Mapping Flood-Prone Areas Using GIS and Morphometric Analysis in the Mantaro Watershed, Peru: Approach to Susceptibility Assessment and Management

Abstract

Floods represent one of the most significant climate-related hazards, particularly in regions with complex topographies and variable precipitation patterns. This study assesses flood-prone areas within the Mantaro watershed, Peru, using Geographic Information Systems (GISs) and morphometric analysis. The methodology integrates digital elevation models (DEMs) with hydrological parameters, applying weighted sum analysis to classify 18 sub-watersheds into different flood priority levels. Morphometric parameters, including basin relief, drainage density, and slope, were analyzed to establish correlations between watershed morphology and flood susceptibility. The results indicate that approximately 74.38% of the watershed exhibits high to very high flood risk, with the most vulnerable sub-watersheds characterized by steep slopes, high drainage densities, and compact morphometric configurations. The correlation matrix confirms that watershed topography significantly influences surface runoff behavior, underscoring the necessity of incorporating geospatial analysis into flood risk assessment frameworks. The classification of sub-watersheds into priority levels provides a scientific basis for optimizing resource allocation in flood mitigation strategies. This study highlights the importance of integrating advanced geospatial technologies, such as GISs and remote sensing, into hydrological risk assessments. The findings emphasize the need for proactive watershed management, including the use of real-time monitoring and digital tools for climate adaptation. Future research should explore the influence of land-use changes and climate variability on flood dynamics to enhance predictive modeling. These insights contribute to evidence-based decision-making for disaster risk reduction, reinforcing resilience in climate-sensitive regions.

1. Introduction

Global warming has driven significant changes in temperature and precipitation patterns, disrupting the natural hydrological cycle and intensifying extreme events such as droughts and floods [1,2,3] Among these, floods are particularly responsible for widespread ecological, social, and economic losses due to intense rainfall and inadequate drainage management [4,5]. The increasing urbanization and population density in vulnerable areas further amplified flood exposure [6].

Historically, flood susceptibility assessment relied on qualitative observations and the manual interpretation of topographic maps [7,8]. The development of quantitative geomorphology enabled the mathematical evaluation of watershed characteristics, with early studies introducing key morphometric parameters based on shape, area, and relief [9,10].

The subsequent advent of Geographic Information Systems (GISs) and remote sensing (RS) revolutionized watershed studies, allowing for high-resolution, spatially explicit assessments of hydrological and geomorphological variables [11,12,13,14] Digital elevation models (DEMs), in particular, provided reliable, freely available topographic data essential for accurate morphometric analysis and susceptibility mapping [5,15]. Despite these advancements, some processes remained computationally intensive or required specialized software [16].

In recent years, cloud computing platforms such as Google Earth Engine (GEE) have significantly enhanced watershed analysis by providing access to extensive preprocessed datasets and robust computational capacity for handling large-scale studies [17]. The integration of RS, GISs, and GEE has proven highly effective in flood susceptibility assessments, particularly when combined with morphometric analysis techniques [1,4,18].

Within the context of flood susceptibility, watersheds are critical spatial units where the interaction of topography, drainage networks, and precipitation patterns determines surface runoff behavior [11,19]. Morphometric analysis quantitatively characterizes these landform attributes, providing insights into hydrological processes and flood-prone conditions [1,12,13].

In Peru, 159 major hydrographic units have been delineated across the Pacific, Amazon, and Titicaca regions, with a hierarchical classification system implemented by the National Water Authority [20]. The country’s complex topography and climatic variability, exacerbated by phenomena such as the El Niño–Southern Oscillation (ENSO), contribute to high flood susceptibility, particularly in the Andean and Amazonian regions [21,22].

The Mantaro watershed, located in the central Peruvian Andes, holds strategic importance due to its role in agricultural production, hydroelectric generation, and urban water supply [23]. However, its complex terrain and seasonal rainfall patterns increase vulnerability to floods, highlighting the need for advanced susceptibility assessments to inform water resource management and disaster prevention efforts.

Accordingly, this study aims to identify flood susceptibility prone areas within the Mantaro watershed by integrating morphometric analysis with geospatial tools such as GISs and GEE. By classifying sub-watersheds into susceptibility levels based on quantitative topographic and hydrological indicators, this research provides scientific support for watershed management, climate adaptation, and regional planning initiatives.

2. Materials and Methods

2.1. Study Area

The Mantaro watershed is located in the central area of the Peruvian Tropical Andes (PTA) located between 10°33′52.66″ and 13°32′31.39″ south latitude, and between 73°55′10.88″ and 76°39′16.01″ west longitude [24] covering an area of 34,363 km². This area is sorted by Pasco (6.56%), Ayacucho (11,21%), Huancavelica (25.43%) and Mantaro (56.80%) local water administration [25]. Furthermore, according to the Pfafstetter delimitation and coding of hydrographic units in Peru, the Mantaro watershed is classified as level 4 and consists of 18 sub-watersheds, 9 of which are at level 5 with the remaining 9 at level 6 [26], as shown in Figure 1 and

Table 1. Sub-watershed general characteristics.

