1 Inroduction

Floods, one of the most destructive and costly disasters of the twenty-first century, have prompted a reevaluation of flood risk management (Ahmed et al., 2024; Chen et al., 2023; de Brujin et al., 2022; Hussain et al., 2021). Flood risk poses a threat to over 5% of the total area (Bezak et al., 2023). The occurrence of floods is a dynamic process resulting from the complex interaction among various factors, like watershed management, hydro-meteorological, hydrogeological, and geomorphological conditions (Rahman et al., 2021). Rapid population growth has spurred an exponential increase in urban sprawl,that led to the decline in agricultural land, floodplains, waterbodies, and wetlands. These transformations have profound consequences for flood dynamics (Rahman et al., 2021). The development and deployment of effective flood risk management techniques are critical to mitigating the adverse impacts of flooding (Deepak et al., 2020). A deep understanding of long-term river flood dynamics is crucial for developing preparedness, adaptation, and disaster management strategies, as well as improving predictions of future flood frequency change (Yu et al., 2023). In the last three decades, floods have cost the global economy close to USD 386 billion. According to the Intergovernmental Panel on Climate Change, a substantial percentage of the globe showed an increase in unanticipated disasters, such as droughts, very high temperatures, and average precipitation of different magnitudes (Seneviratne et al., 2021). Contemporary climate research projects a marked increase in both the frequency and intensity of floods under future climate scenarios (Bodoque et al., 2020). Furthermore, the process of rapid urbanization (Miranda & Ferreira, 2019), population increase and economic development would enhance flood hazards (Hoque et al., 2019). Consequently, inhabitants, their properties, and the ecosystem would be perpetually at danger in the future (Jhan et al., 2020).

The unique landscape of the Himalayas provides the perfect conditions for flash floods, often triggered by cloudbursts due to the steep and unstable terrain (Dimri et al., 2017). Generally, flash floods are caused by two factors: meteorological conditions and the topographical/surface characteristics of drainage basins. For the evaluation and control of flash floods, geomorphological information has been studied in conjunction with other topographical variables such as land use, soil type and slope (Zzaman et al., 2021). Performing an assessment and mapping of vulnerability could provide a comprehensive understanding of the actual situation and the level of influence that a hazard may have on individuals, capital, assets, and location (Fernandez et al., 2016). The establishment of suitable vulnerability indices, indicators, and their integration are crucial to generating an actual vulnerability scenario for flood risk assessment (Jamshed et al., 2017; Roy & Blaschke, 2015). Lawmakers could make use of the available data and mapping information on several aspects of vulnerability, such as coping capability and socioeconomic vulnerability, in order to create administrative plans that are effective in reducing risks and implementing preventative measures (Lummen & Yamada, 2014; Menoni et al., 2012). Recent trends and advancements in geospatial technology for mapping flood-prone areas have provided the ability to monitor (Haq et al., 2012), analyse the damage (Suriya et al., 2012), and mitigate the threat of flooding (Sanyal & Lu, 2005). Incorporating geospatial technology permits the monitoring of real-time flood control (Bhat et al., 2022; Rautela et al., 2022; Walsh et al., 1998). Machine learning algorithms can serve as a complement to hydrological and hydraulic models (Zhao et al., 2023).Multi-criteria Decision Analysis (MCDA) that utilizes Geographic Information System (GIS) is a assemblage of techniques that are designed to evaluate and merge geographical data and user preferences, with the aim of facilitating decision-making processes (Malczewski, 2006; Mann & Gupta, 2023). As a result of its adjustability in analysing various and contradictory criteria, the MCDA method is generally applicable to flood vulnerability assessments (Feizizadeh & Kienberger, 2017; Giupponi et al., 2013). MCDA presents a potential solution for tackling complex issues such as flood risk, which can be approached from multiple perspectives, including economic, physical, social, and environmental angles. This approach also takes into account the interests of various stakeholders by incorporating weighted criteria. A common approach for multi-criteria decision making is the Analytical Hierarchy Process (AHP), which provides a useful way to deal with challenging challenges (Vijith & Dodge-Wan, 2019).

