Morphometric Assessment of Wadi Wala Watershed, Southern Jordan Using ASTER (DEM) and GIS

Morphometric analysis is of vital concern to understand hydromophological processes in a given watershed, and thus, it is a priority for assessing water resources in drainage basins. A morphometric analysis was conducted to identify the drainage properties of Wadi Wala and the 23 fourth-order sub-basins. ASTER DEM data was employed to compile slope, elevation, and aspect maps. Arc GIS software was used to measure and calculate basic, derived and shape morphometric parameters. W. Wala is found to be a sixth-order drainage basin, and the drainage pattern is trellis to sub-trellis in the central and lower part of the catchment, whereas it is dendritic to sub-dendritic pattern in the southern and northern parts. The slopes of the catchment vary from 0° - 5° to >35° in slope categories. Tectonic uplifting and tilting, lithology, structure and rejuvenation are the major factors controlling morphological variation over the watershed. The recognized fault systems are chiefly controlling the drainage pattern, and the elongated shape of the sub-basins is attributed to dense lineaments in the central and eastern parts of the watershed. The Rb values for the entire catchment and the sub-catchments range from 2 to 7, with a mean of 4.55, which indicates the distortion of drainage pattern by geological structure. Hypsometric integral values are high for the W. Wala watershed and the sub-basins, where it ranges from 70% to 89%. High HI values indicate that drainage basins are at the youth-age stage of geomorphic development, and they are affected by tectonic uplifting, tilting, and the dominance of hillslope process. Variation in HI values is apparent between sub-basins located at the western part, or, the rejuvenated belt where HI values range from 85% to 89%. Whereas the HI values of the sub-basins located at the eastern part of the watershed, vary from 70% to 84%. Regression analysis reveals that R2 values, which represent the degree of control of driving parameters on HI are reasonably high for the height of local base level (m) and the mean height of sub-basins (m). Both parameters contribute 0.42 and 0.39 respectively (where the F-value is significant at 0.1% and 0.5% levels). Such results imply that the height of local base level (m), and the mean height (m) are the only morphometric driving parameters which have significant control on HI values in the W. Wala watershed. High annual soil loss and sediment load estimated recently, denote that the catchment is highly susceptible to surface erosion at present. Hence, the present study, and the resultant information would help to plan for efficient soil and water conservation measures to reduce soil erosion rates, conserve water, and to control sediment into W. Wala dam.


