FLUVIAL EROSION RISK ANALYSIS: AN AMAZON STUDY CASE 1 REVISTA GEONORTE, V.12, N.39, p.01-25, 2021. (ISSN 2237 - 1419) DOI: 10.21170/geonorte.2021.V.12.N.39.01.25 FLUVIAL EROSION RISK ANALYSIS: AN AMAZON STUDY CASE Análise de risco a erosão fluvial: um estudo de caso na Amazônia Iris Celeste Nascimento Bandeira Geological Survey of Brazil iris.bandeira@cprm.gov.br Raimundo Almir Costa da Conceição Geological Survey of Brazil raimundo.conceicao@cprm.gov.br Milena Marília Nogueira de Andrade Federal Rural University of the Amazon. milena.andrade@ufra.edu.br Sheila Gatinho Teixeira Geological Survey of Brazil sheila.teixeira@cprm.gov.br Dianne Danielle Farias Fonseca Geological Survey of Brazil dianne.fonseca@cprm.gov.br Joao Batista Marcelo de Lima Geological Survey of Brazil joao.marcelo@cprm.gov.br Andressa Macedo Silva de Azambuja Geological Survey of Brazil andressa.azambuja@cprm.gov.br Aderson Manoel da Silva Gregorio Centro de Instrução Almirante Braz de Aguiar aderson.gregorio@marinha.mil.br Luciana de Jesus Penha Pamplona Miyagawa Geological Survey of Brazil luciana.miyagawa@cprm.gov.br Silvio Tadeu Teles da Silva Estácio Nazaré College, Belém-Pará, Brazil silvio.teles@estacio.br Paulo Afonso Pereira Aguiar Geological Survey of Brazil paulo.aguiar@cprm.gov.br Recebido em 26/11/2020 Aceito em 30/04/2021 ABSTRACT: In the Amazon region, there are more than 69.000 people living in areas at risk of fluvial erosion processes. In addition to the large number of people impacted, studies have shown that the erosion patterns identified on the margins of mega rivers in the Amazon region are distinct due to the fact they are related the mass movement leading to great soil displacement known as ‘Terras Caidas’. In this context, this study aims to evaluate quantitatively the degrees of risk in areas subject to fluvial erosion in three communities: Itanduba, São Braz, and Fátima de Urucurituba. The methods include hazard attributes, as well as vulnerability aspects, through the Analytic Hierarchy Process (AHP). A multitemporal analysis were made to validated the marginal erosion at the studied areas. The results indicated a high risk of fluvial erosion on these areas. The local families lives under high and very high social vulnerability in conditions with little infrastructure and very close to the susceptible erosive riverbank. The riverbank is composed of poorly consolidated sediments, show instability indicators, and are usually associated to drainages with flow rates above 100.000m3/s. The results and methodology brings an important contribuition to territorial planning of the region. Keywords: Analytic Hierarchy Process - AHP, Hazard, Vulnerability, Pará, Brazil. mailto:iris.bandeira@cprm.gov.br mailto:raimundo.conceicao@cprm.gov.br mailto:milena.andrade@ufra.edu.br mailto:sheila.teixeira@cprm.gov.br mailto:dianne.fonseca@cprm.gov.br mailto:joao.marcelo@cprm.gov.br mailto:andressa.azambuja@cprm.gov.br mailto:aderson.gregorio@marinha.mil.br mailto:luciana.miyagawa@cprm.gov.br mailto:silvio.teles@estacio.br FLUVIAL EROSION RISK ANALYSIS: AN AMAZON STUDY CASE 2 REVISTA GEONORTE, V.12, N.39, p.01-25, 2021. (ISSN 2237 - 1419) DOI: 10.21170/geonorte.2021.V.12.N.39.01.25 RESUMO: Na região amazônica, existem mais de 69.000 pessoas vivendo em áreas de risco a processos de erosão fluvial. Além do grande número de pessoas impactadas, estudos têm mostrado que os padrões de erosão identificados nas margens dos mega rios da região amazônica são distintos, por estarem relacionados ao movimento de massa que causam grandes deslocamentos de terra, denominados como “Terras Caidas”. Nesse contexto, o presente estudo tem como objetivo avaliar quantitativamente os graus de risco de áreas sujeitas à erosão fluvial, em três comunidades: Itanduba, São Braz e Fátima de Urucurituba. O método utilizado inclui avaliação dos atributos do perigo, assim como aspectos de vulnerabilidade que foram hierarquizados através do Processo Analítico Hierárquico (AHP). Uma análise multitemporal foi feita para validar a erosão marginal nas áreas estudadas. Os resultados indicaram alto risco de erosão fluvial nessas áreas. As famílias locais vivem em situação de vulnerabilidade social alta e muito alta, em condições de pouca infraestrutura e muito próximas à margem de rio susceptível à erosão. Essas margens são compostas por sedimentos pouco consolidados, apresentando indicadores de instabilidade que geralmente estão associados a drenagens com vazões acima de 100.000m3/s. Os resultados e metodologia trazem uma importante contribuição para o planejamento territorial da região. Palavras-chave: Processo Analítico Hierárquico, Perigo, Vulnerabilidade, Pará, Brasil. INTRODUÇÃO Erosion is the third more common disaster in Brazil (TOLEDO; LANA, 2019) and is the third more common in the Amazon (TEXEIRA et al., 2019). In addition to being susceptible to erosion, the margins of the Amazon river also have a high potential for mass movement events such as land falls and landslides. The erosive process that occurs due to the high speed of the river flow modifies the geomorphology and increases the inclination and height of the riverbanks (CHARMAN; GRIFFITHS, 1993; OMNR, 2002; THORNE; OSMAN, 1988). Depending on the geometry of the margins (height and inclination) (MACFALL et al., 2014) and the geotechnical characteristics there can be a higher or lower potentiality for landslides (THORNE, 1991). Also, according to Aly El-Dien et al. (2014) and Bandeira et al. (2018) mass movement might also occur either when the river is in the flood period causing slumps or when it is in the ebb period, associated to lowering water levels of the river leading to landslides and creeps. The riverbank erosion is considered as a geological type of natural disaster according to the Brazilian Classification for Disaster (COBRADE). After majors Brazilian municipalities affected by mass movement, inundation, floods and erosion processes in 2011, the Federal Law 12.608/2012 was created. It instituted the National Policy on Protection and Civil Defence - PNPDEC as well as the National Plan for Risk Management and Natural Disaster Response (BRAZIL, 2012) aiming to prevent and, as a consequence, reduce social and economical losses related to natural disasters. Such plan instituted the execution of the mapping, description and classification of areas of high and very high geological risk in municipalities from all units of the federation. This activity has been officially performed by Geological Survey of Brazil – CPRM resulting in 1605 municipalities mapped throughout Brazil (GEOLOGICAL SURVEY OF BRAZIL - CPRM, 2020). From that amount two hundred and twenty-three municipalities were mapped in the hydrographic region of the Amazon river and 74 of them host areas with high and very high risk of fluvial erosion. On the mapped areas there are more than 69.000 people living along the margin of rivers at risk of experiencing material losses and physical damage related to fluvial erosion processes. The main losses refer to home appliances, properties and sometimes losses of lives (ANDRADE; SZLAFSZTEIN, 2019; BANDEIRA et al., 2018a). To aggravate the https://www.jstage.jst.go.jp/search/global/_search/-char/en?item=8&word=Ahmed+ALY+EL-DIEN FLUVIAL EROSION RISK ANALYSIS: AN AMAZON STUDY CASE 3 REVISTA GEONORTE, V.12, N.39, p.01-25, 2021. (ISSN 2237 - 1419) DOI: 10.21170/geonorte.2021.V.12.N.39.01.25 situation, this region shows low economic and social development indexes in comparison to other regions of Brazil (INSTITUTO DE PESQUISA ECONÔMICA APLICADA, 2019). In order to map such risks, the methodology of qualitative analysis and risk characterization (SUN et al., 2020) of the Ministry of Cities and the Institute for Technological Research (CARVALHO; GALVÃO, 2006; CARVALHO et al., 2007; PIMENTEL et al., 2020) has been applied to the entire national territory. This methodology was elaborated to satisfactorily assist the cases of mass movement and inundations. However, it still poses a challenge to objectively apply it on the Amazon hydrological and geomorphological systems and dimensions (SZLAFSZTEIN, 2018). Particularities from large rivers system are: extensive alluvial deposits susceptible to fluvial erosion, high precipitation index, fluvial hydrodynamics related to big and mega rivers (LATRUBESSE, 2008; ASHWORTH; LEWIN, 2012), and localised erosion as well as large soil displacements (slumps, creep and landslides) regionally known as Terras Caidas (BANDEIRA et al., 2018). Despite the existence of the studies on the fluvial erosional processes, their triggering factors and the risk faced by people who occupy the margins of the Amazon rivers, such