The Kashmir Valley is located in the northwest part of the Himalayas, nestled between the slopes of two mountain ranges: Zanskar, which rises to approximately 6,000 m above mean sea level to the north-northeast, and Pir Panjal, reaching about 5,000 m above mean sea level to the south-southwest (Bhat et al., 2019a, b). The Kashmir valley ranks among the most flood-prone places in India (Sen, 2010), with records of severe flooding dating back to 883 A.D (Bilham & Bali, 2014; Bilham et al., 2010; Khoihami, 1885; Koul, 1925; Lawrence, 1895; Singh & Kumar, 2013). The valley of Kashmir has seen a number of floods, the most famous of which occurred in 879 A.D., 1841, 1893, 1903, 1929, 1948, 1950, 1957, 1959, 1992, 1996, 2002, 2006, 2010, and 2014 (Bhat et al., 2019a, b). The Kashmir Valley was hit by a catastrophic flood in 2014 that lead to the loss of over 100 lives and an economic loss of INR 1 trillion (World Bank, 2015). The incessant rains that occurred from September 1 to September 7, 2014, resulted in extreme flooding in various parts of South Asia, with approximately 356 mm of rainfall. The 2014 flood event was particularly devastating as it inundated even the inactive parts of the floodplain, encompassing 3.7% of the Kashmir Valley’s total area (Bhatt et al., 2016). In addition, the specific geomorphic orientation, characterized by a bowl-shaped basin, is considered a contributing factor to waterlogging and recurring flooding in the Kashmir (Alam et al., 2018; Meraj et al., 2015). The flood danger is further increased by human settlement growth, alteration of the floodplains, erosion/degradation by perennial rivers, and ensuing alluvial deposition in the water bodies that severely damages wetlands and waterways. In conclusion, while the AHP model represents a robust framework for flood vulnerability assessment, its application within the Sindh watershed reveals significant knowledge gaps. These include challenges related to data scarcity and quality, the need to account for the region’s complex spatial variability, inadequate integration of community and stakeholder insights, limited consideration of climate change impacts, and the necessity for enhanced model validation and adaptation. Building on the knowledge gaps identified in flood vulnerability assessments, particularly in flood-prone areas like the Sindh watershed, this study was undertaken to improve the accuracy and applicability of such assessments in environmentally sensitive and economically vital region (Sofi et al., 2023). In this context, the research focused on assessing the topography, basin features, morphology, and identification of flood-vulnerable zones within the Sindh watershed, a significant tributary of the Jhelum River in Kashmir.

1.1 Study area

Kashmir is located between latitudes 73°55′ and 75°35′ east and longitudes 33°25′ and 34°30′ north. The physical characteristics of Kashmir Valley are significantly impacted by the fact that it was carved out by tectonic activity and has a strong tectonic connection to the north-western Himalayan mountain chain (Nisar, 2012). The Sindh watershed is located between 34° 11′ 17″ and 34° 46′ 30″ North and 74° 57′ 10″ and 75° 63′ 4″ East. Its elevation ranges from 1563 to 5375 m (Fig. 1). The average annual rainfall of the Sindh watershed is 619 mm during the period 1981–2019. The average maximum and minimum temperatures during the year 2017–2019 have been recorded as 26° C and -12 °C respectively (Data collected from AWS located in Sonamarg). The Köppen-Geiger climate classification is Dfb. The average temperature in Sonamarg is -1.7 °C. The Sindh River, a major tributary of the River Jhelum, features a rapid, torrential current in its upper sections. It flows for about 116 km, draining a basin area of 1,683.24 km2 (Sofi et al., 2021). The Sindh River receives the meltwater from Zoji-La (3256 m), Kolahoi (5425 m), Panjtarni snowfields and small tributaries from the Amarnath snowfields (5270 m) also feed the river. The variation in the altitude of the basin ranges from more than 5000 m a.s.l in the extreme eastern region to less than 1600 m a.s.l. near Anchar Lake and Narain Bagh, where the river discharges into the Jhelum River (Sofi et al., 2022). The catchment area mostly consists of alluvium, Cambrio, Ord, Silurian, and Panjal traps. Volcanic and Triassic limestone and alluvium are spread over a vast area compared to other rock formations in the river basin. According to Dada et al. (2013), around 13% of the entire catchment area of the Sindh River features slopes with a range of 45°–90°. The region has a rugged and complex topography with steep slopes and high relief (Bhat et al., 2011).

