Next Article in Journal
Assessment of Climate Change in Angola and Potential Impacts on Agriculture
Previous Article in Journal
Nexus of Foreign Direct Investment (FDI) and Environmental Emissions in South Africa: A Markov-Switching Regression
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Climate
Volume 13
Issue 1
10.3390/cli13010011
climate-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Identifying the Areas at Risk of Huaico Occurrences in the Department of Lima, Peru

by
Geise Macedo dos Santos
1,
Vania Elisabete Schneider
2,3,
Gisele Cemin
4 and
Matheus Poletto
1,*
1
Postgraduate Program in Process Engineering and Technologies, University of Caxias do Sul, Caxias do Sul 95070-560, RS, Brazil
2
Graduate School, Trujillo Catholic University (UCT), Moche 13001, Peru
3
Postgraduate Program in Civil Engineering, Federal University of Sergipe, Aracaju 49060-108, SE, Brazil
4
Postgraduate Program in Environmental Engineering and Sciences, University of Caxias do Sul, Caxias do Sul 95070-560, RS, Brazil
*
Author to whom correspondence should be addressed.
Climate 2025, 13(1), 11; https://doi.org/10.3390/cli13010011
Submission received: 16 November 2024 / Revised: 17 December 2024 / Accepted: 24 December 2024 / Published: 5 January 2025
Browse Figures
Versions Notes

Abstract

:
Because of local climate, a phenomenon called huaico occurs in the coastal regions of Peru, configured by an alluvial flow of surface runoff caused by precipitation and accompanied by the transport of solid particles. A total of 24% of the huaicos recorded in Peru from 2003 to 2019 were concentrated in the Department of Lima alone and affected 38,000 people. Thus, the aim of this study was to use Maxent to identify the areas at risk of huaicos in this department. To this end, a georeferenced database was created that included the locations of these events for modeling. We used variables suggested by Peru’s Geological, Mining, and Metallurgical Institute (INGEMMET)—geology, geomorphology, DEM, slope, and precipitation—which returned extremely high kappa coefficients. Approximately 42% of Lima’s area is likely to have a huaico occurrence. The most crucial variable for the models was the geomorphological classification characterized by the accumulation of mobilized material, as was the case in previous huaico models. In addition, the monthly approach should have been more effective at determining the differences in the precipitation levels. Thus, new models for the coastal departments of Peru using Maxent algorithms should take a new approach related to precipitation, although the use of Maxent proved satisfactory.

1. Introduction

The coastal formation of Peru is basically a strip of plain formed by aeolian and alluvial deposits dating from the Cenozoic Era and Quaternary Period. Further inland are volcanic and sedimentary mountain formations with pyroclastic characteristics from the Mesozoic Era of the Cretaceous Period [1,2,3]. Vegetation density is low or nonexistent in areas covered by woodlands and coastal Andean deserts, respectively, exposing the soil or fragmented rock formations. This contributes to the transport of these materials down the mountains [4]. This configuration favors a specific type of natural disaster locally called a huaico, which corresponds to alluvial flows. A huaico is characterized by the surface runoff flows caused by precipitation in areas with steep slopes and is accompanied by the erosion and transport of solid particles from poorly cohesive soils or the fall and landslide of rocks [5,6]. The risk associated with huaicos lies in the location of the occurrence and the affected areas. Usually, huaicos reach the urbanized areas in the coastal zone of Peru, a region where around 65% of the country’s population is concentrated [7].
Between 2003 and the first quarter of 2021 alone, 2421 huaicos were recorded in the country [8], affecting approximately 450,000 people, damaging approximately 37,000 homes, and causing 141 deaths. Although these are recurrent events, their location varies significantly over time. In the first quarter of 2021 alone, the department with the highest number of huaicos was the Department of Cusco, with 41 recorded. In the first quarter of 2019, 83 huaicos were recorded in the Lima Department, the region with the highest recurrence that year [8,9]. Despite the spatial unpredictability of occurrences, the Peruvian department most affected by huaicos is the Department of Lima, where the country’s capital is located, which registered 24% of the huaicos between 2003 and 2021. During this period, 584 huaicos affected 38,100 people [8]. Figure 1 presents some examples of the areas impacted by huaicos in the Department of Lima.
The occurrence of huaicos is associated primarily with precipitation, which in the El Niño years reaches unusual levels in the country. During the recent coastal El Niño event in 2017, 570 huaicos were recorded—around 24% of the occurrences between 2003 and 2021. In 2019, another El Niño year, 492 huaicos were recorded in Peru [9].
With the recurrence of natural disasters, policies and practices must be adopted to reduce the risks resulting from natural hazards. The Peru National Disaster Risk Management Plan, valid from 2022 to 2030, identifies the country as vulnerable to huaicos. Furthermore, in adopting the guidelines of the Sendai Framework for Disaster Risk Reduction 2015–2030, which has, as its first guiding action, an obligation to minimize disaster risks by increasing our understanding of them [14], the plan also points out a deficiency in the country’s familiarity with all the dimensions of disaster risks. One of the issues highlighted is the need for specific investigations with the potential for practical applications. Inseparable from the risks is the human occupation of the areas subject to natural hazards. In addition to their identification, structural and structuring actions must be followed to minimize the vulnerability of the resident population [6].
In recent years, machine learning methods have effectively mapped areas susceptible to natural hazards. In the literature, some reviews have categorized the use of machine learning by hazard typology, such as landslides [15,16,17], floods [18,19], natural fires [20,21], and earthquakes [22]. For Peru, studies have evaluated only fires in the Amazon region [23], earthquake risk in Pisco [24], and susceptibility to landslides [25]. The machine learning algorithms most frequently used in these studies are the artificial neural network (ANN) and multilayer perceptron (MLP), both approaches using neurons, support vector machine (SVM), and random forest (RF).
Although Maxent was initially designed to model species distributions, the machine learning technique called maximum entropy modeling [26] has recently been used to map natural hazards. According to Phillips [26], the program calculates the probability of an event occurring by considering the input data characterizing the area and the occurrence records from regions with characteristics compatible with those that have undergone similar events. Javidan et al. [27] describe Maxent as a machine learning technique that enables investigating the relationship between a dependent variable (such as a landslide, flood, and others) and several independent variables (driving factors/geoenvironmental factors).
In risk mapping, Maxent determines the hazard distribution by finding the distribution with the highest degree of randomness within the assigned constraints, defined by the relationship between hazard circumstances at the input and their environmental covariates [28].
It is noteworthy that most studies using Maxent for mapping risk are in the Middle East and Asia. In these studies, as seen in Table 1, the recurrence of relief elements generated from a digital elevation model (DEM), such as slope, aspect, height above the drainage network, terrain ruggedness index (TRI), and topographic wetness index (TWI), in addition to the DEM itself, is observed. Geology and information related to the anthropic use of the areas, such as distance from roads and land use, are also recurring elements in the models. The climate is also addressed, especially when assessing fires, landslides, and floods.
Since a huaico results from specific climatic and geomorphological variables, identifying the environmental variables that are most likely to drive this type of hazard is necessary to build a model. Identifying areas where these events are likely to occur on a large scale is an essential planning tool. Such a model can be replicated for Peru and other Latin countries affected by huaicos, such as Colombia, Ecuador, and Venezuela.
Thus, this work aims to use the Maxent machine learning algorithm, which is already used to identify potential natural hazard areas, to identify areas at risk of huaicos in the Department of Lima. Lima has been heavily affected by huaicos in recent years, making it ideal for testing the potential of using this algorithm. In addition, it is essential to evaluate the changes in the models related to the different average maximum precipitation levels. Thus, an approach using monthly models was adopted. A set of models was created that considered the driving variables described in the literature and other recurrent variables to identify their correlation with the occurrence of huaicos.
The use of machine learning in general and Maxent specifically to model huaicos is novel, as it is a monthly approach for hazardous area identification. There are no studies related to modeling huaico-prone areas and no specifications for precipitation data.

