Earth Sci Inform DOI 10.1007/s12145-016-0284-0

METHODOLOGY ARTICLE

QMorphoStream: processing tools in QGIS environment for the quantitative geomorphic analysis of watersheds and river networks Carlo Tebano 1 & Francesco Pasanisi 1 & Sergio Grauso 2

Received: 22 June 2016 / Accepted: 6 December 2016 # Springer-Verlag Berlin Heidelberg 2016

Abstract The quantitative geomorphic analysis is a powerful tool for the study of geomorphology and landforms, as it provides objective methods to describe the main properties of drainage basins by means of an appropriate set of parameters. Over the last decades, GIS techniques and processing tools have been widely applied to the geomorphic analysis, and specific applications were developed, essentially using commercial software. In the present paper, the first experimental version of QMorphoStream, an originally developed set of processing tools for quantitative geomorphic analysis in QGIS environment, is presented. Besides the obvious advantage in terms of cost reduction, the choice of an open source development environment allowed us to integrate original algorithms with both QGIS built-in functions and processing tools available in the developers’ community. Keywords River system . Quantitative geomorphic analysis . QGIS . Python

Introduction Based on the principles formulated by Horton (1945) and Strahler (1952a, 1957), the quantitative geomorphic analysis Responsible editor: Hassan A. Babaie * Francesco Pasanisi [email protected]

1

ENEA, Italian National Agency for New Technologies, Energy and Sustainable Economic Development, Piazzale Enrico Fermi, 1, 80055 Portici (NA), Italy

2

ENEA, Italian National Agency for New Technologies, Energy and Sustainable Economic Development, Via Anguillarese, 301, 00123 Santa Maria di Galeria (Roma), Italy

is very popular among researchers and practitioners dealing with watersheds and drainage networks. The main advantage of this method is that it describes the fundamental features of watersheds and drainage networks by means of a standard set of quantitative parameters, avoiding qualitative considerations. The above mentioned parameters are representative of the linear, areal and relief aspects of the investigated basin. Since the 1950s, a number of studies have shown the existence of significant correlations between geomorphic parameters (along with hydrologic and climatic parameters) and the hydro-sedimentary response of the watershed (Anderson 1957; Stall and Bartelli 1959; Avena et al. 1967; Ciccacci et al. 1980; Lupia Palmieri 1983; Grauso 1986; Ciccacci et al. 1987; Lupia Palmieri et al. 1995; Grauso et al. 2008; Wuttichaikitcharoen and Babel 2014; Genchi et al. 2016). These correlations can be expressed in terms of predictive formulae aiming to estimate the sediment yield at river basin scale when direct measures are not available. Other significant applications of the geomorphic quantitative analysis are also reported in literature, dealing with different subjects, as flood hazard evaluation (Della Seta et al. 2005; Karymbalis et al. 2011) or understanding of tectonic processes (Pérez-Peña et al. 2010; Mahmood and Gloaguen 2012). Over the last decades, Geographic Information Systems (GIS) have proved to be a very useful and effective tool for the knowledge, description and analysis of the environment. In particular, studies performed since the1990s (Blasi et al. 1990; Gigli et al. 1991) have shown how the powerful capabilities of GIS in terms of data management and processing, spatial analysis and mapping could be successfully applied to geomorphic analysis. A review of early literature shows that the processing tools developed for the automatic or semi-automatic computation of geomorphic parameters as well as reported applications and case studies are mostly based on commercial software. For

Earth Sci Inform

example, a series of applications have been developed in ESRI ArcView/ArcGIS environment (ESRI 2016): De Bonis et al. (2002) and Sarangi et al. (2003) developed tools and user interfaces to estimate the main geomorphic parameter from vectorialised drainage networks and watershed digital elevation models (DEM); Pérez-Peña et al. (2009) presented an ArcGIS extension for the hypsometric analysis of a watershed from DEMs; Langen and Griffith (2012) proposed a method for automatic coding of stream order on stream network data, based on structural query language (SQL). In latest years, free and open source GIS software have been largely developed and applied with the aim to offer the users the possibility to customize functions and tools and sharing solutions with the scientific community. This approach allows a continuous improvement and updating of GIS platforms thanks to the contribution of different scientists and users. In particular, applications to watersheds and stream networks analysis are reported. Among them, Dorillo (2010), and Jasiewicz and Metz (2011) proposed processing toolsets, based on raster analysis of DEMs, running under GRASS GIS (Neteler et al. 2012): the first estimates main geomorphic parameters, while the latter focuses on the extraction and analysis of stream network according to different criteria. Abera et al. (2014) presented a specialized toolset for topographic, hydrologic and geomorphic analysis running under the uDig application framework (uDig 2016). A discussion on the application of open source GIS to earth and environmental sciences, including watershed analysis, can be found in Noti (2014). QGIS is an open source Geographic Information System, released under the GNU general public license (QGIS 2016). Initially conceived as a simple GIS data viewer, QGIS experienced a very rapid growth over the last years, and its built-in functions, plugins and processing tools are nowadays widely used in many fields. Besides native algorithms, third-party processing tools and functionalities as GRASS (Neteler et al. 2012) and SAGA (Conrad et al. 2015) tools are also integrated in QGIS current releases. The original toolset QMorphoStream, developed under the QGIS processing framework (formerly SEXTANTE) is presented in the following. The toolset was designed and developed with the aim to integrate in a single logical frame all the steps of geomorphic quantitative analysis of a drainage system. An overview of the followed methodology and a description of the computation procedures are provided in the next sections. Results are finally discussed along with planned future works.