Code	Name	Area (km2)	Perimeter (km)	Level
MSW_1	San Juan	941.340	193.890	6
MSW_2	UH 499697	995.152	196.414	6
MSW_3	Conocancha	717.748	159.869	6
MSW_4	Junin	1719.858	228.375	6
MSW_5	UH 499695	1719.262	252.923	6
MSW_6	Yauli	691.189	165.109	6
MSW_7	UH 499693	943.353	186.564	6
MSW_8	Pachacayo	821.464	160.430	6
MSW_9	UH 499691	2113.261	285.442	6
MSW_10	Conas	1718.361	228.759	5
MSW_11	Medio alto mantaro	891.774	172.394	5
MSW_12	Vilca	2957.000	301.270	5
MSW_13	Interna Huarmicocha	88.372	53.667	5
MSW_14	Medio mantaro	611.624	136.441	5
MSW_15	Bajo mantaro	8139.947	571.542	5
MSW_16	Medio bajo mantaro	1289.140	214.897	5
MSW_17	Ichu	1382.734	219.196	5
MSW_18	Huarpa	6804.950	478.875	5

.

These hierarchical levels, established by the National Water Authority (ANA), represent the administrative subdivision of the Mantaro watershed into progressively smaller hydrological units for water resource management purposes. Specifically, level 5 corresponds to larger sub-watersheds, while level 6 identifies more localized drainage units. However, the flood susceptibility classification presented in this study is based on independent morphometric analysis and does not rely on the pre-existing flood assessment levels assigned to these administrative units.

Additionally, agriculture is the dominant employment sector in the Mantaro watershed, as indicated by the macroeconomic variable of the economically active employed population (54.6%). Agriculture spans approximately 339,065 hectares of agricultural land, 29% of which is at risk. This is used for key crops including potatoes, maize, carrots, barley, alfalfa, and artichokes [27].

Finally, according to Thornthwaite’s climate classification, the Mantaro River Basin encompasses nine distinct climate types, reflecting its considerable altitudinal and geographic variability. These include warm and very humid climates such as A(r)A’H4 and B(r)A’H4, characterized by abundant precipitation throughout the year and high relative humidity; temperate and rainy zones like A(r)B’2H3 and C(o,i)B’2H3, which experience seasonal rainfall deficits yet maintain high humidity levels; and colder climates such as B(i)D’H3, B(o,i)C’H3, and B(o,i)D’H3, defined by semi-frigid to cold conditions, autumn and winter rainfall deficiencies, and humid atmospheres. Additionally, the classification includes a warm-temperate and humid-type (B(r)B’1H4), and a region of permanent snow cover (N), highlighting the basin’s extreme climatic diversity [28].

2.2. Data

In this research, required data comes from two main sources. On the one hand, the Peruvian watershed shapefile at levels 4, 5, and 6 were retrieved from the ANA database [29] to identify the location of 18 sub-watersheds (Figure 1) and obtain an individual layer per sub-watershed. On the other hand, a digital elevation model (DEM) limited to the area of the 18 sub-watersheds was obtained from Google Earth Engine (GEE) cloud platform [30] to analyze the drainage/stream network through morphometric parameters.

It is crucial to mention that the digital elevation dataset was produced by the Shuttle Radar Topography Mission (SRTM) [31]. Currently, SRTM Plus is provided by the National Aeronautics and Space Administration (NASA) at a spatial resolution of 1 arc-second (approximately 30 m) [30]. Furthermore, unlike previous versions that contain voids or are void-filled using commercial sources, this dataset has undergone a gap-filling process utilizing open access data sources, including ASTER GDEM2, GMTED2010, and NED, ensuring greater continuity and accuracy in terrain representation [30].

Finally, the DEM in GEE was obtained from the USGS SRTMGL1 30 m resolution dataset via Google Earth Engine. The elevation band was extracted and clipped to the defined area of interest. The resulting elevation image was visualized and subsequently exported at a 30 m spatial resolution using the UTM Zone 18S coordinate reference system (EPSG:32718) for further analysis.

It is important to clarify that this study focuses exclusively on flood susceptibility derived from morphometric characteristics obtained from the 30 × 30 m resolution DEM data. The exclusion of dynamic climatic variables, such as rainfall anomalies, ENSO-related events, or climate change projections, reflects both theoretical considerations and current limitations in accessing high-resolution, basin-wide climatic datasets applicable to this scale of analysis.

2.3. Methodology

This research is composed of different phases, which are explained in Figure 2.

2.3.1. ArcGIS Processing

Once the Mantaro watersheds are uploaded into the software with the DEM, the applied commands are (i) extract by mask; (ii) fill; (iii) flow direction; (iv) flow accumulation; (v) raster calculation; (vi) stream order—Strahler [8]; and (vii) stream to features. Each step uses the information generated by previous commands until raster calculation as step vi combines the information from iv and v, and vii requires iii and vi as inputs.