2. Materials and Methods

This section identifies the study area and describes the input data and the procedures applied to develop the models in Maxent.

2.1. Study Area

The Department of Lima, where Peru’s capital is located, experienced 24% of the huaicos registered between 2003 and 2019 [8]. In addition to the study area location, the map in Figure 1 shows 1650 huaico locations, as inventoried by Peru’s Geological, Mining, and Metallurgical Institute (INGEMMET) [50], during the period from 2003 to 2021.
The Department of Lima is classified as Hot Desert Climate by the Köppen–Geiger classification [51]. The National Meteorological and Hydrological Service of Peru (SENAMHI) sub-characterizes five different climates in the Department of Lima, varying from arid with moisture deficiency throughout the year in the coastal zone to rainy with a dry autumn and winter in the most continental area of the department. In the arid climate, precipitation is 5 mm annually; in the rainy climate, precipitation can reach 1200 mm. The average maximum temperature varies from 19 °C to 31 °C [52].
Figure 2 illustrates that the data are not randomly allocated, with the concentration of huaicos near the capital, Lima. This work identifies the environmental variables coinciding with historical records.

2.2. Input Data for the Algorithm

2.2.1. Data from the Inventory of Geological Hazards in Peru

The Geological, Mining, and Metallurgical Institute (INGEMMET) maintains metadata catalog records of geological hazards in Peru, including inventories of falls, landslides, flows, complex movements, overturns, and others. This inventory identifies huaicos as debris flows (INGEMMET, 2023) [50]. The records are in point format. However, there is no description of how the points were selected when the phenomenon described occurs over a large area.
The geological hazards file was filtered to encompass the Department of Lima using QGIS software 3.30.2. Then, using the WGS 1984 system, the geographic coordinates of the occurrences were extracted and converted into CSV format (comma-separated values) to be input as data into Maxent. Note that in any case, Maxent only accepts points as input data.
It is worth noting that the Maxent algorithm works only in the presence of an event, and it is unnecessary to indicate areas where the event did not occur. However, the model must consider some pixels as a practical absence to calculate model statistics [30,48].
Maxent relies on a machine learning response that makes predictions from incomplete data. This results in a Maxent output in ASCII format, which is a continuous prediction of specific presence ranging from 0 to 1 [27,30].

2.2.2. Environmental Variables

The characterization of areas that experience huaicos in the Department of Lima, derived from INGEMMET data, is based on the driving factors for the occurrence of these hazards. These include the precipitation in areas where ephemeral water courses form, the geological and geomorphological formation itself, and the relief of the region, both the altitudinal conformation of the terrain and its slope [53,54,55,56,57].
Therefore, these characteristics were expected to be the most prominent in the models for identifying areas at risk of huaicos occurrence. However, other environmental variables already cited in the literature were also added to the modeling, as described in the following sections.
It was also essential to evaluate the correlation between environmental variables before applying them to the models. This evaluation was carried out using a specific tool within the QGIS environment.

Digital Elevation Model

Kornejady, Ownegh, and Bahremand [29] treat altitude as the most straightforward relief feature, and they agree with Pourghasemi et al. [39] regarding its importance in landslides. In general, topography contributes to the formation of hazard-prone environments. Topographic factors extend beyond the altitude itself to slope and aspect. The topographic wetness index (TWI) quantifies the impact of topography on hydrological processes and is related to landslides [38]. The terrain roughness index (TRI) measures the terrain’s roughness, and is therefore another index associated with landslides.
Although the literature does not refer to huaicos, a phenomenon that could be understood as debris flows, but to landslides and floods, Rusk et al. [24] determined that topography and its covariates influence soil stability and water movement, among other factors.
This work used the Shuttle Radar Topography Mission (STRM) DEM with 30 m resolution pixels, dating from 2000, obtained through the United States Geological Survey (USGS).
The digital elevation model of the study area in QGIS served as input data for modeling. From the DEM, slope data in degrees were also generated in QGIS using the slope tool. The aspect measuring the orientation of each cell relative to north; the topographic wetness index (TWI), which indicates the tendency of the terrain to accumulate water; and the terrain roughness index (TRI) were used. Specific tools in QGIS also generated the aspect data and the terrain roughness index (TRI). The topographic wetness index corresponds to the natural logarithm of the ratio between the runoff accumulation area that drains into a pixel and the tangent of the pixel’s slope. The TWI was obtained through QGIS’s mathematical operations with the DEM.

Climate Data

Sheikh, Kornejady, and Ownegh [32] used the average annual precipitation to assess landslide and flood risk. Arabameri et al. [41] also used yearly average precipitation to assess erosion risk. In assessing fire risk, Adab et al. [44] considered the average annual rainfall and the average precipitation of the driest month. For this study, using monthly precipitation rather than yearly precipitation was more credible owing to prior knowledge of the area, which distinguished between rainy months (especially in El Niño years) and dry months. For this reason, since precipitation is usually low in the region, historical maximum values were used because an average value would underestimate precipitation volumes. Again, according to the climate classification created by SENAMHI in the Department of Lima, precipitation averages 5 mm annually and in the rainy climate. Although more continental areas of the department can reach precipitations of 1200 mm, it is possible to associate the occurrence of huaicos with arid regions [52]. Therefore, in the arid climate, all months are virtually the driest, while in the rainy environment, winter and autumn are dry [52]. Nevertheless, the average monthly precipitation was computed to compare the average and maximum data.
The National Meteorological and Hydrological Service of Peru (SENAMHI) provides historical rainfall data for Peru. The Department of Lima has 46 rainfall stations, 36 of which have at least 40 years of historical data. Information on maximum and historical monthly averages for each station was generated using electronic spreadsheets.
Subsequently, the maximums were spatially distributed through Thiessen polygons in a GIS environment, generating a set of precipitation coverage polygons per season for each month. The use of Thiessen polygons was presented as the first evaluation option. It is worth noting that other approaches to spatialization of precipitation can be tested and compared.

Geomorphological Data

The literature frequently asserts that the area’s geology has a remarkable influence on the stability of rock formations when assessing the risk of landslides and erosion. In Peru, INGEMMET makes geological data available at a scale of 1:50,000 and was updated in 2022 [58].
Although geomorphological features are related to the occurrence of huaicos, geomorphological and hydrogeological data have not been used in the literature. Hydrogeology, in turn, was evaluated because it is related to soil saturation, which can also influence soil/rock stability. Geomorphological data at a scale of 1:250,000, dating from 2016, and hydrogeological data at a scale of 1:2,000,000, also from 2016, were obtained from INGEMMET [59].
Soil typology was obtained from the Ministry of the Environment (MINAM) at a scale of 1:2,000,000 and dates from 2009 [60]. Soil properties, such as shear strength and permeability, are related to the propensity for sliding and water movement in the surface portion [28].
Using QGIS software, all data were filtered to include only the Lima Department.

Land Use Data

Files were prepared to characterize land use, showing distances from urbanized areas, watercourses, and ecosystems. The hydrography was obtained from the Geo Peru National Georeferenced Data Platform at a scale of 1:100,000, prepared by the National Geographic Institute.
Urbanized areas were obtained by classifying Landsat 2023 images provided by the United States Geological Survey (USGS). The classification was conducted in QGIS using an unsupervised method, pixel by pixel.
The linear distance of each cell from the point of interest was then generated in QGIS for the hydrography and urbanized areas. Although huaicos occur due to the generation of ephemeral surface flows and in natural environments, the distances between urbanized areas and hydrography can suggest a possible indirect relationship between these factors. Although huaicos are characterized by the generation of ephemeral flows, these occur both in basins completely devoid of surface flows and in basins with perennial watercourses. If this region has the potential for huaicos to occur, the flow generated by the huaicos will contribute to the watercourse. It is worth noting that it is during this gravitational movement that the flows generated by the huaicos reach the urbanized areas that are in their path.
The 2019 MINAM ecosystem information was used on a scale of 2:200,000. Although at first glance it is not directly related to the occurrence of huaicos, this information may allow us to identify a correlation.

