern Arava in Israel, derived
from isobase lines. Amsterdam: Geomorphology, 7: 305-315.
3 - Hasuy, Y. 1990. Neotectnica e aspectos fundamentais da tectnica ressurgente no Brasil. In:
Workshop Neotecto. Sedim. Cont. Cenoz. SE Brasil, 1, Belo Horizonte. Anais... Belo Horizonte, SBG-MG.
Boletim 11, p.1-31.
4 - Hobson, R.D. 1972. Suface roughness in topography: quantitative approach. In: Chorley, R.J.
(ed) Spatial analysis in geomorphology. Harper and Row, p.225-245.
5 - Horton, R.E. 1945. Erosional development of streams and their drainage basins: hydrophysical
approach to quantitative morphology. Geol.Soc.Am.Bull. 56(3):275-370.
6 - King, L. 1956. Geomorfologia do Brasil Oriental. Rev. Bras. Geog., 18(2):147-266.
7 - Lipski, M.; Endo, I.; Castro, P.T.A.; Trzazkos-Lipski, B. 2001. Estudo do campo de tenses
Cenozicas no Quadriltero Ferrfero, Minas Gerais. VIII Simp. Nac. Est. Tec, UFPE
8 - Neteler, M; Mitasova, H. 2002. Open Source GIS: A GRASS GIS Approach. Boston, Kluwer
Academic Publishers, 464 p. (software available at http://grass.itc.it)
9 - Riccomini, C.; Crsta, A.P. 1988. Anlise preliminar de lineamentos em imagens de sensores
remotos aplicada prospeco mineral na rea dos granitides Mandira, SP. Bol.Inst.Geoc. USP. Srie
Cientfica. 19:23-37
10 - Saadi, A. 1993. Neotectnica e tectnica recorrente na poro sul do Crton do So Francisco.
In: Simp.Crat. S. Franc., 2, Salvador. Anais...Salvador - SBG/BA. p.230-232
11 - Varajo, C.A.C. 1991. A questo da correlao das superfcies de eroso do Quadriltero Ferrfero,
44000'
43 30'
43000'
Nova Era
78
Caet
Caet dome
Sabar
00
Itacolomi / Espinhao Group
lin
Raposo
42030'
nc
44 30'
sy
Nova Lima
80
Serra
do Cu
rral
Rio das Velhas Supergroup
77
an
da
re
Ibirit
Minas Supergroup
20000'
la
Belo Horizonte
dome
Fig.1 - Simplified Geology of Southeastern

So Francisco Craton, Minas Gerais, Brazil
JOO MONLEVADE
BELO HORIZONTE
Alvinpolis
Moe
Brumadinho
da sy
SE So Francisco Craton Gneiss
nclin
60
Neoproterozoic Belt
Rio Casca
77
ITABIRITO
60
Bonfim dome
Dom Silvrio Group
Bao dome
Acaiaca
MARIANA
77
PONTE NOVA
40
Mantiqueira Gneisses
OURO PRETO
Belo Vale
80
10
Juiz de Fora Granulites
CONGONHAS
00
Ouro Branco
20030'
Study area
Santa Rita dome
40 km
77
20
6
20
40
60
80
00
20
40
60
Modified from Dorr (1969), Silva,Baltazar and Brandalise (1991), Raposo (1991).
740
660
760
680
720
740
720
640
700
700
740
760
1300
1200
700
740
900
1800
800
780
720
760
720
1240
1300
860
820
680
840
700
760
800
900
Curves
1
1.05
1.1
1.15
1.2
1.25
1.3
740
780
940
720
800
860
740
680
780
820
780
720
600
580
680
600
Fig.2 - Surface roughness map
Fig.3 - Map of isobase
N
NE
Frequency
10%
E
SE
S
10%
SW
W
NW
Length
Fig.4 - Interpreted lineaments
Fig.5 - Aspect map
Frequency
10% 5
10%
Length
Frequency
5
10%
10%
Length
Frequency
10%
10%
Length
Frequency
10%
10%
Length
Frequency
20%
10
10
20%
Length
Fig.6 - Morphological scenario proposed
ANEXO 7
Anlise morfomtrica em Sistemas de Informao Geogrfica:

aplicao dos softwares livres GRASS e R
Resumo apresentado ao VIII Simpsio de Geologia do Sudeste