Fig. 1
figure 1

Location map of the Sindh watershed in Kashmir Himalaya

Full size image

2 Material and methods

2.1 Data collection

The daily precipitation data of the region has been acquired from Indian Meteorological Department gridded data and installed weather station (CR1000XSeries) of University of Kashmir at Sonamarg, Kashmir. The land use land cover, Digital Elevation model data of the study area has been acquired from Earth data. The shape files of soil and geological data sets has been downloaded from FAO and USGS. This study employs a methodology that accomplishes two key objectives:

  1. (a)

    Computation of Causative Flood Parameters: It identifies and analyzes the factors that contribute to flooding within the study region.

  2. (b)

    Flood Susceptibility Zonation: It evaluates and categorizes different zones within the study area based on their susceptibility to flooding.

2.2 Land use land Change (LULC)

Sindh watershed faces a double threat: increasing flood vulnerability and alarming land degradation. Forest cover has significantly declined (14.02% between 2005–2013 and 3.8% further by 2017) (Fig. 2). Additionally, barren and wasteland areas have expanded exponentially (70.13% and 74.94% increases during two consecutive periods). The empirical findings of the study conducted by Saleem et al. 2021 indicate a substantial increase in agricultural and residential zones over the preceding quarter-century. This increase can be ascribed to demographic growth and the escalating requisites for agricultural yield and urban development. Such transformations bear significant ramifications for both the ecosystem and the environment, in addition to the sustenance of the indigenous populace. The change of natural lands, encompassing forests and grasslands, into agricultural or urban sectors can precipitate habitat depletion and deterioration, soil attrition, and an increase in runoff and contamination. Land use and land cover (LULC) changes can significantly influence river systems. During the rainy season, these changes may exacerbate flooding and lead to increased damage to riverbanks. Conversely, during the dry season, they may lead to reduced water availability in the area (Sugianto et al., 2022). A prominent example is the expansion of urban areas, often achieved by converting forests to agricultural land or settlements. This real-world change demonstrably impacts water disposal processes. In developed areas, the increase in impervious surfaces, like concrete, is linked to a higher rate of surface runoff (Zeiger & Hubbart, 2018).This prevents the natural holding capacity of water and changes the subsoil layer or groundwater movement, leading to an increase in flood development and the volume of flood discharge (Maskrey et al., 2022). The land degradation exacerbates flood vulnerability by reducing vegetation, which naturally absorbs rainwater, stabilizes slopes, and maintains healthy river ecosystems. Consequently, sustainable land management practices like reforestation, soil conservation, and controlled grazing are crucial for flood management in the Sindh watershed.

Fig. 2
figure 2

Land use land Change in Sindh watershed from 2015 to 2017

Full size image

2.3 Flood vulnerability assessment

This study utilizes the Analytical Hierarchy Process (AHP) to identify Flood Vulnerability Zones (FVZ) (Fig. 3). The AHP involves several steps: defining the multifaceted problem, establishing a hierarchical model, conducting pairwise comparisons of factors, assigning values to relevant classes, calculating weights for each variable to ensure consistency, and ultimately selecting the optimal solution (Dandapat & Panda, 2017; Danumah et al., 2016; Roy & Blaschke, 2015). The AHP model, despite its advantages in flood vulnerability assessment for the Sindh watershed, has limitations. Subjectivity in weight assignment, difficulty in incorporating dynamic factors, limited representation of complex interactions, and challenges in validating subjectivity can potentially bias the results. To minimize these limitations, this study employed a MCDM framework that considered flood influencing factors for the computation of relative weights (or priorities) to develop a flood hazard model. The model evaluates the relative influence of ten chosen flood initiation factors on overall flood potential by assigning them weighted contributions (Chakraborty & Mukhopadhyay, 2019; Lawal et al., 2012). Subsequent subdivisions of these influencing factors were defined according to their respective rankings (Danumah et al., 2016). Based on the Saaty, 1980 preference descriptor (Table 3), relative ranks of 1 to 5 were assigned for each of the five thematic layers. Factors assigned the highest weightage during the analysis are considered the most significant contributors to flood vulnerability, followed by those with high, moderate, low, and very low values. A weighted linear combination (WLC) method, implemented within a geospatial platform, integrated these factors to create a FVZ map.

