1 Introduction

Volcán de Colima is one of the most active volcanoes in Mexico (Fig. 1a). At present (September 2014), after 3 years of dome growing, the explosive activity renewed with small explosions and the emplacement of rock falls and small pyroclastic density currents (PDCs) down to 3 km from the summit dome. The 1913 Plinian eruption is the largest event in the last century and was characterized by the emplacement of several scoria/pumice flows and fallout dispersal to the NE (Saucedo et al. 2010). Other Plinian eruptions were reported in ad 1818 and probably in ad 1690 (Luhr and Carmichael 1982). After the ad 1913 eruption, the volcano experienced several volcanic crises initiated by dome growth and successive collapses. The Holocenic tephro-stratigraphic record is incomplete due to low preservation of tephra in tropical climate (Luhr et al. 2010), but also for the erosive effects of the subsequent volcanic events (i.e., lahars). In contrast, the Holocene and Late Pleistocene eruptive records activity of Volcán de Colima and its ancestral edifice, the Paleofuego Volcano, indicate that volcanic collapses represented a recurrent activity, with partial or total destruction of the edifice and the emplacement of large volume (1–18 km3) debris avalanche deposits, including the 7,500 year. bp debris avalanche on top of which the cities of Colima and Villa de Alvarez are edified (Komorowski et al. 1997).

Fig. 1
figure 1

a Sketch map of the Trans-Mexican Volcanic Belt where the Colima Volcanic Complex and other main stratovolcanoes are located: 1 San Juan; 2 Ceboruco; 3 Tancitaro; 4 Jocotitlán; 5 Nevado de Toluca; 6 Popocatépetl; 7 Ixtacíhuatl; 8 La Malinche; 9 Pico de Orizaba; 10 Cofre de Perote; 11 San Martín; 12 Chichón. b spot image (1, 2 and 3 bands in RGB combination, 10 m of resolution) showing the Colima Volcanic Complex, with the Volcán de Colima edifice that consists of the active Volcán de Fuego cone built inside the Paleofuego scarp (white line). Aerial view of the southern sector of the Volcán de Colima showing the main towns (c, Cuahutémoc and d, Quesería and Alcaraces) which are surrounded by extensive cultivated fields

Full size image

To date, two main hazard maps have been published (Del Pozzo et al. 1996; Navarro and Cortés 2003), both based on the stratigraphic record of past eruptions. Since then, a detailed analysis of eruptive activity of the Volcán de Colima have been presented by several authors, including the detailed stratigraphic reconstruction of past events, definition of magnitude and aerial distribution of deposits, and hazard assessment based on numerical simulations for different eruptive scenarios (Saucedo et al. 2005; Dávila et al. 2007; Saucedo et al. 2008, 2010; Cortés et al. 2010a, b; Sulpizio et al. 2010; Bonasia et al. 2011; Borselli et al. 2011; Capra et al. 2014). The purpose of this paper is to present a revised volcanic hazard zonation for the Volcán de Colima that benefits from all these new works including the geological map of the volcano (Cortés et al. 2010a), and the hazard zonation based on numerical simulations (i.e., Saucedo et al. 2005; Dávila et al. 2007; Sulpizio et al. 2010; Bonasia et al. 2011). These previous works are here synthesized and adapted to propose: (1) hazard zonation for different eruptive scenarios (sub-Plinian/Plinian eruption, dome grown/collapse, edifice failure, rain-triggered lahars); (2) a qualitative hazard map where all these hazards are superimposed; and finally (3) a susceptibility map, where zone more prone to be affected by a volcanic event are identified, and hazard is ranked using a color scale.