Furthermore, in the current section, specifically in step v, the streams were defined using a threshold value representing approximately 1% of the DEM cell count.

2.3.2. Google Earth Engine Processing

Despite the prior use of GEE to obtain the DEM, it was appropriate for this study to analyze slopes using the data that were generated in the previous phase (ArcGIS) and subsequently uploaded into the GEE environment.

A sub-watershed (MSW_1–18) was extracted from a previously classified feature collection using Google Earth Engine (GEE). The Shuttle Radar Topography Mission (SRTM) digital elevation model at a 30 m resolution was clipped to the study area to derive the terrain slope. A terrain analysis module was applied to calculate the slope raster, which was subsequently visualized using a custom color palette. Descriptive statistics (maximum, minimum, mean, median, standard deviation, and variance) were computed for the slope values within the watershed boundary using the reduced region function with appropriate reducers. Additionally, pixel count was evaluated to estimate the number of valid data points within the region. All geospatial operations, including data masking, visualization, and statistical analysis, were executed through the GEE JavaScript API.

2.3.3. Morphometric Analysis

This research has sorted the morphometric parameters into three groups per sub-watershed, as presented in Table 2, Table 3 and Table 4:

Table 2. Preliminary and linear morphometric parameter.

N°	Parameter	Unit	Value from	Source
1	Area (A)	km2	ArcGIS	-
2	Perimeter (P)	km	ArcGIS	-
3	N° of streams order 1	nominal	ArcGIS	-
4	N° of total streams	nominal	ArcGIS	-
5	Stream order (u)	nominal	ArcGIS	[8]
6	Stream length (Lu)	km	ArcGIS	[7]
7	Basin length (Lb)	km	1.312 × Area0.568	[32]

Table 2. Preliminary and linear morphometric parameter.

N°	Parameter	Unit	Value from	Source
1	Area (A)	km2	ArcGIS	-
2	Perimeter (P)	km	ArcGIS	-
3	N° of streams order 1	nominal	ArcGIS	-
4	N° of total streams	nominal	ArcGIS	-
5	Stream order (u)	nominal	ArcGIS	[8]
6	Stream length (Lu)	km	ArcGIS	[7]
7	Basin length (Lb)	km	1.312 × Area0.568	[32]

Table 3. Morphometric parameter related to area.

N°	Parameter	Unit	Value from	Source
1	Stream frequency (Fs)	km−2	$\Sigma Nu/A$	[7]
2	Drainage density (Dd)	km−1	$\Sigma L/A$	[7]
3	Form factor (Ff)	nominal	$\Sigma A/{Lb}^2$	[7]
4	Circularity ratio (Cr)	nominal	$Cr=4\pi A/P^2$	[10]
5	Texture ratio (Tr)	nominal	$Tr=Nt/P$	[7]
6	Elongation ratio (Er)	nominal	$Er=2\sqrt{\left({\displaystyle \frac{A}{\pi }}\right)}/Lb$	[9]
7	Shape factor (Sf)	nominal	$Sf={Lb}^2/A$	[7]

Table 3. Morphometric parameter related to area.

N°	Parameter	Unit	Value from	Source
1	Stream frequency (Fs)	km−2	$\Sigma Nu/A$	[7]
2	Drainage density (Dd)	km−1	$\Sigma L/A$	[7]
3	Form factor (Ff)	nominal	$\Sigma A/{Lb}^2$	[7]
4	Circularity ratio (Cr)	nominal	$Cr=4\pi A/P^2$	[10]
5	Texture ratio (Tr)	nominal	$Tr=Nt/P$	[7]
6	Elongation ratio (Er)	nominal	$Er=2\sqrt{\left({\displaystyle \frac{A}{\pi }}\right)}/Lb$	[9]
7	Shape factor (Sf)	nominal	$Sf={Lb}^2/A$	[7]

Table 4. Morphometric parameter related to relief.

N°	Parameter	Unit	Value from	Source
1	Max elevation	m	DEM	-
2	Min elevation	m	DEM	-
3	Basin relief (R)	m	R = H-h	[9]
4	Relief ratio (Rr)	nominal	R = H/Lb	[9]
5	Average slope (As)	Degrees	Google Earth Engine	-

Table 4. Morphometric parameter related to relief.

N°	Parameter	Unit	Value from	Source
1	Max elevation	m	DEM	-
2	Min elevation	m	DEM	-
3	Basin relief (R)	m	R = H-h	[9]
4	Relief ratio (Rr)	nominal	R = H/Lb	[9]
5	Average slope (As)	Degrees	Google Earth Engine	-

2.3.4. Weighted Sum Analysis

Firstly, this phase necessitates the development of an organized matrix encompassing each of the 18 sub-watersheds along with their respective morphometric values. Subsequently, as indicated in [1], the morphometric parameters were categorized into two groups. The first group comprises parameters with a direct influence, including basin relief (R), relief ratio (Rr), drainage density (Dd), stream frequency (Fs), circulatory ratio (Cr), texture ratio (Tr), and average slope (As). The second group consists of parameters with an indirect influence, which are associated with the elongation ratio (Er), form factor (Ff), and shape factor (Sf).