2.3. Maxent

To insert the data into Maxent, all environmental variables were transformed into a matrix format (Geotiff), ensuring that all data had the same size, number of pixels, and positioning. The data compatibility process was executed in QGIS software, where the initial vector format data were transformed into matrix data. The matrix data underwent reprojection of the coordinate system and datum (WGS UTM 18S) and were filtered to display only the Department of Lima.
Of the records of huaico occurrences in the Department of Lima, 70% were randomly selected, with the remaining 30% reserved for model validation, according to the methodology proposed by Arabameri et al. [41] and Javidan et al. [27].
Maxent has a user-friendly interface where it is simple to input and configure data. The samples (70% of the records) comprising the data on the occurrence of huaicos in the Department of Lima were input into CSV format.
After inserting the environmental variables into ASCII format, each ecological variable must be considered as either continuous or categorical data. The remaining information corresponds to the model configuration. The model configuration was performed according to the guidelines established by Dalapicolla [61], suggesting the adoption of a random test percentage of 30%, bootstrap model replica-running, 15 replicas, and a maximum number of 10,000 iterations.
The data insertion process in Maxent followed the steps shown in Figure 3.
A set of monthly models was organized, that is, one model for each month, considering all the environmental variables (DEM, slope, aspect, TRI, TWI, geology, geomorphology, hydrogeology, soils, distance from water courses, distance from urbanized areas, ecosystems, and monthly precipitation) and another set of monthly models (one model for each month) for only the variables described by INGEMMET (DEM, slope, geology, geomorphology, and precipitation) as drivers in the huaico formation process.
The data from Maxent were analyzed with the kappa index for each model and the overlap of each area with the risk of occurrence in each monthly model for the other months. The continuous probability model (0 to 1) was reclassified based on the logistic threshold of a training presence of 10 percentiles.

3. Results and Discussion

3.1. Environmental Variables

The results obtained from the correlation between environmental variables in a GIS environment showed the maximum correlation for each variable related to itself, resulting in the value of 1. The only high correlations, above 0.7 [29], were between the average maximum precipitations. The correlations did not exceed 0.4 among the other variables, indicating acceptable correlation levels below 0.7. A model containing more than one mean maximum precipitation will include a high correlation between these variables, which, according to Kornejady, Ownegh, and Bahremand [29], should be avoided to minimize the complexity of the model. In any case, Maxent remains minimally affected by the correlation of variables, even with small samples [48].
Regarding the physical characteristics of the study area (Figure 4), the altitude, according to the digital elevation model of the Department of Lima, ranges from 0 to 6401 m. The variation from 0 to 1000 m is located in the coastal region of the department, with approximately 30% of the Department of Lima within this altitude range. Moving inland, the altitude of the department increases. However, a closer look at the department’s relief reveals the “quebradas”, which are micro basins where significant variations in altitude occur on a small scale and where superficial flows are formed. These are found along the coastal region and the valleys of the perennial rivers of the department. The slope is also steeper in these regions, becoming flatter at the bottoms of the valleys and in the coastal region, which can be called a floodplain. Throughout the department, the slope varies from 0° to 89.95°, with 21% of the areas having a slope up to 10° and 83% having a slope up to 50°. The aspect ratio in the Department of Lima varies from 0° to 360°, with visual orientation patterns in the “quebradas”. The TRI (terrain ruggedness index) varies from 0 to 454, being lower in the lower and flatter areas. This TRI ranges from level to moderately rough, with the index reaching levels of extreme ruggedness. The TWI (topographic wetness index) varies from 0 to 20.01, with the highest values in the lower areas and less steep portions of the terrain, thus indicating that these areas are the most suitable for water accumulation and possible flooding.
The Department of Lima has 38 geological classifications, which are grouped into 14 macro-classifications in Figure 5.
On the other hand, the Department of Lima has 40 hydrogeological categories. Similar to the geology, these categories were grouped into classifications to facilitate visualization. Regional geomorphology has 45 classifications. It should be noted that the department’s 38 geological classifications, 40 hydrogeological classifications, and 45 geomorphological classifications were used for modeling. There are only seven pedological classifications in the Department of Lima.
The caption in Figure 5 presents the correspondence between the code for each class and the class denomination as a whole.
Figure 6 illustrates that the Euclidean distance from urbanized areas varies from 0 to 35.7 km, and the perennial watercourses in the department vary from 0 to 25.6 km. Regarding ecosystems, the Peruvian Pacific Coastal Desert classifications are observed. These are characterized by a hyper-arid climate devoid of vegetation, consisting of sandy soils and rocky outcrops. In addition, the classes of Serranía Esteparia (mountain range steppe) and Puna stand out for their herbaceous vegetation.
Figure 7 shows that some meteorological stations experienced historical precipitation maximums of as little as 1 mm in June, July, and August. In contrast, there are regions in the department where, during these same months, the average maximums reached 308 mm. In any case, February has historically had the highest average maximums for the stations with the lowest average maximums (52 mm). For the stations with the highest average maximums, precipitation during February was much larger (974 mm). March, October, and December also have quite pronounced historical average maximums. These observations are validated by the data shown in Figure 8, which indicates the average values for the 36 rainfall stations analyzed. They highlight the comparison between the averages and the maximums, where the average levels represent a maximum of 44% of the maximum historical levels in March, reaching only 13% of the maximum levels historically recorded in the department during May.
Regarding the increase in precipitation levels in Peru, the El Niño phenomenon responsible for warming the Pacific waters stands out [62]. For example, in the cycles of 1982–1983, 1997–1998, and 2017, precipitation levels were extraordinary, exceeding 1000 mm over a three-month cycle, triggering huaicos and floods that led to considerable economic losses and even deaths [62].
Gonzales and Ingol [41] observed the particular formation of the 2017 Coastal El Niño that directly and powerfully struck Peru’s coastal region. The authors concluded that it was a succession of events. In their own words, the Pacific Decade Oscillation in its positive phase gave rise to coupled ocean–atmosphere processes generating the ENSO Modoki and La Niña Modoki, and the coastal ENSO was a sub-process of the La Niña Modoki, the lower part of 2017’s ENSO Modoki. The Modoki event represents a unique pattern of tripolar pressure at sea level [63]. Strong El Niño events result in higher levels of precipitation and areas affected by precipitation, which consequently increase the occurrence and recurrence of huaicos.
Events such as the 2017 Coastal El Niño are difficult to predict and prepare for. Including the data in the modeling allows for an evaluation of how the Coastal El Niño intensity influences the occurrence of the huaicos.