2 Eruptive history of Volcán de Colima

Volcán de Colima is the youngest and active volcano of the Colima Volcanic Complex and consists of the ancestral edifice named Paleofuego, and the active cone, also known as Volcán de Fuego, that is built inside the Paleofuego caldera, a scarp formed from an edifice failure event (Fig. 1b). The age of the active cone has been estimated at about 2,500 year bp that corresponds with the age of the last edifice failure (Komorowski et al. 1997) (Table 1). One of the main debates is about the origin and age of the Paleofuego caldera. Luhr and Prestegaard (1988) and Robin et al. (1987), both consider that this scarp is the product of a single event, occurred 4,280 ± 180 year bp or 9,370 ± 400 year bp, respectively. Komorowski et al. (1997) point out the possibility that the scarp is the result of several discrete collapses, based on the evidence of the presence of at least nine different debris avalanche deposits that can be recognized on the southern sector of the volcano (Table 1). Roverato et al. (2011) proposed that the Paleofuego scarp is probably Pleistocenic in age. These uncertainties are mainly related to the difficulty in correlating debris avalanche outcrops around the volcano, mostly because of the lack of outcrops exposing stratigraphic relations, the homogenous composition of the lava fragments, being all andesitic, and finally the difficulty in correlating the deposits with their source area. Despite these uncertainties, it is clear that edifice failure has been a recurrent event at Colima. Minor eruptive episodes have not been identified in the Late Pleistocene eruptive records of Volcán de Colima probably eroded and buried by the voluminous debris avalanche deposits.

Table 1 Edifice collapse events identified at the Volcán de Colima
Full size table

A detailed study about the Holocene stratigraphic record of Volcán de Colima (Luhr et al. 2010) identified a period of high explosive activity between 3,000 and 4,000 year. bp and between 500 and 1,500 year. bp, but the low preservation of the studied deposits has made it very difficult to reconstruct the characteristics and magnitude of these eruptions. In contrast, a quite good historic record for the Volcán de Fuego active cone for the last 500 years is available (Table 2) (De la Cruz-Reyna 1993; Breton et al. 2002). During the last 5 centuries, the most violent Plinian eruptions occurred in ad 1576, ad 1818 and ad 1913, sub-Plinian events were also reported in ad 1622 and ad 1711 (Table 2). For the older episodes, no detailed data are available, but the 1913 activity was well documented and studied (Saucedo et al. 2010; Gavilanes-Ruiz and Cuevas-Muñiz 2014). The ad 1869 activity is quite distinctive; it started on the NE flank of the volcano and gave place to the formation of El Volcancito scoria cone and associated lava flows that lasted up to ad 1878 and was accompanied by several eruptive columns up to 4,000 m in height. The ad 1913 Plinian eruption represents the largest known historic eruption of this volcano, with the formation of a 23-km-high eruptive column, whose products directly affected the western and central regions of Mexico. After this cataclysmic event, Volcán de Colima experienced several volcanic crises. Examples are the ad 1991, 1994, 1998–1999 and 2004–2005 events, characterized by a rapid dome growth (few months) and subsequent gravitational collapse or associated with strong explosions (Merapi and Soufriere type). The associated block-and-ash PDCs emplaced on main ravines down to a maximum of 6.1 km from the summit. The more diluted ash clouds traveled up to few hundred meters farther downslope. Lava flows were also erupted. In ad 1975, three lava flows emplaced on the SE flank of the volcano, down to a distance of 4,500 m, and in ad 2004 a lava flow extruded from the crater and partially filled the El Playón, the depression confined by the Paleofuego scarp. Lahars have been recognized all along the stratigraphic record of the volcano, from very large debris flows linked to major edifice collapse events (Cortés et al. 2010a, b) to post-eruptive lahars associated to Plinian or dome-collapse eruptions during rainy seasons (Dávila et al. 2007; Vázquez et al. 2014).

Table 2 Chronology of the eruptive history of the Volcán de Colima during the last ~500 years
Full size table

Based on the eruptive history here described, four main different eruptive scenarios can be identified: (1) edifice failure; (2) Plinian eruption; (3) dome collapse (Merapi, Soufriere or Vulcanian eruptions); (4) rain-triggered lahars. Here, we present the hazard zonation for each of these different scenarios and a final qualitative hazard map from the superimposition of the different maps, highlighting the most exposed areas to volcanic hazards at Volcán de Colima. Even if lava flow occurred during the last decades during exogenous event of lava dome growth, they are not considered here in the hazard evaluation since they are limited to few km from the source. Only Navarro and Cortés (2003) reported the lava flow hazard zonation in their map, but simply depicting the extension of past lava flows.