Secondly, for parameters with direct influence, a hierarchical ranking was assigned, with the highest value receiving a rank of 1 and continuing sequentially up to 18, corresponding to each sub-watershed. Conversely, for parameters with indirect influence, the highest value was assigned a rank of 18, decreasing sequentially to 1 across the sub-watersheds [33]. It is important to note that, although [33] applied this methodology in the context of soil erosion risk, several of the selected morphometric parameters such as slope, drainage density, and relief are widely recognized as influencing surface runoff and flood susceptibility conditions. Therefore, this approach has been adapted to identify flood-prone areas based on geomorphometric characteristics, without representing a direct flood hazard or inundation model.

Thirdly, the correlation matrix of the morphometric parameters was computed using RStudio version 2024.04.2 + 764, incorporating the previously established ranking. The weight of each morphometric parameter was determined by dividing the sum of its respective column by the total sum of the correlation matrix [33].

Lastly, the weighted sum analysis composite parameter (WSAcp) is computed by multiplying the assigned weight of each morphometric parameter by its hierarchical ranking for each sub-watershed.

2.3.5. Flood Susceptibility Classification Based on Morphometric Analysis

The final methodological phase of this research involves summing the weighted sum analysis composite parameters (WSAcp) for morphometric parameters with direct influence (+) and indirect influence (−) for each sub-watershed. Consequently, sub-watersheds with lower composite values are identified as the most flood-prone areas.

The classification identifies sub-watersheds with different levels of flood susceptibility based on the composite weighted sum of morphometric parameters. Although the spatial units correspond to pre-existing sub-watersheds delineated by the National Water Authority of Peru, the flood susceptibility levels were independently derived using quantitative geomorphometric indicators. This approach ensures that the classification reflects physical susceptibility conditions, rather than administrative or pre-existing designations.

The division into four susceptibility categories, very high, high, medium, and low, follows a consistent threshold-based scheme, where sub-watersheds with the lowest composite scores exhibit the most unfavorable morphometric characteristics, such as steep slopes, high drainage density, and compact shapes, which are known to enhance surface runoff and increase flooding potential. This classification method is adapted from [1] and widely used in hydrological susceptibility assessments. By employing relative ranking within the watershed, the approach provides a systematic and replicable framework for prioritizing sub-watersheds based on their intrinsic susceptibility to flooding.

Specifically, given that this study examines 18 sub-watersheds within the Mantaro watershed, the five sub-watersheds with the lowest composite values are classified as very high flood-prone areas, the next five as high, the following four as medium, and the remaining four as low flood-prone areas.

3. Results and Discussions

A preventive approach to risk assessment and management must be a cornerstone of national policy to address various natural hazards, particularly in the context of climate change, which has intensified their frequency and severity [34]. This is especially crucial for Peru, recognized as one of the most vulnerable developing countries [35]. Therefore, this research identifies flood-prone areas (sub-watersheds) in one of the most valuable sectors of Peru due to its capability of food and energy production [27].

3.1. Sub-Watershed and Morphometric Parameters Values

The morphometric analysis of the 18 sub-watersheds in the Mantaro watershed, as detailed in

Table 5. Morphometric parameters values per sub-watershed.

	Basin Relief (R)	Relief Ratio (Rr)	Drainage Density (Dd)	Stream Frequency (Fs)	Circulatory Ratio (Cr)	Texture Ratio (Tr)	Average Slope (As)	Elongation Ratio (Er)	Form Factor (Ff)	Shape Factor (Sf)
MSW_1	1194	0.082	0.391	0.116	0.315	0.289	7.323	0.540	0.229	4.368
MSW_2	1108	0.077	0.370	0.111	0.324	0.290	10.258	0.538	0.227	4.401
MSW_3	1371	0.097	0.316	0.111	0.353	0.256	12.059	0.550	0.238	4.210
MSW_4	754	0.054	0.468	0.124	0.414	0.473	6.931	0.518	0.211	4.741
MSW_5	1714	0.060	0.332	0.113	0.338	0.387	14.675	0.518	0.211	4.741
MSW_6	1881	0.104	0.320	0.106	0.319	0.224	17.691	0.551	0.239	4.188
MSW_7	2171	0.089	0.336	0.130	0.341	0.332	16.872	0.540	0.229	4.369
MSW_8	2174	0.096	0.350	0.145	0.401	0.374	14.313	0.545	0.233	4.288
MSW_9	2002	0.051	0.377	0.123	0.326	0.462	13.249	0.511	0.205	4.876
MSW_10	1733	0.054	0.352	0.131	0.413	0.503	12.526	0.518	0.211	4.741
MSW_11	2546	0.089	0.370	0.130	0.377	0.342	15.976	0.542	0.231	4.336
MSW_12	2305	0.043	0.325	0.116	0.409	0.571	15.935	0.499	0.196	5.104
MSW_13	482	0.308	0.274	0.124	0.386	0.112	10.452	0.634	0.316	3.166
MSW_14	2043	0.097	0.309	0.129	0.413	0.293	19.369	0.556	0.243	4.119
MSW_15	5258	0.026	0.331	0.115	0.313	0.819	30.214	0.466	0.171	5.857
MSW_16	2596	0.062	0.314	0.126	0.351	0.382	25.524	0.528	0.219	4.559
MSW_17	2392	0.065	0.335	0.125	0.362	0.397	16.074	0.526	0.217	4.603
MSW_18	3122	0.027	0.354	0.120	0.373	0.854	17.431	0.472	0.175	5.717