Methodology QMorphoStream consists of four main algorithms performing different functionalities for the quantitative geomorphic

analysis that are added as processing tools to the QGIS processing toolbar (Fig. 1): 1. PLANO-ALTIMETRIC ANALYSIS: computing a first set of parameters describing the basic size and geometric features of the catchment, including the relief aspects; 2. ASSIGN STRAHLER ORDER: describing the topological organization of the drainage network through the stream order designation as suggested by Strahler (1957); 3. GEOMORPHIC ANALYSIS: deriving a generalised set of standard parameters according to the state of the art of quantitative geomorphic analysis; 4. STREAM SLOPE ANALYSIS: deriving the stream slope gradients.

The processing tools have been implemented in Python. The Python development environment was a natural choice since it is automatically integrated in QGIS and no external compilation is needed. The following input data in the QGIS map are required: 1. 2. 3. 4.

stream network (vector layer); outlet point (vector layer); basin boundary (vector layer); digital elevation model of watershed (raster layer).

Since the geomorphic parameters are mostly based on measures of lengths and areas, a projected metric coordinate system (e.g. UTM) must be used for the input layers and map. The logical structure and data flow of QMorphoStream modules are depicted in Fig. 2. It is to be observed that both stream network and basin boundary must be provided in the QGIS map as input data to QMorphoStream. This information is normally derived either from digital or analogical available cartographic data or automatically extracted via DEM-based GIS tools (e.g. Tarboton et al. 1991) or from a combination of both methods. The philosophy of QMorphoStream is mainly based on externally generated basin and stream network layers. This choice derives from different considerations. The first is that automatically extracted stream networks can considerably differ from reality, especially in flat areas or where the stream channels have been artificially modelled. Moreover, the algorithms for raster-based watershed analysis normally require to specify a number of parameters that significantly affect the result; in particular, the so-called Barea threshold^ parameter, i.e. the minimum number of cells required for the detection of streams, basically defines the resolution of the output drainage network and its optimal choice is a hard task. Actually, this implies that a number of different basins and stream networks could be generated from a given DEM, depending on the choice of processing parameters, and a series of attempts

Earth Sci Inform

Fig. 1 A sample QGIS map for quantitative geomorphic analysis through QMorhoStream toolset. The required input layers are highlighted on the left (blue dashed line). The new tools are added in the processing tools on the right (red dashed lines)

and errors must be repeated until the optimal solution is found. Lastly, DEM resolution itself can significantly affect the quality of derived hydrological features (López-Vicente and Álvarez 2016). In any cases, a validation of the resulting network is required and this is often performed by comparison with the Bblue lines^ of official cartography. All these considerations convinced us to retain the official cartography as suitable working basis. Nonetheless, QMorphoStream processes the river basin and drainage network layers regardless of how they have been derived, ensuring the repeatability of the Fig. 2 Logical scheme and data flow of QMorphoStream modules

analysis and univocal results, with the trivial assumptions that the meaning of results depends on the quality of input data.

Procedures and output In the present section the main computation procedures of the QMorphoStream modules are synthetically described. A list and a short description of the computed parameters is also provided for each of the processing tools.