as: Carvalho (2006), Passos; Soares (2017), Magalhães and Vieira (2018), Bandeira et al. (2018), Souza (2019), Silva; Andrade (2019) and on the sectorization of the risk from 2012 to 2019 conducted by Geological Survey of Brazil - CPRM (2020), there are no studies on risk assessment mapping taking into consideration Amazon geological, hydrological and socioeconomic features in the fluvial erosion processes, in their analysis. In this context, the study aim to analyse riverbank risk erosion in three areas in the Amazon. STUDY AREA Location The study area includes three communities located at the lower course of the Amazon river: (i) Itanduba, (ii) São Braz and (iii) Fátima de Urucurituba (Figure 1). These areas are in the rural zone of Prainha, Porto de Moz and Santarém, respectively, located in the state of Pará. The criterio for selection this study areas where due to their locations at the margin of large rivers, and because it presents a high or very high risk of river erosion. FLUVIAL EROSION RISK ANALYSIS: AN AMAZON STUDY CASE 4 REVISTA GEONORTE, V.12, N.39, p.01-25, 2021. (ISSN 2237 - 1419) DOI: 10.21170/geonorte.2021.V.12.N.39.01.25 Figure 1. Location of areas at risk of fluvial erosion in the lower Amazon river. Source: Autor (2021). Physiography Geomorphologically, the three areas of study are included in the Amazon Plain and in the hydrographic basin of the Amazon river, the greatest river in the world with the greatest flow rate (ALBERT et al., 2018). The fluvial system displays river channel with anabranching pattern whereas its tributaries show meandering to straight forms; its margins are straight, concave and convex; 3 to 8 km wide; average flow rate between 59.000m³/s and 160.000m³/s and an approximate average depth of the river channel between 25 and 70m (MARINHA, 2020). The fluctuation of the water level in the period of lower river flow (September to February) is that of 3m whereas in the period of higher water level (March to August) it is 1,50m (AGÊNCIA NACIONAL DE ÁGUAS, 2019). The study area is slightly influenced by the dynamic tide with amplitude variations between 10 to 50cm/ day depending on the ebb and flow periods of the river (GALLO; VIZON, 2005). Such fluvial hydrodynamics are associated to a tropical climate whereas the highest precipitation indexes occur from February to May with a monthly average above 300 mm whereas the lowest average rates occur from August to November, sometimes below 40 mm (AGÊNCIA NACIONAL DE ÁGUAS, 2020). The areas bordering the Amazon river on all three stretches focused on this study, are geologically characterized by extensive alluvial deposits (VASQUEZ; ROSA-COSTA, 2008) composed of sandy and silty-clayey sediments of low geomechanical resistance and low to moderate cohesion. The Land Use and Land Cover includes pasture lands where there is low to moderate forest formation overall the studt areas (MAPBIOMAS, 2019). FLUVIAL EROSION RISK ANALYSIS: AN AMAZON STUDY CASE 5 REVISTA GEONORTE, V.12, N.39, p.01-25, 2021. (ISSN 2237 - 1419) DOI: 10.21170/geonorte.2021.V.12.N.39.01.25 Socio-economic aspects The municipality of Santarém is the third largest by population in the state of Pará with a population of 294.580 where 73.2% are in the urban area IBGE (2010a). Approximately 3.340 live along the margins of the Amazon river near areas where several cases of fluvial erosion have occurred. The economy of Santarem is based on services, commerce, agropecuary, agroindustry and ecotourism (IBGE, 2010a). Porto de Moz is that of 33.956 inhabitants where 57% in the rural zone (IBGE, 2010b) noting that approximately 5.247 people live near the margins of the river with high susceptibility to fluvial erosion processes. The economic activities of the municipality are based on farming, pecuary, plant extractivism and subsistence fishing (IBGE, 2010b). Around 29.349 inhabitants lives in Prainha is that of where 69.5% in the rural zone (Brazilian Institute of Geography and Statistics, 2010c) noting that approximately 531 people live near the margins of the river. The economy of Prainha is linked to agropecuary, industry and general services (IBGE, 2010c). METHODOLOGY The study was divided into the following steps: (1) Definition of the hazard and vulnerability criteria and attributes for the study area; (2) Application of the multi-criteria analysis (AHP); (3) Assessment and classification of the risk of fluvial erosion and (4) Validation of the degrees of risk related to the erosion rate. Also the number of affected people in each community were counted by visual analysis. Criteria and attributes used in the hazard and vulnerability assessment The criteria for fluvial erosion hazard assessment considered the particularities for the event in the Amazon region according to Bandeira et al. (2018). Therefore, the following criteria were selected: lithology of the margin, morphology of the margin, height of the riverbank slopes, instability indicators, flow rate, waves and vegetation (Table 1). According to Couper (2003) the composition of the lithology plays a very important role in relation to the stability of the riverbank where the rate of erosion increases as the silt/clay content along the sandy margins submitted to the same flow speed decreases. For the study area, the expansive clay poses a greater risk of erosion than sediments that, in turn, are more susceptible than clay-bearing sediments. The latter posing more risk of erosion than rocks. The lithology on the river margins is one of the factors that distinguishes the fluvial erosion in Amazon and associates this process to the mass movement processes (BANDEIRA et al., 2018). The flow rate of the rivers is higher in the river channel and plays a role in the erosion rate when is near the margin riverbank. The highest flow rates occur during the flood period and are related to areas where the most intense erosional processes occur (BANDEIRA et al., 2018). Once the prossess is in course, instability indicators tend to occur. The morphology of the drainage influence in the resistance to fluvial erosion (AVILA et al., 2014). According to these authors different susceptibility to marginal erosion dependes on the hydrodynamics of the river, as well as curvier forms, straighter or the width of the drainage. According to Aly El-Dien et al. (2014) and Bandeira et al. (2018) mass movement might also occur either when the river is in the flood period causing slumps or when it is in the ebb period. The authors alo associated to lowering water levels of the river leading https://www.sciencedirect.com/science/article/abs/pii/S0169555X03000485#! https://www.jstage.jst.go.jp/search/global/_search/-char/en?item=8&word=Ahmed+ALY+EL-DIEN FLUVIAL EROSION RISK ANALYSIS: AN AMAZON STUDY CASE 6 REVISTA GEONORTE, V.12, N.39, p.01-25, 2021. (ISSN 2237 - 1419) DOI: 10.21170/geonorte.2021.V.12.N.39.01.25 to landslides and creeps related to fluctuations of the piezometric surface causing changes in the physical and hydraulic features, the pore pressure, decrease in the matrix suction, and decrease of the resistance to shear. Since the inclination of the riverbank in the study area is usually higher than 700, therefore only its height was used as one of the geometrical criteria to be added to the hazard assessment. For waves attribute, this study classification criteria uses waves associated to the wind the Beaufort wind force scale (MARINHA, 2019). As for the vegetation, according to Coppin; Richards (1990) and field observations revealed that the existence of vegetation along the margins of large rivers can cause positive or negative impacts on the stability of the riverbanks. But it will have little influence on the erosional process when compared to other conditioning factors analysed. The criteria for vulnerability assessment consider type of housing, distance of the riverbank, retaining works and family income. The type of housing, wooden or bricken houses, influences how the material responds to the hazard pressure (MINISTÉRIO DAS CIDADES, 2007). The distance of the riverbank takes into account the erosion rates analysed in this study and the data achieved by Teixeira (2017) and Teixeira et al. (2018). Proximity of the houses and facilities to the riverbank maximize the consequences caused by the contact of the population and hazards (GHOSH; SAHU, 2019). Retaining works are included as structutral measure to prevent or adapt facing erosion (SILVA; ANDRADE, 2019). For Amazon, the income has been considered in previous studies for vulnerability analysis in natural hazard context (ANDRADE et al., 2010; ANDRADE; SZLAFSZTEIN, 2018). FLUVIAL EROSION RISK ANALYSIS: AN AMAZON