3.2. Modeling in Maxent

Table 2 shows the quality of information of the Maxent models. All models have adequate quality, both in the “Test AUC”, which must be greater than 0.7, and in the “10 percentile training presence test omission”, which should not exceed 0.2. The area under the curve ranges from 0 to 1, with 1 indicating the highest accuracy. The AUC ranges of 0.5–0.6, 0.6–0.7, 0.7–0.8, 0.8–0.9, and 0.9–1 classify the models as poor, average, good, very good, and excellent, respectively, according to Yesilnacar [64], with our models classified as very good.
The kappa coefficient represents five classes—k < 0.4, 0.4 < k < 0.55, 0.55 < k < 0.85, 0.85 < k < 0.99, respectively—with poor, average, good, excellent, and very excellent validation [65]. In the kappa test, the models generated with all variables performed excellently. However, the models generated were based only on the environmental variables described by INGEMMET as influential in the formation of huaicos, and the kappa index was almost maximum, classified as very excellent.
Figure 9 and Figure 10 present the results of the models, reclassified to represent the areas with a probability of huaico occurrence.
On average, 17% of the areas in the Department of Lima have a probability of huaico occurrence when considering the models using all the variables. July and August had the lowest probability of huaicos occurring, while June and September had the highest likelihood of huaicos occurring. However, these higher areas in June and September seem inconsistent with the average maximum precipitation of these months, which is lower than those with the highest precipitation.
When analyzing the models generated using the INGEMMET variables identified as drivers of huaico occurrence, the models’ quality was better, and the areas with a probability of huaico occurrence were higher. On average, 42% of the Department of Lima has a likelihood of occurrence of huaicos. Using variables that do not effectively contribute to the occurrence of huaicos made the models complex, causing the algorithm to search for nonexistent relations between the sample data and the environmental variables. This resulted in lower model quality than those using influential contributing variables.
Regarding the coincidence between the monthly models, Table 3 shows the coincidence matrix of the areas with a probability of occurrence for each model, taking the months horizontally as a reference. On average, there is a coincidence of 84% between the monthly models. The greater the correspondence between the models throughout the year, the greater the chance that even with precipitation variation, the areas with a probability of occurrence will be basically the same.
Regarding the variables that contribute the most to the models, Figure 11 shows that the distance from water courses is the factor that contributes most to the monthly models, followed by geomorphology, digital elevation model, distance from urbanized areas, slope, and geology.
In general, there is a predominance of the probability of occurrence of huaicos up to 5 km from watercourses. The distance from urbanized areas coincides with the distance from rivers, as they are located primarily in the bottoms of valleys, close to waterways. Thus, the occurrence of huaicos is associated mainly with a distance of up to 2.5 km from urbanized areas.
The geomorphology representing the probability of occurrence includes mountains in intrusive rocks, mountains in volcanic rocks, mountains in volcanic-sedimentary rocks, and alluvial–torrential slopes or foothills.
Mountains in intrusive rock correspond to outcrops of intrusive rocks, igneous rocks with high degrees of fracturing due to tectonic processes that, together with the steep slopes associated with this formation, increase rockfalls [54]. Mountains in volcanic rocks are outcrops of volcanic rocks with steep slopes susceptible to rockfalls and collapses, and that may present intense fracturing and weathering [66]. Mountains in volcanic–sedimentary rocks also have steep slopes associated with large landslides, huaicos, and rock avalanches [54]. The geomorphological classification of alluvial–torrential slope or foothill is characterized by the accumulation of material mobilized by previous huaicos, which locally modify the direction of river courses and are located at the mouths of streams toward the main rivers [55].
Regarding the relief, the areas where huaicos are likely to occur are in the valleys, which corroborates precisely with the shorter distances from water courses. Regarding slope, the occurrences are related to inclinations of up to 30 degrees, which in themselves represent extremely high slopes of approximately 65%.
However, it is important to note that the variable contributing the most also has the greatest distance from a normal distribution.
In the monthly models that consider only the variables specified by INGEMMET as relevant to the occurrence of huaicos, all months presented a similar distribution of contributions from environmental variables, with the most significant contribution from geomorphology in the classes described above. The second most important contributing factor is relief, which also follows the ranges described in the characterization of the results of the previous model. The other factors, geology, slope, and average monthly precipitation, have the least influence in the models and similar influences (Figure 12).
Unlike the models using all variables, in the model using only the variables considered influential in the occurrence of huaicos according to INGEMMET, all variables have a normal distribution, contributing to their better quality and representativeness.
Kornejady, Ownegh, and Bahremand [29] and Phillips [26] warn against interpreting the contributions of factors in models, since Maxent can find the best solution through different paths, which may result in various factor contributions. In other words, the contribution cannot be considered an indicator of the importance of the environmental variable. The jackknife test provides more significant information regarding the importance of variables. The test requires removing a factor from the modeling, using a variable alone, and using all variables [29]. The jackknife tests responded almost equally to the contributions of the variables in all the models. Thus, in this work, it is safe to regard the relations of the variable contributions as a measure of the importance of the variables in predicting the occurrence of huaicos.
Although identifying areas where huaicos occur does not correspond to identifying areas where landslides and floods occur, Rusk et al. [28] evaluated both hazards and adopted variables similar to those in this work. For landslides, factors related to relief and geomorphology had a high contribution. However, unlike this work, for Rusk et al. [28], precipitation contributed significantly to the model for both types of hazards: approximately 35% for floods and 25% for landslides. The results for landslides by Javidan et al. [27] point in the same direction, where relevant geomorphological factors, such as slope and altitude alone, did not interfere with the model.
Although the occurrence of huaicos does not directly depend on the distance from watercourses, there is a concentrated probability of huaico occurrence in the vicinity of rivers in the models that used all variables, unlike the results of Javidan et al. [28]. They found the distance from watercourses to be an essential predictive factor due to its importance in the speed and magnitude of flooding, given the high flow concentration in the drainage network, increasing the occurrence of floods near streams. However, the relation is comprehensible because some “quebradas” belong to perennial water course watersheds.
Chen et al. [30] also indicated a relationship between landslides and the proximity of watercourses and roads in more significant portions of the terrain. The authors also highlighted the distance from roads in their models, an anthropic factor that modifies the natural stability of the studied areas and threatens the infrastructure, significantly decreasing the quality of the model. Anthropic factors do not influence huaico occurrences. This is due to the fact that the formation of huaicos does not occur in inhabited areas. The areas where huaicos form are areas of steep slope, difficult access, few inputs, that is, of low general attractiveness for the spread of urbanization and crops. In short, they are areas with desert characteristics. Still, the impacts are intensified by the anthropic factors of urbanized areas and infrastructure. Since, the expansion of urbanization in the flatter areas corresponds precisely to the areas where the flows generated by the huaicos are directed by gravity.
Similar to the results of Javidan et al. [27], there is a lower incidence of huaicos in high-altitude locations. This finding was based on the results of the elevation response curves, which demonstrate that huaicos, like floods, occur in low-lying areas and plains. It should be noted that the runoff from the basins that form the huaicos converges typically into a single channel, where it can either reach an urbanized area or follow a floodplain to the ocean and then reach an urbanized area.
Javidan et al. [27] found that higher precipitation levels increase landslides (650–690 mm annually) and lower rainfall levels cause erosion. For huaicos, an increase in precipitation should result in a larger area with a probability of huaicos occurring. However, the rise in precipitation did not respond linearly in the models. In general, precipitation did not contribute significantly to this phenomenon. Therefore, the model constitution does not permit us to infer the levels of rainfall that trigger the huaicos.
Giráldez et al. [67] evaluated the extreme climate events of precipitation in metropolitan Lima from 1965 to 2013. The authors found that 16 mm of precipitation occurred in 1970, which was sufficiently strong to flood the streets, causing power failures, damaging the Jorge Chávez International Airport facilities, destroying about 2000 homes, and triggering landslides and rivers to overflow. Although there are no specific data on huaicos, their occurrence during this extreme event is almost certain. This could mean that almost all the precipitation data used in this work would be enough to trigger a huaico occurrence.
In contrast, Rusk et al. [28] point out that climate change can alter model responses to increase the potential for disasters. A further step in taking action to prevent and mitigate the impacts of disasters is identifying the social vulnerability of the affected populations.