Introduction
A drainage basin is recognized as a fundamental hydro-geomorphic unit for watershed management [1]. Therefore geomorphometric indices and parameters have been widely employed to investigate the sustainable development of natural resources. Morphometric characteristics of a watershed are significant for assessing surface water resources and groundwater potential [2] [3]. Geomorphometric properties are also essential for proper utilization of land and water resources of a catchment for optimum production with minimal environmental hazards (i.e. severe soil erosion, high sediment yield rates, landslide activity and flooding)to protect the people who live across the catchment, or in occupied areas near the outlet of a watershed [3] [4] [5] [6]. Morphometry refers to the measurement and evaluation of the configuration of the earth's surface, including the shape and dimensions of its landforms, and different aspects of drainage basins [7]. Morphometric analysis is performed through the measurement and calculation of basic parameters, derived parameters, and shape parameters of drainage basins using DEM's, GIS tool, and mathematical equations developed for this purpose [8] [9] [10] [11]. The measured bifurcation ratio (R b ) for example refers to the degree in which geological structure controls the drainage network, whereas, a high value of mean bifurcation ration (R bm ) of a drainage system indicates the runoff and other external agents that contribute to the formation of drainage networks [12] [13]. Assessment of geo-environmental hazards especially flash floods was carried out for arid watersheds which occasionally threaten small and large areas of human settlement [14]- [21]. Watershed prioritization for soil and water conservation, and site selection for water harvesting, were conducted recently based on morphometric analysis, sediment yield estimation, land use/cover, and soil erosion modeling using GIS and remote sensing [4] [6] [22]- [27]. Other applications of morphometric analysis have been conducted worldwide such as: studying the imprints of Quaternary active tectonics over structures and drainage basins [28] [29] [30] [31] [32], tectonic control on Y. Farhan geomorphic processes in shaping drainage networks [33], and landslides coupled with their triggering mechanisms [34]. Geology (lithology and structure), morphology (relief and slope), and climate (precipitation and evaporation) constitute a major complex of physical factors controlling the drainage pattern, density, and geometry of the fluvial system [35]. The relative influence of each factor on fluvial activity varies from one region to another, and subsequently there are noticeable differences exhibited in morphometric properties between drainage basins. Furthermore, the hydrological descriptors of drainage basins (including arid catchments) are positively correlated with morphometric parameters of a watershed, such as: slope, shape, size, drainage density, elevation, basin length, maximum stream length, and total length of stream segments etc. [36]. The hydrological behavior of a drainage basin is largely determined by its geomorphic, geologic, climatic, and morphometric characteristics as defined by linear, areal and relief aspects of the basins [37] [38]. Quantitative morphometric analysis of drainage networks and other properties is traditionally tackled by geomorphologists, hydrologists, and civil engineers. In this context, Strahler [39] argued that morphometric analysis is considered a simple study approach, thus, enabling assessment of basin morphology and processes, and morphometric comparison of different basins developed in different environments. Such approaches enhance our understanding of the geomorphic evolution of drainage basins. It was also concluded in the recent past that any significant changes affecting any environmental component of the watershed, will influence other components especially those located downstream, denoting that any natural or anthropogenic geomorphic and hydrologic changes taking place, will instantly affect certain areas, and may spread to other parts of the watershed [40]. The development of powerful and cost-effective GIS and remote sensing techniques enables us to measure, calculate, and process with high accuracy basic, derived, and shape morphometric parameters of drainage basins. Further, the availability of free access digital elevation data (i.e., STRM and ASTER DEMs) of high resolution, have enhanced rapid quantification of drainage networks, morphometric thematic mapping, and thus, have expanded the applications of morphometric analysis to other fields of research. The main objectives of the present study are: 1) Analyze morphometric properties of W. Wala watershed, and the related 23 fourth-order sub-watersheds, using GIS and ASTER DEM data.
2) Explore the physical behavior and interrelations between morphometric parameters in arid watersheds and sub-watersheds using regression analysis.
3) Statistical evaluation was carried out (R L ), and the number of streams and stream lengths in relation to stream order.
Considering W. Wala as an agricultural watershed, and a promising catchment for future water resources development, morphometric analysis and the resultant information are significant for proper planning of soil and water conservation measures, to minimize soil erosion rates and sediment load, exploration of groundwater potential, and surface water management. Moreover, the present results can also help other investigations that may be carried out in the Y. Farhan watershed study.

Study Area
The W. Wala catchment occupies the upper part of W. Mujib-Wala watershed, covering a triangular shaped catchment of 2063.6 km 2 . It lies between 35˚65'E to 36˚30'E Longitudes, and 31˚55'N to 31˚90'N latitudes ( Figure 1). Terrain elevation ranges from-327 m (b.s.l) at the point where W. Wala merges with W. Mujib (3 km before the wadi system discharge into the Dead Sea) to 1007 m (a.s.l) northwest of the watershed (Figure 2(a) and Figure 2(b)). Flat/undulating terrain (0˚ -5˚) and (5˚ -10˚) dominates the eastern part of the watershed, whereas, steep slopes (>35˚) and dissected terrain characterize the western parts ( Figure 3).
The climate is classified as dry Mediterranean, with relatively cold winters and hot summers, while the canyons downstream close to the Dead Sea are arid. Mean annual rainfall ranges from 346 mm at Madaba (several kilometers to the northwest of the watershed) to 282 mm at Dhiban, and 266 mm at W. Wala weather station. The average annual rainfall for the entire watershed ranges between 100 and 200 mm. Rainfall is concentrated in winter (October to March). Large seasonal variations in temperature are evident, where daily temperatures range from a maximum of >40˚C in August to a minimum of −5˚C in January. The mean annual potential evaporation at the outlet close to the Dead Sea is 2200 mm, with a mean that increases from 1600 mm in the western highlands to 2000 mm in the eastern part of the watershed. Cretaceous carbonate rocks outcrop in most of W.Wala catchment. The oldest rocks exposed in the study area are the Massive limestone unit of Turonian age. The lower part is composed of marl, marly limestone, sand, and chert nodules, while the upper part is composed   This formation is 90 m thick. The chalk-marl formation reposes on top of the phosphorite formation, and ranges in thickness from 20 to 450 m. It consists of marl and chalk with chalk limestone. The chert-limestone formation overlies the chalk-marl member. Massive chalk limestone, alternating thin bedded limestone and chert layers, and range in age from early Paleocene to middle Eocene [41].
Basaltic flows of the Pleistocene age are exposed in the upper reaches of W. Wala. Additionally, superficial deposits of Fluviatile and Lacustrine Gravels of the separated by rocky benches. The wadi profile display well-defined discontinuities which probably represent some form of rejuvenated points. In this regard, four or five rejuvenation stages can be recognized [44] [45]. Rejuvenation processes have resulted in a "poly-cyclic" drainage basin as concluded earlier by Chorely [46]. It is certain that geomorphic development, rejuvenation, and in-Y. Farhan tense incision are responsible for the presence of sharp convex upward hypsometric curve and a high HI value (88.14%). The shape of the HC and high HI value denotes that W. Wala watershed and the sub-watersheds are at the youth-age stage of geomorphic evolution. Thus, they are of high susceptibility to soil erosion, deep incision, landslides activity and flooding [47]. Open rangelands constitute 47% of the catchment area. Rainfed cultivation of cereals (wheat and barley) is practiced in 38% of the total area of the watershed, whereas, 7% of the catchment is urban [43].