STUDY CASE 7 REVISTA GEONORTE, V.12, N.39, p.01-25, 2021. (ISSN 2237 - 1419) DOI: 10.21170/geonorte.2021.V.12.N.39.01.25 Table 1. Criteria and attributes used in the Risk Criteria Atributes Obtaining the data Justification References H a z a rd L it h o lo g y Sedimentary saprolite or profile of alteration, Clay sediments, Silt sediments, Sandy sediments, Expansive clay Field observations, granulometric analysis using the laser light scattering method, sieving and x- ray diffraction assays were conducted. Also, indirect calculations of the soil erodibility factor were carried out. The composition of the lithology plays the most important role in relation to the stability of the riverbank where the rate of erosion increases as the silt/clay content along the sandy margins submitted to the same flow speed decreases. Thorne (1991); OMNR (2002); Maskare (2015); Abidin et al. (2017); Couper (2003); Bandeira et al. (2018). F lo w r a te 500 to 999m3/s, 1.000 to 9.999m3/s, 10.000 to 49.999m3/s, 50.000 to 99.999m3/s, above 100.000m3/s. Field data collected using a 600KHz ADCP - Acoustic Doppler Current Profile and databank of agência Nacional das Águas Agency (2019) and information Molinier et al. (1994). The flow rate besides being responsible for the transportation of sediments it defines, to a greater or lesser degree, both the depositional and erosional rates of the margins and increases the excess shear stress on the riverbank. FEMA (1999); OMNR (2002); Knighton (1998); Stevaux; Latubresse (2017); Nardi et al. (2013). In s ta b ili ty i n d ic a to rs Small stepped landslides called terracettes, cracks in the soil and in the houses, failure scars, erosional features, inclined trees and utility posts. Observation and recorded on site. The instability indicators also contribute to the risk assessment of a given area mainly the fluvial erosion in large rivers. Thus, the greater the number of indicators the greater the hazard degree. However, it has been noted there exist areas that show no indicators and yet they are affected by land slumps. Carvalho et al. (2007); Bandeira et al.(2018). H e ig h t o f th e r iv e rb a n k Height is subdivided into the following categories: below 2m, between 3 and 6m, between 7 and 10m and above 10m. Preliminarily identified based on bathymetric charts (MARINHA, 2020) and obtained during field work. Depending on the geometry of the margins (height and inclination) there can be a higher or lower potentiality for erosion and landslides. Charman; Griffiths (1993); OMNR (2002); Thorne; Osman (1988); Macfall et al., (2014); Thorne (1991). FLUVIAL EROSION RISK ANALYSIS: AN AMAZON STUDY CASE 8 REVISTA GEONORTE, V.12, N.39, p.01-25, 2021. (ISSN 2237 - 1419) DOI: 10.21170/geonorte.2021.V.12.N.39.01.25 Criteria Atributes Obtaining the data Justification References H a z a rd M o rp h o lo g y River confluences, concave margins, convex and straight morphologies. Was obtained by visually analysing Geoeye imagery from years 2000, 2017 and 2018. The morphology of the drainage is intrinsically related to erosion. Areas that have the same resistance to fluvial erosion might show different susceptibility to marginal erosion depending on the hydrodynamics of the river, as well as curvier forms, straighter or the width of the drainage. It is also noted that at places of drainage confluence there is a high potential to erosion. Thorne (1991); Curran; Mcteague (2011); Derruau (1966); Avila et al. (2014). W a v e s Categories: grade 0 to 3 (Calm to gentle breeze) associated to mirrored water until 1m high undulations; 4 to 6 (Moderate breeze to very fresh wind) corresponding to 1m high undulations up to waves approximately 3.6m high; and above 7 (Strong wind to storms) relating to waves higher than 3.6 m. Was gathered on site based on the analysis of the Beaufort scale (MARINHA, 2019). The fluvial erosion is also affected by the impact of waves breaking at the margin of the river. In the study area where the rivers are very wide, kilometric at times, such waves exist due to two factors: intense wind and the transit of vessels. This study considers only waves caused by the wind as there are no studies evaluating waves generated by vessels. Bhowmik et al. (1982); Analysis of the Beaufort scale MARINHA (2019). V e g e ta ti o n Areas with no vegetation and areas with large to medium-sized vegetation. Field observation and visual analysis of satellite imagery (WorldView from 2014 and Geoeye from 2018) according to Monteiro et al. (2016). Large to medium-sized shrub species represent the most hazard as they can overload the soil and cause landslides, as well as trees exposed to the wind transmit dynamic forces to the soil favouring the destabilization of the marginal riverbank. However, it is noted that in areas with no vegetation the susceptibility to fluvial erosion is also high due to the increase of the surface runoff. However, the small- sized species accrete more resistance to the soil due to their light weight in relation to large- sized ones as well as roots. Coppin; Richards (1990); Sutili (2007); Mulyono et al. (2018); Gray; MacDonald (1989); Monteiro et al. (2016); Luna (2016); Peixoto et al. (2009); Maffra et al. (2017). FLUVIAL EROSION RISK ANALYSIS: AN AMAZON STUDY CASE 9 REVISTA GEONORTE, V.12, N.39, p.01-25, 2021. (ISSN 2237 - 1419) DOI: 10.21170/geonorte.2021.V.12.N.39.01.25 Criteria Atributes Obtaining the data Justification References V u ln e ra b il it y Type of housing Wooden houses Bricken houses Field observation and visual analysis. The quality of the material used to build houses is important as the structure has to withstand pressure and threat. Carvalho et al. (2007) Distance of the riverbank Edifications built more than 50 m away from the riverbank; distances between 20 and 50m; distances between 10 and 20m and very high vulnerability - edifications built less than 10m away from the riverbank Field observation and measuring. The distance between the houses and the marginal riverbank was also included, taking into consideration that the closer to the eroded margin the higher the vulnerability in occupied areas. Mollah; Ferdaush (2015) Retaining works no retaining works; existing and inefficient retaining works; and existing and efficient retaining works. Field observation and visual analysis. The retaining works criterion lessens the vulnerability and, consequently, the risk, as it minimizes the impacts on the terrain. Andrade; Szlafstein (2019). Family Income Less than the amount of 1 monthly minimum wage in Brazil; between 1 and 2 times the amount of the monthly minimum wage in Brazil, between 2 and 3 times the amount of the monthly minimum wage in Brazil and more than 4 times the amount of the monthly minimum wage. Information provided by inhabitants who occupy the risk area. The income is related to the ability to recover during the post-disaster phase. Cutter et al. (2003), Andrade et al. (2010) Source: Autor (2021). FLUVIAL EROSION RISK ANALYSIS: AN AMAZON STUDY CASE 10 REVISTA GEONORTE, V.12, N.39, p.01-25, 2021. (ISSN 2237 - 1419) DOI: 10.21170/geonorte.2021.V.12.N.39.01.25 Analytic hierarchy process (AHP) After establishing criteria and attributes, the degree of absolute relevance that rests on each one of them in the processes of risk of fluvial erosion and vulnerability was conducted using the AHP methodology by assigning grades and weights. This method is used to organise the criteria, subcriteria and alternatives of an analysis, in a descending order of importance. The given hierarchy contains relevant details that define the problem as completely as possible. This methods provides a general overview of the complexity related to the situation as well as assisting the decision maker analyse whether the problems have the same order of magnitude (SAATY, 1988). The Hierarchical Decision Process was built on three hierarchical levels, adapted according to Bathrellos; Skilodimou (2007): Level 1, object of the classification that is the risk assessment of fluvial erosion; Level 2, identification and relation between the criteria established to reach the goal; Level 3, identification and relation between the attributes or subcriteria. To perform that it was used the AHP online calculator from BPMSG (BUSINESS PERFORMANCE MANAGEMENT SINGAPORE). It is an online free tool used to implement the AHP available at https://bpmsg.com/ahp/ahp-calc.php. The hierarchy of relevance is created by a pairwise comparison square matrix n x n where the intensity of the relations between the rows and columns of such matrix is determined (VAIDYA; KUMAR, 2006). The comparisons are carried out following a scale that indicates how many times