3.3. Uncertainty

It is important to note that the records of huaico occurrences themselves represent a degree of uncertainty in the model. Since the occurrence data used as input for the model were in point format, as were the available occurrence data, there are no records of the discrete areas affected by huaicos, and there is no way to insert areas as input data in Maxent. This limits the generation of a model that reflects reality as closely as possible. There is also no guarantee that the points identified as areas of huaico occurrence were located correctly.
Furthermore, the other input data have different scales of elaboration. The hydrogeology and soil data, in particular, at scales of 1:2,000,000, present generic data about the spatiality of this information. This does not prevent us from evaluating a large area, such as the Department of Lima, but in the models where these data are inserted, they can decrease the quality of the models. For this same reason, in the models considering only the variables indicated as influential by INGEMMET in the occurrence of huaicos, the absence of hydrogeology and soil values reduces the uncertainty in these models. The existence of more detailed data could alter both the final response of the models and the relevance of these data within the analysis. This could improve the quality levels of the models, although, it could have no effect on the overall quality of these as well.

4. Conclusions

Using variables not considered to be drivers of huaicos made the models complex and underestimated the probability of occurrence. Moreover, the monthly approach was ineffective when distinguishing differences among the precipitation levels. In general, precipitation did not contribute significantly to this phenomenon. Thus, the constitution of these models does not permit inference of the precipitation levels that trigger the huaicos. Therefore, new models for the coastal departments of Peru, in Maxent algorithms and other machine learning algorithms, will only use the INGEMMET driver variables, and a single precipitation datum will be used. The question remains as to whether the annual precipitation data can fit the models better than a month’s average data. Another alternative yet to be tested is using other machine learning algorithms in huaico occurrence susceptibility.
Inevitably, climate change in this region points toward an increase in temperature that will likely lead to an increase in precipitation events, thereby increasing the number of huaicos and affected areas, although the models in this work were not yet able to identify the contribution of precipitation in the huaico occurrences. Because these models use past data, a priori predictions of future changes would not alter the construction of the model. However, future work could incorporate projected precipitation data as an input variable for comparison with those based on historical data. Firstly, the use of precipitation data should be adjusted as suggested previously.
Maxent returned extremely high levels of correspondence between the models and the actual registered occurrence data. In this panorama, almost 42% of the area of the Department of Lima exhibits a probability of huaico occurrence. There is no reasonable expectation of reducing the areas with the potential for huaicos to occur. These areas could be zoned according to public policies in which disaster preparations could be adopted. Furthermore, these areas could be evaluated to avoid occupying or even vacate them. Regarding public policies, residents should be continuously educated about the risks of huaicos, and structural measures, such as systems to dam the flow, can be implemented to minimize the impacts of huaicos.

Author Contributions

G.M.d.S. was responsible for conceptualization, methodology, data development, formal analysis, and writing—original draft preparation. V.E.S. and G.C. were responsible for formal analysis and writing—review and editing. M.P. was responsible for writing—review and editing and supervision. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data Availability Statement