, reveals significant variations in key hydrological parameters. The basin relief (R) ranges from 482 m (MSW_13) to 5258 m (MSW_15), indicating substantial topographic differences that influence runoff velocity and erosion potential [36]. Similarly, the relief ratio (Rr) exhibits notable variability, with MSW_13 presenting the highest value (0.308), suggesting steeper slopes and increased risk to rapid water discharge [37], while MSW_15 and MSW_18 display the lowest values (0.026 and 0.027, respectively), indicative of more gradual terrain.

Hydrological connectivity, as reflected by drainage density (Dd) and stream frequency (Fs), also varies across sub-watersheds. MSW_4 has the highest drainage density (0.468 km⁻¹), implying an extensive stream network and lower infiltration potential [38], whereas MSW_13 records the lowest value (0.274 km⁻¹), suggesting a reduced drainage efficiency. Likewise, stream frequency is highest in MSW_8 (0.145) and lowest in MSW_6 (0.106), reinforcing the spatial heterogeneity of surface water dynamics. Moreover, average slope (As) shows a strong correlation with relief characteristics, with MSW_15 exhibiting the steepest terrain (30.214°), potentially accelerating runoff and increasing flood risk, while MSW_4 has the lowest slope value (6.931°), favoring slower water movement and greater infiltration. These findings highlight the complexity of watershed hydrology and underscore the necessity of region-specific flood management strategies.

These morphometric variations, summarized in Table 5, directly influence the susceptibility classification presented in subsequent analyses. Sub-watersheds characterized by steep slopes, high basin relief, and dense drainage networks, such as MSW_15 and MSW_4, exhibit physical conditions that accelerate surface runoff, reduce infiltration capacity, and consequently increase flood susceptibility [37,39]. In contrast, sub-watersheds with more elongated shapes, lower drainage densities, and gentler slopes, such as MSW_2 and MSW_13, demonstrate greater potential for water retention and reduced flood susceptibility. These morphometric indicators serve as the foundation for the weighted sum analysis applied in this study, reinforcing the relevance of geomorphological parameters in understanding spatial flood susceptibility patterns within the Mantaro watershed.

3.2. Preliminary Hierarchy Order

The hierarchical ranking of morphometric parameters across the 18 sub-watersheds, as presented in

Table 6. Hierarchical ranking assigned for direct and indirect morphometric parameter per sub-watershed.

	Basin Relief (R)	Relief Ratio (Rr)	Drainage Density (Dd)	Stream Frequency (Fs)	Circulatory Ratio (Cr)	Texture Ratio (Tr)	Average Slope (As)	Elongation Ratio (Er)	Form Factor (Ff)	Shape Factor (Sf)
MSW_1	15	8	2	13	17	15	17	12	12	7
MSW_2	16	9	4	17	15	14	16	10	10	9
MSW_3	14	3	15	16	10	16	14	15	15	4
MSW_4	17	14	1	9	1	5	18	5	5	14
MSW_5	13	12	11	15	13	8	10	6	6	13
MSW_6	11	2	14	18	16	17	4	16	16	3
MSW_7	8	7	9	3	12	12	6	11	11	8
MSW_8	7	5	8	1	5	10	11	14	14	5
MSW_9	10	15	3	10	14	6	12	4	4	15
MSW_10	12	13	7	2	3	4	13	7	7	12
MSW_11	4	6	5	4	7	11	8	13	13	6
MSW_12	6	16	13	12	4	3	9	3	3	16
MSW_13	18	1	18	8	6	18	15	18	18	1
MSW_14	9	4	17	5	2	13	3	17	17	2
MSW_15	1	18	12	14	18	2	1	1	1	18
MSW_16	3	11	16	6	11	9	2	9	9	10
MSW_17	5	10	10	7	9	7	7	8	8	11
MSW_18	2	17	6	11	8	1	5	2	2	17