3 Hazard zonation maps

3.1 Edifice failure scenario (debris avalanche and associated secondary debris flow)

Edifice failure can be considered a common occurrence at Volcán de Colima when compared with other Mexican stratovolcanos (Capra et al. 2002b). Based on the stratigraphic record, these collapses were accompanied by explosive activity, as testified by the presence of pyroclastic layers in direct contact with the debris avalanche deposits (i.e., Komorowski et al. 1997; Cortés et al. 2010b; Luhr et al. 2010; Roverato et al. 2011). The structural regime acting on the volcano, with a N–S component of the maximum horizontal stress and coupled with a spreading effect to the south, has been also considered as a controlling parameter of the volcano instability (Norini et al. 2010). The studied events indicate that the volcano experienced different flank failures with volumes between 1 and 2 km3, and major sector collapses with volumes up to 18 km3 (Table 1). Based on a statistical analysis of the events occurred during the Holocene, Borselli et al. (2011) recognized a possible recurrence time of 2,698 years, suggesting that the next collapse could happen in ~300 years from present. Borselli et al. (2011) also defined that, at present, the more instable portion of the volcano is the SW flank, with an involved mass of around 1 km3, used to simulate inundation limits with the Titan 2D code (Borselli et al. 2011). This scenario represents a good estimation of a possible flank failure, and here it is used to define the hazard zonation (Fig. 2). For sector collapses of larger volume, in the range of 5 km3, which would correspond to the total collapse of the active Volcán de Fuego cone (calculated from an altitude of 2,700 m a.s.l. up to its summit), the hazard zonation is based on the distribution of debris avalanche deposits as observed in the field (Cortés et al. 2010a). Even if deposits of larger volume have been reported (Stoopes and Sheridan 1992; Komorowski et al. 1997; Table 1), they are not realistic due to the actual volume of the Volcán de Fuego cone. Most of the past events were associated to explosive eruptions, and diluted (blast) and concentrated PDC deposits were described on top of debris avalanche deposits (Komorowski et al. 1997; Cortés et al. 2010b; Roverato et al. 2011). We are not able to propose a hazard zonation for such a type of phenomena, since field data on maximum runout are very scarce, but they should be taken into account in case of the occurrence of an edifice failure. Moreover, it is worth mentioning that concentrated PDCs will affect the area already devastated by the debris avalanche itself (i.e., 1980 Mt St Helens eruption, i.e., Waitt 1981). A further complication, as indicated by past events, is that debris avalanche deposits are able to reach the two main rivers that flank the volcano, Naranjo and Armería. The avalanche deposits can dam the river flows, inducing the formation of temporary lakes whose failure would generate large debris flows of the order of 0.1 km3 (Capra and Macías 2002; Cortés et al. 2010b). In order to take into account this eventuality, the hazard zonation (Fig. 2) includes inundation limits for debris flows on these rivers based on LAHARZ simulations (Schilling 1998), including the possible extension of temporary lakes, estimated considering a maximum debris avalanche thickness of 80 m into the river flows (Cortés et al. 2010b).

Fig. 2
figure 2

Hazard assessment for edifice failure (1 and 5 km3) and associated secondary effects such as debris flow from dam-break

Full size image

3.2 Plinian eruptive scenario

The ad 1913 eruption has been classified as VEI 4 (Saucedo et al. 2010; Bonasia et al. 2011), and based on a statistical analysis, the probability of occurrence of at least one eruption exceeding a VEI 3 over a time period of 100 years is 0.8 (Mendoza-Rosas and De la Cruz-Reyna 2008).