Earth Sci Inform Table 1 Geometric parameters computed by the plano-altimetric analysis tool

Parameter

Unit

Description

Pm A Hmax Hmin Hmean

km

Basin perimeter length Basin surface area Maximum elevation of the basin Minimum elevation of the basin Mean elevation of the basin

Hrange

m km km

Le Lnet

km2 m m m

Difference in height: Hrange = Hmax-Hmin; Euclidean distance between the outlet point and the farthest point on the basin boundary Distance between the outlet and the main river source point measured along the river network

Plano-altimetric analysis The BPlano-altimetric analysis^ tool calculates a first set of parameters describing the basic size and geometric features of the catchment. These parameters are added as new fields in the attribute table of the basin boundary layer as listed in Table 1: The basin perimeter and surface area are directly computed from the input layer. The maximum, minimum and mean elevations are derived from the DEM recalling the QGIS plugin BZonal Statistics^. A specific algorithm was implemented to calculate the Euclidean distance Le. Namely, the lengths of all ideal lines joining the outlet point with the vertices of the basin boundary

are measured. Then, the longest distance is automatically selected and returned as Le value. In order to estimate the distance Lnet, the outlet and source points are to be specified. The outlet point is given as input map layer, as already stated. As for the source point, the user can either select a supplementary point layer where a known river source is identified and mapped or ask the processing tool to automatically derive it. In this latter case, the source point is estimated as the highest point of the stream network, through a specific algorithm performing the following steps: 1. Rasterization of the stream network layer using the GRASS tool Bv.to.rast.value^;

Fig. 3 Results of the Bplano-altimetric analysis^ tool. The source and outlet point, and their distance along the stream network can be observed. The table reporting the hypsometric data is added to the map

Earth Sci Inform

0),0/0,B)^, being A = rasterized stream network, and B = DEM; 3. Computation of the maximum elevation zmax in the raster layer generated at previous step, using the GRASS tool Br. info^; 4. Creation of a new raster layer using the SAGA tool Braster calculator^, in which the value is set to 1 in the cell with z = zmax, and is null elsewhere. The following formula is used in the Braster calculator^ tool: Bifelse(eq(A,zmax),1,0/0)^, being A = rasterized stream network; 5. Vectorization of the layer generated in the previous step to derive a single point layer representing the source point of the stream network. In this case, the GRASS tool Br.to.vect^ is used.

Fig. 4 Example of hypsometric curve derived from the Bplano-altimetric analysis^ tool. In the case here shown, the shape of the curve and the value of the integral are both indicative of a morphologically Bmature^ basin

2. Creation of a new raster layer using the SAGA tool Braster calculator^, interpolating the rasterized stream network and the DEM. The following formula is used in the Braster calculator^ tool: Bifelse(eq(A,0/

Once the source point is added to the map (Fig. 3), the distance Lnet from the outlet to the source point along the network is evaluated recalling the GRASS tool Bv.net. distance^. The plano-altimetric analysis tool also performs the hypsometric analysis of the drainage basin, i.e., the investigation on how the catchment area is distributed with respect to elevation (Strahler 1952b). In terms of absolute units of measure, imagining to divide the whole basin in cross-sectional areas by elevation steps, the analysis provides a curve in which the elevation z is plotted on the ordinate and the area a of the part of the

Fig. 5 Input form of the BAssign Strahler order^ processing tool. The only input layer required is the stream network. The outlet segment must be selected by the user

Earth Sci Inform

Fig. 6 The layer BStrahler order^ is added to the map and categorized on the basis of the BSTRAHLER^ field added to the attribute table. The calculated stream length is also added as a new field

basin lying above the elevation z is plotted on the abscissa. Normally, a percentage hypsometric curve is derived, in which the non-dimensional variables x and y are defined as follows:

Finally, the hypsometric integral can be derived: Z 1 Hy ¼ x dy

–

The shape of the hypsometric curve, that can be parameterized in terms of statistical moments or density functions (Harlin 1978), as well as the value of the hypsometric integral, are indicative of the morphological Bage^ of the basin. In general, Byoung^ basins are typically characterized by convex upward hypsometric curves and higher values of integral (say, greater than 0.60), being Hy = 1 at a theoretical initial configuration (Strahler 1952b; Singh et al. 2008). On the contrary, Bmature^ and Bold^ basins generally show a S-shaped curve

–

min y ¼ z−H H range is the ratio of the height of the contour level z

above the minimum elevation to the difference between the maximum and minimum elevations; x ¼ Aa is the ratio of the area above the contour level z to the total basin area ;

From the above definitions, both x and y vary between 0 and 1, being y = 1 for x = 0, and y = 0 for x = 1.