Materials and Methods
Topographic maps with a scale of 1:50,000 (20 m contour interval) were pur-  (H I ). Shape parameters are elongation ratio (R e ), circularity ratio (R c ), and form factor (R f ). The stream ordering of the entire W. Wala watershed and the 23 subwatersheds was implemented according to Strahler [39], and the W. Wala catchment was found to be of sixth order. Derivatives of DEM were also slope categories, aspect and elevation maps using the spatial analyst tool available in Arc GIS. The methods adopted for calculation of morphometric parameters are illustrated in Table 1, and results of computation are illustrated in Table 2, and Table S1. Regression analysis is employed to assess the interrelationship between the area of sub-watersheds and other morphometric parameters, where the basin area is considered an independent variable, and other morphometric parameters are dependent variables. In addition, the scale dependency of HI values for 10 sub-watersheds was conducted to evaluate the effect of different driving parameters (i.e., stream order, basin area (km 2 ), height of local base level (m), elongation ratio, form factor, and mean height (m) on hypsometric integral. The value of R 2 represents and indicator of the degree of control of these parameters on HIs.

Morphometric Assessment of W. Wala Watershed
Quantitative analysis was performed for W. Wala catchment and the 23 fourthorder sub-watersheds in order to evaluate the morphometric properties of the drainage networks. Twenty-one morphometric parameters were considered to characterize the watershed and to improve our understanding of drainage basin development with reference to intrinsic controlling factors such as lithology, structure and tectonics geomorphic processes and rejuvenation stages. The results of morphometric analysis for the entire catchment and the 23 sub-basins are illustrated in Table 2, and Table S1. The drainage pattern is trellis to sub-trellis in the central and lower parts of the watershed, whereas it is dendritic to a sub-dendritic pattern in the southern and northern parts. The W. Wala catchment is classified as a sixth-order basin ( Figure 5). Referring to the ratio between basin area (A) and perimeter (P) (5.127:1), the borderline of W. Wala is a relatively irregular water divide.

Basic Morphometric Parameters
The basic morphometric parameters calculated for W. Wala and the 23 sub-basin consists of basin area (A), basin perimeter (P), basin length (L b ), stream order (u), stream length (L u ) and mean stream length (R bm ), and maximum and minimum heights of basin (H and h).