The ad 1913 eruption occurred after one decade of low activity (Table 2), and started on January 18, with an opening phase producing gravitation collapse of the summit dome and forming block-and-ash flows that reached a distance down to 4 km from the summit. On January 20, a second eruptive phase started with strong Vulcanian explosions generating block-and-ash flow deposits with up to 20 % of scoria fragments that extended down to 12 km from the crater. After less than 2 h, the paroxysmal phase took place and an eruptive column of 23 km in height formed, lasting 9 h. Stratospheric winds dispersed the tephra to the NW, and ash deposition has been reported as far as on Saltillo city, 700 km away from Volcán de Colima (Saucedo et al. 2010). During the Plinian column formation, partial collapses emplaced pumice/scoria-rich PDCs on main ravines down to 15 km from the vent. A dilute PDC preceded the column formation, and its deposits have been recognized up to the north flank of Nevado de Colima volcano, just at the base of the fallout layer (Saucedo et al. 2012). Based on the direct observation of the ad 1913 activity, two major hazards have to be considered during Plinian activity: fallout and PDC generation.

3.2.1 Fallout deposits

Bonasia et al. (2011) reconstructed the Plinian phase of the 1913 eruption and constrained the volcanological parameters applying a best-fit inversion method using field data measured by Saucedo et al. (2010). The obtained values for total mass and column height, together with a statistical set of re-analysis wind profiles, were used as input parameters for the HAZMAP code (Macedonio et al. 2005) to construct tephra loading probability maps for two different thresholds (100 and 200 kg/m2) which, considering a deposit density of about 1,000 kg/m3 (typical for basalt-andesite fallout deposit), correspond to a thickness of 10 and 20 cm, respectively. These ash loading thresholds are considered critical for low to medium quality buildings. Based on these maps we selected the 1 and 10 % probability curves of exceeding a deposit thickness of 10 cm. Its oblate shape reflects seasonal changes in wind direction (Fig. 3).

Fig. 3
figure 3

Hazard assessment for Plinian eruption, including distribution of column collapse pumice flow deposits, dilute pyroclastic density current (H/L 0.2), 10 cm fallout probabilistic curves (1 and 10 %)

Full size image

3.2.2 Column collapse-generated PDCs

Saucedo et al. (2005, 2010) evaluated the distribution and maximum extent of PDCs originated during the Plinian phase of the ad 1913 eruption, and proposed the hazard zonation based on H/L estimation and using the FLOW3D code. They proposed three different hazard zonations for PDCs generated by Merapi-type dome collapse, Vulcanian activity and partial collapse of the Plinian eruptive column. The Merapi type, along with moderate Soufriere-type activity represents a very common eruptive style at Volcán de Colima during dome grown episodes. These eruptive scenarios have been recently evaluated by Sulpizio et al. (2010) and they will be described in the next paragraph. The work published by Saucedo et al. (2005) represents the only effort to analyze the hazard related to strong Vulcanian explosions and partial column collapse producing block-and-ash flows with a content up to 20 % of juvenile fragments (pumice and scoria) and pumice/scoria flow, respectively. This zonation is here used and it shows that main ravines will be filled with dense material down to distances of 15 km from de source (Fig. 3). A major problem is to define the possible extent of the ash cloud accompanying dense PDCs, from which ash fall or dilute PDCs can originate (e.g., Sulpizio et al. 2014). For this purpose and based on direct observation just after the ad 1913 eruption, an energy cone with a H/L value of 0.2 was used to trace the possible limit of affectation. Even if this approach defines only a rough approximation, the area enclosed on the H/L = 0.2 line is in agreement with the observed field data, and it was modified to the north to consider the extent of the surge deposit as observed at the base of the pumice fall layer near the summit of the Nevado de Colima edifice (Saucedo et al. 2012).