Fig. 3
figure 3

Methodology framework for finding out flood vulnerable zones

Full size image

2.4 Flood-inducing factors

The methodology presented in this study focuses on assessing the vulnerability of the Sindh watershed in Jammu and Kashmir region to floods. Western disturbances (WDs), which bring water vapor mainly from the tropical Atlantic Ocean, Mediterranean Sea, Caspian Sea, and Black Sea, are considered one of the primary sources of winter precipitation for the region. The study uses Inverse Distance Weighted (IDW) interpolation to calculate the spatial distribution of rainfall in the region, which was reclassified into 5 classes based on average annual rainfall (Fig. 4). Slope and drainage density are also assessed as important factors in flood vulnerability. The study generates slope and Drainage density maps classified into 5 classes each (Figs. 5 and 6). The LULC map of the basin, downloaded from Earth data, is classified into 7 categories, and grouped into 4 classes (Fig. 7). The surface runoff and infiltration rate of the basin are found to be directly correlated with soil texture, which was classified into 3 categories based on the FAO/UNESCO soil map of India (Fig. 8). The geological map of the study area was created using the World Geologic Maps (USGS) of India and was tailored specifically for the study region. Total 7 types of rock formations are present in the study area, which has been classified into 4 groups (Fig. 9). Finally, the study uses a Digital Elevation Model (DEM) to posturize the elevation of the study area, which was reclassified into 5 categories (Fig. 10). The methodology presented in this study can be useful for assessing the vulnerability of other regions to floods by considering the relevant factors that affect their susceptibility to flooding.

Fig. 4
figure 4

Rainfall map of the Sindh watershed

Full size image
Fig. 5
figure 5

Slope map of the Sindh watershed

Full size image
Fig. 6
figure 6

Drainage Density map of the Sindh watershed

Full size image
Fig. 7
figure 7

Landuse/Landcover map of the Sindh watershed

Full size image
Fig. 8
figure 8

Soil map of the Sindh watershed

Full size image
Fig. 9
figure 9

Geological map of the Sindh watershed

Full size image
Fig. 10
figure 10

Elevation classification of the Sindh watershed

Full size image

2.5 Analytical Hierarchy Process (AHP)

The Analytic Hierarchy Process (AHP), developed by Thomas L. Saaty in the 1970s, is a multi-criteria decision-making tool. It facilitates structuring complex problems by allowing different stakeholders to weigh in on various criteria. Through pairwise comparisons, the AHP helps prioritize these criteria and arrive at the most suitable course of action (Dandapat & Panda, 2017). Over time, the integration of AHP with GIS has become increasingly popular for flood mapping (Moazzam et al., 2022). Despite inherent uncertainties, the Analytical Hierarchy Process (AHP) remains a widely accepted method for establishing criteria weights through pairwise comparisons (Rao et al., 1991). Its success in various sectors highlights its strengths (Althuwaynee et al., 2014; Feloni et al., 2020; Subbarayan & Sivaranjani, 2020). Researchers have recognized AHP’s strong theoretical foundation when combined with weighted linear combination in Geographic Information Systems (GIS) (Ghosh et al., 2022; Hasanuzzaman et al., 2022). AHP offers advantages like parameter normalization (Chandra et al., 2023), facilitating the expression of decision risk (Fernandez & Lutz, 2010), and demonstrating versatility in site-specific vulnerability assessment (Hughes, 1986; Mishra & Sinha, 2020; Saaty, 1980).To construct the comparison matrix, pairwise comparisons of variables from the thematic maps were performed for five vulnerability zones (Table 1). A significant contribution of the AHP methodology is its ability to assign different weights for the seven elements and rank them based on their relative importance using a technique that ranges from 1 to 9 (Table 2) depending on their relevance, as proposed by Saaty (1980). AHP weights are determined in a methodical manner by means of normalization and pairwise comparisons. Decision-makers first weigh each consideration against all others to ascertain its relative preference or importance, which is represented by numerical numbers. To ensure that all of the comparisons are on the same scale and comparable to one another, these pairwise comparison values are then normalized by dividing each value by the total of its corresponding column. The associated significance score, which is usually the total of the normalized values in each row, is then multiplied by each normalized value. The weighted total for each component is obtained by adding the resultant products. At last, the weighted totals are normalized by dividing each by the total of all the weighted totals. This results in the ultimate weights, which indicate the proportional significance of every element in the process of making decisions. As the size of the flood rises, so does the significance of the flood-influencing factors. Each matrix element was divided by the total of its columns to determine the priority scales. The row average was used to determine the priority vector. As suggested by Griot (2007), the AHP findings are reported as a score of 1 (Table 3). The following formula is used to estimate each factor’s weight and relative relevance in the hierarchy, and then vulnerability is evaluated (Eq. 1).