(outubro 2003)
Anlise morfomtrica em Sistemas de Informao Geogrfica:

aplicao dos softwares livres GRASS e R
Carlos Henrique Grohmann de Carvalho guano@usp.br IGc-USP - Bolsista CNPq
A confeco e interpretao de mapas morfomtricos uma importante ferramenta em estudos ligados

neotectnica e geomorfologia, onde a resposta da paisagem natural frente dinmica interna do planeta
muitas vezes mascarada pela rpida ao dos agentes intempricos, notadamente em regies tropicais e
sub-tropicais.
A utilizao de Sistemas de Informao Geogrfica (SIG), permite maior rapidez e preciso neste
processo, porm a metodologia empregada depende das ferramentas disponveis e do grau de conhecimento
de cada pesquisador com relao aos programas de computador utilizados. Este trabalho busca uma breve
avaliao dos procedimentos necessrios para a confeco e correlao de mapas morfomtricos com a
utilizao integrada dos programas livres GRASS (http://grass.itc.it) e R (www.r-project.org).
Os parmetros mais utilizados neste tipo de anlise so declividade, orientao de vertentes, densidade
de drenagens e de lineamentos, rugosidade de relevo, superfcies de base e gradiente hidrulico.
Inicialmente, deve-se gerar o modelo numrico de terreno (MNT), a partir da interpolao de pontos
cotados ou de curvas de nvel (em formato matricial ou vetorial), que ser utilizado para calcular os mapas
de declividade e orientao de vertentes. A rede de drenagem deve estar em formato vetorial e devero ser
preparadas duas verses diferentes, uma com as drenagens de 2 e 3 ordens para extrao das linhas de
base e outra com as drenagens de 2 ordem desde suas cabeceiras para o clculo do gradiente hidrulico. Os
lineamentos podem ser identificados e digitalizados diretamente no mdulo de edio vetorial, sobre imagens
de relevo sombreado.
Para determinar as densidades de drenagens e lineamentos, estes elementos devem ser convertidos
para formato raster (matricial) e em seguida para um arquivo de pontos, que ser usado na contagem do
comprimento acumulado em cada cela, de acordo com a resoluo escolhida. Para a rugosidade de relevo,
a rea real das celas obtida das relaes geomtricas entre a grade regular e a declividade.
O mtodo das isobases pode ser aplicado ao se converter em formato raster o arquivo de drenagens
preparado anteriormente, e multiplic-lo com as curvas de nvel, tambm em formato raster. O conjunto de
pontos obtido pode ser interpolado linearmente ou pelo mtodo de splines regularizados com tenso.
O mapa de gradiente hidrulico envolve a transformao da drenagem vetorial para raster, seu
cruzamento com o MNT e o uso de uma rotina especfica desenvolvida em linguagem Object Pascal para
o clculo automtico dos vrtices extremos de cada drenagem, sua posio espacial (x,y,z) e consequente
determinao dos gradientes que so atribudos ao ponto mdio de cada segmento.
A anlise dos parmetros obtidos pode ser feita com o pacote estatstico R, atravs de uma interface
de comunicao com o GRASS, o que possibilita tratar os mapas raster e de pontos como variveis para a
confeco de histogramas e anlises de regresso.
A integrao entre Sistemas de Informao Geogrfica e [geo]estatstica na anlise morfomtrica
permite maior agilidade e preciso na determinao dos parmetros necessrios. A utilizao de ferramentas
livres e de cdigo aberto garante o acesso a todos, e sua crescente popularizao abre novas perspectivas de
desenvolvimento neste campo.