Basin Area (A), Basin Length (Lb), and Basin Perimeter (P)
Drainage area (A) is a fundamental morphometric parameter for hydrological data processes, analysis and interpretation. Larger basins and sub-basins with high relative relief are generally characterized by greater discharge, and directly influenced the peaks and runoff magnitudes. Thus, the basin area is an essential component in hydrological processes [21]. In this context, Chorley et al. [50] concluded that the maximum discharge of flood per unit area, is inversely related to the size of the drainage basin. The total drainage area for W. Wala is Bifurcation ratio (R b ) where N u = total no. of stream segments of oder "u", N u + 1 = no. of segments of the next high order [59] 10 Mean bifurcation ratio (R bm ) R bm = average of bifurcation ratio of Strahler all order. [49] 11 Stream Length ratio (R L ) where L u = the total stream length of order "u", L u − 1 = No. of segment of the next lower order.
[39]   (Table S1). Sub-watershed 23 represents the shortest, but with longest perimeter, while sub-watershed 13 is the longest, but with the Y. Farhan

Stream Length (Lu)
Stream length is measured from the origin of a stream to the drainage divide. L u is a dimensional parameter employed to understand the characteristics of the elements of the drainage network and its contributing basin surfaces [39].  (Table S1), and the L sm value for any given order is greater than that of the lower order and less than that of its next higher order. For the 23 sub-watersheds, the Lsm values range from 0.5999 for the first-order streams, to 24.734 for the fourth-order stream.

Maximum and Minimum Heights (H, h)
The values vary for the sub-watersheds, but they are substantially high.

Slope (Sb)
Slope of drainage basins as a morphometric factor is considered to be of hydrological significance [8]. and low infiltration rates, which in turn accelerate soil erosion. Thus, sediment load production tends to be high especially in over-grazed barren slopes [53].
According to Mesa [8], the catchment slope was computed applying the following formula.

Bifurcation Ratio (Rb)
The bifurcation ratio (R b ) is defined as the ratio of the number of streams of a given order (N u ) to the number of streams in the next higher order (N u + 1). The

Stream Length Ratio (RL)
The watershed and watershed ratios refer to the ratio between the mean length of streams of a given order (L u ) to the mean length of streams in the next lower

RHO Coefficient (ρ)
RHO coefficient is defined as the ratio between the stream length ratio (R L ) and the bifurcation ratio (R b ) [51]. It is affected by geologic, geomorphic, climatic, biologic, and anthropogenic factors [8]. RHO is a significant variable that determines the relationship between D d and the geomorphic evolution of a drainage basin. Therefore, it permits the assessment of the storage capacity of the drainage network [51]. High RHO value of a drainage network is indicative of high hydric storage during flooding, thus, the erosion effect is decreased during the raised discharge [8]. The RHO value for the entire W. Wala 1.22, and for the 23 sub-basin varies from 0.092 to 0.466.

Stream Frequency (Fs)
Stream frequency is defined as the ratio of the total number of streams (N u ) of all orders in a catchment and the basin area [51]. Thus, any increase in stream population is associated with that of drainage density [3]. Low F s values denotes that a relatively low infiltration rate of surface water is achieved; thus, the groundwater potential is relatively low [54]. Melton [55] stated that low value of stream frequency (1.0 to 3.5) indicates that the stream is controlled by fractures, and high stream frequency (4 to 10) denotes low impermeability and more surface runoff. The value of stream frequency for W. Wala watershed is 1.20 km −2 , and for the 23 sub-basins, it ranges from 1.064 km −2 to 1.771 km −2 . Such values imply that the W. Wala catchment is relatively of high runoff.

Drainage Density (Dd)
D d is defined as the total length of streams per unit area divided by the area of drainage basin [51]. It refers to the closeness of spacing of channels, and it is therefore used as a measure of topographic dissection and runoff potential of a given watershed. High D d value indicates high runoff, and consequently a low infiltration rate. By contrast, low drainage density of a basin implies low runoff and high infiltration [56]. Additionally, Strahler [39]

Drainage Texture (Dt)
Drainage texture (D t ) represents the relative channel spacing in a fluvially dissected topography. D t refers to the total number of stream segments of all orders per perimeter of that basin [51]. It depends on a number of physical factors, such as lithology, relief, soil, vegetation, infiltration-capacity, climate, rainfall, and stage of drainage basin development. The D t value for W. Wala is 1.7, and for the 23 sub-watersheds varies from 1.36 to 2.753. However, the variation in D t values is slight between the upper sub-basins ( x = 1.734). According to Smith [57], the drainage texture for W. Wala and the sub-basins ranges from very coarse to coarse texture. It is postulated that high drainage texture values indicated the presence of fragile slope materials and soft rocks. Although the drainage intensity is low for W. Wala, the deterioration of vegetation cover, overgrazing, and high basin relief caused serious soil erosion, thus, high sediment yield was recorded recently [58].

Basin Relief (Bh)
Basin relief (B h ) is the difference in elevation between the highest and the lowest point of a given watershed [59]. B h parameter significantly controls stream gradient, thus, influencing flooding patterns, and the amount of sediments that can be transported. Basin relief in this regard, is a measure of the potential energy of the drainage system present by virtue of elevation above a given datum [60].  (Table 2 and Table S1). High B h values imply a high potential erosional energy of the drainage basin. As a result of progressive lowering of the Dead Sea base level, and tectonic uplifting, the wadi retained rapid down cutting and incision through its geomorphic history, giving rise to remarkable canyons (300 -500 m of depth) downstream, dissected rough terrain in the lower reaches, interrupted valley-side slopes, and a noticeable rejuvenation stages (1, 2, 3, and 4 on Figure 7(b)) appearing on the superimposed and projected cross profiles

Relief Ratio (Rr)
Relief ratio (R r ) is elaborated as a dimensionless height-length ratio between the basin relief (B h ) and the basin length (L b ) [59]. The R r parameter allows compar-

Ruggedness Number (Rn)
Ruggedness number (R n ) is a dimensionless parameter expressing the product of basin relief (B h ) and drainage density [39] [48]. The R n parameter has been introduced to measure the flash flood potential of streams [63], and is also employed to express the geometric characteristics of drainage basins [10]. High Wala catchment is considered to be of pronounced morphology [45]. Watersheds having high R n values are subjected to dynamic geomorphic processes, with long and steep slopes interrupted by sharp breaks of slope due to rejuvenation. Further, the catchment is of high susceptibility to soil erosion, sediment load production, mass movements, and of high response to an as increase in peak discharge.

Hypsometric Integral (HI)
Hypsometric analysis refers to the relative proportion of an area at different elevations of the earth's surface [64]. This approach has been developed to interpret the geomorphic stage of landscape development, denudational processes acting over drainage basins, and to analyze and explain the impact of tectonic activity over a region. The hypsometry of a drainage basin can be evaluated graphically through the "hypsometric curve" (HC) and quantitatively as an integral termed "hypsometric integral" (HI) which are both analyzed with reference to the degree of drainage basin dissection and the relative age landforms. Thus, hypsometric analysis is essential for establishing the relative age of landforms. Consequently, hypsometric analysis is essential tool to assess the impact of lithology, tectonics and climate on landform change, and to evaluate the interaction between tectonic uplift and erosion over an area or watershed. The hypsometric curve represents the volume of rock mass in the watershed against the remaining mass [65] [66] [67], whereas, the hypsometric integral is calculated from the area under a hypsometric curve and expressed as a percentage, where its value varies from 0 to 1 [68]. The hypsometric curves of W. Wala and the 10 sub-watersheds selected for the purpose of further analysis are convex upward, which is indicative of the youth-age stage of geomorphic development. Two categories of hypsometric integrals are identified. The values of the first category range from 85% to 89%, and represent the rejuvenated belt ( Figure 8) characterizing the western part of the watershed. Whereas, the second category describes the HI values which vary from 70% to 84% and pertained to the eastern zone of the watershed (Figure 8) which is less impacted by rejuvenation. Elongation ratio (R e ) is defined as the ratio between the diameter of a circle of the same area as the basin area (A) and basin length (L b ) [59]. Strahler [39] claimed that the values of R e range from 0.6 to 1.0 over a wide range of geological and environmental conditions.

Elongation Ratio (Re)
Values close to 1.0 are characteristic of areas with very low relief, while values in the range of 0.6 to 0.8 are typical of drainage basins with high relief and steep slopes. Low values of R e denote that basins are more elongated, and whenever values approach 1.0, the shape of drainage basin approaches a circle [59]. A circular basin is more efficient in runoff than is an elongated one [69]. The elongation ratio of W. Wala is 0.577, and for sub-watersheds, it ranges from 0.475 to 1.0 (one sub-basin is only circular), which implies that most of the sub-basin are more elongated and elongated in shape.