3.3 Dome collapse

During the last 20–30 years, the eruptive activity of Volcán de Colima has been characterized by repetitive growth and collapse of summit domes, which generated block-and-ash flows that inundated the upper slopes of the volcano and engulfed the deep and incised valleys at the base of the main cone. The last major block-and-ash flow-generating explosion occurred in June 2005 (Macías et al. 2006). It closed the 2004–2005 period of activity of the volcano and completed the dismantling of the summit dome. Since early 2007, a new dome has slowly grown in the funnel-shaped crater on the volcano summit (Stevenson and Varley 2008). Two main dome collapse behaviors have been recognized, gravitational (Merapi type) and explosive (Soufriere type). Gravitational dome collapses generated BAF deposits mostly on the SW ravines, down to 4 km of distances, except for the ad 2004 event that generated a BAF along the La Lumbre ravine down to 6.1 km from summit cone. In contrast, dome explosions generated more mobile block-and-ash flows along all main ravines, which easily reached distances down to 5.5 km from the source. No stronger Vulcanian explosions generating longer runout (>10 km) of BAFs such as those reported during the 1913 have been observed during these dome growth episodes. Sulpizio et al. (2010) proposed a hazard zonation for both Merapi- and Soufriere-type dome collapses, including the possible extent for ash cloud deposition accompanying the dense basal avalanche estimated with a H/L of 0.3 (Fig. 4).

Fig. 4
figure 4

Hazard assessment for summit dome collapse, including Merapi-, Soufriére- and Vulcanian-type eruptive scenarios and possible extension of the ash cloud inundation zone (H/L 0.3)

Full size image

3.4 Rain-triggered lahars

Lahars are very common occurrences at Volcán de Colima, including events that followed major edifice failure (Cortés et al. 2010b; Roverato et al. 2011) and Plinian activity (Saucedo et al. 2010). More commonly, they are initiated by heavy rains that occur each rainy season (from June to October), being more voluminous in correspondence of hurricanes or very extreme rain events (Capra et al. 2013, 2014). Rain-triggered lahars are very variable in volume, from a minimum of 3 × 105 m3 (Dávila et al. 2007; Capra et al. 2010; Vázquez et al. 2014) to 3 × 106 m3, which corresponds to the extraordinary event occurred in ad 1955 (Saucedo et al. 2008). This event was originated on the SE slope of Nevado de Colima, where several discrete pulses combined to form the main debris flow in the Atenquique main drainage, killing 23 people in the homonymous village and emplacing a 3.2 × 106 m3 deposit as far as 25 km from the source, reaching also the Tuxpan River (Saucedo et al. 2008). After the ad 1913 Plinian eruption, several lahars occurred along main ravines and reached distances as far as 20 km from the summit cone, where they diluted into the Tuxpan River. No volume estimations are available for these events, but their runout is of the same order of magnitude of that of the Atenquique event (Saucedo et al. 2008). Based on LAHARZ simulation (Schilling 1998), Dávila et al. (2007) performed the hazard evaluation for a low volume (0.5 × 106 m3) rain-triggered lahars along main ravines (Fig. 5). The same methodology was used for lahars with a volume of 5 × 106 m3, value that includes events such as the ad 1955 Atenquique episode or the post-eruptive lahars that followed the ad 1913 Plinian eruption that were able to reach the two main rivers in the area, the Armería and Tuxpan-Naranjo (Gavilanes-Ruiz 2004; Capra et al. 2014).

Fig. 5
figure 5

Hazard assessment for post-eruptive rain-triggered lahars (5 × 105 m3) and syn-eruptive episodes (5 × 106 m3) such as those associated to the 1913 Plinian eruption

Full size image

4 Discussion: qualitative hazard map

Based on the single eruptive scenarios presented in the previous sections, we propose here a map comprehensive of all different type of activities (Fig. 6). Since no probability of occurrence is here defined, this map represents a qualitative hazard map where maximum extent of the areas that can be affected by different eruptive scenarios is defined. Edifice failures represent the major hazard for the southern sector of the volcano. A discrete flank failure (~1 km3) would threat an area included in a radius of 20 km from the volcano summit, filling major ravines where subsequent lahars will form reaching the Armería and Tuxpan-Naranajo Rivers. In case of the more voluminous sector collapse (5 km3), the southern region of the volcano, down to more than 40 km from the summit, would be affected and buried, directly threatening more than 300,000 people and including the Colima city. In addition, the emplacement of the debris avalanche deposits in Naranjo and Armería rivers would induce the formation of temporary lakes, whose ruptures would trigger voluminous debris flows that would inundate areas far away from the volcano, reaching the Pacific coast. Based on the evidences of past events, edifice collapses were associated with magmatic activity with high probability to be accompanied by a blast (Belousov et al. 2007).