Data are contained within the article.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Moncayo, O.P. Rasgos Generales Sobre la Geología del Peru; INGEMMET: Lima, Peru, 1998.
  2. Instituto Geológico, Minero y Metalúrgico—INGEMMET. Estudio de Los Recursos Minerales del Perú; Franja n° 4: Lima, Peru, 2005. [Google Scholar]
  3. Medina, L.; Villacorta, S.; Zavala, B.; Vílchez, M.; Núñez, S.; Luque, G.; Calderón, E. Caracterización geomorfológica del Norte peruano. In Boletín de la Sociedad Geológica del Perú; INGEMMET: Lima, Peru, 2015; Volume 110, pp. 205–208. [Google Scholar]
  4. Ministerio del Ambiente del Gobierno del Perú—MINAM. Mapa Nacional de Ecosistemas del Perú; Memoria: Louisville, KY, USA, 2019. [Google Scholar]
  5. Rivera, M.; Vilchez, M.; Vela, J. Peligros por huaicos en la ciudad de Arequipa. In Fortalecimiento de Capacidades Para Mitigar Los Impactos de Huaicos en Perú; Porras, M.R., Phillips, J.C., Armijos Burneo, M.T., Contreras, R.A., Eds.; INGEMMET: Lima, Peru, 2018; pp. 45–49. [Google Scholar]
  6. Perú. Plan Nacional de Gestión del Riesgo de Desastres (PLANAGERD) 2022–2030; Instituto Nacional de Defensa Civil—INDECI: Lima, Peru, 2022.
  7. French, A.; Mechler, R.; Arestegui, M.; Macclune, K.; Cisneros, A. Root causes of recurrent catastrophe: The political ecology of El Niño-related disasters in Peru. Int. J. Disaster Risk Reduct. 2020, 47, 101539. [Google Scholar] [CrossRef]
  8. Instituto Nacional de Defensa Civil—INDECI. Compendio Estadístico del Instituto Nacional de Defensa Civil; INDECI: Lima, Peru, 2021.
  9. Instituto Nacional de Defensa Civil—INDECI. Compendio Estadístico del Instituto Nacional de Defensa Civil; INDECI: Lima, Peru, 2019.
  10. Federación de Periodistas del Perú. Lluvias y Huaicos: Al Menos 13 Distritos de Lima Sufren de Inundaciones. Available online: https://fpp.org.pe/__trashed-91/ (accessed on 10 December 2024).
  11. Diario La Otra Cara. Lima Este Inundada por los Huaicos y las Intensas Lluvias. Available online: https://diariolaotracara.com/lima-este-inundada-por-los-huaicos-y-las-intensas-lluvias/ (accessed on 10 December 2024).
  12. Willax. Huaicos en Lima: Alcalde de Punta de Hermosa Señala que “No hay Presupuesto” para Evitar el Colapso de Casa de Playa. Available online: https://willax.pe/actualidad/huaicos-en-lima-alcalde-de-punta-de-hermosa-senala-que-no-hay-presupuesto-para-evitar-el-colapso-de-casa-de-playa#google_vignette (accessed on 10 December 2024).
  13. CanalB pe. Lima Fueron los Más Golpeados Nuevamente por Huaicos. Available online: https://canalb.pe/noticias/actualidad/distritos-del-este-y-el-sur-de-lima-fueron-los-mas-golpeados-nuevamente-por-huaicos (accessed on 10 December 2024).
  14. United Nations—UN. Sendai Framework for Disaster Risk Reduction 2015–2030; United Nations Office for Disaster Risk Reduction: Geneva, Switzerland, 2015. [Google Scholar]
  15. Tehrani, F.S.; Calvello, M.; Liu, Z.; Zhang, L.; Lacasse, S. Machine learning and landslide studies: Recent advances and applications. Nat. Hazards 2022, 114, 1197–1245. [Google Scholar] [CrossRef]
  16. Ado, M.; Amitab, K.; Maji, A.K.; Jasińska, E.; Gono, R.; Leonowicz, Z.; Jasiński, M. Landslide susceptibility mapping using machine learning: A literature survey. Remote Sens. 2022, 14, 3029. [Google Scholar] [CrossRef]
  17. Bao, H.; Zeng, C.; Peng, Y.; Wu, S. The use of digital technologies for landslide disaster risk research and disaster risk management: Progress and prospects. Environ. Earth Sci. 2022, 81, 446. [Google Scholar] [CrossRef]
  18. Mosavi, A.; Ozturk, P.; Chau, K. Flood prediction using machine learning models: Review. Water 2018, 10, 1536. [Google Scholar] [CrossRef]
  19. Li, J.; Fang, Z.; Zhang, J.; Huang, O.; He, C. Mapping basin-scale supply-demand dynamics of flood regulation service—A case study in the Baiyangdian Lake Basin, China. Ecol. Indic. 2022, 139, 108902. [Google Scholar] [CrossRef]
  20. Bot, K.; Borges, J.G. A systematic review of applications of machine learning techniques for wildfire management decision support. Inventions 2022, 7, 15. [Google Scholar] [CrossRef]
  21. Alkhatib, R.; Sahwan, W.; Alkhatieb, A.; Schütt, B. A brief review of machine learning algorithms in forest fire science. Appl. Sci. 2023, 13, 8275. [Google Scholar] [CrossRef]
  22. Ridzwan, N.S.M.; Yusoff, S.H.M. Machine learning for earthquake prediction: A review (2017–2021). Earth Sci. Inform. 2023, 16, 1133–1149. [Google Scholar] [CrossRef]
  23. Ferreira, I.J.M.; Campanharo, W.A.; Barbosa, M.L.F.; da Silva, S.S.; Selaya, G.; Aragão, L.E.O.C.; Anderson, L.O. Assessment of fire hazard in Southwestern Amazon. Front. For. Glob. Change 2023, 6, 1107417. [Google Scholar] [CrossRef]
  24. Izquierdo- Horna, L.; Zevallos, J.; Yepez, Y. An integrated approach to seismic risk assessment using random forest and hierarchical analysis: Pisco, Peru. Heliyon 2022, 8, e10926. [Google Scholar] [CrossRef]
  25. Kumar, C.; Walton, G.; Santi, P.; Luza, C. An Ensemble Approach of Feature Selection and Machine Learning Models for Regional Landslide Susceptibility Mapping in the Arid Mountainous Terrain of Southern Peru. Remote Sens. 2023, 15, 1376. [Google Scholar] [CrossRef]
  26. Phillips, S.J.; Anderson, R.P.; Dudík, M.; Schapire, R.E.; Blair, M. Opening the black box: An open-source release of Maxent. Ecography 2017, 40, 887–893. [Google Scholar] [CrossRef]
  27. Javidan, N.; Kavian, A.; Pourghasemi, H.R.; Conoscenti, C.; Jafarian, Z.; Rodrigo-Comino, J. Evaluation of multi-hazard map produced using MaxEnt machine learning technique. Sci. Rep. 2021, 11, 6496. [Google Scholar] [CrossRef] [PubMed]
  28. Rusk, J.; Maharjan, A.; Tiwari, P.; Chen, T.H.K.; Shneiderman, S.; Turin, M.; Seto, K.C. Multi-hazard susceptibility and exposure assessment of the Hindu Kush Himalaya. Sci. Total Environ. 2022, 804, 150039. [Google Scholar] [CrossRef]
  29. Kornejady, A.; Ownegh, M.; Bahreman, A. Landslide susceptibility assessment uses a maximum entropy model with two data sampling methods. Catena 2017, 152, 144–162. [Google Scholar] [CrossRef]
  30. Chen, W.; Pourghasemi, H.R.; Kornejady, A.; Zhang, N. Landslide spatial modeling: Introducing new ensembles of ANN, MaxEnt, and SVM machine learning techniques. Geoderma 2017, 305, 314–327. [Google Scholar] [CrossRef]
  31. Mokhtari, M.; Abedian, S. Spatial prediction of landslide susceptibility in Taleghan basin, Iran. Stoch. Environ. Res. Risk Assess. 2019, 33, 1297–1325. [Google Scholar] [CrossRef]
  32. Sheikh, V.; Kornejady, A.; Ownegh, M. Application of the coupled TOPSIS–Mahalanobis distance for multi-hazard-based management of the target districts of the Golestan Province, Iran. Nat. Hazards 2019, 96, 1335–1365. [Google Scholar] [CrossRef]
  33. Yusri, S.; Retnowati, E.; Prastowo, M. Combining Participatory Mapping, Cloud Computing and Machine Learning for Mapping Climate Induced Landslide Susceptibility in Lembeh Island, North Sulawesi. In IOP Conference Series; Earth and Environmental Science: New York, NY, USA, 2019; Volume 363. [Google Scholar] [CrossRef]
  34. Innocenti, C.; Battaglini, L.; D’Angelo, S.; Fiorentino, A. Submarine landslides: Mapping the susceptibility in European seas. Q. J. Eng. Geol. Hydrogeol. 2020, 54, qjegh2020–qjegh2027. [Google Scholar] [CrossRef]
  35. Kim, H.G.; Park, C.Y. Landslide susceptibility analysis of photovoltaic power stations in Gangwon-do, Republic of Korea. Geomat. Nat. Hazards Risk 2021, 12, 2328–2351. [Google Scholar] [CrossRef]
  36. Liu, Y.; Xu, P.; Cao, C.; Shan, B.; Zhu, K.; Ma, Q.; Zhang, Z.; Yin, H. A comparative evaluation of machine learning algorithms and an improved optimal model for landslide susceptibility: A case study. Geomat. Nat. Hazards Risk 2021, 12, 1973–2001. [Google Scholar] [CrossRef]
  37. Liu, Y.; Zhao, L.; Bao, A.; Li, J.; Yan, X. Chinese high-resolution satellite data and gis -based assessment of landslide susceptibility along Highway G30 in Guozigou Valley using logistic regression and maxent model. Remote Sens. 2022, 14, 3620. [Google Scholar] [CrossRef]