Circularity Ratio (Rc)
The circularity ratio is defined as the ratio of the basin area (A) and the area of a circle with the same perimeter (P) as that of the basin [39]. R c is controlled by the length and frequency of the streams, lithology and structure, land use/ cover, climate, relief and slope for streams of different orders. The R c parameter is indicative of basin shape, the rate of infiltration, and the time needed for the excess water to reach the basin outlet. Miller [70] described drainage basins of different circularity ratios ranging from 0.4 to 0.5 and concluded that they are strongly elongated, with homogeneous geological materials, and a uniform rate of infiltration; therefore, the excess runoff takes a longer time to reach the basin outlet.

Form Factor (Rf)
The form factor (R f ) is defined as the ratio between the area of the basin (A) and the square of the basin length ( 2 b L ). It was elaborated by Horton (1945) to predict the flow intensity of a drainage basin of a defined area. It reveals an inverse relationship with square of the axial length and a direct relation with peak dis-Y. Farhan charge [37]. For a perfectly circular basin, the value of R f should always be <0.75 [12]. The smaller the value of form factor (<0.45), the more the basin will be elongated. Watersheds with high R f values experience high peak flows of short duration. By contrast, an elongated watershed with low form factor, has a low peak flows of longer duration. The R f value for W. Wala is 0.268, and for the 23 sub-basins it ranges from 0.086 to 0.81, which indicates that the sub-basins are often elongated basins; consequently, low peak flows of long duration predominate [3].  [73]. The use of the ASTER DEM and GIS software package enables rapid, precise, and inexpensive tools to extract and analyze morphometric parameters for W. Wala and the 23 sub-watersheds. R b values of stream order I is higher than those of stream order II and III (Table S1)   The higher values of R L values indicate the young geomorphic landforms development across the watershed [13]. Moreover, the highest minimum and maximum lengths characterize stream order I (Figure 9 reveal an average R r value of 19.12. The R r values explain the intensity of the stream channel gradient, therefore, it is an important factor in assessing erosion processes, soil loss rates and sediment load. Consequently, the peak of discharge and runoff intensity can be predicted [21]. Further, the western sub-watersheds have a higher mean slope, thus providing favorable topographic conditions for a

Conclusions
The present study verifies the efficiency of remote sensing data (ASTER DEM) and Arc GIS tool for hydro-morphometric analysis at sub-watershed level for W.
Wala, a rejuvenated rift catchment. Morphometric analysis was employed to compute basic, derived, and shape parameters of the entire watershed and the 23 sub-watersheds. A strong relationship has been identified between stream order   [23.2], can be considered an important factor in the evaluation of soil erosion and sediment load in W. Wala catchment. Consequently, the peak of discharge and runoff intensity can be predicted. The hypsometric curves for W. Wala catchment and the ten sub-watersheds are of convex upward shape,

Slope
which is indicative of a youth-age stage of geomorphic development. The calculated hypsometric integral values for the western sub-basins (the rejuvenated belt) vary from 85.0% to 89.0%. By contrast, HI values for the eastern sector sub-basins range from 70% to 84.0%. High HI values indicate that these drainage basins are affected by tectonic activity, uplifting, and active hill-slope processes.
Low HI values can be interpreted that the eastern basins are extend 30 -50 km east of the main base level, and thus they are less impacted by rejuvenation processes compared to the western sub-watersheds. The degree of control of driving morphometric parameters over HI values was assessed using regression analysis. The results show that R 2 values (which represent the degree of control of driving parameters over HI) are generally low, except for the height of local base level (m) parameter which accounts for 0.42 (F-value is significant at 0.1% level), and for the mean height (m) variables of sub-basins which contributes 0.39. It is perceptible that the height of local base level (m) and the mean height (m) of sub-basins have a significant control over HI. The development of the stream segments, and the elongated shape of the sub-watersheds are attributed to tectonic and structure, and to morphological controls. The recognized fault groups and the resultant dense lineaments, high slope and relative relief contribute to an overall variation in morphometric and hydrological properties of sub-basins drainage (i.e., runoff and infiltration, soil erosion, landslide activity, sediment load, and flooding). Furthermore, the shape parameters resulted in elongated or less elongated sub-watersheds with low flood peaks and longer flood flows. Flood characteristics, soil erosion rates, and recently recorded high sediment load are considered intrinsic factors in watershed management, rural land use planning, range management, prioritization of sub-basins for soil and water conservation measures, and flood risk assessment.  (Total)  I  II  III  IV  I  II  III  IV  II  III  IV  I  II  III   1