Table 1 Pair-wise comparison of 7 × 7 decision matrix
Full size table
Table 2 Fundamental AHP judgment scale with integers 1 to 9 and their definition
Full size table
Table 3 Determination of the weight of each criterion
Full size table

2.6 Consistency check

The constructed pairwise comparison matrix and its weightings were validated using the Consistency Ratio (CR), ensuring a value below the recommended threshold of 0.1 (Saaty, 1980). The calculated consistency index of 0.083 indicates an acceptable level of consistency within the decision-making process (Rautela et al., 2023).

$$CR=\frac{CI}{RI}$$
(1)

Where, CR is the consistency ratio, CI is the consistency index and RI is the random index.

3 Results

3.1 Assessment of Flood Vulnerability Zones (FVZ)

In the present study, the Analytic Hierarchy Process (AHP), a reliable and flexible multi-criteria mathematical approach has been employed to develop the FVZ map for the Sindh watershed. To construct the decision matrix, a 7 × 7 matrix was created, where all diagonal elements are equal to one (Kazakis et al., 2015). The induced variables, namely slope, rainfall, drainage density, land use and land cover (LULC), geology, soil, and elevation, were included in the decision matrix. The derived consistency ratio (CR) value of 0.089 was obtained, which was found to be lower than the acceptable threshold of 0.1, indicating acceptable consistency of the model. The Consistency Ratio (CR) is a crucial aspect of ensuring reliable decision-making in the AHP. Thus, a CR value less than 0.1 generally indicates an acceptable level of consistency within the pairwise comparison matrix (Mokhtari et al., 2023; Rautela et al., 2023; Saaty, 1980). In the Analytic Hierarchy Process (AHP), a Consistency Ratio (CR) less than or equal to 0.1 indicates acceptable consistency within the pairwise comparison matrix. However, a CR exceeding 0.1 suggests potential inconsistencies in the judgments. If this occurs, revisiting and refining the pairwise comparisons might be necessary to ensure reliable decision-making (Saaty, 1980). The AHP analysis results indicate that the Consistency Ratio (CR) values obtained for the different flood factors, namely slope (0.386), rainfall (0.192), drainage density (0.129), land use and land cover (LULC) (0.096), geology (0.077), soil (0.064), and digital elevation model (DEM) (0.055), are within acceptable limits (as presented in Table 5). These weights can be used in conjunction with the Weighted Linear Combination (WLC) approach to prepare the Flood Vulnerability Zone (FVZ) map. Consequently, to establish priorities between the various parameters under consideration, a normalized pair-wise comparison was conducted. This involved dividing each element of the 7 × 7 decision matrix by the total sum of the rankings of priority among the parameters (Saaty, 1980) (see Table 4). The normalization procedure facilitated the scaling of the seven parameters within a range of 0 to 1.

Table 4 Rankings of flood inducing parameters of different sub-classes using AHP comparison matrix
Full size table

The vulnerability zones were evaluated by means of FVI (Flood Vulnerability Index) values that fall within the range of 1 to 3, specifically: 1 (indicating low vulnerability), 2 (indicating moderate vulnerability), and 3 (indicating high vulnerability). Table 5 presents the evaluation outcomes for zones that are susceptible to flooding. The corresponding areas and percentages have been computed using the AHP methodology to analyze the flood-vulnerable zones. The valleys that are completely shaped by rivers are particularly prone to flooding, particularly in the presence of steep slopes, as well as during rainfall events. The land use and land cover (LULC) area is also a significant factor that influences the level of vulnerability to flooding. The analysis reveals that approximately about 128.51 km2, is characterized as highly vulnerable to flooding. The higher vulnerability is mainly due to the presence of steep slopes, barren land, water bodies, and built-up land, as well as the occurrence of extreme rainfall events. To validate the present study, data from last flood events have been collected for the Sindh watershed (Table 6). The result of the present study is well correlated with the past floods that have been occurred in the valley (Fig. 11). Snow Avalanches and Flash Floods are the two major water related disasters that have been reported in the past. These events can be extremely dangerous and can cause significant damage to critical infrastructure in the surrounding areas.

Table 5 Area-wise flood vulnerability status of the Sindh watershed
Full size table
Table 6 Flood vulnerable villages in the Gund and Lar Tehsil located in the Sindh watershed (https://ganderbal.nic.in/disaster-management/)
Full size table
Fig. 11
figure 11

Flood Vulnerability Zonation of Sindh watershed and villages that are affected with recent floods

Full size image

The topographic situation has tremendous influence on the flow regimes of the watersheds to make the area more vulnerable to flooding (Meraj et al., 2015). In contrast, a relatively smaller proportion of the watershed area of 2.96 km2 and 620.12 km2, is categorized as very low and low vulnerable zones respectively. The less vulnerable zone is predominantly composed of high-altitude areas. The analysis also indicates that the moderately vulnerable zone covers a substantial portion of the land area. This area is mainly characterized by mild slopes and adequate vegetation cover. Due to the dense vegetation and forest cover, extensive plains that act as a natural barrier to settlements, and minimal development activities, the probability of flooding is relatively lower in these regions and tends to reduce the extent of flood damage (Bhattacharjee & Behera, 2018).

4 Discussion

While the focus of this research lies on the Sindh watershed, it draws broader implications for Indian Himalayan region (IHR). Historically, urbanization, decreasing forest cover, increasing grazing and barren land in these areas has been driven by colonial expansion, economic growth, and population rise (Kuniyal et al., 2021). Consequently, most of the Himalayan cities now face challenges stemming from unchecked growth and encroachment in flood-prone zones (Rautela et al., 2023). Also, historical flood events due to prolonged rainfall, snow or rock avalanches, cloud burst proves that these encroached regions are highly susceptible to floods (Allen et al., 2016; Kumar & Acharya, 2016; Sattar et al., 2019; Shugar et al., 2021; Sinha et al., 2022; Ziegler et al., 2014). With climate change impacts, the vulnerability to extreme hydro-climatic events are becoming more apparent. This necessitates regular revaluation of flood risk in these areas (Cea & Costabile, 2022). However, the urban development plans of many Himalayan cities, acknowledge the risk of flooding in certain built-up locations. However, the presence of existing structures in these vulnerable areas poses a complex dilemma for government agencies (Ouma & Tateishi, 2014). Balancing the rights of landowners who have already invested in these areas with the need for flood mitigation measures presents a significant challenge. Decisions regarding flood mitigation investments can also be politically contentious due to their potentially high socio-economic costs. Thus, addressing flood risk in highly susceptible areas requires careful consideration of both technical and socio-political factors (GWIR, 2017).

As discussed above numerous cities across the riverbanks in the IHR are grappling with the compounding challenges posed by prolonged rainfall, cloud burst, a trend expected to intensify in the coming decades. To address this, the present study employs an Analytic Hierarchy Process (AHP) to analyze flood-influencing parameters, offering valuable insights for vulnerability assessments in a river basin scale (Rautela et al., 2023, Mokhtari et al., 2023). The change of LULC leads to increase surface runoff and suspended sediments in the river channel and affect the ecological functionality of the pastures in the region (Romshoo & Fayaz, 2019; Rautela et al., 2024). The alteration of the hydrology, morphology, and ecology in the watershed exacerbate the vulnerability to flooding. The analysis of flood locations across various parameters reveals intriguing patterns, with areas characterized by low-lying slopes, urban development, inadequate drainage infrastructure, and proximity to river channel exhibiting heightened flood susceptibility (Kumar et al., 2023). Addressing these challenges requires a multifaceted approach that integrates ecological conservation, sustainable urban planning, and improved drainage infrastructure to mitigate the growing threat of floods in a river basin scale.