  38. Shahzad, N.; Ding, X.; Abbas, S. A comparative assessment of machine learning models for landslide susceptibility mapping in the rugged terrain of northern pakistan. Appl. Sci. 2022, 12, 2280. [Google Scholar] [CrossRef]
  39. Pourghasemi, H.R.; Yousefi, S.; Kornejady, A.; Cerdà, A. Performance assessment of individual and ensemble data-mining techniques for gully erosion modeling. Sci. Total Environ. 2017, 609, 764–775. [Google Scholar] [CrossRef] [PubMed]
  40. Botero-Acosta, A.; Chu, M.L.; Guzman, J.A.; Starks, P.J.; Moriasi, D.N. Riparian erosion vulnerability model based on environmental characteristics. J. Environ. Manag. 2017, 203, 592–602. [Google Scholar] [CrossRef] [PubMed]
  41. Arabameri, A.; Asadi Nalivan, O.; Saha, S.; Roy, J.; Pradhan, B.; Tiefenbacher, J.P.; Thi Ngo, P.T. Novel ensemble approaches of machine learning techniques in modeling the gully erosion susceptibility. Remote Sens. 2020, 12, 1890. [Google Scholar] [CrossRef]
  42. Bosino, A.; Giordani, P.; Quénéhervé, G.; Maerker, M. Assessment of calanchi and rill –interrill erosion susceptibilities using terrain analysis and geostochastics: A case study in the Oltrepo Pavese, Northern Apennines, Italy. Earth Surf. Process. Landf. 2020, 45, 3025–3041. [Google Scholar] [CrossRef]
  43. Vacchiano, G.; Foderi, C.; Berretti, R.; Marchi, E.; Motta, R. Modeling anthropogenic and natural fire ignitions in an inner-alpine valley. Nat. Hazards Earth Syst. Sci. 2018, 18, 935–948. [Google Scholar] [CrossRef]
  44. Adab, H.; Atabati, A.; Oliveira, S.; Gheshlagh, A.M. Assessing fire hazard potential and its main drivers in Mazandaran province, Iran: A data-driven approach. Environ. Monit. Assess. 2018, 190, 670. [Google Scholar] [CrossRef] [PubMed]
  45. Martín, Y.; Zúñiga-Antón, M.; Mimbrero, M.R. Modeling temporal variation of fire occurrence towards the dynamic prediction of human wildfire ignition danger in northeast Spain. Geomat. Nat. Hazards Risk 2018, 10, 385–411. [Google Scholar] [CrossRef]
  46. Lim, C.H.; Kim, Y.S.; Won, M.; Kim, S.J.; Lee, W.K. Can satellite-based data substitute for surveyed data to predict the spatial probability of forest fire? A geostatistical approach to forest fire in the Republic of Korea. Geomat. Nat. Hazards Risk 2019, 10, 719–739. [Google Scholar] [CrossRef]
  47. Darabi, H.; Choubin, B.; Rahmati, O.; Haghighi, A.T.; Pradhan, B.; Kløve, B. Urban flood risk mapping using the GARP and QUEST models: A comparative study of machine learning techniques. J. Hydrol. 2019, 569, 142–154. [Google Scholar] [CrossRef]
  48. Norallahi, M.; Kaboli, H.S. Urban flood hazard mapping using machine learning models: GARP, RF, MaxEnt, and NB. Nat. Hazards 2021, 106, 119–137. [Google Scholar] [CrossRef]
  49. Rajabi, A.; Shabanlou, S.; Yosefvand, F.; Kiani, A. Exploring the sample size and replication scenarios effect on spatial prediction of flood, using MARS and MaxEnt methods case study: Saliantape catchment, Golestan, Iran. Nat. Hazards 2021, 109, 871–901. [Google Scholar] [CrossRef]
  50. Instituto Geológico, Minero y Metalúrgico—INGEMMET. Mapa de Inventario de Peligros Geológicos; INGEMMET: Lima, Peru, 2019.
  51. Peel, M.C.; Finlayson, B.L.; McMahon, T.A. Updated world map of the Köppen-Geiger climate classification. Hydrol. Earth Syst. Sci. 2007, 11, 1633–1644. [Google Scholar] [CrossRef]
  52. Servicio Nacional de Meteorología e Hidrología del Perú—SENAMHI. Mapa Climático del Perú. Available online: https://www.senamhi.gob.pe/?p=mapa-climatico-del-peru (accessed on 15 December 2024).
  53. Instituto Geológico, Minero y Metalúrgico—INGEMMET. Zonas Críticas Por Peligros Geológicos en la Cuenca Hancay-Lambayeque; INGEMMET: Lima, Peru, 2005.
  54. Instituto Geológico, Minero y Metalúrgico—INGEMMET. Peligros Por Movimientos en Masa en la Quebrada Tinaja (Dist. Pachacamac-Cieneguilla, Provincia y Región Lima); INGEMMET: Lima, Peru, 2012.
  55. Instituto Geológico, Minero y Metalúrgico—INGEMMET. Evaluación de Peligros Geológicos Por Movimientos en Masa en el Anexo de Colpa; INGEMMET: Lima, Peru, 2019.
  56. Instituto Geológico, Minero y Metalúrgico—INGEMMET. Inspección Geológica Realizada en la Quebrada San Lázaro—Arequipa; INGEMMET: Lima, Peru, 2020.
  57. Instituto Geológico, Minero y Metalúrgico—INGEMMET. Evaluación de Peligros Geológicos Por Flujo de Detritos (Huaicos) en la Quebrada Zaparo; INGEMMET: Lima, Peru, 2021.
  58. Instituto Geológico, Minero y Metalúrgico—INGEMMET. Carta Geológica Nacional a Escala 1:50000; INGEMMET: Lima, Peru, 2022.
  59. Instituto Geológico, Minero y Metalúrgico—INGEMMET. Mapa Geomorfológico del Perú; INGEMMET: Lima, Peru, 2016.
  60. Ministerio del Ambiente del Gobierno del Perú—MINAM. Mapa Nacional de Suelos del Perú; INGEMMET: Lima, Peru, 2009.
  61. Dalapicolla, J. Tutorial de Modelos de Distribuição de Espécies: Guia Prático Usando o Maxent e o Arcgis 10. In Laboratório de Mastozoologia e Biogeografia; Universidade Federal do Espírito Santo: Vitória, Brazil, 2016. [Google Scholar]
  62. Servicio Nacional de Meteorología e Hidrología del Perú—SENAMHI. Ministerio del Ambiente El Niño: Histórico de lluvias; Lima, Peru. 2021. Available online: https://www.senamhi.gob.pe/?&p=escenarios-lluvia (accessed on 17 September 2024).
  63. Gonzales, E.; Ingol, E. Determination of a New Coastal ENSO Oceanic Index for Northern Peru. Climate 2021, 9, 71. [Google Scholar] [CrossRef]
  64. Yesilnacar, E.K. The Application of Computational Intelligence to Landslide Susceptibility Mapping in Turkey. Ph.D Thesis, Department of Geomatics, University of Melbourne, Parkville, VIC, Australia, 2005; p. 423. [Google Scholar]
  65. Monserud, R.A.; Leemans, R. Comparing global vegetation maps with the Kappa statistic. Ecol. Model. 1992, 62, 275–293. [Google Scholar] [CrossRef]
  66. Instituto Geológico, Minero y Metalúrgico—INGEMMET. Evaluación de Peligros Geológicos en el Sector de Octuyoj-Manás; INGEMMET: Lima, Peru, 2021.
  67. Giráldez, L.; Silva, Y.; Flores-Rojas, J.L.; Trasmonte, G. Diagnosis of the Extreme Climate Events of Temperature and Precipitation in Metropolitan Lima during 1965–2013. Climate 2022, 10, 112. [Google Scholar] [CrossRef]
Climate 13 00011 g001
Figure 1. Photographic records of huaicos in the Department of Lima. (a) Huaico that occurred in 2023 in an unspecified area of Lima [10]. (b) Huaico that also occurred in 2023 in Chaclayo [11]. (c) Huaico that occurred in 2023 in Punta Hermosa [12]. (d) Huaico that occurred in 2023 in an unspecified area of Lima [13].
Figure 1. Photographic records of huaicos in the Department of Lima. (a) Huaico that occurred in 2023 in an unspecified area of Lima [10]. (b) Huaico that also occurred in 2023 in Chaclayo [11]. (c) Huaico that occurred in 2023 in Punta Hermosa [12]. (d) Huaico that occurred in 2023 in an unspecified area of Lima [13].
Climate 13 00011 g001
Climate 13 00011 g002
Figure 2. Map of the study area and locations of huaico occurrences.
Figure 2. Map of the study area and locations of huaico occurrences.
Climate 13 00011 g002
Climate 13 00011 g003
Figure 3. Flowchart of the modeling process.
Figure 3. Flowchart of the modeling process.
Climate 13 00011 g003
Climate 13 00011 g004
Figure 4. Environmental relief variables.
Figure 4. Environmental relief variables.
Climate 13 00011 g004
Climate 13 00011 g005