one variable is more important or more dominant than another, with respect to a given property on which they are compared. Based on this scale, 1 indicates an equal importance where both variables contribute equally to the objective and 9 indicates the maximum importance where a variable is that of the highest order of importance in relation to another (SAATY, 2008). The first step of the analysis is the fulfilment of the comparative judgements, therefore, the calculation of the Consistency Index (C.I) becomes necessary and represents the degree of reliability of the assessment, given by: I.C. = λmax − n n−1 (1). Where n is the order of the matrix and λmax is the maximum eigenvalue (comparative judgement scale) that can be reached by multiplying the judgement matrix by the priorities column vector. Exemplifying the AHP calculations applied to the previous case (Table 2). Table 2. Exemplifying the AHP calculation Lithology Flow rate Lithology 1 2 Flow rate 0,5 1 Source: Autor (2021). Applying the mathematical principle to calculate the eigenvalue on a second order matrix: [ 1 − 𝜆 2 0,5 1 − 𝜆 ] = (1 − 𝜆). (1 − 𝜆) − 2. (0,5) = 𝜆² − 2𝜆 Then, λ=0 or λ=2 are eigenvalues (comparative judgement scale) that correspond to consistency numbers. Considering v1 = Lithology and v2 = Flow rate. For λ=0, we have: [ 1 2 0,5 1 ] . [ 𝑣1 𝑣2 ] = [ 0 0 ] 1. 𝑣1 + 2. 𝑣2 = 0 0,5. 𝑣1 + 𝑣2 = 0 So: 𝑣1 = −2𝑣2 https://bpmsg.com/ahp/ahp-calc.php FLUVIAL EROSION RISK ANALYSIS: AN AMAZON STUDY CASE 2 REVISTA GEONORTE, V.12, N.39, p.01-25, 2021. (ISSN 2237 - 1419) DOI: 10.21170/geonorte.2021.V.12.N.39.01.25 For λ=2, we have: [ −1 2 0,5 −1 ] . [ 𝑣1 𝑣2 ] = [ 0 0 ] −1. 𝑣1 + 2. 𝑣2 = 0 0,5. 𝑣1 − 1𝑣2 = 0 So: 𝑣1 = 2𝑣2 Therefore, the eigenvalues obtained correspond to the priority order corroborating the analysis of the decision maker, i.e., the lithology (v1) is twice more important than the Flow rate (v2). Based on that, one can say that: I.C. = 2 − 2 2−1 = 0 ou 0,00% In percentage: 𝑣1 + 𝑣2 = 100% 2𝑣2 + 𝑣2 = 100% 3𝑣2 = 100% 𝑣2 = 33,33% (weight) Then, 𝑣1 = 66,66% (weight) The second and third criteria are associated to the assemblage of the pairwise comparison matrix assigning values from the comparative judgement scale in a pair to pair relation aiming to establish grades (for the criteria) and weights (for the attributesto prioritize the alternatives. The AHP online calculator from BPMSG (BUSINESS PERFORMANCE MANAGEMENT SINGAPORE) was applied to this study. It is an online free tool used to implement the AHP available at https://bpmsg.com/ahp/ahp- calc.php Risk assessment and classification In order to classify the risk, the grades and weights of the criteria and attributes obtained from the multicriterial analysis were applied to the Hazard, Vulnerability and Risk equations (Table 3). Such equations were included in the software used for calculations (Excel) and the analysis of the results for the Hazard criteria. Table 3. Formulas for hazard, vulnerability and risk. Equation 1 Equation 2 Equation 3 𝑯𝒊𝒋 = ∑(𝑾𝒌 𝒂 𝒌=𝟏 × 𝑮𝒌) Where: 𝑯 = Hazard 𝒂 = Number of threat criteria used 𝑾𝒌 = Weight of each attibute within the Threat criterion 𝑮𝒌 = Grade of the threat criterion 𝑉𝑖𝑗 = ∑(𝑊𝑧 𝑣 𝑧=1 × 𝐺𝑧) Where: 𝑉 = Vulnerability 𝑣 = Number of vulnerability criteria used 𝑊𝑧 = Weight of each attibute within the vulnerability criterion 𝐺𝑧 = Grade of the vulnerabilidade criterion Risk = 𝐻xV Source: Rebelo (2001). Validation of the risk degree associated to the erosion rate The erosion rate was applied to validate the results for the following reasons: (i) the eroded areas in the Lower Amazon river undergo a continuous erosional process with an average recess greater than 1,4km/year according to multitemporal analysis of satellite LANDSAT for a period of 31 years (TEIXEIRA, 2017; TEIXEIRA et al., 2018); (ii) it allows for an estimate of how much the river is eroding as its channels are continuously adjusting their morphology by means of erosional and depositional processes associated the effects of tectonism, climate change and sea level change https://bpmsg.com/ahp/ahp-calc.php https://bpmsg.com/ahp/ahp-calc.php FLUVIAL EROSION RISK ANALYSIS: AN AMAZON STUDY CASE 11 REVISTA GEONORTE, V.12, N.39, p.01-25, 2021. (ISSN 2237 - 1419) DOI: 10.21170/geonorte.2021.V.12.N.39.01.25 (MERTES; DUNNE, 2007) ; and (iii) unavailability of high spacial and temporal resolution imagery in the entire. A timeline analysis from 3 to 10 years based on Maxar Technologies and Digital Globe imagery, available on Google Earth PRO, on a 1:5000 scale was applied to the three study areas. The erosion rate was estimated from graphical inferrence analysis using the software Digital Shoreline Analysis System – DSAS 5.0 adapting the methods from Thieller et al. (2005). DSAS amplifies the use of software ArcGIS10.7 automating great part of the tasks related to the quantitative analysis of the evolution of the erosional and depositional trends from statistics on time series and multiple positions of the shoreline. This tool applies the analytical method End of Point Rate – EPR to calculate vector rate-of-change data between two shoreline horizons, by dividing the distance of shoreline movement between two distinct periods by the time period elapsed, thus, providing the annual rate of shoreline change (HIMMELSTOSS et al., 2018) (Table 4) . Table 4. Images used to analyze the fluvial erosion rate at the margins of the Amazon river. Municipality Satellite Year Spacial resolution Porto de Moz WorldView-1 08/2003 50cm WorldView-2 07/2014 47cm Prainha WorldView-2 10/2010 47cm WorldView-3 09/2018 37cm Santarém WorldView-2 08/2015 50cm Geoeye-1 07/2018 42cm Source: Autor (2021). The transects between the shoreline and the baseline to calculate the erosion rate were 5m apart in the Porto de Moz and Prainha studies and 20m apart in the Santarém studies. Based on the results on the shoreline changes on the investigated areas the erosion rate was subdivided into 4 intervals: lower than 1m/year, low index; 1 to 2m/year, moderate index; 2 to 3m/year, high index and higher than 3m/year, a very high index. RESULTS AND DISCUSSION Analytic hierarchy process (AHP) value Considering the attributes for risk analyses the pairwise comparison matrix and assigning values from the comparative judgement scale in a pair to pair, it was possible to establish grades (for the criteria) and weights (for the attributes) to prioritize the alternatives (Tables 5 and 6). Table 5. Pairwise comparison matrices for Hazard and Vulnerability to fluvial erosion criteria. Hazard criteria Grade Vulnerability criteria Grade Lithologyof the margin 0,183 Type of housing 0,35 Flow rate 0,169 Instability indicators 0,158 Distance house/ riverbank 0,27 Morphology of the margin 0,143 Height of the riverbank 0,136 Retainment works 0,2 Waves 0,111 Vegetation 0,1 Family income 0,18 Source: Autor (2021). FLUVIAL EROSION RISK ANALYSIS: AN AMAZON STUDY CASE 12 REVISTA GEONORTE, V.12, N.39, p.01-25, 2021. (ISSN 2237 - 1419) DOI: 10.21170/geonorte.2021.V.12.N.39.01.25 Table 6. Pairwise comparison matrices for Hazard and vulnerability to fluvial erosion attributes Hazard Vulnerability Lithology of the margin Weight Type of housing Weight Expansive clay-bearing sediments 0,32 Brick house 0,55 Sediments or soils with higher sand content 0,3 Sediments or soils with higher silt content 0,2 Wooden house 0,45 Sediments or soils with with higher clay content 0,15 Sedimentary saprolite or profile of alteration 0,03 Flow rate Distance house/riverbank Above100.000m3/s 0,35 0 to 10 m 0,4 50.000 to 99.999m3/s 0,3 10.000 to 49.999 m3/s 0,2 1.000m to 9.999 m3/s 0,1 10 to 20 m 0,3 Between 500 e 999 m3/s 0,05 Instability indicators 20 to 50 m 0,2 More than 3 indicators 0,4 2 indicators 0,3 More than 50 m 0,1 1 indicator 0,2 0 indicator 0,1 Morphology of the margin Retaining works Confluence of drainages 0,4 No retaining works 0,5 Concave drainage 0,3 Convex drainage 0,2 Straight drainage 0,1 Existing and Inefficient retaining works 0,3 Height of the riverbank Higher than 10m 0,4 7 to 10m 0,3 Existing and efficient retaining works 0,2 3 to 6m 0,2 Lower than 2m 0,1 Waves Family income Higher than 3,6m 0,5 Less than the amount of 1 monthly minimum wage in Brazil 0,4 1m high undulations to 3.5m high waves 0,3 Mirrored water to 1m high ondulations 0,2 Between 1 and 2 times the amount of the monthly minimum wage in Brazil 0,3 Vegetation Inexistent 0,35 Between 2 and 3 times the amount of the monthly minimum wage in Brazil 0,2 Large size – taller than 3m 0,3 Medium size - 1,5 to 3m tall 0,25 More than 4 times the amount of the minimum wage in Brazil 0,1 Small size – shorter than 1,5m 0,1 Source: Autor (2021). The results