Shaded relief map of Cotopaxi volcano and surrounding areas with main drainage systems and locations indicated in the text. The main geographic locations are shown in the bottom inset. The black box indicates the quarry site. Black striped area refers to the zone of intense farming and cultivation. In the upper inset, Cotopaxi volcano seen from the north. In the foreground the Limpiopungo plain, with blocks carried by recent lahars. 

Contexts in source publication

Context 1

... this work, we concentrate on lahar hazard assessment in the Río Cutuchi system, which conveys all the main drainages of west and south of the Cotopaxi cone (Fig. 1). The Cutuchi valley hosts several human settlements, most of which lie on recent lahar deposit terraces, only a few meters above the present river bed. Further south, the Río Cutuchi crosses the large urban settlement of Latacunga (43 km SE of the volcano, Fig. 1), with a population of 52,000 inhabitants. Its path is generally parallel ...

Context 2

... which conveys all the main drainages of west and south of the Cotopaxi cone (Fig. 1). The Cutuchi valley hosts several human settlements, most of which lie on recent lahar deposit terraces, only a few meters above the present river bed. Further south, the Río Cutuchi crosses the large urban settlement of Latacunga (43 km SE of the volcano, Fig. 1), with a population of 52,000 inhabitants. Its path is generally parallel to the Pan American Highway, the major state road connecting the country from north to south. Many small-scale lahars have followed this drainage in the past (b 100 million m 3 ), as well as some large-scale historical flows that severely impacted the ...

Context 3

... is one of the highest active volcanoes of the world: its per- fect ice-capped, 2200 m-high cone reaches an elevation of 5897 m above sea level with a basal diameter of 22 km. The volcano is located 60 km south of Quito, the capital of Ecuador, and it is surrounded by vil- lages, national lines of communication and other infrastructure (Fig. 1). Its steep flanks (30°-35°) culminate in a crater hosting a tephra cone ( Hradecka et al., 1974), which in turn harbors a smaller crater, the active vent of historic eruptions. The volcano has three main drainage systems ( Fig. 2): 1) to the north is the Pita-Guayallabamba river system that flows through the Los Chillos and Tumbaco ...

Context 4

... (small, steep-sided valleys): the Barrancas and Saquìmala, whose streams eventually join with the Río Cutuchi and flow through the center of the Latacunga valley. The city of Latacunga and settlements along the entire valley grew rapidly in recent decades within the city and along the river, in particular with greenhouses for growing roses ( Fig. 1), resulting presently in a very densely populated area with high industrial investment where houses, factories, hospitals, schools, the Pan American Highway and other infrastructure have been built on top of lahar deposits of less than 1000 years old. The population in Latacunga was estimated at 63,800 inhabitants in 2012, living in an ...

Context 5

... Cutuchi is the main river southwest of Cotopaxi. Its headwaters flow to the south of Rumiñahui (Fig. 1), a deeply eroded extinct volcano situated to the west of Cotopaxi, and its tributaries cover the entire western sector of Cotopaxi. Stratigraphic studies along the Río Cutuchi valleys revealed an impressive sequence of lahar deposits Pistolesi et al., 2013). Historical accounts also describe that since the Spanish Conquest (XVI ...

Context 6

... to the preservation of tephra beds intercalated between the lahar deposits, and these were directly related to their generating explosive eruptions (Fig. 3). Debris flow deposits were also informally grouped into three main categories (Types A, B and C) by combining sedimentological data obtained from a large quarried area along the Río Cutuchi (Fig. 1). These three categories of deposits largely vary in terms of dispersal, thickness, average and maximum size of transported blocks increasing from A to C, and were related to lahars with different sizes and trigger mechanisms ( Pistolesi et al., 2013) (Fig. 4A, B, C). The stratigraphy of debris flow deposits in the Cutuchi quarries ( ...

Context 7

... the Río Cutuchi (Fig. 2). In simulation α, for each drainage system, different volumes were used and then recombined in order to obtain a volume comprised between 100 and 150 million m 3 (Fig. 5A, Table 1). A final lahar volume of 120 million m 3 , corre- sponding to the sum of Río Cutuchi (60 million m 3 ), Río Saquìmala (30 million m 3 , with 90 million m 3 where Cutuchi and Saquìmala join together) and Río Barrancas-Alaquez (30 million m 3 ) contributions, was finally chosen (Figs. 5B, 7). The final target volume (120 million m 3 ) ...

Context 8

... inundation depths along the 73-km stretch of the Río Cutuchi are reported in Fig. 10: along the first 10 km, apparent flow depths range between 10 and 50 m for simulation α, consistent with the range (12-33 m) derived from historical data and the computed maximum flow depth for the 1877 lahar ( Barberi et al., 1992). Apparent flow depths for simulation β resulted between 10 and 65 m, with a maximum difference of 18 m ...

Context 9

... a-b (6700 m-long) drawn in proximity to Latacunga (Figs. 8, 11) shows that the apparent height of the flow for an event of 360 million m 3 is double (35 m vs. 18 m) with respect to the 120 million m 3 case. In simulation β, the inundated area reaches the elevated areas on the east side of the river, where the main part of the city is located. The value obtained for simulation α is consistent with ...

Context 10

... on the east side of the river, where the main part of the city is located. The value obtained for simulation α is consistent with the 10 m flow depth for a 90 million m 3 lahar obtained by Barberi et al. (1992). The majority of the city of Latacunga could be damaged also in this case: most of the major structures, facilities including schools, (Fig. 10). Pierson et al., 1990), mainly because the bulking rate is considered proportional to the power dissipated by the fluid (Finnie, 1972). By using the same regression line obtained by Aguilera et al. (2004), we plot our data to calculate the theoretical bulking values in two reaches of Río Cutuchi (Fig. 12). The first is a 3 km-long ...

Context 11

... structures, facilities including schools, (Fig. 10). Pierson et al., 1990), mainly because the bulking rate is considered proportional to the power dissipated by the fluid (Finnie, 1972). By using the same regression line obtained by Aguilera et al. (2004), we plot our data to calculate the theoretical bulking values in two reaches of Río Cutuchi (Fig. 12). The first is a 3 km-long section (θ = 1.5°), 10 km from the onset of the simulation. We used apparent peak flow depths of simulations α and β (10 and 16 m, respectively) which corre- spond to Sx * h of 0.26 m and 0.41 m, respectively. These convert to bulking values of 190 and 300 m 3 /s, corresponding to total bulking of 6 × 10 5 and ...

Context 12

... and β (10 and 16 m, respectively) which corre- spond to Sx * h of 0.26 m and 0.41 m, respectively. These convert to bulking values of 190 and 300 m 3 /s, corresponding to total bulking of 6 × 10 5 and 9 × 10 5 m 3 calculated for 3 km of the section. In the area of Latacunga, we used a 10 km-long section (θ = 0.6°) with flow depths of 18 and 35 m (Figs. 10, 11), corresponding to Sx * h of 0.20 and 0.38, respectively. These convert to bulking values of 140 and 280 m 3 /m, and to total bulking values of 1.4 × 10 6 and 2.8 × 10 6 m 3 , respectively, for the 10 km-long section. If we assume an average bulking value, and we use it for the total length of 39 km of the simulation up to Latacunga, ...