, highlights distinct variations. The basin relief (R) ranking indicates that MSW_15 holds the first position (rank 1), reinforcing its significant elevation difference, while MSW_13 exhibits the last rank (18), suggesting a relatively lower relief variation. Similarly, the relief ratio (Rr) ranking aligns with this pattern, where MSW_13 is ranked first due to its steep slopes, increasing its propensity for rapid runoff [37], whereas MSW_15 and MSW_18 occupy the lowest ranks, indicative of gentler terrain and reduced runoff acceleration. Regarding drainage density (Dd) and stream frequency (Fs), MSW_4 and MSW_14 are ranked first, confirming their dense and well-connected stream networks, which contribute to faster hydrological responses and increased surface runoff [37]. In contrast, MSW_13 and MSW_3, which hold the lowest ranks, exhibit lower drainage densities, implying reduced runoff connectivity and potentially higher infiltration. The indirect influence parameters, including elongation ratio (Er), form factor (Ff), and shape factor (Sf), further differentiate sub-watersheds, with MSW_15 demonstrating the most elongated shape (rank 1) and MSW_13 exhibiting a more compact configuration (rank 18).

Table 6 summarizes the integrated flood susceptibility ranking for each sub-watershed based on the combined evaluation of morphometric parameters presented in this section. While Table 1 includes the physical characteristics of the sub-watersheds, such as the area, perimeter, and administrative level, these values are not directly comparable to the susceptibility classification, which reflects geomorphometric conditions influencing surface runoff potential.

The relevance of a preliminary hierarchical ranking in morphometric studies lies in its capacity to systematically evaluate watershed characteristics and their influence on hydrological behavior [40]. By prioritizing parameters that directly and indirectly affect surface runoff, this ranking enhances the accuracy of predictive models and provides a robust foundation for decision-making in watershed management [41]. Furthermore, it enables the comparative analysis of multiple watersheds, ensuring that mitigation strategies are aligned with the specific geomorphological conditions of each sub-watershed.

3.3. Morphometric Parameter Weight

The correlation matrix presented in

Table 7. Correlation matrix of morphometric parameters.

	Basin Relief (R)	Relief Ratio (Rr)	Drainage Density (Dd)	Stream Frequency (Fs)	Circulatory Ratio (Cr)	Texture Ratio (Tr)	Average Slope (As)	Elongation Ratio (Er)	Form Factor (Ff)	Shape Factor (Sf)
Basin Relief (R)	1.000	−0.406	−0.139	0.276	−0.063	0.552	0.822	0.399	0.399	−0.399
Relief Ratio (Rr)	−0.406	1.000	−0.406	0.090	0.057	−0.926	−0.129	−0.988	−0.988	0.988
Drainage Density (Dd)	−0.139	−0.406	1.000	0.040	−0.088	0.309	−0.498	0.406	0.406	−0.406
Stream Frequency (Fs)	0.276	0.090	0.040	1.000	0.606	0.166	0.117	−0.148	−0.148	0.148
Circulatory Ratio (Cr)	−0.063	0.057	−0.088	0.606	1.000	0.207	−0.162	−0.108	−0.108	0.108
Texture Ratio (Tr)	0.552	−0.926	0.309	0.166	0.207	1.000	0.249	0.913	0.913	−0.913
Average Slope (As)	0.822	−0.129	−0.498	0.117	−0.162	0.249	1.000	0.123	0.123	−0.123
Elongation Ratio (Er)	0.399	−0.988	0.406	−0.148	−0.108	0.913	0.123	1.000	1.000	−1.000
Form Factor (Ff)	0.399	−0.988	0.406	−0.148	−0.108	0.913	0.123	1.000	1.000	−1.000
Shape Factor (Sf)	−0.399	0.988	−0.406	0.148	0.108	−0.913	−0.123	−1.000	−1.000	1.000
sum	2.442	−1.707	0.623	2.147	1.449	2.471	1.521	1.598	1.598	−1.598
Weight (w)	0.232	−0.162	0.059	0.204	0.137	0.234	0.144	0.152	0.152	−0.152

quantifies the interrelationships among key morphometric parameters, providing insight into their combined influence on flood risk. Basin relief (R) exhibits a strong positive correlation with average slope (As) (0.822), indicating that sub-watersheds with higher elevation differences tend to have steeper slopes, which can accelerate surface runoff and increase erosion potential [42]. Additionally, texture ratio (Tr) shows a moderate correlation with basin relief (0.552), suggesting that areas with higher relief tend to develop more intricate drainage patterns, potentially enhancing runoff efficiency [43].

Conversely, relief ratio (Rr) demonstrates a strong negative correlation with texture ratio (−0.926) and elongation ratio (Er) (−0.988), implying that steeper terrains are associated with less elongated watershed shapes and lower textural complexity; however, [44], states that slopes ranging from moderate to steeper ground are associated with elongated watershed shapes. Similarly, elongation ratio and form factor (Ff) share a near-perfect inverse relationship with basin relief (−0.988), confirming that more elongated sub-watersheds correspond to lower elevation variations, which can lead to slower runoff and increased infiltration [45]. These correlations highlight the intricate balance between watershed topography and hydrological response, reinforcing the importance of integrating multiple morphometric parameters into flood risk assessments.