for the Hazard criteria showed values ranging from 0,08984, representing the lowest value on the hazard scale that indicating irrelevant to null danger, to 0,38301 FLUVIAL EROSION RISK ANALYSIS: AN AMAZON STUDY CASE 13 REVISTA GEONORTE, V.12, N.39, p.01-25, 2021. (ISSN 2237 - 1419) DOI: 10.21170/geonorte.2021.V.12.N.39.01.25 representing the maximum hazard degree where all attributes are potentialized. In terms of vulnerability criteria, values vary between 0,067, indicating the lowest vulnerability degree in the assessment and 0,208, the maximum vulnerability. Finaly the risk yielding results that can vary between 0,0060193 indicating the lowest or inexistent risk degree and 0,07966608, the maximum risk verified in the assessment. Applications of AHP in the Itanduba community, municipality of Prainha The community of Itanduba islocated on a fluvial island between Amazon and Pará do Uruara rivers closer to the margin of the latter. The margin is convex in shape, associated to a subvertical riverbank (80 to 85°), approximately 7m high, composed of sandy-silty material and covered by grass- like vegetation. On this stretch of the river, the flow rate is that of 147.000m3/ s showing waves up to 1m high that form erosion features and terracettes on the riverbank. The area is occupied by 15 (fifteen) wooden stilt houses placed between 1 and 12m away from the margin of the drainage. This is a hazardous and vulnerable area with high risk of erosion (Table 7). Such result is confirmed by the multitemporal analysis conducted on the risk area at the Itanduba community, that revealed a continuous and accelerated erosional process as well as the margin retreat at a rate of 0,5 to 3m/year (Figure 2). The result obtained is consistent with the risk level mapped by Bandeira; Costa (2019). Table 7. Classification of risk of fluvial erosion for the study area in the Itanduba community. Source: Autor (2021). Hazard criteria Grade Attributes Weight Lithology of the margin 0,183 Sediments or soils with higher sandy content 0,30 Flow rate of the drainage 0,169 Above 100.000m3/s 0,35 Instability indicators 0,158 2 indicators 0,3 Morphology of the river margin 0,143 Convex 0,2 Height of the riverbank 0,136 7 to10m high 0,3 Waves 0,111 Mirrored water to 1m waves 0,2 Vegetation 0,1 Nonexistent 0,35 Hazard Classification 0,28805 (High) Vulnerability criteria Grade Attributes Weight Type of housing 0,35 Wooden house 0,45 Distance margin/riverbank 0,27 10 to 20 m 0,3 Retaining Works 0,2 No retaining works 0,5 Family income 0,18 Less than the amount of 1monthly minimum wage in Brazil 0,4 Vulnerability Classification 0,181 (Very High) Risk 0,05213705 (High) FLUVIAL EROSION RISK ANALYSIS: AN AMAZON STUDY CASE 15 REVISTA GEONORTE, V.12, N.39, p.01-25, 2021. (ISSN 2237 - 1419) DOI: 10.21170/geonorte.2021.V.12.N.39.01.25 Figure 2. Map showing the placement variation of the river margin associated to its retreat caused by erosional processes in the Itanduba community, in the municipality of Prainha, state of Pará. Source: Organized by Authors (2021). Applications of AHP in the São Braz community, municipality of Porto de Moz The socio-economic vulnerability of São Braz community is very high as the income of the 10 families that occupy the area stays under the amount of 1 monthly minimum wage in Brazil leading them to subsistence fishing and buffalo raising. These families live in wooden stilt house very close to the marginal riverbank which has no form of retainment. Considering the hazard criteria presented, the margin in the study area is concave, covered by low to medium-sized vegetation, lithologically composed of silty-clayey material, high erodibility rate and no expansive clay. The inclination and height of the marginal riverbank are approximately 80° and 3.5m, respectively. It stays submerged during the flood period of the Amazon river (March to August), inundating the entire community area, whereas during the ebb period of the river (September to February) the riverbank emerges 1.6m. In this region, the flow rate has reached 109.000m3/s in the flood period and small wind and vessel-generated waves were noted. The greatest indicator of instability observed in the terrain refers to an event that occurred very close to the risk area in August, 2016 when a massive amount of earth (relative to an area 400m long x 1000m wide) collapsed submerging 14 edifications that were dragged by the river stream. Due to this event, many personal belongings were lost and part of the community disappeared, in a few hours, in the water. There were no deaths or injured inhabitants only the loss of material goods and the relocation of 31 people (Bandeira et al., 2019). By assigning grades and weights to the hazard and vulnerability criteria it was possible to identify a high-risk condition (Table 8). Such result is confirmed by the multitemporal analysis conducted on the risk area at the São Braz community that revealed a continuous and accelerated erosional process as well as the margin retreat at a rate FLUVIAL EROSION RISK ANALYSIS: AN AMAZON STUDY CASE 16 REVISTA GEONORTE, V.12, N.39, p.01-25, 2021. (ISSN 2237 - 1419) DOI: 10.21170/geonorte.2021.V.12.N.39.01.25 of 0.5 to 3m/year (Figure 3). The result obtained is consistent with the risk level identified by Bandeira; Simões (2017). Table 8. Classification of risk of fluvial erosion presented for the São Braz community. Source: Autor (2021). Hazard criteria Grade Attributes Weight Lithology of the margin 0,183 Sediments or soils with higher silt content 0,2 Flow rate of the drainage 0,169 Above 100.000m3/s 0,35 Instability indicators 0,158 1 indicator 0,2 Morphology of the river margin 0,143 Concave 0,3 Height of the riverbank 0,136 3 to 6m 0,2 Waves 0,111 Mirrored water to 1m waves 0,2 Vegetation 0,1 Small size – shorter than 1,5m 0,1 Hazard Classification 0,22965 (High) Vulnerability criteria Grade Attributes Weight Type of housing 0,35 Wooden house 0,45 Distance margin/riverbank 0,27 0 to 10 m 0,4 Retaining Works 0,2 No retaining works 0,5 Family income 0,18 Less than the amount of 1monthly minimum wage in Brazil 0,4 Vulnerability Classification 0,208 (Very High) Risk 0,0477672 (Very High) FLUVIAL EROSION RISK ANALYSIS: AN AMAZON STUDY CASE 17 REVISTA GEONORTE, V.12, N.39, p.01-25, 2021. (ISSN 2237 - 1419) DOI: 10.21170/geonorte.2021.V.12.N.39.01.25 Figure 3. Map showing the placement variation of the river margin associated to its retret caused by erosional processes in the São Braz community, in the municipality of Porto de Moz, state of Pará. Source: Organized by Authors (2021). Applications of AHP in the Fátima de Urucurituba community, municipality of Santarém The study area is a riverside community with high socio-economic vulnerability where approximately 46 families live on an income between 1 and 2 times the amount of the monthly minimum wage in Brazil. These families live in wooden stilt houses at a distance of 40 to 100m away from the marginal riverbank that has no type of retaining works. The region is defined as a rectilinear side bar or a marginal dyke that separates the Amazon river from Pacoval lake near its confluence with the Tapajos river. At this location the marginal riverbank is subvertical, approximately 35m deep, composed of interlayered silty-sandy sediments and expansive clay-bearing material, covered by small to medium-sized vegetation. The area is affected by small undulations associated to the frequent transit of large vessels causing turbulence in the water, as well as wind-generated waves up to 2m high. During the ebb period (mainly between September and November) the silty-sandy deposits and water-saturated clay slide as the water level gets lower generating subaqueous mass movement followed by high energy waves that hold great destructive power. This phenomenon is accelerated by the action of wind-generated waves and the transit of large vessels, the moon tide and mainly by the high flow rate of the Amazon river that registered 157.366m3/s for this stretch. Instability indicators such as cracks or fractures due to traction on the soil were found as well as terracettes and the consequent reduction of exposed land. In addition to FLUVIAL EROSION RISK ANALYSIS: AN AMAZON STUDY CASE 18 REVISTA GEONORTE, V.12, N.39, p.01-25, 2021. (ISSN 2237 - 1419) DOI: 10.21170/geonorte.2021.V.12.N.39.01.25 these indicators, the inhabitants of the region have already registered significant material losses such as the destruction of the local school and several houses in Ouctuber, 2010, due to the erosion of an extensive area (LUZARDO; TEIXEIRA, 2012). The degree of risk is high (Table 9) and it is the same verified by Luzardo