Context 13

... high-concentration, sediment-loaded fl ows that occur on volcanic terrains. They may be a primary phenomenon directly triggered by eruptive activity (e.g. the 1985 Nevado del Ruiz event in Colombia; Pierson et al., 1990) or they may result from the post-eruptive mobilization (secondary lahars) of volcanic debris (e.g. the 1991 eruption of Mt. Pinatubo, Philippines; Rodolfo et al., 1996). Lahar generation requires a combination of three main factors including: (i) a triggering mechanism that rapidly makes available an adequate water source; (ii) the availability of abundant, un- consolidated debris; and (iii) steep slopes (Vallance, 2000). Due to water incorporation and volume increase, lahars can easily over fl ow lateral banks and spread over areas of low gradient. This can produce catastrophic consequences for the communities living along the areas, which can be inundated unexpectedly by lahars — as recently shown by the Pinatubo, Mayon and Nevado del Ruiz events (Pierson et al., 1990; Voight, 1990; Rodolfo, 1995; Newhall and Punongbayan, 1996; Tanguy et al., 1998). According to Varnes (1978), we include here with the term ‘ lahars ’ not only debris fl ows but also hyperconcentrated mud fl ows (Smith and Lowe, 1991). Although the term has sometimes been used to refer to the deposits of such fl ows, Smith and Fritz (1989, p. 375) dismiss this meaning stating that “ lahar is an event that can refer to one or more discrete processes, but does not refer to a deposit ” . Small debris fl ows occur daily on the fl anks of many active and inactive volcanoes; although traditionally the term ‘ lahar ’ has generally been restricted to those events which can generate a hazard to populations (Scott, 1988), nowadays lahar is fairly broadly used regardless if the event poses hazard to population or not (Manville et al., 2009). In the following, we will use lahar to refer to the process, and as a synonym of debris fl ow. Lahars can be generated at crater lakes, following crater failures (Bornas et al., 2003; Manville and Cronin, 2007; Manville, 2010; Massey et al., 2010), or explosive expulsion of water (Zen and Hadikusumo, 1965; Nairn et al., 1979; Suryo and Clarke, 1985; Thouret et al., 1998; Németh et al., 2006; Kilgour et al., 2010) or by heavy rainfall on freshly deposited tephra or loose material on volcano slopes, as in the case of the 1991 eruption of Mt. Pinatubo (Rodolfo, 1989; Pierson et al., 1992; Arboleda and Martinez, 1996; Rodolfo et al., 1996). Lahars are also generated by the rapid melting of snow or ice on ice clad volcanoes, due to the interaction of eruptive products with the ice cap (Major and Newhall, 1989). As demonstrated by the 1985 Nevado del Ruiz event in Colombia, even small volcanic eruptions can trigger catastrophic lahar events when pyroclastic material interacts with a summit glacier (Pierson et al., 1990; Voight, 1990; Tanguy et al., 1998), especially if combined with seismic shaking and intense scouring of the ice (with consequent channeling of the lahar). During the past 30 years, lahar-related disasters have been docu- mented worldwide in the volcanological literature (Voight, 1990; Hall, 1992; Rodolfo, 1995; Newhall and Punongbayan, 1996; Tanguy et al., 1998; Wood and Soulard, 2009). In the last century, about 30,000 casualties were reported in relation to the occurrence of lahar events (Witham, 2005); about 80% of this terri fi c quota is related to a single event (the devastation of the village of Armero, following the 1985 eruption of Nevado del Ruiz). Continuous growth of buildings and settled areas in lahar fl ow paths demonstrates that better risk perception, land-use planning and rapid evacuation plans may combine in reducing damage and loss of life from future lahars (Wood and Soulard, 2009). Cotopaxi volcano is well known for the potential destructiveness of its lahars, and models of lahar invasion have been recently applied to the northern and southern drainages (Barberi et al., 1992; Aguilera et al., 2004). Inundation areas for future lahars have also been partially explored by several authors (Miller et al., 1978; Hall and von Hillebrandt, 1988a, 1988b; Mothes et al., 2004; Mothes, 2006) who presented maps of lahar inundation areas related to the last (1877) eruptive event and for future lahar-generating events of similar size. Modeling procedures were not presented in detail in these papers so that a more in-depth discussion on the potential of lahar inundation at Cotopaxi and the mapping of the related hazard are elements of con- siderable importance. In this work, we concentrate on lahar hazard assessment in the Río Cutuchi system, which conveys all the main drainages of west and south of the Cotopaxi cone (Fig. 1). The Cutuchi valley hosts several human settlements, most of which lie on recent lahar deposit terraces, only a few meters above the present river bed. Further south, the Río Cutuchi crosses the large urban settlement of Latacunga (43 km SE of the volcano, Fig. 1), with a population of 52,000 inhabitants. Its path is generally parallel to the Pan American Highway, the major state road connecting the country from north to south. Many small-scale lahars have followed this drainage in the past ( b 100 million m 3 ), as well as some large-scale historical fl ows that severely impacted the population. We present lahar simulations with different starting volumes performed with a GIS-based semi-empirical model (LAHARZ; Iverson et al., 1998) in order to evaluate the inundating potential of Cotopaxi lahars and to assess the inundation hazard in the southern part of the volcano towards the city of Latacunga. The manuscript is arranged with a section (Section 2) in which we brie fl y describe the geological data relevant to this study (recent Cotopaxi activity, lahar deposits, lahar triggering mechanisms, and the factors which control lahar volume). In Section 3 we introduce and discuss lahar modeling approaches and strategies (limitations, model used for simulations); used input parameters and results are presented in Sections 4 and 5; and fi nally, we conclude with a discussion of the results and their implications on lahar hazard. Cotopaxi is one of the highest active volcanoes of the world: its per- fect ice-capped, 2200 m-high cone reaches an elevation of 5897 m above sea level with a basal diameter of 22 km. The volcano is located 60 km south of Quito, the capital of Ecuador, and it is surrounded by villages, national lines of communication and other infrastructure (Fig. 1). Its steep fl anks (30° – 35°) culminate in a crater hosting a tephra cone (Hradecka et al., 1974), which in turn harbors a smaller crater, the active vent of historic eruptions. The volcano has three main drainage systems (Fig. 2): 1) to the north is the Pita – Guayallabamba river system that fl ows through the Los Chillos and Tumbaco valleys, both lying east of Quito and 400 m lower in elevation; 2) to the east is the Tambo – Tamboyacu river system, which forms part of the Río Napo and fl ows through the sparsely populated Amazonian lowlands; and 3) to the west – southwest, the drainage includes 2 main ‘ quebradas ’ (small, steep-sided valleys): the Barrancas and Saquìmala, whose streams eventually join with the Río Cutuchi and fl ow through the center of the Latacunga valley. The city of Latacunga and settlements along the entire valley grew rapidly in recent decades within the city and along the river, in particular with greenhouses for growing roses (Fig. 1), resulting presently in a very densely populated area with high industrial investment where houses, factories, hospitals, schools, the Pan American Highway and other infrastructure have been built on top of lahar deposits of less than 1000 years old. The population in Latacunga was estimated at 63,800 inhabitants in 2012, living in an area of about 10 km 2 with a population density of 6.4/km 2 (INEC — Instituto Nacional de Estadistica y Censos, 2012); the previous census of 2001 estimated the population at 52,000, with an increase of more than 20% in the past 10 years. The recent eruptive activity of Cotopaxi consists of Plinian outbursts of VEI 3 – 4 (at least 20 episodes in the last 2000 years; Barberi et al., 1995; Pistolesi et al., 2011), with associated lava fl ows, long-lasting ash eruptions and minor Strombolian activity. Pistolesi et al. (2011) upgraded the information made available by previous studies (Barberi et al., 1995; Mothes et al., 2004; Hall and Mothes, 2008a) and recog- nized 21 continuous tephra beds subdivided in three eruptive periods, from the emplacement of the regional tephra marker of the Quilotoa volcano ash (1150 AD; Hall and Mothes, 2008b) to the present (Fig. 3). The fi rst eruptive period (1150 to 1534) was characterized by two mid-intensity explosive eruptions (layers B L and S W , Fig. 3), apparently separated and followed by quiescent periods a few centuries long. Activity of the second period coincided with two Plinian eruptions in 1742 – 44 (layer M T ) and in 1766 – 68 (layer M B ). The latter, in particular, was associated with the formation of scoria fl ows, which were dispersed over large areas on the northern side of the cone. After the M B eruption, activity at Cotopaxi passed into a period of long- lasting, moderate-intensity ash generation punctuated by the occurrence of low-magnitude, short-lived sub-Plinian eruptions (P D , P L and P E; third period, Fig. 3). ‘ Boiling-over activity ’ and generation of scoria fl ow deposits were always associated with these events. Tephra deposits of the last large eruptive event of Cotopaxi, which occurred in 1877, are recorded by the P E layer. January 2001 saw increased level of activity at Cotopaxi volcano (Molina et al., 2006) with an increase of long-period (LP) seismic events accompanied by very long-period (VLP) events occurring directly under the volcano; fumarolic activity at the summit and a thermal anomaly suggesting injection of new magma ...