Fig. 6
figure 6

Qualitative volcanic hazard map of the Volcán de Colima

Full size image

The fallout hazard zonation covers a wide area. The lower probability curve (1 %) of exceeding a deposit thickness of 10 cm could affect an area of 535 km2, including several towns (~180,000 of inhabitants) such as Cuahutémoc, San Marcos, Zapotiltic, Tuxpán and Ciudad Guzmán where, during the ad 1913 eruption, 10–13 cm of ash were reported (Saucedo et al. 2010). The Colima city is outside of this hazard zonation, which indicates that it could be affected by less than 10 cm of ash due to stratospheric wind direction in the area (mainly to N, NE or NW). The Colima national airport located near Cuahutémoc town would also suffer from the presence of ash in the atmosphere, with interruption of flight operations. Apart from the possible damage to infrastructure, heavy ash fall would threat all the cultivated areas that are extensively distributed on the volcano slopes, with severe loss for the sugarcane industry, which is located in Quesería village (Fig. 6). It is important to take into account that, in case of lower eruptive columns, winds can probably disperse ash to the S, as occurred on ad 2003, after a strong Vulcanian explosion, or as reported in the historic record (Table 2). Even in that case, reduced amount of ash would settle in Colima city.

PDCs associated to Plinian or sub-Plinian eruptions and strong Vulcanian explosions can affect several villages in a radius of 15 km from the summit cone. Column collapse-generated PDCs would fill major ravines, with the possibility to affect la Yerbabuena (evacuated during the ad 1999 and ad 2002 volcanic crises), Becerrera, Tonila and San Marcos villages, for a total of more than 7,000 of inhabitants including several ranches. In particular, the southern sector of the volcano is largely cultivated for sugarcane, and during daytime, tens of workers are dispersed on these fields, especially on the S-SE sector. Finally, lahars along the Monte grande ravines would be able to damage the sugarcane refinery at Quesería town. BAFs from dome collapse are threatening a more reduced area, down to a distance of 8 km from the crater. The dense basal avalanche would be confined to major ravines, without direct affectation of inhabited areas. In contrast, the ash cloud would be able to affect a larger area overrunning topographic highs, as observed in other volcanoes (Merapi, Indonesia, Itoh et al. 2000; Unzen, Japan, Fujii and Nakada 1999) provoking deaths and injuries in up to five ranches, including El Borbollón, located besides La Lumbre ravine. Finally, lahars, with variable magnitude, can inundate main ravines, with major damage to towns that are settled along active drainages such as Atenquique, San Marcos and Quesería (Fig. 1d). The new map here proposed differs substantially from the previous published maps (Navarro and Cortés 2003; Del Pozzo et al. 1996), since it includes the hazards zonation for different types of PDCs related to different eruptive scenarios, lahars and debris avalanche inundation areas depending on the flow volumes, and ash fall distribution based on a statistical evaluation of wind direction. In addition, secondary effects such as the formation of temporary dam along the main rivers are also included. Based on these scenarios, a susceptibility map is presented, showing the most exposed areas to volcanic threat. The map was constructed by a simple arithmetic sum of rasters of all hazard zonation previously described, to obtain the maximum number of event that affects each pixel (Fig. 7a). The sector collapse of the active cone (5 km3) is not here considered since at present does not represent a realistic scenario and in case of occurrence it will affect all the area (Fig. 6). The map was then classified in four main levels of hazards (Fig. 7b) based on the combination of different hazardous occurrences that can affect a given area (Bartolini et al. 2014). The high-level area extends down to 8.5 km from the crater, and it is exposed to all of the considered hazardous events. In particular, it corresponds with the area that can be affected by PDCs from Merapi- and Soufriere-type dome collapse, and lahars (<5 × 105 m3) along main ravines. These episodes represent the most common and frequent eruptive scenarios of dome collapses and rain-triggered lahars as observed during the last 20 years (Table 2). The moderate-level includes the area from 8.5 to 12 km from the vent and corresponds to the emplacement area of PDCs associated to strong Vulcanian explosions and associated ash clouds (H/L = 0.3), PDCs from Plinian column collapses and associated ash cloud (H/L = 0.2), including more than 10 cm of fallout and lahars (>5 × 106 m3). This eruptive scenario corresponds to the Plinian phase of the ad 1913 eruption (Saucedo et al. 2010) and represents a threat for more than 7,300 of people. The low-level zone includes the area that could be affected by debris avalanches from flank collapses (1 km3), more than 10 cm of fallout, and dilute PDCs on top of the Nevado de Colima volcano. Finally, the very low-level correspond with the rest of the area that can be threatened by more than 10 cm of fallout, which can cause several damage to infrastructures at several towns (~160,000 of inhabitants) such as Tuxpan and Ciudad Gúzman as occurred during the 1913 Plinian eruption. This type of map could be considered as a qualitative volcanic hazard map, which can be useful to Civil Protection authorities in defining different actions depending on the most probable eruptive scenario during a volcanic crisis. The classified map could be used to establish different level of volcanic alert, such as the system used at Popocatépetl volcano, where a “traffic light” with three main phases as green, yellow and red, and each of them with different sub levels is considered for alert purposes (Centro Nacional de Prevención de Desastres, http://www.cenapred.gob.mx/es/Instrumentacion/InstVolcanica/MVolcan/Semaforo/).