The correlation patterns observed in Table 7 confirm that flood susceptibility within the Mantaro watershed is governed by the complex interactions between relief, drainage network characteristics, and watershed geometry. These interdependencies justify the inclusion and relative weighting of specific morphometric parameters in the weighted sum analysis, ensuring that the final susceptibility classification reflects both the direct and indirect influence of watershed morphology on hydrological behavior. Therefore, the correlation matrix not only validates the parameter selection but also strengthens the scientific basis of the susceptibility mapping framework applied in this study.

3.4. Weighted Sum Analysis Composite Parameters

The weighted sum analysis composite parameter (WSAcp) provides an integrated evaluation of flood susceptibility by combining the hierarchical rankings of morphometric parameters with their respective weightings.

Table 8. WSAcp values per sub-watershed.

	Basin Relief (R)	Relief Ratio (Rr)	Drainage Density (Dd)	Stream Frequency (Fs)	Circulatory Ratio (Cr)	Texture Ratio (Tr)	Average Slope (As)	Elongation Ratio (Er)	Form Factor (Ff)	Shape Factor (Sf)
MSW_1	3.47	−1.30	0.12	2.65	2.34	3.52	2.45	1.82	1.82	−1.06
MSW_2	3.71	−1.46	0.24	3.46	2.06	3.28	2.31	1.52	1.52	−1.36
MSW_3	3.24	−0.49	0.89	3.26	1.37	3.75	2.02	2.27	2.27	−0.61
MSW_4	3.94	−2.27	0.06	1.83	0.14	1.17	2.60	0.76	0.76	−2.12
MSW_5	3.01	−1.94	0.65	3.05	1.79	1.87	1.44	0.91	0.91	−1.97
MSW_6	2.55	−0.32	0.83	3.66	2.20	3.98	0.58	2.42	2.42	−0.45
MSW_7	1.85	−1.13	0.53	0.61	1.65	2.81	0.87	1.67	1.67	−1.21
MSW_8	1.62	−0.81	0.47	0.20	0.69	2.34	1.59	2.12	2.12	−0.76
MSW_9	2.32	−2.43	0.18	2.04	1.92	1.41	1.73	0.61	0.61	−2.27
MSW_10	2.78	−2.10	0.41	0.41	0.41	0.94	1.88	1.06	1.06	−1.82
MSW_11	0.93	−0.97	0.30	0.81	0.96	2.58	1.15	1.97	1.97	−0.91
MSW_12	1.39	−2.59	0.77	2.44	0.55	0.70	1.30	0.45	0.45	−2.42
MSW_13	4.17	−0.16	1.06	1.63	0.82	4.22	2.16	2.73	2.73	−0.15
MSW_14	2.08	−0.65	1.01	1.02	0.27	3.05	0.43	2.58	2.58	−0.30
MSW_15	0.23	−2.91	0.71	2.85	2.47	0.47	0.14	0.15	0.15	−2.73
MSW_16	0.69	−1.78	0.95	1.22	1.51	2.11	0.29	1.36	1.36	−1.52
MSW_17	1.16	−1.62	0.59	1.43	1.24	1.64	1.01	1.21	1.21	−1.67
MSW_18	0.46	−2.75	0.35	2.24	1.10	0.23	0.72	0.30	0.30	−2.58

presents the WSAcp values for each sub-watershed, revealing previous significant variations that directly correlate with the flood-prone areas.

3.5. Flood-Prone Areas

The final classification of flood-prone sub-watersheds, based on the WSAcp values, is presented in

Table 9. Final ranking for flood-prone sub-watersheds and risky level.

Sub-Watershed	WSAcp(+)	WSAcp(−)	Value (+) + (−)	Ranking Level	Priority
MSW_1	13.25	2.58	10.67	17	Low
MSW_2	13.60	1.67	11.93	18	Low
MSW_3	14.04	3.94	10.11	16	Low
MSW_4	7.47	−0.61	8.07	11	Medium
MSW_5	9.88	−0.15	10.03	15	Low
MSW_6	13.48	4.39	9.08	14	Medium
MSW_7	7.19	2.12	5.07	8	High
MSW_8	6.11	3.49	2.62	2	Very high
MSW_9	7.16	−1.06	8.22	12	Medium
MSW_10	4.72	0.30	4.42	6	High
MSW_11	5.76	3.03	2.73	3	Very high
MSW_12	4.56	−1.52	6.08	9	High
MSW_13	13.91	5.30	8.60	13	Medium
MSW_14	7.21	4.85	2.37	1	Very high
MSW_15	3.96	−2.42	6.39	10	High
MSW_16	4.99	1.21	3.78	4	Very high
MSW_17	5.44	0.76	4.69	7	High
MSW_18	2.36	−1.97	4.33	5	Very high

and Figure 3. This ranking categorizes the 18 sub-watersheds into four priority levels: very high, high, medium, and low. MSW_14 (2.37), MSW_8 (2.62), and MSW_11 (2.73) emerge as one of the highest priority sub-watersheds, classified under the category of “very high”. These sub-watersheds exhibit a combination of high drainage density (Dd), steep average slopes (As), and compact morphometric characteristics, which accelerate surface runoff and increase flood susceptibility. Their prioritization in flood risk management strategies is essential to mitigate potential damage.