Context 14

... or volcanic debris fl ows, consist of high-concentration, sediment-loaded fl ows that occur on volcanic terrains. They may be a primary phenomenon directly triggered by eruptive activity (e.g. the 1985 Nevado del Ruiz event in Colombia; Pierson et al., 1990) or they may result from the post-eruptive mobilization (secondary lahars) of volcanic debris (e.g. the 1991 eruption of Mt. Pinatubo, Philippines; Rodolfo et al., 1996). Lahar generation requires a combination of three main factors including: (i) a triggering mechanism that rapidly makes available an adequate water source; (ii) the availability of abundant, un- consolidated debris; and (iii) steep slopes (Vallance, 2000). Due to water incorporation and volume increase, lahars can easily over fl ow lateral banks and spread over areas of low gradient. This can produce catastrophic consequences for the communities living along the areas, which can be inundated unexpectedly by lahars — as recently shown by the Pinatubo, Mayon and Nevado del Ruiz events (Pierson et al., 1990; Voight, 1990; Rodolfo, 1995; Newhall and Punongbayan, 1996; Tanguy et al., 1998). According to Varnes (1978), we include here with the term ‘ lahars ’ not only debris fl ows but also hyperconcentrated mud fl ows (Smith and Lowe, 1991). Although the term has sometimes been used to refer to the deposits of such fl ows, Smith and Fritz (1989, p. 375) dismiss this meaning stating that “ lahar is an event that can refer to one or more discrete processes, but does not refer to a deposit ” . Small debris fl ows occur daily on the fl anks of many active and inactive volcanoes; although traditionally the term ‘ lahar ’ has generally been restricted to those events which can generate a hazard to populations (Scott, 1988), nowadays lahar is fairly broadly used regardless if the event poses hazard to population or not (Manville et al., 2009). In the following, we will use lahar to refer to the process, and as a synonym of debris fl ow. Lahars can be generated at crater lakes, following crater failures (Bornas et al., 2003; Manville and Cronin, 2007; Manville, 2010; Massey et al., 2010), or explosive expulsion of water (Zen and Hadikusumo, 1965; Nairn et al., 1979; Suryo and Clarke, 1985; Thouret et al., 1998; Németh et al., 2006; Kilgour et al., 2010) or by heavy rainfall on freshly deposited tephra or loose material on volcano slopes, as in the case of the 1991 eruption of Mt. Pinatubo (Rodolfo, 1989; Pierson et al., 1992; Arboleda and Martinez, 1996; Rodolfo et al., 1996). Lahars are also generated by the rapid melting of snow or ice on ice clad volcanoes, due to the interaction of eruptive products with the ice cap (Major and Newhall, 1989). As demonstrated by the 1985 Nevado del Ruiz event in Colombia, even small volcanic eruptions can trigger catastrophic lahar events when pyroclastic material interacts with a summit glacier (Pierson et al., 1990; Voight, 1990; Tanguy et al., 1998), especially if combined with seismic shaking and intense scouring of the ice (with consequent channeling of the lahar). During the past 30 years, lahar-related disasters have been docu- mented worldwide in the volcanological literature (Voight, 1990; Hall, 1992; Rodolfo, 1995; Newhall and Punongbayan, 1996; Tanguy et al., 1998; Wood and Soulard, 2009). In the last century, about 30,000 casualties were reported in relation to the occurrence of lahar events (Witham, 2005); about 80% of this terri fi c quota is related to a single event (the devastation of the village of Armero, following the 1985 eruption of Nevado del Ruiz). Continuous growth of buildings and settled areas in lahar fl ow paths demonstrates that better risk perception, land-use planning and rapid evacuation plans may combine in reducing damage and loss of life from future lahars (Wood and Soulard, 2009). Cotopaxi volcano is well known for the potential destructiveness of its lahars, and models of lahar invasion have been recently applied to the northern and southern drainages (Barberi et al., 1992; Aguilera et al., 2004). Inundation areas for future lahars have also been partially explored by several authors (Miller et al., 1978; Hall and von Hillebrandt, 1988a, 1988b; Mothes et al., 2004; Mothes, 2006) who presented maps of lahar inundation areas related to the last (1877) eruptive event and for future lahar-generating events of similar size. Modeling procedures were not presented in detail in these papers so that a more in-depth discussion on the potential of lahar inundation at Cotopaxi and the mapping of the related hazard are elements of con- siderable importance. In this work, we concentrate on lahar hazard assessment in the Río Cutuchi system, which conveys all the main drainages of west and south of the Cotopaxi cone (Fig. 1). The Cutuchi valley hosts several human settlements, most of which lie on recent lahar deposit terraces, only a few meters above the present river bed. Further south, the Río Cutuchi crosses the large urban settlement of Latacunga (43 km SE of the volcano, Fig. 1), with a population of 52,000 inhabitants. Its path is generally parallel to the Pan American Highway, the major state road connecting the country from north to south. Many small-scale lahars have followed this drainage in the past ( b 100 million m 3 ), as well as some large-scale historical fl ows that severely impacted the population. We present lahar simulations with different starting volumes performed with a GIS-based semi-empirical model (LAHARZ; Iverson et al., 1998) in order to evaluate the inundating potential of Cotopaxi lahars and to assess the inundation hazard in the southern part of the volcano towards the city of Latacunga. The manuscript is arranged with a section (Section 2) in which we brie fl y describe the geological data relevant to this study (recent Cotopaxi activity, lahar deposits, lahar triggering mechanisms, and the factors which control lahar volume). In Section 3 we introduce and discuss lahar modeling approaches and strategies (limitations, model used for simulations); used input parameters and results are presented in Sections 4 and 5; and fi nally, we conclude with a discussion of the results and their implications on lahar hazard. Cotopaxi is one of the highest active volcanoes of the world: its per- fect ice-capped, 2200 m-high cone reaches an elevation of 5897 m above sea level with a basal diameter of 22 km. The volcano is located 60 km south of Quito, the capital of Ecuador, and it is surrounded by villages, national lines of communication and other infrastructure (Fig. 1). Its steep fl anks (30° – 35°) culminate in a crater hosting a tephra cone (Hradecka et al., 1974), which in turn harbors a smaller crater, the active vent of historic eruptions. The volcano has three main drainage systems (Fig. 2): 1) to the north is the Pita – Guayallabamba river system that fl ows through the Los Chillos and Tumbaco valleys, both lying east of Quito and 400 m lower in elevation; 2) to the east is the Tambo – Tamboyacu river system, which forms part of the Río Napo and fl ows through the sparsely populated Amazonian lowlands; and 3) to the west – southwest, the drainage includes 2 main ‘ quebradas ’ (small, steep-sided valleys): the Barrancas and Saquìmala, whose streams eventually join with the Río Cutuchi and fl ow through the center of the Latacunga valley. The city of Latacunga and settlements along the entire valley grew rapidly in recent decades within the city and along the river, in particular with greenhouses for growing roses (Fig. 1), resulting presently in a very densely populated area with high industrial investment where houses, factories, hospitals, schools, the Pan American Highway and other infrastructure have been built on top of lahar deposits of less than 1000 years old. The population in Latacunga was estimated at 63,800 inhabitants in 2012, living in an area of about 10 km 2 with a population density of 6.4/km 2 (INEC — Instituto Nacional de Estadistica y Censos, 2012); the previous census of 2001 estimated the population at 52,000, with an increase of more than 20% in the past 10 years. The recent eruptive activity of Cotopaxi consists of Plinian outbursts of VEI 3 – 4 (at least 20 episodes in the last 2000 years; Barberi et al., 1995; Pistolesi et al., 2011), with associated lava fl ows, long-lasting ash eruptions and minor Strombolian activity. Pistolesi et al. (2011) upgraded the information made available by previous studies (Barberi et al., 1995; Mothes et al., 2004; Hall and Mothes, 2008a) and recog- nized 21 continuous tephra beds subdivided in three eruptive periods, from the emplacement of the regional tephra marker of the Quilotoa volcano ash (1150 AD; Hall and Mothes, 2008b) to the present (Fig. 3). The fi rst eruptive period (1150 to 1534) was characterized by two mid-intensity explosive eruptions (layers B L and S W , Fig. 3), apparently separated and followed by quiescent periods a few centuries long. Activity of the second period coincided with two Plinian eruptions in 1742 – 44 (layer M T ) and in 1766 – 68 (layer M B ). The latter, in particular, was associated with the formation of scoria fl ows, which were dispersed over large areas on the northern side of the cone. After the M B eruption, activity at Cotopaxi passed into a period of long- lasting, moderate-intensity ash generation punctuated by the occurrence of low-magnitude, short-lived sub-Plinian eruptions (P D , P L and P E; third period, Fig. 3). ‘ Boiling-over activity ’ and generation of scoria fl ow deposits were always associated with these events. Tephra deposits of the last large eruptive event of Cotopaxi, which occurred in 1877, are recorded by the P E layer. January 2001 saw increased level of activity ...