and Teixeira (2012). The result is consistent with the erosion rate that varies from 0.5 to 77m with some accretion spots up to 3m (Figure 4). Table 9. Classification of risk of fluvial erosion for the Fátima de Urucurituba community. Hazard criteria Grade Attributes Weight Lithology of the margin 0,183 Expansive clay - bearing sediments 0,32 Flow rate 0,169 Above 100.000m3/s 0,35 Instability indicators 0,158 More than 3 indicators 0,4 Morfology of the margin 0,143 Concave 0,3 Height of the riverbank 0,136 Higher than 10m 0,4 Waves 0,111 1m high ondulations up to 3,5m high waves 0,3 Vegetation 0,1 Short size – shorter than 1,5m 0,1 Hazard Classification 0,321151 (Very High) Vulnerability criteria Grade Attributes Weight Type of housing 0,35 Wooden house 0,45 Distance margin/ riverbank 0,27 20 to 50 m 0,2 Retaining works 0,2 No retaining works 0,5 Family income 0,18 1 to 2 times the amount of the monthly minimum wage in Brazil 0,3 Vulnerability Classification 0,154 (High) Risk 0,04951254 (High) Source: Autor (2021). FLUVIAL EROSION RISK ANALYSIS: AN AMAZON STUDY CASE 19 REVISTA GEONORTE, V.12, N.39, p.01-25, 2021. (ISSN 2237 - 1419) DOI: 10.21170/geonorte.2021.V.12.N.39.01.25 Figure 4. Map showing placement variation of the river margin associated to its retreat caused by erosional processes in the Fátima de Urucurituba community in the municipality of Santarém, state of Pará. Source: Organized by Authors (2021). These regions at risk of river erosion are also affected by seasonal floods, due to the elevation of the level of the Amazon River. Mainly in the period from May to July this region reaches its highest level. Even being on stilts with high heights, in periods of great floods, as the events that occurred in 2009, the residents use diverse measures to cope and adapt. Some of these measures are temporary wooden structures that elevate the floor of the houses and also help to protect animals (locally called as marombas). Other measures are access ramps in front of houses, Suspended vegetable gardens, and temporary bridges on the street (ANDRADE; SZLAFSZTEIN, 2019). However, even so, water enters homes, causing material damage to families, such as animal deaths, destruction of floodplain plantations, and destruction of homes (SOUZA et al., 2010). FINAL CONSIDERATIONS The methodology applied to the risk mapping of fluvial erosion in the lower Amazon river in the state of Pará was considered satisfactory. The conditioning factors such as: lithology, flow rate, instability indicators, morphology of the margin, height of the riverbank, waves, vegetation, type of housing, distance house/ riverbank, type of retaining works and socio-economic status of people, made it possible to differentiate degrees of risk. The degree of risk assessed by the methodology applied to this study, taking into consideration quantitative criteria, comes close and many times is equal to the degrees of risk mapped by Geological Survey of Brasil - CPRM from 2012 to 2019. It is important to highlight that the using AHP induces to stricter systematization of risk assessment as it considerably reduces the subjectivity the analyst. Once it is done, enables a more appropriate comparison, and consequently generating data, between different risk areas. However, in more complex assessments, the experience of the FLUVIAL EROSION RISK ANALYSIS: AN AMAZON STUDY CASE 20 REVISTA GEONORTE, V.12, N.39, p.01-25, 2021. (ISSN 2237 - 1419) DOI: 10.21170/geonorte.2021.V.12.N.39.01.25 analyst is indispensable. Therefore for refine this methodology there are some suggestiond, such as:  Multitemporal analysis of satellite images from different periods in the entire Amazon region to obtain a history of values that allows the classification, in different levels, of the erosion rate at the margins of the rivers.  Systematic survey on the resistance to shear of the material along the margins and its relation to the geomorphological and hydrological variables (flow rate and fluctuation of the water level).  Quantification of the sedimentary load of the rivers and its influence on the eroded areas.  Evaluation of the tectonics influence on the process and its relation to seisms and the trigger of processes that cause mass movements associated to fluvial erosion.  Research on the potentiality of the waves caused both by the wind and the navigation of vessels, in the fluvial erosion process.  Statistical studies related to the rainfall data to verify the soil saturation during direct precipitations on the area undergoing erosional processes and during the elevation of the water level in the rivers, which, in this region, usually occurs 2 to 3 months, on average, after a heavy rainfall and, thus, evaluate what precipitation levels can influence the erosional process of the margins. Note that this information was not used as a hazard criterion due to the low density of hydrometeorological stations in the lower Amazon region. The ideal scenario for this project would be the construction of at least 3 (three) more stations to be placed in the area between the municipality of Obidos and Porto de Moz as the information generated by these stations would be more specific and representative. The vulnerability vector can also be refined by adding other information such as: the exact number of people composing the local population, since the more populous the area is the higher the vulnerability, as a greater number of people can be affected. Variables such as age criteria, physically disabled people with difficulties to move and educational level are also widely used in studies that take vulnerability into consideration on risk mapping. However, at a local level, these data are obtained using Social Sciences methods and that would require more time and budget to acquire and treat such data. The application of this methodology in other regions can lead to a revaluation of the grades and weights, besides removal, hierarchy changes or addition of new criteria according to the local peculiarities. Acknowledgements The authors thank the Geological Survey of Brazil - CPRM for fomenting research. REFERENCES ABIDIN, R.Z; SULAIMAN, M.S; YUSOFF, N. Erosion risk assessment: A case study of the Langat River bank in Malaysia. International Soil and Water Conservation Research, v. 5, p. 26–35, 2017. ALBERT, J.S; VAL, P.; HOORN, C. The changing course of the Amazon River in the Neogene: center stage for neotropical diversification. Neotropical Ichthyology, v. 16, n. 3, 2018. FLUVIAL EROSION RISK ANALYSIS: AN AMAZON STUDY CASE 21 REVISTA GEONORTE, V.12, N.39, p.01-25, 2021. (ISSN 2237 - 1419) DOI: 10.21170/geonorte.2021.V.12.N.39.01.25 ALY EL-DIEN, A; TAKEBAYASHI, H; FUJITA, M. Stability of Riverbanks Under Unsteady Flow Conditions. International Journal of Erosion Control Engineering, v.7, n. 2, p. 48-55, 2014. ASHWORTH PJ, LEWIN J. How do big rivers come to be different?. Earth-Science Reviews, v. 114, n. 1-2, p. 84–107, 2012. AGÊNCIA NACIONAL DE ÁGUAS - ANA. Flow rate. Available at . Accessed on: 2019, Octuber 03. ANDRADE, M.M.N.; SZLAFSZTEIN, C.F; SOUZA FILHO, P.W.M; ARAUJO, A.R; GOMES, M.K.T. A socioeconomic and natural vulnerability index for oil spills in an Amazonian harbor: A case study using GIS and remote sensing. Journal of Environmental Management, v. 91, n. 10, p.1972-1980, 2010. ANDRADE. M.M.N; SZLAFSZTEIN, C.F. Vulnerability assessment including tangible and intangible components in the index composition: An Amazon case study of flooding and fl ash flooding. Science of the Total Environment, v. 630, n. 15, p. 903-912. 2018. ANDRADE, M.M.N; SZLAFSZTEIN, C.F. Coping and adaptation strategies and institutional perceptions of hydrological risk in an urban Amazonian city. Disasters, v. 44, v. 4. 2019. AVILA, H; VARGAS, G; DAZA, R. Susceptibility analysis of river bank erosion based on exposure to shear stress and velocity combined with hydrologic and geomorphologic variables. World environmental and water resources congress: water without borders. ASCE-American Society of Civil Engineers, 2014. BANDEIRA, I.C.N.; ADAMY, A.; ANDRETTA, E.R.; COSTA, C.R.A; ANDRADE, M.M.N. Terras caídas: Fluvial erosion or distinct phenomenon in the Amazon?. Environmental Earth Sciences, v. 77, n. 222, p.1-16, 2018. BANDEIRA, I.C.N.; SIMÕES, P.M.L. Setorização de áreas de alto e muito alto risco a movimentos de massa, enchentes e inundações: Porto de Moz, Pará. Belém, PA: Serviço Geológico do Brasil - CPRM, 27p. 2017. Available at . Accessed on 2019 June 05. BANDEIRA, I.C.N.; COSTA, C.R.A. Setorização de áreas em alto e muito alto risco a movimentos de massa, enchentes e inundações: Prainha, Pará. Belém, PA: Geological Survey of Brazil - CPRM, 46p. 2019. Available at . Accessed on 2019 June 05. BANDEIRA ICN, SIMÕES PML, MELO JUNIOR HR (2018b) Caracterização geológico- geotécnica dos sedimentos marginais da comunidade de São Braz, município de Porto de Moz – PA e sua relação com suscetibilidade à erosão fluvial (terras caídas).. Revista Brasileira de Geologia de Engenharia e Ambiental – RBGEA, v. 9, n. 1 – 2, p. 19-25, 2019. BATHRELLOS, G.; SKILODIMOU, H. Using the analytic hierarchy process to create an erosion risk map. A case study in Malakasiotiko stream, Trikala perfecture. Bulletin of the Geological Society of Greece. v. 40, n4, 2007. Available at . Accessed on 2018 June 03. BHOWMIK, N.G.; DEMISSIE, M.; GUO, C.Y. Waves Generated by river traffic and wind on the Illinois and Mississippi Rivers. Research Report - University of Illinois at Urbana-Champaign, Water Resources Center, 1982. p.167. BRAZIL. Lei n°. 12.608 de 10 de abril, 2012. Institui a Política Nacional de Proteção e Defesa Civil. Casa Civil. 2012. Available at . Accessed on 2017 August 02. https://www.jstage.jst.go.jp/search/global/_search/-char/en?item=8&word=Ahmed+ALY+EL-DIEN https://www.jstage.jst.go.jp/search/global/_search/-char/en?item=8&word=Hiroshi+TAKEBAYASHI https://www.jstage.jst.go.jp/search/global/_search/-char/en?item=8&word=Masaharu+FUJITA http://www.snirh.gov.br/hidrotelemetria/Mapa.aspx http://lattes.cnpq.br/1930321094483005 http://lattes.cnpq.br/3282736820907252 http://lattes.cnpq.br/6511016682362834 http://lattes.cnpq.br/1930321094483005 http://lattes.cnpq.br/1930321094483005 http://lattes.cnpq.br/9116903517174291 http://lattes.cnpq.br/9116903517174291 http://rigeo.cprm.gov.br/jspui/handle/doc/18358 http://lattes.cnpq.br/9116903517174291 http://lattes.cnpq.br/9116903517174291 https://doi.org/10.12681/bgsg.17205 http://www.planalto.gov.br/ccivil_03/_Ato2011-2014/2012/Lei/L12608.htm http://www.planalto.gov.br/ccivil_03/_Ato2011-2014/2012/Lei/L12608.htm FLUVIAL EROSION RISK ANALYSIS: AN AMAZON STUDY CASE 22 REVISTA GEONORTE, V.12, N.39, p.01-25, 2021. (ISSN 2237 - 1419) DOI: 10.21170/geonorte.2021.V.12.N.39.01.25 CARVALHO, J.A.L. Terras caídas e consequências sociais: Costa de Miracauera - Paraná da Trindade, Municipio de Itacoatiara – AM, Brasil. Dissertação (Mestrado em Sociedade e Cultura na Amazônia), Universidade Federal do Amazonas – UFAM. 2006. 141p. CARVALHO, C.S.; GALVÃO, T. Prevenção de riscos de deslizamentos em encostas: guia para elaboração de políticas municipais. Brasília: Ministério das Cidades; Cities Alliance. 2006. 111 p. Available at . Accessed on 2019 September 5. CARVALHO, C.S.; MACEDO, E.S.; OGURA, A.T. Mapeamento de riscos em encostas e margem de rios. Brasília: Ministério das Cidades; Instituto de Pesquisas Tecnológicas – IPT. 2007. 176 p. Available at Accessed on 2017 June 20. CHARMAN, J.H.; GRIFFITHS, J.S. Terrain evaluation methods for predicting relative hazards from mass movement, fluvial erosion and soil erosion in the developing world. In: Merriman PA, Browitt CWA (eds) Natural disasters: protecting vulnerable communities. Thomas Telford, Londres, 1993, p. 167-183. COPPIN, N.J; RICHARDS, I.G. Use of vegetation in civil engineering. Butterworths, London, UK. 1990. p.310. COUPER, P. Effects of silt–clay content on the susceptibility of river banks to subaerial erosion. Geomorphology, v. 56, n. 1, p. 95-108, 2003. CURRAN, J.H.; MCTEAGUE, M.L. Geomorphology and Bank Erosion of the Matanuska River, Southcentral Alaska: U.S. Geological Survey Scientific Investigations Report 2011-5214, 2011. 64 p. CUTTER, S.L.; BORUFF, B.J.; SHIRLEY, W.L. Social vulnerability to environmental hazards. Social Science Quarterly, v. 84, n. 2, 2003. DERRUAU, M. Geomorfología. Barcelona. Ed. Ariel, 1966. 442p. FEDERAL EMERGENCY MANAGEMENT AGENCY – FEMA. Riverine Erosion Hazard Areas: Mapping Feasibility Study. 1999. Available at . Accessed on: 2020 June 28. GALLO, M.N.; VINZON, S.B. Generation of overtides and compound tides in Amazon estuary. Ocean Dynamics, v. 55, n. 5-6, p. 441-448, 2005. GEOLOGICAL SURVEY OF BRAZIL – CPRM. Setorização de risco geológico. 2020. Available at Accessed on 2020 April 20. GHOSH, D.; SAHU, A.S. The impact of population displacement due to river bank erosion on the education of erosion victims: a study in jangipur sub-division of murshidabad district, West Bengal, India. Bulletin of Geography. Socio-economic Series, v. 46, n. 46, p. 103–118, 2019. GRAY, D.H.; MACDONALD, A. The role of vegetation in river bank erosion. In: M. A. Ports (ed). Hydraulic engineering. Proceedings of the 1989 national conference on hydraulic engineering, 1989. 218-223p. HIMMELSTOSS, E.A.; HENDERSON, R.E.; KRATZMANN, M.G.; FARRIS, A.S. Digital Shoreline Analysis System (DSAS) version 5.0 user guide: U.S. Geological Survey, Open-File, Report 2018–1179. 2018. Available at . Accessed on 2019 April 20. https://www.mdr.gov.br/images/stories/ArquivosSNPU/Biblioteca/PrevencaoErradicacao/Livro_Curso_Capacitacao_Tecnicos_Municipais.pdf https://www.mdr.gov.br/images/stories/ArquivosSNPU/Biblioteca/PrevencaoErradicacao/Livro_Curso_Capacitacao_Tecnicos_Municipais.pdf https://www.mdr.gov.br/images/stories/ArquivosSNPU/Biblioteca/PrevencaoErradicacao/Livro_Mapeamento_Enconstas_Margens.pdf https://www.mdr.gov.br/images/stories/ArquivosSNPU/Biblioteca/PrevencaoErradicacao/Livro_Mapeamento_Enconstas_Margens.pdf https://www.sciencedirect.com/science/article/abs/pii/S0169555X03000485#! https://www.sciencedirect.com/science/journal/0169555X https://onlinelibrary.wiley.com/journal/15406237 https://www.fema.gov/media-library/assets/documents/7235 https://www.fema.gov/media-library/assets/documents/7235 http://www.cprm.gov.br/publique/Gestao-Territorial/Prevencao-de-Desastres/Produtos-por-Estado---Setorizacao-de-Risco-Geologico-5390.html http://www.cprm.gov.br/publique/Gestao-Territorial/Prevencao-de-Desastres/Produtos-por-Estado---Setorizacao-de-Risco-Geologico-5390.html https://doi.org/10.3133/ofr20181179 FLUVIAL EROSION RISK ANALYSIS: AN AMAZON STUDY CASE 23 REVISTA GEONORTE, V.12, N.39, p.01-25, 2021. (ISSN 2237 - 1419) DOI: 10.21170/geonorte.2021.V.12.N.39.01.25 INSTITUTO BRASILEIRO DE GEOGRAFIA E ESTATISTICA – IBGE. Panorama de Santarém, Estado do Pará. 2010a. Available at . Accessed on 2019 Octuber 20. INSTITUTO BRASILEIRO DE GEOGRAFIA E ESTATISTICA - IBGE. Panorama de Porto de Moz, Estado do Pará. 2010b. Available at . Accessed on 2019 Octuber 20. INSTITUTO BRASILEIRO DE GEOGRAFIA E ESTATISTICA - IBGE. Panorama de Prainha, Estado do Pará. 2010c. Available at . Accessed on 2019 Octuber 20. INSTITUTO BRASILEIRO DE GEOGRAFIA E ESTATISTICA - IBGE. 2020: Bases para download. Available at: Accessed on 2020 Octuber 20. INSTITUTO BRASILEIRO DE GEOGRAFIA E ESTATISTICA - IBGE. 2017: Bases para download. Available at: Accessed on 2020 Octuber 20. INSTITUTO DE PESQUISA ECONÔMICA APLICADA - IPEA, PROGRAMA DAS NAÇÕES UNIDAS PARA O DESENVOLVIMENTO - PNUD, FUNDAÇÃO JOÃO PINHEIRO – FJP. Radar IDHM: evolução do IDHM e de seus índices componentes no período de 2012 a 2017. Brasília, DF: IPEA, 2019. KNIGHTON, D. Fluvial forms and process: A new perspective. London: E. Arnold, 1998. 400p. LATRUBESSE, E.M. Patterns of anabranching channels: The ultimate end-member adjustment of mega rivers. Geomorphology, v.101, n. 1-2, p. 130-145, 2008. LUNA, A.V. Roles of vegetation on river bank accretion. Dissertation, National University of Colombia, Bogotá, Colombia, 2016. 249p. LUZARDO, R.; TEIXEIRA, S. Setorização de áreas de alto e muito alto risco a movimentos de massa, enchentes e inundações – Municipio de Santarém - PA. Belém, PA: Serviço Geológico do Brasil - CPRM. 2012. Available at . Accessed on 2019 June 05. MAFFRA, C.R.B.; MORAES, M.T.; SOUZA, R.S.; SUTILI, F.J.; PINHEIRO, R.J.B.; SOARES, J.M.D. Métodos de avaliação da influencia e contribuiçao das plantas sobre a estabilidade de taludes. Revista Scientia Agraria, Curitiba, v. 18, n. 4, p.129-143, 2017. MAGALHÃES, R.C.; VIEIRA, A.F.S.G. As características hidrológicas do solo de várzea e sua vulnerabilidade ao processo de terras caídas na Amazônia Central (BR). Revista Brasileira de Geografia Física, v. 11, n. 3, p. 773-788, 2018. MAPBIOMAS. Uso e ocupação do solo. 2019. Available at: . Access on: 2019 November 10. MARINHA. Atlas of Pilot Charts. Board of Hidrography and Navigation of the Brazilian Navy. Rio de Janeiro. 2019. Available at Accessed on 2020 March 10. MARINHA. Cartas rasters. 2020. Available at . Accessed on 2020 June 12. MACFALL, J.; ROBINETTE, P.; WELCH, D. Factors influencing bank geomorphology and erosion of the Haw River, a high order river in North Carolina, Since European Settlement. PLoS One, 2014. https://cidades.ibge.gov.br/brasil/pa/santarem/panorama https://cidades.ibge.gov.br/brasil/pa/porto-de-moz/panorama https://cidades.ibge.gov.br/brasil/pa/porto-de-moz/panorama https://cidades.ibge.gov.br/brasil/pa/prainha/panorama https://www.marinha.mil.br/chm/sites/www.marinha.mil.br.chm/files/u1974/cp.pdf https://www.marinha.mil.br/chm/dados-do-segnav/cartas-raster https://www.marinha.mil.br/chm/dados-do-segnav/cartas-raster FLUVIAL EROSION RISK ANALYSIS: AN AMAZON STUDY CASE 24 REVISTA GEONORTE, V.12, N.39, p.01-25, 2021. (ISSN 2237 - 1419) DOI: 10.21170/geonorte.2021.V.12.N.39.01.25 MASKARE, B.B. Lithological control on channel morphology: a study of kas river Gorge in Ahmednagar district, Maharashtra, Índia. Global Journal of Engineering Science and Research Management, v. 2, 2015. MERTES, L.A.K.; DUNNE, T. Effects of Tectonism, Climate Change, and Sea‐level Change on the Form and Behaviour of the Modern Amazon River and its Floodplain. In: A. GUPTA (Ed.). Large Rivers: Geomorphology and Management, Wiley, Chichester, 2007. Cap. 8, p. 115-144, 2007. MOLLAH, T.H.; FERDAUSH, J. Riverbank Erosion, Population Migration and Rural Vulnerability in Bangladesh (A Case Study on Kazipur Upazila at Sirajgonj District). Environment and Ecology Research, v. 3, n. 5, p.125-131, 2015. MONTEIRO, J.S.; PADILHA, D.G.; CRUZ, J.C. Protocol of factors assessment influential in susceptibility to erosion of river slopes. Revista Árvore, Viçosa-MG, v. 40, n. 5, p. 815-823, 2016. MULYONO, A.; SUBARDJA, A.; EKASARI, I.; LAILATI, M.; SUDIRJA, R.; NINGRUM, W. The Hydromechanics of Vegetation for Slope Stabilization. IOP Conference Series: Earth and Environmental Science, v. 118, 2018. MOLINIER, M.; GUYOT, J.L.; OLIVEIRA, E.; GUIMARAES, V.; CHAVES, A. Hidrologia da bacia do Rio Amazonas. A água em Revista. Rio de janeiro: Geological Survey of Brazil - CPRM. p. 31-36, 1994. NARDI, L.; CAMPO, L.; RINALDI, M. Quantification of riverbank erosion and application in risk analysis. Natural Hazards, v. 69, n. 1, p. 869–887, 2013. ONTARIO MINISTRY OF NATURAL RESOURCES – OMNR. Technical Guide – River and Stream Systems: Erosion Hazard Limit. Queen’s Printer for Ontario, 2002. PASSOS, M.S.; SOARES, E.A.A. Multitemporal analysis of the Solimões-Amazonas river system between the Purus and Negro tributaries, Western Amazon, Brazil. Geology USP, Scientific Series, São Paulo, v. 17, n. 1, p. 61-74, 2017. PEIXOTO, J.M.A; NELSON, B.W.; WITTMANN, F. Spatial and temporal dynamics of river channel migration and vegetation in central Amazonian white-water floosplains by remote- sensing techniques. Remote Sensing of Environment, v. 113, p. 2258-2266, 2009. PIMENTEL,J.;DUTRA,T.; RIBEIRO,R.S; PFALTZGRAFF, P.A.S.; BRENNY, M.E.RPEIXO TO, D.; SILVA, D.R.; IWANAMI, H.; NISHIMURA, T. Risk Assessment and Hazard Mapping Technique in the Project for Strengthening National Strategy of Integrated Natural Disaster Risk Management. International Journal of Erosion Control Engineering, v. 13, n. 1, p. 35-47, 2020. REBELO, F. Natural Risks and anthropic actions: studies and considerations. 2nd ed. Coimbra: [s.n.], 2001. p.289. SAATY, T.L. What is the Analytic Hierarchy Process?. In: MITRA G., GREENBERG H.J., LOOTSMA F.A., RIJKAERT M.J., ZIMMERMANN H.J. (eds) Mathematical Models for Decision Support. NATO ASI Series (Series F: Computer and Systems Sciences). Springer, Berlin, Heidelberg, v. 48, p. 109-121, 1988. SAATY, T.L. Decision making with the analytic hierarchy process. International journal of services sciences, v. 1, n. 1, p.83-98, 2008. SILVA, A.B.N.; ANDRADE, M.M.N. Identificação de Risco à Erosão Fluvial na cidade de Cametá (PA), Brasil. DELOS: DESARROLLO LOCAL SOSTENIBLE, v. 12, n. 35, p. 1-16, 2019. https://onlinelibrary.wiley.com/action/doSearch?ContribAuthorStored=Mertes%2C+Leal+A+K https://onlinelibrary.wiley.com/action/doSearch?ContribAuthorStored=Dunne%2C+Thomas https://onlinelibrary.wiley.com/doi/book/10.1002/9780470723722 https://www.jstage.jst.go.jp/search/global/_search/-char/en?item=8&word=Jorge+PIMENTEL https://www.jstage.jst.go.jp/search/global/_search/-char/en?item=8&word=Thiago+DUTRA https://www.jstage.jst.go.jp/search/global/_search/-char/en?item=8&word=Rafael+Silva+RIBEIRO https://www.jstage.jst.go.jp/search/global/_search/-char/en?item=8&word=Pedro+Augusto+dos+Santos++PFALTZGRAFF https://www.jstage.jst.go.jp/search/global/_search/-char/en?item=8&word=Maria+Em%C3%ADlia+Radomski++BRENNY https://www.jstage.jst.go.jp/search/global/_search/-char/en?item=8&word=Dario+PEIXOTO https://www.jstage.jst.go.jp/search/global/_search/-char/en?item=8&word=Dario+PEIXOTO https://www.jstage.jst.go.jp/search/global/_search/-char/en?item=8&word=Diogo+Rodrigues+da++SILVA https://www.jstage.jst.go.jp/search/global/_search/-char/en?item=8&word=Hideyuki+IWANAMI https://www.jstage.jst.go.jp/search/global/_search/-char/en?item=8&word=Tomohiro+NISHIMURA FLUVIAL EROSION RISK ANALYSIS: AN AMAZON STUDY CASE 25 REVISTA GEONORTE, V.12, N.39, p.01-25, 2021. (ISSN 2237 - 1419) DOI: 10.21170/geonorte.2021.V.12.N.39.01.25 SOUZA, E.S. Caracterização e Modelagem Geotécnica do fenômeno erosivo Amazônico. Dissertação (Mestrado em engenharia civil). Universidade Federal do Amazonas, faculdade de tecnologia programa de pós-graduação em engenharia civil. Manaus-AM, 2019. 185p. SOUZA, J.C.R.; ALMEIDA, R.A. Vazante e enchente na Amazônia brasileira: impactos ambientais, sociais e econômicos. In: VI SEMINÁRIO LATINO-AMERICANO DE GEOGRAFIA FÍSICA. Universidade de Coimbra-Portugal. 2010 STEVAUX, J.C.; LATRUBRESSE, E.M. Geomorfologia fluvial. Oficina de Texto, São Paulo, 2017. SUN, R.; GAO, G.; GONG, Z.; WU, J. A review of risk analysis methods for natural disasters. Natural Hazards, v. 100, n. 8, p. 571-593, 2020. SUTILI, F.J. Bioengenharia de solos no âmbito fluvial do sul do Brasil. Ph.D. Departamento de Engenharia Civil e Perigos Naturais, Instituto de Bioengenharia de Solos e Planejamento da Paisagem, Universidade Rural de Viena, Viena, 2007. 94p. SZLAFSZTEIN, C.F. DESAFIOS E DEMANDAS EM GEOLOGIA DE ENGENHARIA E AMBIENTAL NA AMAZÔNIA BRASILEIRA, BRASIL. Revista Brasileira de Geologia de Engenharia e Ambiental, v 8, n° 1 - 2, p. 52-59, 2018. TEIXEIRA, S.G. Análise Multitemporal da Dinâmica fluvial do Baixo Rio Amazonas no período de 1985 à 2016. In: Anais do 15° Simpósio de Geologia da Amazônia. Belém-Pará, 2017. TEIXEIRA, S.G.; SILVA, R.L.L.; LOPES, D.F. Influência de parâmetros hidrológicos no processo de erosão fluvial na região de Santarém-PA. In: 16º Congresso Brasileiro de Geologia de Engenharia e Ambiental. São Paulo-SP. Anais do Congresso Brasileiro de Geologia de Engenharia e Ambiental: ABGE. São Paulo-SP, 2018. THIELLER, E.R.; MARTIN, D.; ERGUL, A. The Digital Shoreline Analysis System, version 2.3. Shoreline change measurement software extension ArcMap 9.1. USGS Open-File. 2005. THORNE, C.R.; OSMAN, A.M. Riverbank stability analysis II: Applications. ASCE Journal of Hydraulic Engineering, v. 114, n. 2, p.151-172, 1988. THORNE, C.R. Bank erosion and meander migration of the Red and Mississippi Rivers, USA. In: VAN-DE-VEN, F.H.M., GUTKNECHT, D., LOUCKS, D. P., AND SALEWICZ K. A. (eds), Hydrology for the Water Management of Large River Basins (Proceedings of the Vienna Symposium, August 1991). International Association of Hydrological Sciences, 1991, p. 301- 313. TEIXEIRA, G.T.; BANDEIRA, I.C.N.; FONSECA. D.D.F. Caracterização das tipologias de risco geológico identificadas no Estado do Pará no período de 2012 a 2018. In: Contribuições à Geologia da Amazônia - Volume 11. Sociedade Brasileira de Geologia, Núcleo Norte, 2019. p. 253-269. TOLEDO, V.L.; LANA, J.C. Panorama dos desastres naturais no Brasil: Base de dados nacional do risco geológico. In: 16° Simpósio de Geologia do Sudeste, Campinas-SP. 2019. Geosudeste 2019. Campinas-SP, Sociedade Brasileira de Geologia – Núcleo São Paulo, 2019, p. 224. VAIDYA, O.S.; KUMAR, S. Analytic hierarchy process: An overview of applications. European Journal of Operational Research, v. 169, n. 1 p. 1-29, 2006. VASQUEZ, M.L.; ROSA-COSTA L.T (org.). Geologia e recursos minerais do estado do Pará: texto explicativo do mapa geológico e de recursos minerais do estado do Pará. Belém: CPRM – Serviço Geológico do Brasil. 328 p. Escala 1:1.000.000. Sistema de Informações Geográficas - SIG; Mapas Geológicos Estaduais; Programa Geologia do Brasil - PGB. 2008. https://www.researchgate.net/profile/Claudio_Szlafsztein?_sg%5B0%5D=nESWvkj9dOSSMG8TSL1SnPAmY3VQDEFtNTSi1e_GSCxJCVavUMn_wxPqyKEWDRSm0xJUSrY.ptMHw2hupeLAO5aJUqWJNcx8G-x9JOGFdX6CxclZF3zstP6lryFmMNjZW_7mVYlFv7Tp-w_gpzCDw0HMwlZC2A&_sg%5B1%5D=oWS0AfmZCHa7KhdYSCXwqqZQqlYXs2QwMqJIH8sDM7ISPOMFamybht6aqYnRLNB289bHaTE.FfwWJm8atYHbLzlNM7jCodjzjb4_apBpr-_i-XhEvefHO-12C8ASJGj1pNcmsKPt95K_52WXm7z0fnGmMfg1LA