Dissertacao Mestrado Grohmann PDF

Transféré par

Informations du document

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Dissertacao Mestrado Grohmann PDF

Transféré par

Droits d'auteur :

Formats disponibles

UNIVERSIDADE DE SO PAULO

Carlos Henrique Grohmann

Orientador: Prof. Dr. Mario da Costa Campos Neto

Todas as cincias exatas so dominadas pela idia da aproximao

Figura 1 - Localizao da rea de estudo ..............................................................................5

Anexo 2 - Trend-surfaces analysis of morphometric parameters

Anexo 5 - Serra do Caraa (Quadriltero Ferrfero, MG): parmetros morfomtricos e

Anexo 6 - Serra do Caraa (Quadriltero Ferrfero, MG): morphometric parameters

Anexo 7 - Anlise morfomtrica em Sistemas de Informao Geogrfica: aplicao dos

1.1 Apresentao ao tema

portanto, a necessidade de integrao dessas ferramentas, visando, alm de maior simplicidade

As unidades basais do Super Grupo Minas (Grupos Tamandu, Caraa, Itabira e

principais lineamentos precambrianos brasileiros, como o caso da Descontinuidade Crustal

Figura 2 Relevo sombreado da rea de estudo. Iluminante a N45, inclinao de 30

Nc - Fm. Chapada de Canga:

MPei - Grupo Espinhao/Itacolomi

Arv - Supergrupo Rio das Velhas:

Falha ou zona de cisalhamento

Contato, contato aproximado

Sta Rita Duro

Projeo Universal Transversa de Mercator

Pd - Diques e corpos bsicos intrusivos

(a declinao magntica cresce -5,5 a cada ano)

Figura 3 - Mapa geolgico simplificado da rea de estudo

2.2 O Modelo Digital de Terreno

As Redes Triangularizadas Irregulares (Triangulated Irregular Network TIN) uma

3 Anlise morfomtrica em Sistemas de Informao Geogrfica: aplicaes dos

4 Anlise de superfcies de tendncia de parmetros morfomtricos.

Figura 4 Lineamentos interpretados e diagrama de roscea.

Figura 5 Perfis em varredura; localizao sobre imagem de relevo sombreado.

Com a anlise morfomtrica, diversas estruturas puderam ser inferidas na rea de

Figura 6 Mapas morfomtricos com estruturas inferidas sobrepostas.

As linhas de isobase delimitam superfcies erosionais, portanto as superfcies de base

Figura 7 Vista tri-dimensional da superfcie de base, a partir de ESE.

O mapa de gradiente hidrulico (Figura 6E) ressalta principalmente as escarpas da Serra

reconhecimento de armadilhas estruturais, feies paleogoegrficas ou quebras no estilo

Figura 8 Superfcies de tendncia e mapas de resduos para os

Artigo submetido ao peridico internacional Computers & Geosciences

Morphometric analysis in Geographic Information Systems:

Ouro Branco syncline

Santa Rita dome

2. Geological and geomorphological context

3.1. Lineament Analysis

3.2. Hypsometry, slope and aspect

Figure 4 A) Slope map.

3.3. Swath profiles

Figure 6 Swath profiles. Localization of profiles on the shaded relief image.

3.4. Drainage and lineament density

Figure 7 A) Drainage density map.

B) Lineament density map.

3.5. Surface roughness

a = grid cell size

Figure 8 A) Surface roughness map, for cells with 1000x1000m.

B) Geometric relations between regular grid and slope.

Figure 9 A) Automatically generated isobase map.

B) Manually interpolated isobase map.

y=74.3 + 0.919x R=0,96

Figure 9 - C) Linear regression between automatic and manual isobase.

3.7. Hidraulic Gradient

Figure 10 Hydraulic gradient map.

y=1.01 + 0.21x R=0,3

Figure 11 Linear regression between hydraulic gradient and surface roughness.

by Hamish Bowman 11 April 2003