Context 15

... anks of many active and inactive volcanoes; although traditionally the term ‘ lahar ’ has generally been restricted to those events which can generate a hazard to populations (Scott, 1988), nowadays lahar is fairly broadly used regardless if the event poses hazard to population or not (Manville et al., 2009). In the following, we will use lahar to refer to the process, and as a synonym of debris fl ow. Lahars can be generated at crater lakes, following crater failures (Bornas et al., 2003; Manville and Cronin, 2007; Manville, 2010; Massey et al., 2010), or explosive expulsion of water (Zen and Hadikusumo, 1965; Nairn et al., 1979; Suryo and Clarke, 1985; Thouret et al., 1998; Németh et al., 2006; Kilgour et al., 2010) or by heavy rainfall on freshly deposited tephra or loose material on volcano slopes, as in the case of the 1991 eruption of Mt. Pinatubo (Rodolfo, 1989; Pierson et al., 1992; Arboleda and Martinez, 1996; Rodolfo et al., 1996). Lahars are also generated by the rapid melting of snow or ice on ice clad volcanoes, due to the interaction of eruptive products with the ice cap (Major and Newhall, 1989). As demonstrated by the 1985 Nevado del Ruiz event in Colombia, even small volcanic eruptions can trigger catastrophic lahar events when pyroclastic material interacts with a summit glacier (Pierson et al., 1990; Voight, 1990; Tanguy et al., 1998), especially if combined with seismic shaking and intense scouring of the ice (with consequent channeling of the lahar). During the past 30 years, lahar-related disasters have been docu- mented worldwide in the volcanological literature (Voight, 1990; Hall, 1992; Rodolfo, 1995; Newhall and Punongbayan, 1996; Tanguy et al., 1998; Wood and Soulard, 2009). In the last century, about 30,000 casualties were reported in relation to the occurrence of lahar events (Witham, 2005); about 80% of this terri fi c quota is related to a single event (the devastation of the village of Armero, following the 1985 eruption of Nevado del Ruiz). Continuous growth of buildings and settled areas in lahar fl ow paths demonstrates that better risk perception, land-use planning and rapid evacuation plans may combine in reducing damage and loss of life from future lahars (Wood and Soulard, 2009). Cotopaxi volcano is well known for the potential destructiveness of its lahars, and models of lahar invasion have been recently applied to the northern and southern drainages (Barberi et al., 1992; Aguilera et al., 2004). Inundation areas for future lahars have also been partially explored by several authors (Miller et al., 1978; Hall and von Hillebrandt, 1988a, 1988b; Mothes et al., 2004; Mothes, 2006) who presented maps of lahar inundation areas related to the last (1877) eruptive event and for future lahar-generating events of similar size. Modeling procedures were not presented in detail in these papers so that a more in-depth discussion on the potential of lahar inundation at Cotopaxi and the mapping of the related hazard are elements of con- siderable importance. In this work, we concentrate on lahar hazard assessment in the Río Cutuchi system, which conveys all the main drainages of west and south of the Cotopaxi cone (Fig. 1). The Cutuchi valley hosts several human settlements, most of which lie on recent lahar deposit terraces, only a few meters above the present river bed. Further south, the Río Cutuchi crosses the large urban settlement of Latacunga (43 km SE of the volcano, Fig. 1), with a population of 52,000 inhabitants. Its path is generally parallel to the Pan American Highway, the major state road connecting the country from north to south. Many small-scale lahars have followed this drainage in the past ( b 100 million m 3 ), as well as some large-scale historical fl ows that severely impacted the population. We present lahar simulations with different starting volumes performed with a GIS-based semi-empirical model (LAHARZ; Iverson et al., 1998) in order to evaluate the inundating potential of Cotopaxi lahars and to assess the inundation hazard in the southern part of the volcano towards the city of Latacunga. The manuscript is arranged with a section (Section 2) in which we brie fl y describe the geological data relevant to this study (recent Cotopaxi activity, lahar deposits, lahar triggering mechanisms, and the factors which control lahar volume). In Section 3 we introduce and discuss lahar modeling approaches and strategies (limitations, model used for simulations); used input parameters and results are presented in Sections 4 and 5; and fi nally, we conclude with a discussion of the results and their implications on lahar hazard. Cotopaxi is one of the highest active volcanoes of the world: its per- fect ice-capped, 2200 m-high cone reaches an elevation of 5897 m above sea level with a basal diameter of 22 km. The volcano is located 60 km south of Quito, the capital of Ecuador, and it is surrounded by villages, national lines of communication and other infrastructure (Fig. 1). Its steep fl anks (30° – 35°) culminate in a crater hosting a tephra cone (Hradecka et al., 1974), which in turn harbors a smaller crater, the active vent of historic eruptions. The volcano has three main drainage systems (Fig. 2): 1) to the north is the Pita – Guayallabamba river system that fl ows through the Los Chillos and Tumbaco valleys, both lying east of Quito and 400 m lower in elevation; 2) to the east is the Tambo – Tamboyacu river system, which forms part of the Río Napo and fl ows through the sparsely populated Amazonian lowlands; and 3) to the west – southwest, the drainage includes 2 main ‘ quebradas ’ (small, steep-sided valleys): the Barrancas and Saquìmala, whose streams eventually join with the Río Cutuchi and fl ow through the center of the Latacunga valley. The city of Latacunga and settlements along the entire valley grew rapidly in recent decades within the city and along the river, in particular with greenhouses for growing roses (Fig. 1), resulting presently in a very densely populated area with high industrial investment where houses, factories, hospitals, schools, the Pan American Highway and other infrastructure have been built on top of lahar deposits of less than 1000 years old. The population in Latacunga was estimated at 63,800 inhabitants in 2012, living in an area of about 10 km 2 with a population density of 6.4/km 2 (INEC — Instituto Nacional de Estadistica y Censos, 2012); the previous census of 2001 estimated the population at 52,000, with an increase of more than 20% in the past 10 years. The recent eruptive activity of Cotopaxi consists of Plinian outbursts of VEI 3 – 4 (at least 20 episodes in the last 2000 years; Barberi et al., 1995; Pistolesi et al., 2011), with associated lava fl ows, long-lasting ash eruptions and minor Strombolian activity. Pistolesi et al. (2011) upgraded the information made available by previous studies (Barberi et al., 1995; Mothes et al., 2004; Hall and Mothes, 2008a) and recog- nized 21 continuous tephra beds subdivided in three eruptive periods, from the emplacement of the regional tephra marker of the Quilotoa volcano ash (1150 AD; Hall and Mothes, 2008b) to the present (Fig. 3). The fi rst eruptive period (1150 to 1534) was characterized by two mid-intensity explosive eruptions (layers B L and S W , Fig. 3), apparently separated and followed by quiescent periods a few centuries long. Activity of the second period coincided with two Plinian eruptions in 1742 – 44 (layer M T ) and in 1766 – 68 (layer M B ). The latter, in particular, was associated with the formation of scoria fl ows, which were dispersed over large areas on the northern side of the cone. After the M B eruption, activity at Cotopaxi passed into a period of long- lasting, moderate-intensity ash generation punctuated by the occurrence of low-magnitude, short-lived sub-Plinian eruptions (P D , P L and P E; third period, Fig. 3). ‘ Boiling-over activity ’ and generation of scoria fl ow deposits were always associated with these events. Tephra deposits of the last large eruptive event of Cotopaxi, which occurred in 1877, are recorded by the P E layer. January 2001 saw increased level of activity at Cotopaxi volcano (Molina et al., 2006) with an increase of long-period (LP) seismic events accompanied by very long-period (VLP) events occurring directly under the volcano; fumarolic activity at the summit and a thermal anomaly suggesting injection of new magma (Ramon et al., 2006; Rivero et al., 2006). This evidence and the general behavior of the volcano as portrayed by its geological record (Barberi et al., 1995; Hall and Mothes, 2008a; Pistolesi et al., 2011) make the problem of lahar generation particularly signi fi cant. Río Cutuchi is the main river southwest of Cotopaxi. Its headwaters fl ow to the south of Rumiñahui (Fig. 1), a deeply eroded extinct volcano situated to the west of Cotopaxi, and its tributaries cover the entire western sector of Cotopaxi. Stratigraphic studies along the Río Cutuchi valleys revealed an impressive sequence of lahar deposits (Mothes et al., 2004; Pistolesi et al., 2013). Historical accounts also describe that since the Spanish Conquest (XVI century), most of the lahars produced by Cotopaxi fl owed towards the western and southwest- ern sectors of the volcano along the Río Cutuchi (Mothes et al., 2004; Figs. 2, 3), as reported by Sodiro (1877, p. 4): ( ... ) (in 1742) “ a terrible inundation in near regions destroyed bridges and farms, killing people and livestock ( ... ), and (in 1744) a huge fl ood rapidly arrived in Latacunga which, after the destruction of whatever en- countered along its way, invaded the center of the city, destroying houses ( ... ) ” . Based on the recognition of thin fallout tephra beds within the lahar sequence, Pistolesi (2008) and Pistolesi et al. (2013) proposed a stratigraphic reconstruction of post-XII century debris fl ows and compared their scale and dynamics (Fig. 4A). Well-constrained 14 C ...