Figure 5. Geomorphological environmental variables. Geology map legend: Js—Upper Jurassic marine, JsKi—Upper Jurassic Lower Cretaceous, volcanic sediments, KP—Cretaceous Paleocene, tonalite, granodiorite, Ki—Lower Cretaceous, KG, Kis—Cretaceous, marine and volcanic sediments, Ks—Cretaceous, super units, N—Neogene, Nm—Neogene Miocene, P—Paleozoic, PN—Paleogene Neocene, Pe—Paleogene Eocene, Qh—Quaternary Holocene, Qpl—Quaternary Pleistocene and TsJi—Upper Triassic, Lower Jurassic. Soil map legend: ARh-SCh—Arenosol haplic—Solonchak haplic, FLe-RGe—Fluvisol eutric—Regosol eutric, LPd-ANz—Leptosol dystric—Andosol vitreous, LPd-R—Leptosol districo—Lithic outcrop, LPe-CMe—Leptosol eutric—Cambisol eutric, LPq-R—Lithic Leptosol—Lithic outcrop and RGd-R—Regosol districo—Lithic outcrop. Geomorphology map legend: A—lagoon, Ab—mountain foothill fan, Bo—bofedales (wetlands), C—dune field and coast, Cl—coastal ridge, Dan—anthropogenic deposit, Do—salt dome, F—coastal strip, Lg—water bodies, M—volcanic plateau, Mo—glacial debris, P—mountain slope or foothill, Pl—floodplain, RC—rock hill, RCE—structural hill, RCL—hill and ridge, RCLD—dissected hill and ridge, RM—mountain, RMCE—mountain and structural hill, RME—structural mountain, R×riverbed, Sp—swamp system, T—marine terrace, Ta—alluvial terrace, Ti—undifferentiated terrace, V—slope, and Vll—valley.
Figure 5. Geomorphological environmental variables. Geology map legend: Js—Upper Jurassic marine, JsKi—Upper Jurassic Lower Cretaceous, volcanic sediments, KP—Cretaceous Paleocene, tonalite, granodiorite, Ki—Lower Cretaceous, KG, Kis—Cretaceous, marine and volcanic sediments, Ks—Cretaceous, super units, N—Neogene, Nm—Neogene Miocene, P—Paleozoic, PN—Paleogene Neocene, Pe—Paleogene Eocene, Qh—Quaternary Holocene, Qpl—Quaternary Pleistocene and TsJi—Upper Triassic, Lower Jurassic. Soil map legend: ARh-SCh—Arenosol haplic—Solonchak haplic, FLe-RGe—Fluvisol eutric—Regosol eutric, LPd-ANz—Leptosol dystric—Andosol vitreous, LPd-R—Leptosol districo—Lithic outcrop, LPe-CMe—Leptosol eutric—Cambisol eutric, LPq-R—Lithic Leptosol—Lithic outcrop and RGd-R—Regosol districo—Lithic outcrop. Geomorphology map legend: A—lagoon, Ab—mountain foothill fan, Bo—bofedales (wetlands), C—dune field and coast, Cl—coastal ridge, Dan—anthropogenic deposit, Do—salt dome, F—coastal strip, Lg—water bodies, M—volcanic plateau, Mo—glacial debris, P—mountain slope or foothill, Pl—floodplain, RC—rock hill, RCE—structural hill, RCL—hill and ridge, RCLD—dissected hill and ridge, RM—mountain, RMCE—mountain and structural hill, RME—structural mountain, R×riverbed, Sp—swamp system, T—marine terrace, Ta—alluvial terrace, Ti—undifferentiated terrace, V—slope, and Vll—valley.
Climate 13 00011 g005
Climate 13 00011 g006
Figure 6. Environmental variables of land use and land cover.
Figure 6. Environmental variables of land use and land cover.
Climate 13 00011 g006
Climate 13 00011 g007
Figure 7. Monthly maximum average precipitation (MAP).
Figure 7. Monthly maximum average precipitation (MAP).
Climate 13 00011 g007
Climate 13 00011 g008
Figure 8. Average precipitation behavior at the rain gauge stations in the Department of Lima.
Figure 8. Average precipitation behavior at the rain gauge stations in the Department of Lima.
Climate 13 00011 g008
Climate 13 00011 g009
Figure 9. Probability results considering monthly models.
Figure 9. Probability results considering monthly models.
Climate 13 00011 g009
Climate 13 00011 g010
Figure 10. Areas with and without the probability of huaico occurrences.
Figure 10. Areas with and without the probability of huaico occurrences.
Climate 13 00011 g010
Climate 13 00011 g011
Figure 11. Contribution of variables for each monthly model considering all environmental variables and the behavior of the data compared to normal distribution.
Figure 11. Contribution of variables for each monthly model considering all environmental variables and the behavior of the data compared to normal distribution.
Climate 13 00011 g011
Climate 13 00011 g012
Figure 12. Contribution of variables for each monthly model, considering the environmental variables indicated by INGEMMET and the behavior of the data compared to normal distribution.
Figure 12. Contribution of variables for each monthly model, considering the environmental variables indicated by INGEMMET and the behavior of the data compared to normal distribution.
Climate 13 00011 g012
Table 1. Articles that address the use of Maxent, with a focus on natural hazards risk mapping.
Table 1. Articles that address the use of Maxent, with a focus on natural hazards risk mapping.
Hazard TypeStudy Area Most Used Variables
Landslides
[27,28,29,30,31,32,33,34,35,36,37,38].		Iran (Gorganrood Watershed), Hindu Kush Himalaya, Iran (Ziarat watershed), China (Wanyuan), Iran (Taleghan Basin), Iran (Golestan Province), North Sulawesi (Lembeh Island), European Seas, Republic of Korea (Gangwon-do), China (Xulong Gully), China (Guozigou Valley), North PakistanAltitude, aspect, slope, distance from permanent watercourses, road distance, geology, curvature profile, curvature plane, precipitation, land use, topographic wetness index (TWI), and terrain ruggedness index (TRI)
Erosion
[32,39,40,41,42].		Iran (Golestan am Basin), Iran (Fars Province), United States (Fort Cobb watershed),Italy (Oltrepo Pavese)Altitude, height above nearest drainage, aspect, slope, distance from roads, permanent watercourses, geology, precipitation, temperature, and land use
Fires
[28,43,44,45,46].		Hindu Kush Himalaya, Italy (Aosta Valley), Iran (Mazandaran Province), Northeast Spain, South KoreaAltitude, aspect, slope, precipitation (average, minimum), temperature (average, maximum), land use, and wind
Floods
[27,28,32,47,48,49].		Iran (Gorganrood Watershed), Hindu Kush Himalaya, Iran (Kermanshah City), Iran (Sari City), Iran (Golestan Province), Iran (Saliantapeh Catchment)Altitude, aspect, slope, CN (curve number), distance from permanent watercourses, precipitation, and land use
Table 2. Quality of models.
Table 2. Quality of models.
MonthEnvironmental VariablesAUC TestTen Percentile Training Presence Test OmissionKappa CoefficientEnvironmental VariablesAUC TestTen Percentile Training Presence Test OmissionKappa Coefficient
JanDEM, slope, aspect, TWI, TRI, geology, geomorphology, hydrogeology, soils, ecosystems, distance from watercourses, distance from urbanized areas, precipitation0.88020.12910.79DEM, slope, geology, geomorphology, precipitation0.82890.11420.98
Feb0.88020.12910.800.82210.11400.98
Mar0.87870.13020.790.82570.11870.97
Apr0.87920.12820.790.81810.10720.97
May0.87940.12710.800.82250.11110.98
Jun0.89580.12700.800.81560.12280.98
Jul0.87730.14070.790.82290.12140.98
Aug0.88610.11840.790.82620.11300.98
Sep0.87870.1280.790.81870.1140.98
Oct0.88090.13360.790.83040.08480.98
Nov0.87930.12690.790.81970.12440.98
Dec0.87980.12990.790.82870.11950.98
Table 3. Coincidence between monthly models (%).
Table 3. Coincidence between monthly models (%).
JanFebMarAprMayJunJulAugSepOctNovDec
Jan1008984898867846691869087
Feb8510083858665836492838685
Mar8791100828698986888808479
Apr8989831008866846693878886
May8790878810068826792848687
Jun85861278487100849187848584
Jul90931089088711006994907089
Aug8485898586928310086838484
Sep8488838585638062100838683
Oct8990849087688767941009088
Nov8990848786666565938610087
Dec9091838990688667938891100
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Santos, G.M.d.; Schneider, V.E.; Cemin, G.; Poletto, M. Identifying the Areas at Risk of Huaico Occurrences in the Department of Lima, Peru. Climate 2025, 13, 11. https://doi.org/10.3390/cli13010011

AMA Style

Santos GMd, Schneider VE, Cemin G, Poletto M. Identifying the Areas at Risk of Huaico Occurrences in the Department of Lima, Peru. Climate. 2025; 13(1):11. https://doi.org/10.3390/cli13010011

Chicago/Turabian Style

Santos, Geise Macedo dos, Vania Elisabete Schneider, Gisele Cemin, and Matheus Poletto. 2025. "Identifying the Areas at Risk of Huaico Occurrences in the Department of Lima, Peru" Climate 13, no. 1: 11. https://doi.org/10.3390/cli13010011

APA Style

Santos, G. M. d., Schneider, V. E., Cemin, G., & Poletto, M. (2025). Identifying the Areas at Risk of Huaico Occurrences in the Department of Lima, Peru. Climate, 13(1), 11. https://doi.org/10.3390/cli13010011

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop