The objective of this study was to evaluate the sensitivity of flash floods to future climate change in the Amman–Zarqa Basin, Jordan. Historical daily rainfall and temperature data from 1970 to 2018 were collected, along with projected daily data derived from general circulation models (GCMs) forecast spanning 2019–2060. The methodology involved analyzing historical and model forecast data, conducting trend analysis, mapping changes in land use, estimating runoff volume, selecting indicators, assigning their weights through the analytical hierarchy process, and generating vulnerability maps. Analysis of precipitation trends revealed a 14.61% decrease in total annual rainfall over the past 48 years; however, future projections indicate a 5.26% increase. Downstream sub-catchments in the arid portion are projected to receive higher rainfall, while upstream sub-catchments are expected to experience a substantial decline, resulting in an overall reduction in runoff. Moreover, our findings demonstrate a rising trend in mean temperature, which is expected to persist. Remote sensing data indicate a 14.76% expansion of urban areas, indicative of rapid population growth. Although no highly vulnerable sub-catchments were identified, downstream sub-catchments 8 and 9 exhibited moderate vulnerability to flash floods, which can be attributed to the increase in rainfall and insufficient stormwater infrastructure.

  • Assessment of the hydrologic vulnerability using suitable indicators for urban watersheds.

  • Evaluation of trends in precipitation based on non-parametric statistical tests.

  • Dynamical downscaled GCM data are used to predict future temperature and precipitation.

  • Remote sensing data and GIS applications are used effectively in the hydrologic vulnerability assessment.

  • Frequency analysis for the precipitation used in predicting flash floods characteristics.

Based on the IPCC's Sixth Assessment Report (AR6) released in 2021, it is expected that unless there is a substantial reduction in carbon dioxide (CO2) and other greenhouse gas (GHG) emissions in the coming decades, the global warming targets of 1.5 and 2 °C set in the Paris Agreement will be surpassed within this century (IPCC 2021). This increase in the average global temperature will lead to various climate changes, resulting in a range of impacts. Some regions may experience an increase in intense rainfall storms, while others may face more frequent and severe droughts (IPCC 2007).

Arid areas are characterized by limited rainfall and water resources. Higher temperatures accelerate evaporation, causing a decrease in surface moisture and amplifying the drying of the atmosphere. Climate change and population growth will increase the water demand in arid regions, worsening the issue of water scarcity (Lian et al. 2021). In contrast, a study conducted by Tabari (2020) examined the connection between variations in extreme precipitation and flood intensities using projections from global climate models (GCMs). The findings of this study revealed a significant rise in extreme precipitation events and a potential decrease in overall precipitation in arid regions. Accordingly, flood intensity in semi-arid regions is expected to increase by approximately 3.12% by the end of 21st century. These findings are in line with the conclusions drawn in the IPCC's AR6, which indicated that extreme precipitation events will likely become more frequent and intense over most mid-latitude land masses in a warmer world. Specifically, the report indicates that the current 1-in-20-year annual maximum 24-h precipitation rate is projected to become a 1-in-5 to 1-in-15-year event by the end of the 21st century (Seneviratne et al. 2021). Thus, arid areas are expected to face an elevated risk of flash floods under future climate change conditions. It is important to consider the effects of climate change on hydrologic systems while also considering human interventions in watersheds, such as reservoir construction and changes in land use. Urbanization can further exacerbate the effects of climate change, increase the risks associated with flooding, and degrade water quality (Bates et al. 2008; Chung et al. 2011).

Assessing the vulnerability of hydrologic systems to climate change involves considering both climatic and non-climatic factors. The IPCC defines vulnerability as the susceptibility of a system to cope with and adapt to the adverse effects of climate change, encompassing variability and extremes. The level of vulnerability depends on the system's sensitivity, adaptive capacity, and the nature, extent, and pace of climate variations it experiences (IPCC 2007). Numerous studies explored climate change vulnerability to identify regions highly prone to extreme weather events, including floods. These studies utilized indicators to assess hydrologic vulnerability across three main components. Exposure is typically assessed by analyzing the risks associated with floods and droughts, as well as changes in precipitation, temperatures, and evaporation. Sensitivity is evaluated by examining system characteristics such as population density, forest cover, and reliance on rain-fed agriculture. On the other hand, adaptive capacity is determined by assessing a system's ability to respond to climate change, considering factors such as sanitation services, electricity supply, water accessibility, and poverty levels within the population (Füssel & Klein 2006; Yusuf & Francisco 2009; Rivas et al. 2011; Boori & Voženílek 2014; Binita et al. 2015). Previous research has focused primarily on examining vulnerability to climate change on larger scales, such as large watersheds, countries governorates, and counties. However, there is an urgent need for more localized case studies, particularly in urban regions susceptible to flooding, to assess the social and environmental response to climate change. This need is particularly critical in low-income arid regions with rainfall variability and limited stormwater infrastructure such as the case of Jordan (Mirzabaev et al. 2022).

Jordan, as part of the eastern Mediterranean region, is predominantly arid to semi-arid, experiencing significant rainfall variability (Tarawneh & Kadıoğlu 2003). According to recent climate change projections by Abdulla (2020), Jordan is predicted to witness a temperature increase of 2.5–5 °C and a reduction in annual precipitation of 10–37% by the end of the century. The Amman–Zarqa Basin (AZB), located in the north central part of Jordan, encompasses the two largest cities: Amman and Zarqa (Al Kuisi et al. 2014). Over the past few decades, urbanization in the AZB has been driven by increased migration rates, an influx of refugees, and population growth. This urban expansion resulted in the conversion of over 70% of agricultural land into urban areas (Gharaibeh et al. 2019). The changes in land use, including the conversion of streams into roadways to accommodate traffic, have led to flash floods triggered by intense rainfall storms with high precipitation rates within a short period. Furthermore, the AZB presents an ideal study area for assessing flash flood vulnerability due to its high population density, rainfall variability, topography, and urban coverage. Various approaches, such as morphometric analysis, hydrological modeling, and unit hydrograph, have been employed to assess the risk of flash floods in Jordan (Alhasanat 2014; Farhan & Ayed 2017; Al Azzam & Al Kuisi 2021).

The AZB heavily relies on groundwater as a primary water source. However, the basin's aquifer has faced significant depletion and deterioration in water quality, reaching critical levels due to excessive groundwater extraction (MWI 2017). Previous studies on the impact of climate change on streamflow and rainfall patterns in the AZB have solely focused on the emission scenarios outlined in the Special Report on Emissions Scenarios and the representative concentration pathways (RCPs) (Hammouri et al. 2015; Al-Hasani et al. 2023). However, a comprehensive vulnerability assessment that considers the climatic, social, and physical characteristics of the hydrologic system and following the newest climate change scenarios has not been conducted in Jordan.

Based on the preceding discussion, this study aims to provide a localized and comprehensive assessment of rainfall patterns and related runoff in the southwestern part of the AZB, which is the most urbanized area within the basin, from 1970 to 2018, by considering changes in land-use patterns. For this, data from a regional climate model (RCM) based on the Shared Socioeconomic Pathway 5 (SSP5) scenario were used to examine future rainfall and temperature patterns for the period 2019–2060 and simulated the projected runoff. Additionally, a spatial vulnerability assessment was conducted to measure the risk of flash floods in an urban watershed, using exposure, sensitivity, and adaptive capacity indicators. This study aims to answer three questions: What are the trends in rainfall? What is the likelihood of flash flood occurrences based on climate change projections? And where are the regions, most vulnerable to climate change impacts? As Jordan ranks among the top 10 water-scarce countries globally (UNFCCC 2014), this study is essential for projecting likely water scenarios under a changing climate, enabling policymakers to assess and implement effective environmental protection and adaptation measures. Moreover, the methodologies employed in this study can be transferable to other arid regions facing similar climate change challenges. This study was conducted using the Watershed Modeling System (WMS 11.0) and Geographic Information System (ArcGIS 10.8.2) software packages.

The AZB encompasses an extensive catchment area of approximately 3,900 km2. Within Jordan, the AZB is the most densely populated region, accommodating around 65% of the country's population and over 85% of its industrial activities. The focus of this study centers on the southwestern portion of the AZB, covering an area of 660 km2 (Figure 1). This specific area includes significant locations within Amman, Ruseifa, and Zarqa cities, where more than 6 million people reside (DOS 2020). The study area experiences a Mediterranean climate characterized by hot, dry summers and cool, wet winters. The average mean temperature hovers around 18.4 °C, while the mean annual precipitation ranges from 150 to 400 mm (Hyarat et al. 2022).
Figure 1

Study area.

The study area was selected due to its history of experiencing multiple flash flood incidents. For instance, on February 28, 2019, the Amman downtown area witnessed a flash flood triggered by a heavy rainstorm that lasted for two days, with an estimated intensity of 8–12 mm/h. The storm generated a peak discharge of approximately 658 m3/s, and the flash flood caused damage to vehicles and properties in the flooded area (Engicon 2019) (Figure 2).
Figure 2

Amman downtown flood on February 28, 2019 shows how the Roman theater and streets were covered by the floodwater (Source: Maraya and al Madenah news).

Figure 2

Amman downtown flood on February 28, 2019 shows how the Roman theater and streets were covered by the floodwater (Source: Maraya and al Madenah news).

Close modal
The population density in the study area ranges from 20 to 39,000 persons/km2 (Figure 3(a)) (DOS 2020). The study area catchment is characterized by moderate topographic relief, with elevation ranging from 471 to 1,100 m above sea level (Figure 3(b)). Most of the study area has a moderate slope. The watershed drainage network collects water from the southwestern highlands and the eastern region of Amman, channeling it into the Zarqa River. In addition, the study area is composed of rocks of Cretaceous age divided into the main group of Balqa (B1-B5) and Ajlun (A1-A7) outcrops. The dominant lithologies are mostly limestone, marly limestone, dolomite, chert, phosphate, and chalk, and the main structural feature is the Amman-Halabat structure (Abu Qudaira 2004).
Figure 3

Study area characteristics: (a) population density (b) topographic map, and (c) available rainfall gauge stations.

Figure 3

Study area characteristics: (a) population density (b) topographic map, and (c) available rainfall gauge stations.

Close modal

Daily rainfall for the period of 1970–2018 was obtained for 11 rainfall gauge stations from the Ministry of Water and Irrigation (MWI) (Figure 3(c)). Mean, maximum, and minimum temperatures were acquired for Amman Airport station (AL0019) and Wadi Es-Sir (NRA) station (AL0057) for the period (1988–2016). Data related to population per locality were obtained from the Department of Statistics. The stormwater drainage network was obtained from Greater Amman Municipality and Zarqa Municipality. Table 1 summarizes the sources and other information for the datasets used in this study.

Table 1

Summary of the database used for this study

Data typeData sourceLocationPeriod
Daily rainfall Ministry of Water and Irrigation 11 stations in the study area 1970–2018 
Daily temperature Ministry of Water and Irrigation 2 stations in the study area 1988–2016 
GCM data RICCAR Jordan 2020–2060 
Satellite images Landsat 7 ETM + , Landsat 4–5 and Sentinel-2 Missions Study area 2001, 2011, 2021 
DEM Shuttle Radar Topography Mission (SRTM) Study area 2021 
Soil texture Hunting Technical Services and Soil Survey and Land Research Centre Study area 2021 
Stormwater drainage network Greater Amman Municipality and Zarqa Municipality Amman and Zarqa 2021 
Population per locality Jordanian Department of Statistics Study area 2021 
Data typeData sourceLocationPeriod
Daily rainfall Ministry of Water and Irrigation 11 stations in the study area 1970–2018 
Daily temperature Ministry of Water and Irrigation 2 stations in the study area 1988–2016 
GCM data RICCAR Jordan 2020–2060 
Satellite images Landsat 7 ETM + , Landsat 4–5 and Sentinel-2 Missions Study area 2001, 2011, 2021 
DEM Shuttle Radar Topography Mission (SRTM) Study area 2021 
Soil texture Hunting Technical Services and Soil Survey and Land Research Centre Study area 2021 
Stormwater drainage network Greater Amman Municipality and Zarqa Municipality Amman and Zarqa 2021 
Population per locality Jordanian Department of Statistics Study area 2021 

To evaluate the hydrologic vulnerability, several major processes such as climatic data preprocessing, watershed delineation, rainfall frequency analysis, trend analysis, land-use mapping, hydrological modeling, and vulnerability assessment were employed (Figure 4).
Figure 4

Methodology flowchart.

Figure 4

Methodology flowchart.

Close modal

Watershed and morphometric analysis

The catchment delineation and stream network of the study area were generated with the aid of WMS 11.0 software and ArcMap 10.8.2, using the Digital Elevation Model (DEM) of 12.5*12.5 m resolution (Table 1), a 3D computer graphics representation of the earth's surface elevation. The study area was divided into nine sub-catchments, with 2–6 representing wet to moderate regions with a total rainfall exceeding 200 mm/year. Conversely, sub-catchments 1, 7, 8, and 9 are classified as dry areas, receiving less than 200 mm/year of rainfall (Figure 5). The upstream sub-catchments are numbered 3, 4, 5, and 6, while the downstream sub-catchments are the rest.
Figure 5

Drainage network and sub-catchments in the study area.

Figure 5

Drainage network and sub-catchments in the study area.

Close modal

Since morphometric parameters offer valuable insight into the geometry and hydrological characteristics of a watershed, morphometric analysis was carried out to provide information about the basin's slope, topography, soil condition, runoff characteristics, and surface water potential. In this study, six basic morphometric parameters, seven derived parameters, and three shape parameters were calculated for the study area watershed using equations outlined by Farhan & Ayed (2017) with the aid of ArcMap 10.8.2 and WMS 11.0 software (Table 2).

Table 2

Morphometric analysis

Basic parametersDerived parametersShape parameters
Basin area (A), km2 659.7 Sinuosity (SI) 1.6 Elongation ratio (Re0.84 
Bain perimeter (P), km 177.7 Basin shape index (Ish0.56 Circularity ratio (Rc0.26 
Basin length (Lb), km 34.3 Drainage density (Dd) km/km2 1.38 Form factor ratio (Rf0.56 
Number of streams Nu 600 Stream frequency (Fs0.91   
Stream length (Lu), km 910 Drainage texture (Dt3.38   
Mean stream length (Lsm), km 1.56 Basin relief (Bh) 626   
  Relief ratio (Rr31.94   
Basic parametersDerived parametersShape parameters
Basin area (A), km2 659.7 Sinuosity (SI) 1.6 Elongation ratio (Re0.84 
Bain perimeter (P), km 177.7 Basin shape index (Ish0.56 Circularity ratio (Rc0.26 
Basin length (Lb), km 34.3 Drainage density (Dd) km/km2 1.38 Form factor ratio (Rf0.56 
Number of streams Nu 600 Stream frequency (Fs0.91   
Stream length (Lu), km 910 Drainage texture (Dt3.38   
Mean stream length (Lsm), km 1.56 Basin relief (Bh) 626   
  Relief ratio (Rr31.94   

Based on the morphometric parameter definitions provided by Sukristiyanti et al. (2018), the study area showed certain characteristics. It has a drainage density (Dd) of 1.38, indicating poor drainage, slow hydrologic response, and high susceptibility to flooding. The drainage texture (Dt) is 3.38, classifying it as coarse and in the initial stage of erosion. The form factor ratio (Rf) ranges from 0 for a long, elongated basin shape to 1 for a perfectly circular analog. The area has a form factor (Rf) of 0.78, suggesting a circular shape with high peak flows, strong floods, higher velocities, and greater erosion and transport capacity.

Rainfall data analysis

The inverse distance approach was used to calculate missing rainfall data (Musy et al. 2014). The Mann–Kendall test has been employed extensively in research on hydroclimatic series' spatial variation and temporal trends. This test compares the alternative hypothesis (H1), which states that there is a trend, to the null hypothesis (H0), which contends that there is no trend (the data are independent and randomly arranged). In this context, Sen's slope, a non-parametric measure of trend slope, represents the strength of a diminishing or growing trend by estimating the slope of the trend using a linear model (Da Silva et al. 2015). In this study, the Mann–Kendall test and Sen's slope tests were applied to examine the total annual rainfall and temperature trends for the current and upcoming periods. The null hypothesis was tested at the 95% confidence level (α = 0.05).

The intensity–duration–frequency (IDF) curve illustrates the relationship between rainfall intensity, rainfall duration, and return period. IDF curves are obtained through frequency analysis of rainfall observations (Mahdi & Mohamedmeki 2020). Frequency analysis is used to predict rainfall events that may occur in the future from available historical data. In this study, IDF curves were built for all stations to simulate the runoff and peak discharge resulting from different rainfall storms. Gumbel distribution was the selected statistical method to perform the frequency analysis on the maximum rainfall data due to its effectiveness in modeling maximum values and focusing on extreme events, such as peak rainfalls. Rainfall intensity was calculated for different durations (5, 10, 20, and 30 min and 1, 2, 3, 6, 12, and 24 h) and different return periods (2, 5, 10, 25, 50, 100, and 1,000 years).

Land-use mapping

To detect changes in land use and estimate runoff in the study area, three land-use maps were generated using satellite images from 2001, 2011, and 2021. Due to the limited availability of satellite data before 2,000, the selection of satellite missions was constrained to those with data at the desired time. The ArcMap 10.8.2 image classification tool was used for supervised classification, employing the maximum likelihood algorithm. This process involves comparing pixels in remotely sensed data to create classes that align with the user's specified categories (Perumal & Bhaskaran 2010). The classification accuracy was assessed by utilizing reference land-use data from the Jordan – Land Cover Atlas (Franceschini et al. 2019). The accuracy of the classification was ∼ 77%. To address the different resolutions of the satellite images, the resample tool in ArcGIS software was used for harmonization.

To estimate runoff under future climate change, a predicted land-use pattern is essential. For this purpose, the MOLUSCE plug-in in QGIS 2.16 software was used to predict the future land use for the year 2041. The MOLUSCE plug-in was designed to model and simulate land-use/cover changes based on the Monte Carlo Cellular-Automata modeling approach. In this approach, the system is represented as a grid of cells, and each cell has a state that changes over time according to specific rules. These rules are defined based on a combination of factors and considerations, including neighboring cells' influence, socioeconomic factors, and environmental factors. These rules decide how each cell's state is updated based on its neighboring cells' states, which can include randomness. By repeatedly applying these rules and updating the cell states, the model can simulate how the system behaves over time (Li & Li 2015). The model used land-use maps from 2001 and 2021, as well as DEM and slope raster, as input data. MOLUSCE provides four methods to model potential land-use transition: artificial neural networks (ANNs), logistic regression (LR), multi-criteria evaluation, and weights of evidence. In this study, ANN was chosen as it provided the highest correctness percentage among the other methods. To validate the accuracy of the modeling process, land-use maps for 2001 and 2011 were loaded into the model to simulate the land-use map for 2021. The simulated land-use map was then compared to the actual land-use map of 2021, generated from image classification. The simulation accuracy was 68%.

Runoff analysis

Runoff volumes and peak flows were estimated due to the lack of direct measurement. Runoff estimations need sufficient information about the drainage setting in the area and other surface-related characteristics. The U.S. Soil Conservation Service (SCS) method, also known as the curve number (CN) method, was used in this study to calculate storm runoff volume, peak rate of discharge, and hydrographs. This method employs the following relationship between runoff volume (Q) and rainfall depth (P):
formula
(1)
where Q is the runoff depth in (mm), P is the precipitation depth in (mm), and Ia is the initial abstraction (mm). The initial abstraction (Ia) represents the amount of rainfall water lost before the runoff commences. Therefore, if the amount of rainfall is less than the initial abstraction, then there will be no runoff. It can be computed by
formula
(2)
where S is the total losses of rainfall that are determined by soil type and cover and related to the CN of the land cover. CN is a unitless number that ranges from 0 to 100 and depends on the cover type and hydrological soil group (HSG). The higher the CN, the higher the percentage of the impervious surface of the land cover. SCS divides soil into four groups: A, B, C, and D. HSG categorization is mostly based on the soil's permeability and infiltration capacity. Soil textures of the study area (HTS & SSLRC 1993) were reclassified into HSG. So, to calculate the S, the following equation was used:
formula
(3)

The CN of the different cover types in the study area was determined based on the SCS urban hydrology for the small watershed (TR-55) (SCS 1986).

Rainfall amounts obtained from IDF curves, HSG, and land cover CN were then entered into the HEC-1 simulation model in WMS 11.0 to calculate the runoff volume and peak discharge for each sub-catchment in different duration and return periods. To estimate the runoff amount under future climate change, we utilized rainfall data derived from IDF curves based on GCM data, in addition to land characteristics extracted from the projected land-use map of 2041.

The Nash–Sutcliffe efficiency (NSE) index is a tool used to assess the performance of hydrologic models in simulating runoff (Nash & Sutcliffe 1970). NSE measures the difference between the variance of the measured data (the ‘information’) and that of the residual (the ‘noise’). NSE values range from one to negative infinity, with a value of 1 indicating a perfect model, a value between 0.0 and 1.0 indicating good performance, and a value below 0.0 indicating that the observed mean value is a better predictor than the simulated value. We acquired actual runoff measurements for the Jarash Bridge (AL0060) station from 1997 to 2017 from the Ministry of Water and Irrigation. We compared the actual runoff measurements and the simulated runoff resulting from the HEC-1 model for this station (Figure 6). The resulting NSE value was 0.712, indicating good model performance.
Figure 6

Observed versus simulated runoff was used to validate the HEC-1 model.

Figure 6

Observed versus simulated runoff was used to validate the HEC-1 model.

Close modal

Future climate data

The GCM provides reliable global climate information for historical and future periods. The RCM is applied with higher spatial resolution over a limited area and is driven by GCMs. In this study, climate change data were obtained from the Regional Initiative for the Assessment of Climate Change Impacts on Water Resources and Socio-Economic Vulnerability in the Arab Region (RICCAR). RICCAR established the Mashreq domain where they downscaled climatic data from the Mediterranean (MENA) CORDEX domain. Mashreq domain used six GCMs and one RCM (Table 3) based on a 0.1-degree grid (∼10 km) spatial resolution. The collected data were the predicted daily precipitation and mean temperature for the period of 1970–2060.

Table 3

Obtained climate models

Institution
Global climate model  
CMCC-CM2-SR5 Euro-Mediterranean Centre on Climate Change 
EC-Earth3-Veg European Consortium 
NorESM2-MM Norwegian Meteorological Institute 
MRI-ESM2-0 Meteorological Research Institute, Japan 
MPI-ESM1-2-LR Max Planck Institute for Meteorology, Germany 
CNRM-ESM2-1 National Center for Meteorological Research, France 
Regional climate model 
HCLIM-ALADIN Swedish Meteorological and Hydrological Institute (SMHI) 
Institution
Global climate model  
CMCC-CM2-SR5 Euro-Mediterranean Centre on Climate Change 
EC-Earth3-Veg European Consortium 
NorESM2-MM Norwegian Meteorological Institute 
MRI-ESM2-0 Meteorological Research Institute, Japan 
MPI-ESM1-2-LR Max Planck Institute for Meteorology, Germany 
CNRM-ESM2-1 National Center for Meteorological Research, France 
Regional climate model 
HCLIM-ALADIN Swedish Meteorological and Hydrological Institute (SMHI) 

The data obtained were then subjected to bias correction and represent the latest SSP5 scenario, a part of the Coupled Model Intercomparison Project Phase 6. The selection of the SSP scenarios is crucial for the vulnerability assessment, as it considers non-climatic factors and necessitates the evaluation of both climate change trends and socioeconomic trends. The SSP scenarios provide a framework for exploring how global society, demographics, and economy are likely to evolve over the next century, and how societal actions will impact GHG emissions. The SSP5 scenario represents a fossil-fueled development pathway, characterized by high challenges to mitigation and low challenges to adaptation. In this scenario, radiative forcing reaches levels comparable to those in the RCP 8.5 scenarios (8.5–8.7 W/m2) (8.5–8.7 W/m2) (Riahi et al. 2017).

To determine the most effective model for representing the data, we calculated the root mean square error (RMSE) in annual maximum rainfall between the observed and projected data over the same period (1970–2018). RMSE measures the average magnitude of differences between actual (observed) and predicted (model-generated) data points. A lower RMSE suggests a better match between the model and observed data. After calculating RMSE, we found that the CMCC-CM2-SR5 and EC-Earth3-veg models had the lowest RMSE values. To determine the best model for predicting maximum rainfall between these two, we compared their error distribution curves. The error distribution curve provides valuable insights into how the model's predictions deviate from the observed data and show error distributions. After reviewing the RMSE values and the error distribution curves, we determined that the CMCC-CM2-SR5 model was the most suitable for our study. This model consistently aligned more closely with the observed data, indicating its high performance in predicting rainfall patterns.

Hydrologic vulnerability assessment

Different indicators were assigned for the three components of vulnerability (exposure, sensitivity, and adaptive capacity) (Figure 7) (ACSAD et al. 2017). The indicators were selected based on several factors, including their relevance to the study, data availability, scientific validity, sensitivity to change, and alignment with the local context.
Figure 7

Hydrologic vulnerability assessment components and indicators.

Figure 7

Hydrologic vulnerability assessment components and indicators.

Close modal

The following are more details about indicators:

Exposure: This component evaluated changes in various meteorological and hydrological factors associated with flash flood risk. We examined changes in the total annual rainfall, peak runoff flow, and mean temperature between the historical and future periods.

Sensitivity: For sensitivity, we addressed four key indicators:

  • 1.

    Urban extent: This indicator measured the impact of urbanization on runoff patterns, recognizing its significance in flash flood risk.

  • 2.

    Population density: This indicator indicates the potential risk to human lives and property.

  • 3.

    Distance to roads: This indicator measured the proximity of roads to drainage streams to assess the susceptibility of road infrastructure to flooding during rainfall events.

  • 4.

    Slope: The mean slope was calculated as it influenced the flow of floodwater.

Adaptive capacity: The assessment of adaptive capacity considered the following indicators:

  • 1.

    Vegetation extent: This indicator served as a natural proxy for resilience, reflecting the area's ability to adapt to flash flood risks naturally.

  • 2.

    Stormwater drainage network: This indicator highlighted the role of the stormwater drainage network in managing excess water during rainfall events, providing insights into the area's capacity to adapt to flash flood risks.

Indicators have different scale of measurements, different values, and units. To overcome this problem, the data were transferred into unitless scores with a common scale using the normalization process that normalized data within the range of 0 to 1. Data were then allocated to equal 10 classes. We used geometric aggregation to composite the indicators in each component. Contrary to arithmetic aggregation which adds indicators together, geometric aggregation has partial compensability, meaning that a low value in one indicator cannot be easily offset by high values in other indicators. The formula is as follows:
formula
(4)
where n is the number of indicators, w is the indicator weight with 0 ≤ w ≤ 1 and Σ wi = 1 for each component, and X is the indicator value. The potential impact for each sub-catchment was calculated by aggregating the exposure and sensitivity following the equation:
formula
(5)
where PI is the potential impact, EX is the exposure, and SE is the sensitivity. Equal weights were selected because sensitivity and exposure have the same level of importance.
After that, the vulnerability was calculated by aggregating the potential impact and adaptive capacity:
formula
(6)
where V is the vulnerability, PI is the potential impact, and AC is the adaptive capacity. The trade-offs between potential impact and adaptive capacity are unclear, so equal weight was the preferred option.

Weight assigning

The analytical hierarchy process (AHP) was employed to determine the weights of the indicators used in the vulnerability assessment. In the context of multi-criteria analysis, AHP is a commonly used decision-support model that was first introduced by Saaty (1980). The rationale for using the AHP approach to determine indicator weights lies in its ability to provide a systematic and mathematically rigorous way to assess the relative importance of various factors or criteria in a complex decision-making process. The AHP approach involves breaking down the problem into a hierarchical structure and then ranking the different alternatives based on user judgments. In this study, AHP was implemented as follows:

Step 1: Design paired comparison matrices for sensitivity, exposure, adaptive capacity, and overall matrix as listed in Tables 4 and 5.

Table 4

Overall AHP judgment matrix

ParameterRainfall intensityRunoff amountUrban extentPopulation densityDistance to roadsSlope
Rainfall intensity 
Runoff amount ½ 
Urban extent 1/3 1/3 
Population density 1/5 1/4 1/2 
Distance to roads 1/5 1/4 1/3 1/3 
Slope 1/6 1/3 1/4 1/3 1/3 
ParameterRainfall intensityRunoff amountUrban extentPopulation densityDistance to roadsSlope
Rainfall intensity 
Runoff amount ½ 
Urban extent 1/3 1/3 
Population density 1/5 1/4 1/2 
Distance to roads 1/5 1/4 1/3 1/3 
Slope 1/6 1/3 1/4 1/3 1/3 
Table 5

Judgment matrices in AHP

Sensitivity
Exposure
Adaptive capacity
UEPDSLDtRRIRATCVESN
UE RI VE 
PD 1/2 RA 1/2 SN 
DtR 1/3 1/3 TC 1/7 1/8    
SL 1/4 1/3 1/3        
Sensitivity
Exposure
Adaptive capacity
UEPDSLDtRRIRATCVESN
UE RI VE 
PD 1/2 RA 1/2 SN 
DtR 1/3 1/3 TC 1/7 1/8    
SL 1/4 1/3 1/3        

Step 2: Design pairwise comparisons to determine the relative importance of each indicator over one another, as ‘extreme’ (9), ‘very strong’ (7), ‘essential’ (5), ‘moderate’ (3), or ‘equal’ (1). The relative importance of each pair of indicators was assessed by answering the question: ‘How much more important is indicator A compared to indictor B in causing/contributing to flash flood events?’.

Step 3: Pairwise comparisons should be normalized by adding the values from each matched pair matrix column, then dividing each value by the total of the corresponding columns to get the normalized matrix.

Step 4: Calculate the relative weight of each indicator by taking the average of the normalized values for each row.

Step 5: Calculate the weighted sum value by taking the sum of the product of the degree of importance and relative weight.

Step 6: Calculate the importance of each indicator by dividing the weighted sum value by relative weight.

Step 7: Calculate the maximum eigenvalue (λ) by dividing the total number of importance by the number of indicators.

Step 8: Check the matrix's consistency to ensure that the decision-making judgment is acceptable. To ensure a valid calculation, the consistency ratio (CR) should be less than or equal to 0.1. CR was calculated by:
formula
(7)
where CI is the consistency index and IR is the index random consistency. IR is a constant value depending on the number of elements n.
Consistency index, CI, was calculated by
formula
(8)
where λ is the eigenvalue and n is the number of elements.

The judgment matrix for the indicators that affect flash flood occurrence is listed in Table 4, whereas Table 5 shows the judgment matrices of sensitivity, exposure, and adaptive capacity for weight assigning. The resulting weights of the vulnerability assessment indicators are listed in Table 6.

Table 6

Indictors weights

SensitivityExposureAdaptive capacity
Urban extent 0.4 Change in rainfall 0.56 Vegetation 0.5 
Population density 0.3 Change in runoff 0.37 Stormwater network 0.5 
Distance to road 0.2 Change in temperature 0.06   
Slope 0.1     
CR: 0.06  CR: 0.066 CR: 0 
SensitivityExposureAdaptive capacity
Urban extent 0.4 Change in rainfall 0.56 Vegetation 0.5 
Population density 0.3 Change in runoff 0.37 Stormwater network 0.5 
Distance to road 0.2 Change in temperature 0.06   
Slope 0.1     
CR: 0.06  CR: 0.066 CR: 0 

Assessing changes in flash flood intensity, damage avoidance potential, and risk

To assess the likelihood of flash flood occurrences, evaluate flash flood risk, and devise damage prevention strategies, we adopted the methodologies introduced by Sakib (2023). These methodologies involve three fundamental formulas: Sakib temporal flood intensity ratio (STFIR), Sakib flood damage avoidance potential (SFDAP), and Sakib flash flood risk index (SFFRI). These formulas were applied to all sub-catchments to compute flood potential, potential damage avoidance, and overall risk.

STFIR: This formula examines how flash flood intensity in a specific area is expected to change over time due to climate change. The STFIR provides a ratio comparing the projected flood intensity to the historical flood intensity. The formula is as follows:
formula
(9)
where FIt is the flood intensity during a future time t and FIh is the historical flood intensity. An STFIR value greater than 1 indicates an expected increase in flood intensity in the future, while a value less than 1 suggests a decrease. The magnitude of the STFIR signifies the projected percent change.
SFDAP: This formula assesses the percentage of potential flood damage that can be alleviated through adaptation strategies. It considers the projected change in flood intensity (using the STFIR) and the level of urbanization, which can exacerbate flood damage. The formula is as follows:
formula
(10)
where STFIR is the temporal flood intensity ratio and UH is the urbanization ratio.
SFFRI: This index provides an integrated risk assessment by combining projected changes in flood intensity, urbanization, and road infrastructure density. The formula is as follows:
formula
(11)
where STFIR represents the temporal flood intensity ratio, UH is the urbanization ratio, RD is the road density ratio, and w₁, w₂, and w₃ are the indicator weights. These weights were assigned using the AHP, resulting in weights of 0.6, 0.2, and 0.2 for temporal flood intensity ratio, urbanization ratio, and road density, respectively.

Historical data analysis

In Jordan, the rainfall season typically spans from October to May, with the peak amount of rainfall occurring in January and February across all stations. Figure 8 shows the spatial distribution of rainfall in the study area during the observed period (1970–2018). The precipitation patterns in the study area can be categorized into three groups: dry eastern regions receiving less than 250 mm/year of rainfall, moderate areas with rainfall ranging from 250 to 350 mm/year, and wet western areas receiving over 350 mm/year of rainfall.
Figure 8

Rainfall distribution in the study area in historical period (1970–2018).

Figure 8

Rainfall distribution in the study area in historical period (1970–2018).

Close modal

The number of yearly rainy days was found to vary from decreasing, increasing, or relatively stable among gauge stations during the past decades. This indicates that the frequency of rainy days has not remained constant, but rather, it has shown fluctuations over time. It is noticed that the 11-year moving average mostly behaved as the long-term average line. This finding was observed in all stations.

Analyzing the total annual rainfall data for the period from 1970 to 2018 at 10 out of 11 rainfall gauge stations revealed a decreasing trend. Table 7 provides details of the maximum, mean, and minimum rainfall recorded at all stations. The results of the Mann–Kendall test and Sen's slope are presented in Table 7 and indicate that there is no statistically significant trend in the total annual rainfall across all stations. Nevertheless, a significant decreasing trend was observed in the total monthly rainfall for March at six stations. The Amman Hussein station followed by the Sweileh station has the largest decrease in total annual rainfall by –30.6% and –29.7%, respectively. Overall, the study area experienced a reduction of 14.61% in total annual rainfall during the historical period from 1970 to 2018.

Table 7

Mann–Kendall and Sen's slope tests results

Station nameStation IDMinimum annual rainfall (mm)Maximum annual rainfall (mm)Mean annual rainfall (mm)Standard deviationSen's slopeTrendIncrease or decreasePercentage change (%)
Sukhna AL0012 33.6 292.1 136.97 59.10 −0.538 no significant trend −18.9 
Zarqa AL0015 30.7 270.6 127.02 52.50 −0.035 no significant trend −1.3 
Ruseifa AL0016 40.6 311 143.63 56.88 −0.226 no significant trend −7.5 
Sweileh AL0017 192.8 1,072 471.67 186.79 −3.259 no significant trend −29.7 
Jubeiha AL0018 146 951.6 463.69 174.52 −1.689 no significant trend −17.5 
Amman Airport AL0019 71.2 444.9 235.04 88.09 −0.031 no significant trend −0.4 
Amman Hussein Collage AL0022 142 721.2 384.59 145.11 −2.453 no significant trend −30.6 
Wadi Es-Sir (NRA) AL0057 135.4 800 452.34 166.17 −0.414 no significant trend −2.9 
Wadi Es-Sir AN0022 176.6 980.3 517.28 185.79 −3.450 no significant trend −28.7 
Sahab CD0001 37.7 551.3 237.82 103.00 −0.474 no significant trend −8.6 
El-Muwaqqar CD0003 40.6 296 143.60 60.17 0.454 no significant trend 13.6 
Station nameStation IDMinimum annual rainfall (mm)Maximum annual rainfall (mm)Mean annual rainfall (mm)Standard deviationSen's slopeTrendIncrease or decreasePercentage change (%)
Sukhna AL0012 33.6 292.1 136.97 59.10 −0.538 no significant trend −18.9 
Zarqa AL0015 30.7 270.6 127.02 52.50 −0.035 no significant trend −1.3 
Ruseifa AL0016 40.6 311 143.63 56.88 −0.226 no significant trend −7.5 
Sweileh AL0017 192.8 1,072 471.67 186.79 −3.259 no significant trend −29.7 
Jubeiha AL0018 146 951.6 463.69 174.52 −1.689 no significant trend −17.5 
Amman Airport AL0019 71.2 444.9 235.04 88.09 −0.031 no significant trend −0.4 
Amman Hussein Collage AL0022 142 721.2 384.59 145.11 −2.453 no significant trend −30.6 
Wadi Es-Sir (NRA) AL0057 135.4 800 452.34 166.17 −0.414 no significant trend −2.9 
Wadi Es-Sir AN0022 176.6 980.3 517.28 185.79 −3.450 no significant trend −28.7 
Sahab CD0001 37.7 551.3 237.82 103.00 −0.474 no significant trend −8.6 
El-Muwaqqar CD0003 40.6 296 143.60 60.17 0.454 no significant trend 13.6 

Regarding temperature trends, July and August have the highest mean temperatures, whereas January has the lowest. The mean temperature is about 18.4 °C, the average maximum temperature is 23.9 °C, and the average minimum temperature is 12.8 °C. The minimum temperature has the highest increase trend. According to the results from the Sen's slope test, the increasing percentages for the mean, maximum, and minimum temperatures are 7.3%, 5.8%, and 10.9%, respectively.

Land-use mapping

Land-use maps of the study area in 2001, 2011, and 2021 are shown in Figure 9. Notably, the urban cover is concentrated in the central region along the main drainage stream and tends to extend towards the western portion of the study area. The change in land-use classes between 2001 and 2021 was computed and listed in Table 8.
Table 8

Change in land use between 2001 and 2021

Land-use classArea (km2) 2001Area (km2) 2021Change (km2)Percent (%) 2001Percent (%) 2021Change percent (%)
Urban fabric 178.42 275.67 97.25 27.07 41.83 14.76 
Bare soil 274.41 151.60 −122.81 41.64 23.00 −18.64 
Cultivated soil 158.71 119.03 −39.68 24.08 18.06 −6.02 
Vegetation 44.60 75.40 30.80 6.77 11.44 4.67 
Quarries 3.62 37.38 33.76 0.55 5.67 5.12 
Land-use classArea (km2) 2001Area (km2) 2021Change (km2)Percent (%) 2001Percent (%) 2021Change percent (%)
Urban fabric 178.42 275.67 97.25 27.07 41.83 14.76 
Bare soil 274.41 151.60 −122.81 41.64 23.00 −18.64 
Cultivated soil 158.71 119.03 −39.68 24.08 18.06 −6.02 
Vegetation 44.60 75.40 30.80 6.77 11.44 4.67 
Quarries 3.62 37.38 33.76 0.55 5.67 5.12 
Figure 9

Land-use map of the study area in (a) 2001, (b) 2011, and (c) 2021.

Figure 9

Land-use map of the study area in (a) 2001, (b) 2011, and (c) 2021.

Close modal
Urban fabric has the highest increase in the study area by 14.76% (Table 8). The future land use of 2041 was estimated using the MOLUSCE tool in QGIS (Figure 10). The projected changes in land use between 2021 and 2041 are shown in Table 9. By 2041, it is predicted that urbanization will continue to expand, covering approximately 51% of the study area. However, the projected urban growth rate for the period 2021–2041 is expected to be lower than the growth rate observed in the study area during the period 2001–2021.
Table 9

Land-use change between 2021 and 2041

Land-use classArea (km2) 2021Area (km2) 2041Change (km2)Percent (%) 2021Percent (%) 2041Change percent (%)
Urban fabric 275.67 335.96 60.29 41.83 50.98 9.15 
Bare soil 151.60 152.09 0.49 23.00 23.08 0.07 
Cultivated soil 119.03 74.53 −44.50 18.06 11.31 −6.75 
Vegetation 75.40 62.73 −12.67 11.44 9.52 −1.92 
Quarries 37.38 33.80 −3.58 5.67 5.13 −0.54 
Land-use classArea (km2) 2021Area (km2) 2041Change (km2)Percent (%) 2021Percent (%) 2041Change percent (%)
Urban fabric 275.67 335.96 60.29 41.83 50.98 9.15 
Bare soil 151.60 152.09 0.49 23.00 23.08 0.07 
Cultivated soil 119.03 74.53 −44.50 18.06 11.31 −6.75 
Vegetation 75.40 62.73 −12.67 11.44 9.52 −1.92 
Quarries 37.38 33.80 −3.58 5.67 5.13 −0.54 
Figure 10

Projected land-use map of the study area in 2041.

Figure 10

Projected land-use map of the study area in 2041.

Close modal

Future climate data

According to the GCM data, the study area is expected to experience alterations in its rainfall patterns. Dry areas are projected to receive more rainfall than usual, while wet areas are expected to receive less rainfall. Figure 11 shows the projected total annual rainfall distribution across the study area.
Figure 11

Rainfall distribution in the study area for the projected period (2019–2060).

Figure 11

Rainfall distribution in the study area for the projected period (2019–2060).

Close modal
The study area is projected to experience, on average, a 5.26% increase in total annual rainfall from 1970 to 2060. However, this increase will primarily occur in downstream dry areas. Figure 12 depicts the overall trend in the total annual rainfall for six stations, considering both the observed period (1970–2018) and the predicted period (2019–2070). The rainfall time series illustrated in Figure 12 shows that the dry stations (AL0016, AL0012, and CD0003) are likely to experience a rise in the average annual rainfall, while the wet stations (AL0018, AL0019, and AN0022) are expected to witness a decrease in rainfall. Additionally, the mean annual temperature is expected to rise by 11.7% in the future, reaching 19.4 °C.
Figure 12

Total annual rainfall trend in the recorded (1970–2018) and predicted period (2019–2060) for six stations.

Figure 12

Total annual rainfall trend in the recorded (1970–2018) and predicted period (2019–2060) for six stations.

Close modal

IDF curves and runoff estimation

IDF curves were constructed using the annual maximum rainfall data for all stations. The rainfall storm that caused the last flash flood in the study area had an intensity of 8–12 mm/h and lasted for 2 h. Figure 13 shows IDF curves for wet station AN0022 and dry station AL0016 based on historical data (1970–2018) and GCM data (2019–2060). Based on the IDF curves constructed in this study, this specific storm is projected to occur every 2–5 years in the wet stations (Figure 13(a)), and every 25–100 years in the dry stations (Figure 13(b)). However, considering the future IDF curves derived from GCM data, the same storm is expected to happen approximately every 5–10 years in wet stations and every 10–25 years in dry stations. These findings illustrate that certain storm events may become more frequent in dry regions of the study area.
Figure 13

IDF curves for two stations in the study area.

Figure 13

IDF curves for two stations in the study area.

Close modal
During the historical period (1970–2018), sub-catchment 6 exhibited the highest peak flow and runoff volumes, due to its high CN = 96, while sub-catchment 2 had the lowest values. Upon comparing the hydrographs derived from historical rainfall data with those based on projected future rainfall data, it is evident that the peak flow and runoff volumes are higher in the historical period hydrographs for the same duration of the rainfall storm. This trend was observed not only in the overall hydrograph of the entire study area but also in the upstream sub-catchments situated in regions receiving a higher amount of rainfall (Figure 14(a) and 14(b)). However, downstream sub-catchments, such as 1, are expected to experience a further increase in peak flow (Figure 14(c)). Nevertheless, since the sub-catchments 1, 2, 8, and 9 (see: Figure 5) where the peak flow is likely to increase in the future are located downstream, their contribution to the overall peak flow in the study area is relatively minor. Consequently, these findings suggest that runoff is expected to decrease throughout the study area (Figure 14). For example, the whole study area shows a maximum flood peak of 714 m3/s for a 100-year return period under the current condition, while it will decrease to around 610 m3/year under the climate change conditions.
Figure 14

Unit hydrographs for (a) whole study area, (b) sub-catchment 6, and (c) sub-catchment 1, for historical period and future period, showing the peak flow (m3/s) estimated using HEC-1 and SCS methods for 1-h duration considering various return periods.

Figure 14

Unit hydrographs for (a) whole study area, (b) sub-catchment 6, and (c) sub-catchment 1, for historical period and future period, showing the peak flow (m3/s) estimated using HEC-1 and SCS methods for 1-h duration considering various return periods.

Close modal

Vulnerability assessment

Figure 15 shows the maps displaying exposure indicators and the aggregated map. In the maps, a score of 10 (red color) represents a high exposure to flash floods, indicating areas highly susceptible to climate change impacts. In contrast, a score of 1 (green color) indicates low exposure. Our findings show that sub-catchments 1, 2, 7, 8, and, 9 have the highest exposure score, mainly due to the projected increase in rainfall rate and runoff in these regions. Upstream sub-catchments 3, 4, 5, and 6 have a low exposure score because of the projected decrease in rainfall.
Figure 15

Exposure indicators change in rainfall, runoff, mean temperature, and overall exposure map.

Figure 15

Exposure indicators change in rainfall, runoff, mean temperature, and overall exposure map.

Close modal
Figure 16 shows sensitivity indicator maps and the aggregated map, evaluating the susceptibility of sub-catchments to flash floods. Sensitivity scores range from 1 to 10, with higher scores indicating greater sensitivity to flash flood occurrence. In this regard, sub-catchment 6, followed by sub-catchments 1 and 4, scored the highest due to factors such as dense population, extensive urban development, and roads that were built within stream drainages.
Figure 16

Sensitivity indicators: urban extent, population density, distance to roads, mean slope; and overall sensitivity map.

Figure 16

Sensitivity indicators: urban extent, population density, distance to roads, mean slope; and overall sensitivity map.

Close modal

On the other hand, sub-catchments 2, 5, 8, and 9 have the lowest sensitivity scores. Although sub-catchment 2 possesses the steepest mean slope, its score was offset by its low urban extent. Sub-catchments 3 and 7 have a moderate sensitivity to flash floods.

Figure 17 presents the maps related to adaptive capacity. These maps depict the adaptive capacity of different sub-catchments to cope with flash floods. A score of 10 signifies sub-catchments with a high adaptive capacity, demonstrating their preparedness to handle flash floods. A score of 1 represents areas with the weakest adaptive capacity, suggesting potential difficulties in managing flash flood events. Figure 17 shows that sub-catchments 8 and 9 have the lowest adaptive capacity score due to the absence of proper stormwater drainage infrastructure. Urban sub-catchments 1, 4, 5, 6, and 7 have a moderate adaptive capacity, primarily due to their limited vegetation extent. Sub-catchments 2 and 3 have the highest adaptive capacity score of 9.
Figure 17

Adaptive capacity indicators: stormwater drainage network, vegetation extent, and overall adaptive capacity map.

Figure 17

Adaptive capacity indicators: stormwater drainage network, vegetation extent, and overall adaptive capacity map.

Close modal
Figure 18 presents the potential impact and hydrologic vulnerability map. The potential impact map combines exposure and sensitivity indicators, with scores closer to 10 indicating sub-catchments projected to experience a high potential impact, considering both climatic changes and physical characteristics. On the contrary, scores closer to 1 suggest a smaller potential impact. Our results showed that sub-catchment 1 has the highest potential impact score due to its high exposure and moderate sensitivity, followed by sub-catchments 6 and 2. On the other hand, sub-catchments 7, 8, and 9 scored 3, indicating low to moderate potential impact, whereas sub-catchments 3, 4, and 5 have the lowest scores due to the expected decrease in the rainfall rate.
Figure 18

Potential impact and hydrologic vulnerability to climate change.

Figure 18

Potential impact and hydrologic vulnerability to climate change.

Close modal

The hydrologic vulnerability map combines the potential impact and adaptive capacity indicators. In this context, sub-catchments 8 and 9 exhibit a moderate vulnerability with a score of 6. This vulnerability is attributed to their high exposure, characterized by a higher projected rate of rainfall and runoff than the current rate. However, these sub-catchments have low sensitivity due to their low population density and limited urban extent. Limited stormwater network systems contribute to their moderate vulnerability. Sub-catchment 1 has a vulnerability score of 5, indicating a moderate vulnerability. On the other hand, sub-catchment 6, which experienced previous flash flood events, is expected to have a low to moderate vulnerability with a score of 4. Its vulnerability is mitigated by its lower exposure to increased rainfall, while its adaptive capacity helps in reducing vulnerability. Sub-catchment 2 has the lowest vulnerability score followed by sub-catchments 3, 4, and, 5.

Table 10 presents the scores for exposure, sensitivity, adaptive capacity, vulnerability, and potential impact within each sub-catchment. As depicted in the table, exposure scores are closely tied to adaptive capacity scores, shaping the overall vulnerability scores. For instance, sub-catchment 2 moderate exposure is effectively offset by its high adaptive capacity, resulting in a vulnerability score of 2. Conversely, in sub-catchments 8 and 9, their high exposure is further exacerbated due to the lack of adequate adaptive capacity measures, resulting in a vulnerability score of 6. This highlights the importance of implementing appropriate adaptive strategies to mitigate the effects of climate change.

Table 10

Sub-catchments vulnerability scores

Sub-catchment numberExposure scoreSensitivity scoreAdaptive capacity scorePotential impact scoreVulnerability score
Sub-catchment 1 
Sub-catchment 2 
Sub-catchment 3 
Sub-catchment 4 
Sub-catchment 5 
Sub-catchment 6 
Sub-catchment 7 
Sub-catchment 8 
Sub-catchment 9 
Sub-catchment numberExposure scoreSensitivity scoreAdaptive capacity scorePotential impact scoreVulnerability score
Sub-catchment 1 
Sub-catchment 2 
Sub-catchment 3 
Sub-catchment 4 
Sub-catchment 5 
Sub-catchment 6 
Sub-catchment 7 
Sub-catchment 8 
Sub-catchment 9 

Furthermore, sensitivity scores have a direct impact on potential impact, as in the case of sub-catchment 6. This underscores the role of socioeconomic factors and physical basin attributes in either intensifying or mitigating climate change impacts.

Flash flood intensity, damage avoidance potential, and risk

Table 11 shows the calculated STFIR, SFDAP, and SFFRI indices using the formula mentioned for all sub-catchments.

Table 11

Flood intensity, flood damage, and risk indices

Sub-catchmentsTemporal flood intensity ratio (STFIR)Urbanization ratio (UH)Flood damage avoidance potential (SFDAP)Road density ratio (RD)Temporal flood intensity ratio (SFFRI)
Sub-catchment 1 1.19 0.57 31.8 1.1 1.05 
Sub-catchment 2 1.2 0.18 78.3 1.02 0.96 
Sub-catchment 3 0.57 0.56 67.9 1.26 0.71 
Sub-catchment 4 0.52 0.94 51.0 2.05 0.91 
Sub-catchment 5 0.55 0.84 54.2 2.31 0.96 
Sub-catchment 6 0.81 0.84 31.6 1.08 0.87 
Sub-catchment 7 0.85 0.41 65.3 0.68 0.73 
Sub-catchment 8 1.27 0.14 82.2 0.5 0.89 
Sub-catchment 9 1.9 0.128 75.5 0.47 1.27 
Sub-catchmentsTemporal flood intensity ratio (STFIR)Urbanization ratio (UH)Flood damage avoidance potential (SFDAP)Road density ratio (RD)Temporal flood intensity ratio (SFFRI)
Sub-catchment 1 1.19 0.57 31.8 1.1 1.05 
Sub-catchment 2 1.2 0.18 78.3 1.02 0.96 
Sub-catchment 3 0.57 0.56 67.9 1.26 0.71 
Sub-catchment 4 0.52 0.94 51.0 2.05 0.91 
Sub-catchment 5 0.55 0.84 54.2 2.31 0.96 
Sub-catchment 6 0.81 0.84 31.6 1.08 0.87 
Sub-catchment 7 0.85 0.41 65.3 0.68 0.73 
Sub-catchment 8 1.27 0.14 82.2 0.5 0.89 
Sub-catchment 9 1.9 0.128 75.5 0.47 1.27 

Based on these results, some observations can be drawn:

  • The STFIR values indicate that all sub-catchments except sub-catchments 1, 2, 8, and 9 are expected to experience a decrease in flood intensity in the future due to climate change impacts. Sub-catchment 9 has the highest STFIR value of 1.9, which means its flood intensity is projected to increase compared to historical levels.

  • The UH values reflect the extent of urbanization within each sub-catchment. Sub-catchment 4 has the highest UH value of 0.94, whereas sub-catchment 8 has the lowest urbanization ratio.

  • The RD values represent the ratio of road density in each sub-catchment, which has implications for drainage capacity and the speed of flood during flash flood events. Sub-catchment 5 has the highest RD value of 2.31 which can increase its flash flood risk. It can be noted that sub-catchments 8 and 9 have the lowest RD.

  • The SFDAP values indicate the percentage of potential flood damage that can be averted in the future through the implementation of adaptation strategies that account for the impacts of climate change and urbanization growth. Generally, all sub-catchments show high potential for reducing flood damage through adaption measures. Sub-catchments 6 and 5 have the lowest SFDAP value of 31%, indicating a pressing need for increased attention and focused efforts in implementing flash flood adaptation measures.

  • The SFFRI values provide an overall weighted index of flash flood risk. Sub-catchment 9 has the highest SFFRI value of 1.27, which means it has the highest flash flood risk among all sub-catchments.

Morphometric analysis revealed that the study area watershed exhibits physical characteristics that increase its vulnerability to flash floods. With a circular shape and low drainage density, the watershed is susceptible to experiencing high peak flows and more intense flooding. After analyzing the rainfall trends and characteristics in the study area the average total annual rainfall in the study area is approximately 254 mm. High elevations show high levels of rainfall whereas low elevations are associated with low levels of rainfall. Analysis of rainfall data from 11 rainfall gauge stations indicated the highest total annual rainfall at 517 mm for the Wadi Es-Sir station and the lowest at 127 mm for Zarqa stations. Additionally, the spatial rainfall distribution shows that the rainfall increases towards the west and decreases towards the east. Our results show a decrease of 14.61% in total annual rainfall within the study area between 1970 and 2018 and the maximum daily rainfall and number of rainy days tend to decrease across all stations; this indicates that climate change has affected the area. Furthermore, the average temperatures showed an overall increasing trend in the mean, maximum, and minimum values of 7.3, 5.8, and 10.9%, respectively. In this respect, the average minimum temperature demonstrated the most significant rate of increase, a finding that agrees with a separate study on climate change trends in Jordan (Hamdi et al. 2009).

To assess the characteristics of the study area that contribute to its flash flood vulnerability, the results showed that land-use patterns have changed in the past twenty years. Urban areas expanded by 14.76%, resulting in a net increase of 97.25 km2 during this period. This growth came at the expense of reductions in bare soil and cultivated land cover, as more land was transformed into urban areas. The rise in urbanization was primarily driven by a rapid population increase due to an influx of refugees and internal migration from other Jordanian cities and towns. The projected land use in 2041 shows a dramatic increase in urban land use with 51% of the study area expected to be covered by infrastructure, because of projected population growth. As the study area undergoes more urbanization, the demand for a robust stormwater drainage system becomes increasingly critical to manage the runoff that flows through streets. However, it also poses a challenge to groundwater recharge, as increased impervious surfaces hinder water infiltration. Consequently, the shift towards urbanization could exacerbate the impacts of climate change in the study area, particularly concerning flash floods, as its main drainage streams are covered by roadways, thus increasing the area's susceptibility.

The GCM data were used to analyze the future climate data to reveal the expected trends in rainfall and mean temperature. The GCM results indicate a slight increase in total annual rainfall of 5.26%, although the rate is projected to decline over time. Specifically, the amount of rainfall is expected to increase in 3 of the 11 stations and decrease in the other 8. However, since the stations with increased rainfall are located downstream and only a few in number, their overall impact on runoff is limited. Despite the projected increase in urban cover, which can affect the land's capacity to absorb runoff, the runoff simulation results indicate a notable decrease in both runoff volume and peak discharge across most sub-catchments in the study area due to projected lower rainfall. This situation holds significant implications for water availability in an area already facing water scarcity. The mean temperature is expected to increase from between 18.4° and 19.4° representing a 11.7% increase.

According to the AHP, rainfall is the primary controlling factor influencing flash flood incidents. It carries the highest weight as compared to other factors (runoff, population density, urban extent, distance to roads, and mean slope). Thus, the flash flood vulnerability is highly related to the rainfall rates in the observed and future periods.

We assessed the flash flood vulnerability and located the most vulnerable site in the study area. The assessment of the degree of exposure to climate changes indicated that sub-catchments 9 and 1 are the most exposed to flooding given the expected increase in rainfall and runoff. In contrast, sub-catchments 4 and 5, located upstream, are expected to experience a dramatic drop in rainfall and runoff rates, leaving them susceptible to water shortage with important implications for agriculture activities.

Sub-catchment 6 ranked the highest in terms of sensitivity to flash floods because it has the highest population density. However, given this catchment's higher level of adaptive capacity attributed to the type of infrastructure present and low exposure due to lower rainfall, it is not considered vulnerable to flooding. After combining the exposure and sensitivity, the potential impact map was generated. Sub-catchment 1 scored highest in terms of the potential impact of flash floods. This was attributed to higher recorded levels of rainfall and runoff in a highly urbanized context where there are many roads built along the drainage stream and a high mean slope which impacts the area's capacity to drain water. Further, the adaptive capacity map showed that sub-catchments 8 and 9 have the least capacity to cope with flash floods given the low level of vegetation to absorb excessive runoff and insufficient stormwater drainage infrastructure.

Overall, the vulnerability of flash flood occurrence under future climate change is either low or moderate. Sub-catchments 8 and 9 are the most vulnerable sites because of lack of adaptive capacity due to low-level vegetation cover and poor stormwater drainage, as well as high exposure to flooding due to the expected increase in rainfall and runoff. The index findings indicate that sub-catchment 9 has the highest risk of flash floods. However, when we considered additional indicators beyond just urbanization ratio and road density, such as population density, slope, and adaptive capacity indicators, we found that sub-catchment 8 also experiences a high potential of flash flood risk. Sub-catchments 2, 3, and 4 have low vulnerability as they are projected to be less exposed to flooding over time due to the expected decrease in rainfall. These results are consistent across both the CMCC-CM2-SR5 model and the EC-Earth3-Veg model. The results of this study indicate that while climate change has already affected the study area, heavy rainfall events that lead to flash floods are expected to occur less in the future. At the same time, the study area will suffer greatly from lack of water with ongoing implications for social, economic, and environmental well-being.

The findings of our study were compared to previous studies conducted in the same region and other regions in Jordan. In alignment with prior research, our study confirmed that the southwestern part of the AZB is expected to experience a slight increase in precipitation and a rise in minimum temperatures. Additionally, western upstream sub-catchments will have a noticeable rainfall reduction. Similar conclusions were drawn by Al-Hasani et al. (2023) in their study to assess climate change's impact on surface water in the AZB. However, regional differences emerged within Jordan. While Al-Taani et al. (2023) highlighted the importance of slope in flash flood vulnerability in Aqaba City, our research emphasized the role of rainfall intensity as the primary factor. These differences emphasize that flash flood vulnerability is context-dependent, requiring tailored mitigation strategies.

The methodology utilized in this study can be transferred and applied to evaluate hydrological responses to climate change in urban watersheds within arid regions, particularly in Jordan. It's worth noting that limitations in data availability may hinder a comprehensive evaluation of flood vulnerability. To enhance future assessments, we recommend the inclusion of socioeconomic data and community resilience evaluations. These additions will contribute to a more comprehensive analysis, ultimately improving decision-making regarding site-specific adaptation measures.

This study aimed to assess the hydrologic vulnerability of the AZB, as it represents an important social and economic region in Jordan, to climate change. The study area includes 6 million inhabitants and most of the country's industries; it has also been suffering from flash floods for years. The study was carried out by analyzing climatic data and using GCM to predict climatic conditions. In summary, results from this study indicated a decrease in the overall rainfall for the period 1970–2018 and an increase in temperature for 1988–2016. Land use has also changed in the study area, becoming increasingly urbanized over the past 20 years; this is expected to continue, although at a reduced rate, with approximately half of the study area projected to be urban land by the year 2041.

Our results also revealed that the projected data derived from the CMCC-CM2-SR5 model show that the study area is expected to witness an increase in rainfall of about 5.26%. However, given the expected increase in rainfall will only take place in three stations that are located downstream, whereas the upstream stations will witness a decrease in rainfall and peak runoff discharge is expected to decrease in most of the study area sub-catchments. Moreover, sub-catchments 9 and 8 are the most vulnerable sites to flash floods; however, the highest vulnerability is considered moderate as the most exposed sub-catchments have a low sensitivity given their low population density and urbanization. To mitigate flash flood vulnerability in these two sub-catchments, we recommend enhancing the infrastructure of the stormwater drainage network and constructing streamways to manage excess floodwater.

While the vulnerability to future flash floods in the study area is assessed as moderate, it remains highly susceptible to the effects of climate change. The main populated sub-catchments in the study area are projected to experience a large drop in rainfall and runoff rates, leading to reduced water availability and groundwater recharge. To address these challenges, implementation of adaptation strategies is crucial. This may include the adoption of water harvesting systems, particularly in sub-catchments with the highest rainfall rates, promoting water-saving innovations in irrigation and wastewater treatment, and raising public awareness about rationing water use. It is recommended to take this study's findings into consideration in the national climate change adaptation plans and strategies.

The authors would like to thank the anonymous reviewers for their useful suggestions. The authors thank the Ministry of Water and Irrigation and Greater Amman Municipality for providing the necessary climatic data. We would like to thank RICCAR for providing the climate data. We would like to show our gratitude to Dr Mohmmad Al-Qinna for his valuable suggestions. Special thanks are going to Prof. Dr Ghazi Saffarini and Prof. Dr Jason Rech from Miami University at Oxford for reading and improving the manuscript. This research has been accomplished during the sabbatical leave offered to Prof. Dr Mustafa Al Kuisi from the University of Jordan from October 2021 to September 2022.

Data cannot be made publicly available; readers should contact the corresponding author for details.

The authors declare there is no conflict.

Abdulla
F.
2020
21st century climate change projections of precipitation and temperature in Jordan
.
Procedia Manufacturing
44
,
197
204
.
doi:10.1016/j.promfg.2020.02.222
.
Abu Qudaira
M.
2004
The Geology of Zarqa Area Map Sheet NO.3254-III
.
Al Azzam
N.
&
Al Kuisi
M.
2021
Determination of flash floods hazards and risks for Irbid governorates using hydrological and hydraulic modelling
.
JJEES
81
,
81
9
.
Alhasanat
H.
2014
Flash flood assessment for Wadi Mousa City-Jordan
.
Procedia Economics and Finance
18
,
675
683
.
doi:10.1016/S2212-5671(14)00990-3
.
Al Kuisi
M.
,
Mashal
K.
,
Al-Qinna
M.
,
Hamad
A. A.
&
Margana
A.
2014
Groundwater vulnerability and hazard mapping in an arid region: Sase study, Amman-Zarqa Basin (AZB)-Jordan
.
Journal of Water Resource and Protection
2014
.
doi:10.4236/jwarp.2014.64033
.
Al-Taani
A.
,
Al-husban
Y.
&
Ayan
A.
2023
Assessment of potential flash flood hazards. concerning land use/land cover in Aqaba Governorate, Jordan, using a multi-criteria technique
.
The Egyptian Journal of Remote Sensing and Space Science
26
(
1
),
17
24
.
doi:10.1016/j.ejrs.2022.12.007
.
Arab Center for the Studies of Arid Zones and Dry Lands (ACSAD), GIZ (Deutsche Gesellschaft für Internationale Zusammenarbeit) and United Nations Economic and Social Commission for Western Asia (ESCWA)
2017
Training Manual on the Integrated Vulnerability Assessment Methodology. In Adaptation to Climate Change in the Water Sector in the MENA Region (ACCWaM) Programme. RICCAR Training Manual, Beirut, E/ESCWA/SDPD/2017/RICCAR/Manual
.
Bates
B. C.
,
Kundzewicz
Z. W.
,
Wu
S.
&
Palutikof
J. P.
2008
Climate Change and Water. Technical Paper of the Intergovernmental Panel on Climate Change
.
IPCC Secretariat
,
Geneva
, p.
210
.
Binita
K. C.
,
Shepherd
J. M.
&
Gaither
C. J.
2015
Climate change vulnerability assessment in Georgia
.
Applied Geography
62
,
62
74
.
doi:10.1016/j.apgeog.2015.04.007
.
Boori
M. S.
&
Voženílek
V.
2014
Socio-hydrological vulnerability: A new science through remote sensing and GIS
.
Global Journal of Research in Engineering
14
(
4
),
36
42
.
Chung
E.
,
Park
K.
&
Lee
K. S.
2011
The relative impacts of climate change and urbanization on the hydrological response of a Korean urban watershed
.
Hydrological Processes
25
(
4
),
544
560
.
doi:10.1002/hyp.7781
.
Da Silva
R. M.
,
Santos
C. A. G.
,
Moreira
M.
,
Corte-Real
J.
,
Silva
V. C. L.
&
Medeiros
I. C.
2015
Rainfall and river flow trends using Mann–Kendall and Sen's slope estimator statistical tests in the Cobres River basin
.
Natural Hazards
77
(
2
),
1205
1221
.
doi:10.1007/s11069-015-1644-7
.
Department of Population and Social Statistics
2020
Estimated Population of the Kingdom by Municipality and Sex, at End-Year 2020
.
Available from: http://dosweb.dos.gov.jo/ar/population/population-2/ (accessed 19 September 2021)
.
Engicon
2019
Amman Dountown Flood in 28 Feb. 2019: Preliminary Techincal Report
.
Prime Ministry
,
Amman
,
Jordan
.
Farhan
Y.
&
Ayed
A.
2017
Assessment of flash-flood hazard in arid watersheds of Jordan
.
Journal of Geographic Information System
9
(
06
),
717
.
doi:10.4236/jgis.2017.96045
.
Franceschini
G.
,
De Leo
E.
&
Muchoney
D.
2019
Jordan – Land Cover Atlas
.
FAO
,
Rome
, p.
66
.
Available from: https://www.fao.org/publications/card/ar/c/CA3388EN/ (accessed 28 September 2021)
.
Füssel
H.-M.
&
Klein
R. J. T.
2006
Climate change vulnerability assessments: An evolution of conceptual thinking
.
Climatic Change
75
(
3
),
301
329
.
doi:10.1007/s10584-006-0329-3
.
Gharaibeh
A. A.
,
Zu'bi
A.
,
Esra'a
M.
&
Abuhassan
L. B.
2019
Amman (City of Waters); Policy, land use, and character changes
.
Land
8
(
12
),
195
.
doi:10.3390/land8120195
.
Hamdi
M. R.
,
Abu-Allaban
M.
,
Elshaieb
A.
,
Jaber
M.
&
Momani
N. M.
2009
Climate change in Jordan: A comprehensive examination approach
.
American Journal of Environmental Sciences
5
(
1
),
740
750
.
doi:10.3844/ajessp.2009.58.68
.
Hammouri
N.
,
Abdulla
F.
,
Abu-Qudais
H.
&
Freiwan
M.
2015
Assessing the Impacts of Climate Change on Water Resources of Jordan
.
Hunting Technical Services, Soil Survey and Land Research Centre (HTS and SSLRC)
1993
The Soils of Jordan. Ministry of Agriculture, National Soil Map and Land Use Project, Level 1: Reconnaissance Soil Survey (Scale 1: 250,000), 3 Volumes, Amman
.
Hyarat
T.
,
Al Kuisi
M.
&
Saffarini
G.
2022
Assessment of groundwater quality using water quality index (WQI) and multivariate statistical analysis in Amman-Zarqa area/Jordan
.
Water Practice and Technology
17
(
8
),
1582
1602
.
doi:10.2166/wpt.2022.076
.
IPCC
2007
Climate Change 2007: Synthesis Report. Contribution of Working Groups I, II and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change (Core Writing Team, R. K Pachauri & A. Reisinger, eds.). IPCC, Geneva, Switzerland, 104 pp
.
IPCC
2021
Summary for policymakers. Headline Statements from the Summary for Policymakers (Working Group I)
.
Li
T.
&
Li
W.
2015
Multiple land use change simulation with Monte Carlo approach and CA-ANN model, a case study in Shenzhen, China
.
Environmental Systems Research
4
(
1
),
1
10
.
doi:10.1186/s40068-014-0026-6
.
Lian
X.
,
Piao
S.
,
Chen
A.
,
Huntingford
C.
,
Fu
B.
,
Li
L. Z. X.
,
Huang
J.
,
Sheffield
J.
,
Berg
A. M.
,
Keenan
T. F.
,
McVicar
T. R.
,
Wada
Y.
,
Wang
X.
,
Wang
T.
,
Yang
Y.
&
Roderick
M. L.
2021
Multifaceted characteristics of dryland aridity changes in a warming world
.
Nature Reviews Earth & Environment
2
(
4
),
232
250
.
doi:10.1038/s43017-021-00144-0
.
Mahdi
E. S.
&
Mohamedmeki
M. Z.
2020
Analysis of rainfall intensity-duration-frequency (IDF) curves of Baghdad city
.
IOP Conference Series: Materials Science and Engineering
888
(
1
),
12066
.
doi:10.1088/1757-899X/888/1/012066
.
Minstry of Water and Irrigation
2017
Jordan's Water Sector, Facts and Numbers
.
Available from: https://www.mwi.gov.jo/Default/Ar (accessed 22 March 2023)
.
Mirzabaev
A.
,
Stringer
L. C.
,
Benjaminsen
T. A.
,
Gonzalez
P.
,
Harris
R.
,
Jafari
M.
,
Stevens
N.
,
Tirado
C. M.
&
Zakieldeen
S.
2022
Cross-chapter paper 3: Deserts, semiarid areas and desertification
. In:
Climate Change 2022: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change
(
Pörtner
H.-O.
,
Roberts
D. C.
,
Tignor
M.
,
Poloczanska
E. S.
,
Mintenbeck
K.
,
Alegría
A.
,
Craig
M.
,
Langsdorf
S.
,
Löschke
S.
,
Möller
V.
,
Okem
A.
&
Rama
B.
, eds).
Cambridge University Press
,
Cambridge
,
UK
.
Musy
A.
,
Hingray
B.
&
Picouet
C.
2014
Hydrology: A Science for Engineers
.
CRC Press, Boca Raton, Florida, USA
.
Nash
J. E.
&
Sutcliffe
J. V.
1970
River flow forecasting through conceptual models. 1: Discussion of principles
.
Journal of Hydrology
10
(
3
),
282
290
.
doi:10.1016/0022-1694(70)90255-6
.
Perumal
K.
&
Bhaskaran
R.
2010
Supervised classification performance of multispectral images
.
Journal of Computing
2
(
2
),
124
129
.
Riahi
K.
,
Van Vuuren
D. P.
,
Kriegler
E.
,
Edmonds
J.
,
O'Neill
B. C.
,
Fujimori
S.
,
Bauer
N.
,
Calvin
K.
,
Dellink
R.
,
Fricko
O.
,
Lutz
W.
,
Popp
A.
,
Cuaresma
J.C.
, KC,S.,
Leimbach
M.
,
Jiang
L.
,
Kram
T.
,
Rao
S.
,
Emmerling
J.
,
Ebi
K.
,
Hasegawa
T.
,
Havlik
P.
,
Humpenöder
F.
,
Da Silva
L.A.
,
Smith
S.
,
Stehfest
E.
,
Bosetti
V.
,
Eom
J.
,
Gernaat
D.
,
Masui
T.
,
Rogelj
J.
,
Strefler
J.
,
Drouet
L.
,
Krey
V.
,
Luderer
G.
,
Harmsen
M.
,
Takahashi
K.
,
Baumstark
K.
,
Doelman
J.C.
,
Kainuma
M.
,
Klimont
Z.
,
Marangoni
G.
,
Lotze-Campen
H.
,
Obersteiner
M.
,
Tabeau
A.
&
Tavoni
M.
2017
The shared socioeconomic pathways and their energy, land use, and greenhouse gas emissions implications: An overview
.
Global Environmental Change
42
,
153
168
.
doi:10.1016/j.gloenvcha.2016.05.009
.
Rivas
I.
,
Güitrón
A.
&
Montero
M.
2011
Hydrologic vulnerability to climate change of the Lerma-Chapala Basin, Mexico
.
Sustainability Today
297
308
.
doi:10.2495/ST110271
.
Saaty
T. L.
1980
The Analytic Hierarchy Process: Planning, Priority Setting, Resource Allocation
.
Mc Graw-Hill
,
New York
, p.
19
.
Sakib
S.
2023
S M Nazmuz Sakib's Formulas for Assessing Spatio-Temporal Changes in Flash Flood Intensity, Damage Avoidance Potential, and Risk Under Climate Change
.
doi:10.31219/osf.io/r6dqb
.
Seneviratne
S. I.
,
Zhang
X.
,
Adnan
M.
,
Badi
W.
,
Dereczynski
C.
,
Di Luca
A.
,
Ghosh
S.
,
Iskandar
I.
,
Kossin
J.
,
Lewis
S.
,
Otto
F.
,
Pinto
I.
,
Satoh
M.
,
Vicente-Serrano
S. M.
,
Wehner
M.
,
Zhou
B.
,
2021
Weather and climate extreme events in a changing climate
. In
Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change
(
Masson-Delmotte
V.
,
Zhai
P.
,
Pirani
A.
,
Connors
S. L.
,
Péan
C.
,
Berger
S.
,
Caud
N.
,
Chen
Y.
,
Goldfarb
L.
,
Gomis
M. I.
,
Huang
M.
,
Leitzell
K.
,
Lonnoy
E.
,
Matthews
J. B. R.
,
Maycock
T. K.
,
Waterfield
T.
,
Yelekçi
O.
,
Yu
R.
&
Zhou
B.
, eds).
Cambridge University Press
,
Cambridge
,
UK
, pp.
1513
1766
.
doi:10.1017/9781009157896.013
.
Sukristiyanti
S.
,
Maria
R.
&
Lestiana
H.
2018
Watershed-based morphometric analysis: A review
.
IOP Conference Series: Earth and Environmental Science
118
(
1
),
12028
.
doi:10.1088/1755-1315/118/1/012028
.
Tarawneh
Q.
&
Kadıoğlu
M.
2003
An analysis of precipitation climatology in Jordan
.
Theoretical and Applied Climatology
74
(
1
),
123
136
.
doi:10.1007/s00704-002-0705-5
.
UNFCCC, United Nations Development Program, Ministry of Environment
2014
Jordan's Third National Communication to the United Nations Framework Convention on Climate Change
.
Available from: https://unfccc.int/resource/docs/natc/jornc3.pdf (accessed 30 October 2021)
.
USDA-SCS
1986
Urban Hydrology for Small Watersheds. Technical Release No. 55 (TR-55). USDASCS, Washington, DC
.
Yusuf
A. A.
&
Francisco
H.
2009
Climate Change Vulnerability Mapping for Southeast Asia
.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY-NC-ND 4.0), which permits copying and redistribution for non-commercial purposes with no derivatives, provided the original work is properly cited (http://creativecommons.org/licenses/by-nc-nd/4.0/).