Context 16

... by the rapid melting of snow or ice on ice clad volcanoes, due to the interaction of eruptive products with the ice cap (Major and Newhall, 1989). As demonstrated by the 1985 Nevado del Ruiz event in Colombia, even small volcanic eruptions can trigger catastrophic lahar events when pyroclastic material interacts with a summit glacier (Pierson et al., 1990; Voight, 1990; Tanguy et al., 1998), especially if combined with seismic shaking and intense scouring of the ice (with consequent channeling of the lahar). During the past 30 years, lahar-related disasters have been docu- mented worldwide in the volcanological literature (Voight, 1990; Hall, 1992; Rodolfo, 1995; Newhall and Punongbayan, 1996; Tanguy et al., 1998; Wood and Soulard, 2009). In the last century, about 30,000 casualties were reported in relation to the occurrence of lahar events (Witham, 2005); about 80% of this terri fi c quota is related to a single event (the devastation of the village of Armero, following the 1985 eruption of Nevado del Ruiz). Continuous growth of buildings and settled areas in lahar fl ow paths demonstrates that better risk perception, land-use planning and rapid evacuation plans may combine in reducing damage and loss of life from future lahars (Wood and Soulard, 2009). Cotopaxi volcano is well known for the potential destructiveness of its lahars, and models of lahar invasion have been recently applied to the northern and southern drainages (Barberi et al., 1992; Aguilera et al., 2004). Inundation areas for future lahars have also been partially explored by several authors (Miller et al., 1978; Hall and von Hillebrandt, 1988a, 1988b; Mothes et al., 2004; Mothes, 2006) who presented maps of lahar inundation areas related to the last (1877) eruptive event and for future lahar-generating events of similar size. Modeling procedures were not presented in detail in these papers so that a more in-depth discussion on the potential of lahar inundation at Cotopaxi and the mapping of the related hazard are elements of con- siderable importance. In this work, we concentrate on lahar hazard assessment in the Río Cutuchi system, which conveys all the main drainages of west and south of the Cotopaxi cone (Fig. 1). The Cutuchi valley hosts several human settlements, most of which lie on recent lahar deposit terraces, only a few meters above the present river bed. Further south, the Río Cutuchi crosses the large urban settlement of Latacunga (43 km SE of the volcano, Fig. 1), with a population of 52,000 inhabitants. Its path is generally parallel to the Pan American Highway, the major state road connecting the country from north to south. Many small-scale lahars have followed this drainage in the past ( b 100 million m 3 ), as well as some large-scale historical fl ows that severely impacted the population. We present lahar simulations with different starting volumes performed with a GIS-based semi-empirical model (LAHARZ; Iverson et al., 1998) in order to evaluate the inundating potential of Cotopaxi lahars and to assess the inundation hazard in the southern part of the volcano towards the city of Latacunga. The manuscript is arranged with a section (Section 2) in which we brie fl y describe the geological data relevant to this study (recent Cotopaxi activity, lahar deposits, lahar triggering mechanisms, and the factors which control lahar volume). In Section 3 we introduce and discuss lahar modeling approaches and strategies (limitations, model used for simulations); used input parameters and results are presented in Sections 4 and 5; and fi nally, we conclude with a discussion of the results and their implications on lahar hazard. Cotopaxi is one of the highest active volcanoes of the world: its per- fect ice-capped, 2200 m-high cone reaches an elevation of 5897 m above sea level with a basal diameter of 22 km. The volcano is located 60 km south of Quito, the capital of Ecuador, and it is surrounded by villages, national lines of communication and other infrastructure (Fig. 1). Its steep fl anks (30° – 35°) culminate in a crater hosting a tephra cone (Hradecka et al., 1974), which in turn harbors a smaller crater, the active vent of historic eruptions. The volcano has three main drainage systems (Fig. 2): 1) to the north is the Pita – Guayallabamba river system that fl ows through the Los Chillos and Tumbaco valleys, both lying east of Quito and 400 m lower in elevation; 2) to the east is the Tambo – Tamboyacu river system, which forms part of the Río Napo and fl ows through the sparsely populated Amazonian lowlands; and 3) to the west – southwest, the drainage includes 2 main ‘ quebradas ’ (small, steep-sided valleys): the Barrancas and Saquìmala, whose streams eventually join with the Río Cutuchi and fl ow through the center of the Latacunga valley. The city of Latacunga and settlements along the entire valley grew rapidly in recent decades within the city and along the river, in particular with greenhouses for growing roses (Fig. 1), resulting presently in a very densely populated area with high industrial investment where houses, factories, hospitals, schools, the Pan American Highway and other infrastructure have been built on top of lahar deposits of less than 1000 years old. The population in Latacunga was estimated at 63,800 inhabitants in 2012, living in an area of about 10 km 2 with a population density of 6.4/km 2 (INEC — Instituto Nacional de Estadistica y Censos, 2012); the previous census of 2001 estimated the population at 52,000, with an increase of more than 20% in the past 10 years. The recent eruptive activity of Cotopaxi consists of Plinian outbursts of VEI 3 – 4 (at least 20 episodes in the last 2000 years; Barberi et al., 1995; Pistolesi et al., 2011), with associated lava fl ows, long-lasting ash eruptions and minor Strombolian activity. Pistolesi et al. (2011) upgraded the information made available by previous studies (Barberi et al., 1995; Mothes et al., 2004; Hall and Mothes, 2008a) and recog- nized 21 continuous tephra beds subdivided in three eruptive periods, from the emplacement of the regional tephra marker of the Quilotoa volcano ash (1150 AD; Hall and Mothes, 2008b) to the present (Fig. 3). The fi rst eruptive period (1150 to 1534) was characterized by two mid-intensity explosive eruptions (layers B L and S W , Fig. 3), apparently separated and followed by quiescent periods a few centuries long. Activity of the second period coincided with two Plinian eruptions in 1742 – 44 (layer M T ) and in 1766 – 68 (layer M B ). The latter, in particular, was associated with the formation of scoria fl ows, which were dispersed over large areas on the northern side of the cone. After the M B eruption, activity at Cotopaxi passed into a period of long- lasting, moderate-intensity ash generation punctuated by the occurrence of low-magnitude, short-lived sub-Plinian eruptions (P D , P L and P E; third period, Fig. 3). ‘ Boiling-over activity ’ and generation of scoria fl ow deposits were always associated with these events. Tephra deposits of the last large eruptive event of Cotopaxi, which occurred in 1877, are recorded by the P E layer. January 2001 saw increased level of activity at Cotopaxi volcano (Molina et al., 2006) with an increase of long-period (LP) seismic events accompanied by very long-period (VLP) events occurring directly under the volcano; fumarolic activity at the summit and a thermal anomaly suggesting injection of new magma (Ramon et al., 2006; Rivero et al., 2006). This evidence and the general behavior of the volcano as portrayed by its geological record (Barberi et al., 1995; Hall and Mothes, 2008a; Pistolesi et al., 2011) make the problem of lahar generation particularly signi fi cant. Río Cutuchi is the main river southwest of Cotopaxi. Its headwaters fl ow to the south of Rumiñahui (Fig. 1), a deeply eroded extinct volcano situated to the west of Cotopaxi, and its tributaries cover the entire western sector of Cotopaxi. Stratigraphic studies along the Río Cutuchi valleys revealed an impressive sequence of lahar deposits (Mothes et al., 2004; Pistolesi et al., 2013). Historical accounts also describe that since the Spanish Conquest (XVI century), most of the lahars produced by Cotopaxi fl owed towards the western and southwest- ern sectors of the volcano along the Río Cutuchi (Mothes et al., 2004; Figs. 2, 3), as reported by Sodiro (1877, p. 4): ( ... ) (in 1742) “ a terrible inundation in near regions destroyed bridges and farms, killing people and livestock ( ... ), and (in 1744) a huge fl ood rapidly arrived in Latacunga which, after the destruction of whatever en- countered along its way, invaded the center of the city, destroying houses ( ... ) ” . Based on the recognition of thin fallout tephra beds within the lahar sequence, Pistolesi (2008) and Pistolesi et al. (2013) proposed a stratigraphic reconstruction of post-XII century debris fl ows and compared their scale and dynamics (Fig. 4A). Well-constrained 14 C calendar ages performed on soils and charred grass were assigned to most lahar deposits thanks to the preservation of tephra beds intercalated between the lahar deposits, and these were directly related to their generating explosive eruptions (Fig. 3). Debris fl ow deposits were also informally grouped into three main categories (Types A, B and C) by combining sedimentological data obtained from a large quarried area along the Río Cutuchi (Fig. 1). These three categories of deposits largely vary in terms of dispersal, thickness, average and maximum size of transported blocks increasing from A to C, and were related to lahars with different sizes and trigger mechanisms (Pistolesi et al., 2013) (Fig. 4A, B, C). The stratigraphy of debris fl ow deposits in the Cutuchi quarries (Fig. 3) clearly highlights that the period between 1200 and 1500 AD is characterized by lahar events of large ( N 500 million m 3 ) scale, which carried meter-sized boulders and left deposits more ...

Context 17

... with different starting volumes performed with a GIS-based semi-empirical model (LAHARZ; Iverson et al., 1998) in order to evaluate the inundating potential of Cotopaxi lahars and to assess the inundation hazard in the southern part of the volcano towards the city of Latacunga. The manuscript is arranged with a section (Section 2) in which we brie fl y describe the geological data relevant to this study (recent Cotopaxi activity, lahar deposits, lahar triggering mechanisms, and the factors which control lahar volume). In Section 3 we introduce and discuss lahar modeling approaches and strategies (limitations, model used for simulations); used input parameters and results are presented in Sections 4 and 5; and fi nally, we conclude with a discussion of the results and their implications on lahar hazard. Cotopaxi is one of the highest active volcanoes of the world: its per- fect ice-capped, 2200 m-high cone reaches an elevation of 5897 m above sea level with a basal diameter of 22 km. The volcano is located 60 km south of Quito, the capital of Ecuador, and it is surrounded by villages, national lines of communication and other infrastructure (Fig. 1). Its steep fl anks (30° – 35°) culminate in a crater hosting a tephra cone (Hradecka et al., 1974), which in turn harbors a smaller crater, the active vent of historic eruptions. The volcano has three main drainage systems (Fig. 2): 1) to the north is the Pita – Guayallabamba river system that fl ows through the Los Chillos and Tumbaco valleys, both lying east of Quito and 400 m lower in elevation; 2) to the east is the Tambo – Tamboyacu river system, which forms part of the Río Napo and fl ows through the sparsely populated Amazonian lowlands; and 3) to the west – southwest, the drainage includes 2 main ‘ quebradas ’ (small, steep-sided valleys): the Barrancas and Saquìmala, whose streams eventually join with the Río Cutuchi and fl ow through the center of the Latacunga valley. The city of Latacunga and settlements along the entire valley grew rapidly in recent decades within the city and along the river, in particular with greenhouses for growing roses (Fig. 1), resulting presently in a very densely populated area with high industrial investment where houses, factories, hospitals, schools, the Pan American Highway and other infrastructure have been built on top of lahar deposits of less than 1000 years old. The population in Latacunga was estimated at 63,800 inhabitants in 2012, living in an area of about 10 km 2 with a population density of 6.4/km 2 (INEC — Instituto Nacional de Estadistica y Censos, 2012); the previous census of 2001 estimated the population at 52,000, with an increase of more than 20% in the past 10 years. The recent eruptive activity of Cotopaxi consists of Plinian outbursts of VEI 3 – 4 (at least 20 episodes in the last 2000 years; Barberi et al., 1995; Pistolesi et al., 2011), with associated lava fl ows, long-lasting ash eruptions and minor Strombolian activity. Pistolesi et al. (2011) upgraded the information made available by previous studies (Barberi et al., 1995; Mothes et al., 2004; Hall and Mothes, 2008a) and recog- nized 21 continuous tephra beds subdivided in three eruptive periods, from the emplacement of the regional tephra marker of the Quilotoa volcano ash (1150 AD; Hall and Mothes, 2008b) to the present (Fig. 3). The fi rst eruptive period (1150 to 1534) was characterized by two mid-intensity explosive eruptions (layers B L and S W , Fig. 3), apparently separated and followed by quiescent periods a few centuries long. Activity of the second period coincided with two Plinian eruptions in 1742 – 44 (layer M T ) and in 1766 – 68 (layer M B ). The latter, in particular, was associated with the formation of scoria fl ows, which were dispersed over large areas on the northern side of the cone. After the M B eruption, activity at Cotopaxi passed into a period of long- lasting, moderate-intensity ash generation punctuated by the occurrence of low-magnitude, short-lived sub-Plinian eruptions (P D , P L and P E; third period, Fig. 3). ‘ Boiling-over activity ’ and generation of scoria fl ow deposits were always associated with these events. Tephra deposits of the last large eruptive event of Cotopaxi, which occurred in 1877, are recorded by the P E layer. January 2001 saw increased level of activity at Cotopaxi volcano (Molina et al., 2006) with an increase of long-period (LP) seismic events accompanied by very long-period (VLP) events occurring directly under the volcano; fumarolic activity at the summit and a thermal anomaly suggesting injection of new magma (Ramon et al., 2006; Rivero et al., 2006). This evidence and the general behavior of the volcano as portrayed by its geological record (Barberi et al., 1995; Hall and Mothes, 2008a; Pistolesi et al., 2011) make the problem of lahar generation particularly signi fi cant. Río Cutuchi is the main river southwest of Cotopaxi. Its headwaters fl ow to the south of Rumiñahui (Fig. 1), a deeply eroded extinct volcano situated to the west of Cotopaxi, and its tributaries cover the entire western sector of Cotopaxi. Stratigraphic studies along the Río Cutuchi valleys revealed an impressive sequence of lahar deposits (Mothes et al., 2004; Pistolesi et al., 2013). Historical accounts also describe that since the Spanish Conquest (XVI century), most of the lahars produced by Cotopaxi fl owed towards the western and southwest- ern sectors of the volcano along the Río Cutuchi (Mothes et al., 2004; Figs. 2, 3), as reported by Sodiro (1877, p. 4): ( ... ) (in 1742) “ a terrible inundation in near regions destroyed bridges and farms, killing people and livestock ( ... ), and (in 1744) a huge fl ood rapidly arrived in Latacunga which, after the destruction of whatever en- countered along its way, invaded the center of the city, destroying houses ( ... ) ” . Based on the recognition of thin fallout tephra beds within the lahar sequence, Pistolesi (2008) and Pistolesi et al. (2013) proposed a stratigraphic reconstruction of post-XII century debris fl ows and compared their scale and dynamics (Fig. 4A). Well-constrained 14 C calendar ages performed on soils and charred grass were assigned to most lahar deposits thanks to the preservation of tephra beds intercalated between the lahar deposits, and these were directly related to their generating explosive eruptions (Fig. 3). Debris fl ow deposits were also informally grouped into three main categories (Types A, B and C) by combining sedimentological data obtained from a large quarried area along the Río Cutuchi (Fig. 1). These three categories of deposits largely vary in terms of dispersal, thickness, average and maximum size of transported blocks increasing from A to C, and were related to lahars with different sizes and trigger mechanisms (Pistolesi et al., 2013) (Fig. 4A, B, C). The stratigraphy of debris fl ow deposits in the Cutuchi quarries (Fig. 3) clearly highlights that the period between 1200 and 1500 AD is characterized by lahar events of large ( N 500 million m 3 ) scale, which carried meter-sized boulders and left deposits more than 20 m thick (Figs. 3, 4B, C). Conversely, lahars of the period of activity between 1700 and 1877 (the last eruptive event) show a variable but lower scale in terms of distribution, thickness and size of transported blocks; within this period, the 1877 lahar deposits are smaller than those related to the XVIII century activity. The 1877 eruption (a sub-Plinian event of magnitude VEI 4; Hall and Mothes, 2008a; Pistolesi et al., 2011), in particular, triggered large fl oods along all the main drainages. After exiting the glacier, these water-dominated fl ows mixed with pyroclastic and ice blocks, causing intense erosion and gullying, and rapidly evolved into debris fl ows (Aguilera et al., 2004). Typical valleys along the volcano's fl anks are represented by steep ( N 20°), narrow canyons laterally con- fi ned by vertical stacks of lavas and pyroclastic sequences. The peak discharge of 1877 fl ows was estimated between 50,000 and 60,000 m 3 s − 1 , with velocities in excess of 20 m s − 1 (Mothes et al., 2004). Lahars descending south of Cotopaxi initially fl owed along three main paths (Río Cutuchi, Río Saquìmala and Río Barrancas – Alaquez) and merged at a distance of about 24 km from the crater. The morphology at the foot of the cone rapidly changes into fl at ( b 5°), wide valleys separated by gentle slopes mostly covered by grass, where part of the coarse solid material has been commonly discharged by past debris fl ows. Lahar fl ow depths in the upper Cutuchi course have been estimated to be up to 20 m (Barberi et al., 1992; Mothes et al., 2004), increasing south of Latacunga where the valley gets narrower. Mothes et al. (2004) calculated a total volume of 100 million m 3 for the 1877 lahar drained from the canyons with different discharges and dropping into the Latacunga valley. Barberi et al. (1992), applying equal discharges in all major ‘ quebradas ’ , estimated a volume of 150 million m 3 . In the area of Latacunga, the lahar sequence is today exposed as a stack of debris fl ow deposits in gravel quarries and terrace cliffs. Although too far ( ≈ 40 km) from the volcano to preserve recognizable primary tephra fallout deposits and out of the main dispersal axis of historical Cotopaxi eruptions, the regionally-dispersed Quilotoa ash fall is visible within the sequence (Fig. 4D), separating pre- and post-1150 AD lahar deposits. The role of pyroclastic fl ows in generating rapid melting of snow and ice has been largely discussed in the volcanological literature (Major and Newhall, 1989; Pierson et al., 1990; Zernack et al., 2011). At Cotopaxi, in particular, ice scouring and melting during scoria fl ow events have been observed and described by chroniclers: during the 1877 eruption (and presumably during the 1766 – 68 and 1853 eruptions as well), “ a dark foam-like cloud boiled over the crater, ...

Context 18

... – Tamboyacu river system, which forms part of the Río Napo and fl ows through the sparsely populated Amazonian lowlands; and 3) to the west – southwest, the drainage includes 2 main ‘ quebradas ’ (small, steep-sided valleys): the Barrancas and Saquìmala, whose streams eventually join with the Río Cutuchi and fl ow through the center of the Latacunga valley. The city of Latacunga and settlements along the entire valley grew rapidly in recent decades within the city and along the river, in particular with greenhouses for growing roses (Fig. 1), resulting presently in a very densely populated area with high industrial investment where houses, factories, hospitals, schools, the Pan American Highway and other infrastructure have been built on top of lahar deposits of less than 1000 years old. The population in Latacunga was estimated at 63,800 inhabitants in 2012, living in an area of about 10 km 2 with a population density of 6.4/km 2 (INEC — Instituto Nacional de Estadistica y Censos, 2012); the previous census of 2001 estimated the population at 52,000, with an increase of more than 20% in the past 10 years. The recent eruptive activity of Cotopaxi consists of Plinian outbursts of VEI 3 – 4 (at least 20 episodes in the last 2000 years; Barberi et al., 1995; Pistolesi et al., 2011), with associated lava fl ows, long-lasting ash eruptions and minor Strombolian activity. Pistolesi et al. (2011) upgraded the information made available by previous studies (Barberi et al., 1995; Mothes et al., 2004; Hall and Mothes, 2008a) and recog- nized 21 continuous tephra beds subdivided in three eruptive periods, from the emplacement of the regional tephra marker of the Quilotoa volcano ash (1150 AD; Hall and Mothes, 2008b) to the present (Fig. 3). The fi rst eruptive period (1150 to 1534) was characterized by two mid-intensity explosive eruptions (layers B L and S W , Fig. 3), apparently separated and followed by quiescent periods a few centuries long. Activity of the second period coincided with two Plinian eruptions in 1742 – 44 (layer M T ) and in 1766 – 68 (layer M B ). The latter, in particular, was associated with the formation of scoria fl ows, which were dispersed over large areas on the northern side of the cone. After the M B eruption, activity at Cotopaxi passed into a period of long- lasting, moderate-intensity ash generation punctuated by the occurrence of low-magnitude, short-lived sub-Plinian eruptions (P D , P L and P E; third period, Fig. 3). ‘ Boiling-over activity ’ and generation of scoria fl ow deposits were always associated with these events. Tephra deposits of the last large eruptive event of Cotopaxi, which occurred in 1877, are recorded by the P E layer. January 2001 saw increased level of activity at Cotopaxi volcano (Molina et al., 2006) with an increase of long-period (LP) seismic events accompanied by very long-period (VLP) events occurring directly under the volcano; fumarolic activity at the summit and a thermal anomaly suggesting injection of new magma (Ramon et al., 2006; Rivero et al., 2006). This evidence and the general behavior of the volcano as portrayed by its geological record (Barberi et al., 1995; Hall and Mothes, 2008a; Pistolesi et al., 2011) make the problem of lahar generation particularly signi fi cant. Río Cutuchi is the main river southwest of Cotopaxi. Its headwaters fl ow to the south of Rumiñahui (Fig. 1), a deeply eroded extinct volcano situated to the west of Cotopaxi, and its tributaries cover the entire western sector of Cotopaxi. Stratigraphic studies along the Río Cutuchi valleys revealed an impressive sequence of lahar deposits (Mothes et al., 2004; Pistolesi et al., 2013). Historical accounts also describe that since the Spanish Conquest (XVI century), most of the lahars produced by Cotopaxi fl owed towards the western and southwest- ern sectors of the volcano along the Río Cutuchi (Mothes et al., 2004; Figs. 2, 3), as reported by Sodiro (1877, p. 4): ( ... ) (in 1742) “ a terrible inundation in near regions destroyed bridges and farms, killing people and livestock ( ... ), and (in 1744) a huge fl ood rapidly arrived in Latacunga which, after the destruction of whatever en- countered along its way, invaded the center of the city, destroying houses ( ... ) ” . Based on the recognition of thin fallout tephra beds within the lahar sequence, Pistolesi (2008) and Pistolesi et al. (2013) proposed a stratigraphic reconstruction of post-XII century debris fl ows and compared their scale and dynamics (Fig. 4A). Well-constrained 14 C calendar ages performed on soils and charred grass were assigned to most lahar deposits thanks to the preservation of tephra beds intercalated between the lahar deposits, and these were directly related to their generating explosive eruptions (Fig. 3). Debris fl ow deposits were also informally grouped into three main categories (Types A, B and C) by combining sedimentological data obtained from a large quarried area along the Río Cutuchi (Fig. 1). These three categories of deposits largely vary in terms of dispersal, thickness, average and maximum size of transported blocks increasing from A to C, and were related to lahars with different sizes and trigger mechanisms (Pistolesi et al., 2013) (Fig. 4A, B, C). The stratigraphy of debris fl ow deposits in the Cutuchi quarries (Fig. 3) clearly highlights that the period between 1200 and 1500 AD is characterized by lahar events of large ( N 500 million m 3 ) scale, which carried meter-sized boulders and left deposits more than 20 m thick (Figs. 3, 4B, C). Conversely, lahars of the period of activity between 1700 and 1877 (the last eruptive event) show a variable but lower scale in terms of distribution, thickness and size of transported blocks; within this period, the 1877 lahar deposits are smaller than those related to the XVIII century activity. The 1877 eruption (a sub-Plinian event of magnitude VEI 4; Hall and Mothes, 2008a; Pistolesi et al., 2011), in particular, triggered large fl oods along all the main drainages. After exiting the glacier, these water-dominated fl ows mixed with pyroclastic and ice blocks, causing intense erosion and gullying, and rapidly evolved into debris fl ows (Aguilera et al., 2004). Typical valleys along the volcano's fl anks are represented by steep ( N 20°), narrow canyons laterally con- fi ned by vertical stacks of lavas and pyroclastic sequences. The peak discharge of 1877 fl ows was estimated between 50,000 and 60,000 m 3 s − 1 , with velocities in excess of 20 m s − 1 (Mothes et al., 2004). Lahars descending south of Cotopaxi initially fl owed along three main paths (Río Cutuchi, Río Saquìmala and Río Barrancas – Alaquez) and merged at a distance of about 24 km from the crater. The morphology at the foot of the cone rapidly changes into fl at ( b 5°), wide valleys separated by gentle slopes mostly covered by grass, where part of the coarse solid material has been commonly discharged by past debris fl ows. Lahar fl ow depths in the upper Cutuchi course have been estimated to be up to 20 m (Barberi et al., 1992; Mothes et al., 2004), increasing south of Latacunga where the valley gets narrower. Mothes et al. (2004) calculated a total volume of 100 million m 3 for the 1877 lahar drained from the canyons with different discharges and dropping into the Latacunga valley. Barberi et al. (1992), applying equal discharges in all major ‘ quebradas ’ , estimated a volume of 150 million m 3 . In the area of Latacunga, the lahar sequence is today exposed as a stack of debris fl ow deposits in gravel quarries and terrace cliffs. Although too far ( ≈ 40 km) from the volcano to preserve recognizable primary tephra fallout deposits and out of the main dispersal axis of historical Cotopaxi eruptions, the regionally-dispersed Quilotoa ash fall is visible within the sequence (Fig. 4D), separating pre- and post-1150 AD lahar deposits. The role of pyroclastic fl ows in generating rapid melting of snow and ice has been largely discussed in the volcanological literature (Major and Newhall, 1989; Pierson et al., 1990; Zernack et al., 2011). At Cotopaxi, in particular, ice scouring and melting during scoria fl ow events have been observed and described by chroniclers: during the 1877 eruption (and presumably during the 1766 – 68 and 1853 eruptions as well), “ a dark foam-like cloud boiled over the crater, much like the boiling over of a pot of cooking rice ” , spilling through the sides where the crater rim was lower. Emplacement of these scoria fl ow lobes took place immediately after an “ ash and lapilli rain ” (Wolf, 1878, p. 20). As soon as the lobate pyroclastic cloud overrode the glacier, lahars were triggered (Sodiro, 1877; Wolf, 1878). In most of the cases scoria fl ows and lahars descended the volcano fl anks virtually following the same bottom valley pathways. Pistolesi et al. (2013) suggested three different triggering processes for Cotopaxi lahars, which correspond to the different types of lahar deposits identi fi ed for events occurring in the last 800 years of eruptive activity: 1) Pyroclastic surge events with limited dispersal (run-out slightly be- yond the glacier snout) that can sweep over the entire glacier but with limited snow/ice melting. These pyroclastic density currents generated matrix-rich, block-poor and (almost) valley-con fi ned lahar deposits (Type A deposits). 2) Erosive, pyroclastic scoria- fl ow lobes able to cut deep canyons into the ice cap, leaving 40 – 50 m-high cliffs of exposed ice (Wolf, 1878). Lahars associated with these scoria fl ows are block-rich and produce widespread deposits. The presence of the abundant scoria bombs demonstrates the tight relationship between scoria- fl ows and lahar events. The volume of ice melted during these events is signi fi cant, although it can be related to the localized erosion of the deep ice canyons described above (Types A and B deposits). 3) Lithic- and scoria-rich, highly erosive, radially ...