The results indicated that a significant majority of flood-prone locations are situated in areas of moderate to high vulnerability. While these findings provide valuable insights into flood risk, further investigation is necessary to fully grasp their implications and establish appropriate boundaries. The direct exposure of human populations to flood risk is exacerbated by the destruction of natural barriers to accommodate rapid urban expansion driven by population growth (Hossain et al., 2017). This underscores the urgent need to reconsider existing development patterns that prioritize urban expansion over flood risk mitigation. The challenges faced by most Himalayan cities in planning and flood management are not unique; similar issues are emerging in cities worldwide. Whether in rapidly expanding cities in the Himalayan valley like Srinagar (Uttarakhand), Chamoli (Uttarakhand), Srinagar (Jammu and Kashmir), Mandi (Himachal Pradesh), Kullu (Himachal Pradesh) etc. the impact of urbanization on flood vulnerability is becoming increasingly apparent (Prakasam & Kanwar, 2021). In this context, the methodological framework proposed in this study, particularly the AHP technique, offers a valuable tool for identifying and addressing specific challenges associated with flood vulnerability by considering multiple parameters simultaneously.

5 Recommendations

Based on the results, the current study proposes the following recommendations for future research areas and flood management in the Sindh watershed.

  1. 1.

    Incorporation of Climate Change Scenarios: The current study used the AHP model to assess flood vulnerability considering various parameters like slope, rainfall, drainage density, land use and land cover (LULC), geology, soil, and elevation. However, the impact of climate change in the region on these parameters could be significant. Thus, future research could incorporate climate change scenarios to understand how these parameters might change in the future and how that would affect flood vulnerability.

  2. 2.

    Study of Human Activities and Infrastructure: In the past snow avalanches and flash floods have caused significant damage to infrastructure in the region. Future studies could focus on the impact of human activities and infrastructure on flood vulnerability in the Sindh watershed. This could include studying the impact of deforestation, glacier reduction and changes in land use patterns.

  3. 3.

    Development of Flood Management Strategies: The current study identifies areas of high, moderate, and low vulnerability. Building on this, future research could focus on developing specific flood management strategies for each of these areas. The use of GIS technology is recommended for the comprehensive analysis of spatial relationships between flood vulnerability factors and the identified vulnerable zones. This approach will facilitate a more nuanced understanding of the spatial distribution of flood risks and aid in the development of targeted mitigation strategies. Additionally, having access to discharge information is critical for the verification and adjustment of flood prediction models. Thus, it is essential to set up a comprehensive network of hydrometeorological and river flow monitoring stations across the entire watershed to enhance the accuracy of flood forecasting.

6 Conclusion

This study employed the Multi-Criteria Decision Making (MCDM) Analytic Hierarchy Process (AHP) model to effectively assess flood vulnerability and delineate flood-prone zones within the Sindh watershed. By integrating key factors like slope, rainfall, drainage density, land use/land cover, geology, soil, and elevation, a comprehensive Flood Vulnerability Zone (FVZ) map was generated. The analysis identified the entire stretch of riverbanks as highly vulnerable, while most of the remaining watershed area exhibited moderate susceptibility. An area of approximately 128.51 km2 was categorized as vulnerable, primarily attributed to factors like steep slopes, water bodies, and built-up areas, particularly during extreme rainfall events. The findings resonate with past flood events in the region, highlighting the accuracy of the FVZ map. Considering the historical damage inflicted by snow avalanches and flash floods on infrastructure, proactive flood management strategies are crucial. Tailoring flood management strategies to specific vulnerability zones, coupled with flood-prone area mapping and enforcing construction regulations, are essential steps towards enhancing resilience to flooding within the Sindh watershed. In essence, this study offers valuable insights into flood vulnerability assessments and lays the groundwork for informed decision-making in flood risk management and disaster preparedness efforts.