On the other hand, MSW_2 (11.93), MSW_1 (10.67), and MSW_3 (10.11) are categorized as “low”-priority sub-watershed due to their elongated watershed shapes (Er) and lower drainage densities, which promote greater water infiltration and slower runoff response [39]. Sub-watersheds classified as “high”- and “medium”-priority, such as MSW_7 (5.07) and MSW_4 (8.07), exhibit intermediate characteristics, where localized topographic variations and drainage network structures influence their susceptibility levels. This classification provides a critical basis for developing targeted watershed management plans, emphasizing the implementation of flood control measures in the most vulnerable sub-watersheds while reinforcing resilience in lower-risk areas.

Furthermore, Figure 3, in comparison with Table 1, demonstrates that areas classified as very high and high priority cover 10,418.952 km² and 15,141.395 km², respectively. This indicates that a total of 25,560.347 km² (74.38% of the Mantaro watershed) is highly vulnerable to flooding and requires immediate attention. These critical areas fall under the jurisdiction of the Huancayo, Huancavelica, and Ayacucho local water administrations, highlighting the urgent need for targeted flood management and mitigation strategies.

Although this study provides a technical classification of flood susceptibility based on morphometric analysis, the results have not yet been empirically validated against historical flood events, hydrological records, or field-based observations. Future research should focus on ground-truthing these findings, especially in high-susceptibility areas such as Huancayo, using disaster records, stream gauge data, and community-level surveys. Additionally, incorporating socio-economic vulnerability indicators, such as population density or livelihood dependencies, is essential to align susceptibility mapping with holistic flood management frameworks.

Compared to conventional flood-prone areas assessments relying solely on historical flood records or empirical models, this study demonstrates the advantage of integrating weighted sum analysis (WSA) with geomorphometric parameters derived from high-resolution DEM data. Similar approaches have been successfully applied for erosion or flash flood-prone areas [1,33], but their application to flood susceptibility remains limited, particularly in Andean basins. The methodology adopted here allows for the systematic prioritization of flood-prone areas based on intrinsic watershed characteristics, providing a cost-effective and scalable tool that complements traditional hydrological models, especially where ground data is scarce.

4. Conclusions

This study demonstrates the value of integrating morphometric analysis and GIS-based methodologies to identify flood-prone areas within the Mantaro watershed, Peru. By leveraging high-resolution digital elevation models and hydrological parameters, we detected significant spatial variations in flood susceptibility across the 18 sub-watersheds. Approximately 74.38% of the watershed area falls under categories of high to very high susceptibility, underscoring the need for targeted attention from local water management authorities. The results highlight that sub-watersheds, characterized by steep slopes, dense drainage networks, and compact shapes, present the greatest potential for extreme hydrological events, providing a technical basis for prioritizing mitigation efforts.

Correlation analysis confirms that watershed morphology exerts a direct influence on flood susceptibility. Parameters such as basin relief, average slope, and drainage density show strong associations with flood-prone conditions, reinforcing the importance of incorporating topographic variability into watershed planning. The prioritization of sub-watersheds into four susceptibility levels offers a practical framework for resource allocation, enabling decision-makers to focus interventions on the most critical areas while fostering resilience elsewhere. Moreover, the methodological approach applied here can be replicated in other regions with similar hydrological and geomorphological settings to support adaptive water resource management.

Future research should incorporate rainfall variability, climate change projections, and hydrological simulations to expand upon the susceptibility framework established in this study, thereby achieving a more comprehensive understanding of flood risks in the Mantaro watershed. Integrating dynamic climatic factors will enhance model predictability and improve decision-making at both regional and local scales.

Regarding practical application, it is essential to align flood mitigation strategies with Peru’s governance structures, fiscal realities, and community needs. Beyond traditional infrastructure interventions, this study advocates for implementing nature-based solutions, such as reforestation and slope stabilization, particularly in highly susceptible sub-watersheds. Efforts to strengthen early warning systems through community-based hydrological monitoring and foster collaboration between the National Water Authority (ANA), regional governments, and rural populations are also vital. This tiered, stakeholder-centered approach connects scientific insights with feasible, context-sensitive management actions, contributing to sustainable watershed governance and effective disaster risk reduction.

Given the increasing frequency and severity of extreme weather events, this research further highlights the need for real-time monitoring systems and advanced remote sensing technologies to improve flood forecasting capabilities. Future studies should also address the impacts of land-use changes and climate variability on watershed hydrodynamics to refine predictive models. Ultimately, enhancing interdisciplinary collaboration among scientists, policymakers, and local communities is imperative to develop sustainable, science-based interventions that safeguard vulnerable watersheds and human settlements, reinforcing the region’s climate resilience.