Fig. 7
figure 7

a Volcanic susceptibility map of the Volcán de Colima based on a simple raster operation to identify the areas that are exposed to the highest number of events. b Classified volcanic susceptibility map to identify the areas with different level of hazards. The table indicates the number of inhabitants that can be affected for each level of hazard (demographic data from INEGI, 2010)

Full size image

5 Conclusion

A qualitative volcanic hazard map that shows the area exposed to the highest number of volcanic events is proposed. The hazard map poses in evidence that an eruption of VEI 4, such as the 1913 Plinian eruption, would cause serious consequences not only for the areas directly exposed to the impact of pyroclastic flows and lahars, but for a larger area that can be covered by pumice and ash fall with direct affectation on transportation (terrestrial and aerial), communication, lifelines, agriculture and health for more than 300,000 people, including the cities of Colima and Villa de Alvarez. Commercial activities of the western portion of México would also suffer economic losses since the highway N.54 (Fig. 6), an important connection between the cities of Guadalajara and Manzanillo, being the last the most important harbor of the Pacific Mexican coast, would be affected by the ash fall. In contrast, the most common scenario is restricted to the proximal area with local effects on small ranches and people working in the field. The most catastrophic effect would be related with the flank collapse of the volcano, which represents the less frequent eruptive scenario, but that can cause, in case of occurrences, devastating effects.

The hazard assessment here presented may help decision makers during volcanic crises in defining risk mitigation actions such as evacuation to reduce the loss of life due to the potential impact of volcanic hazards. A study performed by Gavilanes-Ruiz and Cuevas-Muñiz (2014), applied to some inhabitants of the most threatened villages, showed that their knowledge and wisdom regarding the general features of the volcanic activity and hazards are not in contrast to those of authorities and earth scientists. In fact, they generally recognize the lethal power of the volcano, but many of them simply have not found sufficient reasons to participate in evacuations (Gavilanes-Ruiz and Cuevas-Muñiz 2014). Even though the increased public knowledge about volcanic risk does not guarantee the willingness to participate in official risk mitigation strategies, the same studies indicated that the information and risk communication programs as the one performed during 1997–2000 (including the explanation of hazard zoning) improve risk perception and thus should be continued in order to achieve a long lasting impact.