Fig. 7 A schematic drainage network, with indication of false nodes, direct inflows (di), and anomalous inflows (ai,r)

0

Earth Sci Inform

Fig. 8 In order to correctly indentify the stream channels and hydraulic intersections, QMorphoStream assigns a Bgroup^ number to the different features that hydraulically represent a single stream channel

with more or less pronounced upward concavity and lower values of the integral (Fig. 4). The first step of the analysis is performed recalling the QGIS tool BGeoalgorithm-Hypsometric curves^. Then,

percentage areas and non-dimensional parameters x and y of the hypsometric curve are derived for each z value. The computed hypsometric data are automatically written in a table added to the map. The integral Hy is finally calculated from

Fig. 9 A screenshot of the tables plano-altimetric_features, number_of_anomalous_inflows, network_bifurcation_ratios, and average_bifurcation_ ratios added to the map by QMorphoStream

Earth Sci Inform

Fig. 10 A screenshot of the tables anomalous_inflow_indices, drainage_density, and morphometric_indices added to the map by QMorphoStream

x and y interpolated values using numerical routines comprised in the SciPy library (Numpy and Scipy Documentation 2016). Assign Strahler order The concept of stream ordering to describe the hierarchical organization of a drainage network was first introduced by Horton (1945) and then adopted by Strahler (1952a, 1957) with some modifications (Scheidegger 1968). According to Strahler, the 1st order streams are those with no tributaries, generally located upstream in the network. The confluence of two 1st order stream segments gives rise to a 2nd order channel; the confluence of 2nd order streams gives rise to a 3rd order segment and so forth. When two or more stream segments of different order intersect, the order assigned to the downstream segment is the highest among the orders of

upstream segments. The main stem of the river, therefore, is defined as the stream segment of highest order. When a ith order stream channel flows into a (i + 1)th order stream, the confluence is referred to as a Bdirect confluence^, whereas if the order of the recipient segment is greater than (i + 1), an anomalous inflow or Bhierarchical anomaly^ is defined. The QMorphoStream BAssign Strahler order^ tool was developed starting from the code of the BStrahler^ plugin developed by Delahaye (2016) and available on the QGIS repository, that has been extended and adapted to the present algorithm. The outlet arc of the stream network layer must be selected by the user prior to run the tool (Fig. 5). The output of the BAssign Strahler order^ tool is a new vector layer based on the input stream network. In the output layer, the computed Strahler order of each feature of the stream network is added to the attribute table as a new field named ‘STRAHLER’. Another field, reporting the calculated

Table 2 Parameters reported in the network_bifurcation_ratios table. The index i indicates the Strahler order number. For i = imax a single stream segment exists, identified as the main stem of the drainage network, whereas the bifurcation ratios are meaningless and are conventionally set to zero Parameter

Unit

Description

Ni Ndi Rbi

nd nd nd

number of stream segments of ith order in the drainage network number of stream segments of ith order Bdirectly^ flowing into a (i + 1)th order stream segment

Rbdi

nd

direct bifurcation ratio, given by Rbdi ¼ NNdiþ1i

Ri

nd

Bifurcation index, given by the difference Ri = Rbi - Rbdi

bifurcation ratio, given by Rbi ¼ NNiþ1i

Earth Sci Inform Table 3

Parameters reported in the average_bifurcation_ratios table

Parameter

Unit

Rbarit

nd

Rbpon

nd

Rbdarit Rbdpon Rarit Rpon

nd nd nd nd

Description imax−1

∑ Rbi arithmetic average of bifurcations ratios: Rbarit ¼ i¼1imax −1 imax−1

∑ ðN i þ N iþ1 Þ Rbi weighted average of bifurcation ratios: Rbpon ¼ i¼1imax−1 ∑ ðN i þ N iþ1 Þ i¼1

arithmetic average of direct bifurcations ratios Rbdi weighted average of direct bifurcations ratios Rbdi arithmetic average of bifurcations indexes Ri weighted average of bifurcations indexes Ri

stream length (expressed in km), is also added. The output layer is automatically added to the QGIS map and categorized on the basis of Strahler order in the attribute table (Fig. 6). It is to be stressed that the Strahler order numbers are calculated from a purely topological point of view, i.e. based only on the stream network geometry. The relief aspects of the basin, as well as the flow direction and other hydraulic features are not considered in the calculation.

Geomorphic analysis The QMorphoStream BGeomorphic analysis^ processing tool calculates the main geomorphometric parameters of the basin and writes them into a set of tables that are automatically added to the map. The required input layers are listed below: – –

The basin boundary processed by the BPlano-altimetric analysis^ tool, including the basin and network parameters described above; The ordered stream network processed by the BAssign Strahler order^ tool. The outlet arc in the layer must be selected by the user.

Table 4

When processing the drainage network layer, a specific algorithm was implemented to correctly identify and count the stream segments and hydraulic connections. This is a key issue, since a single stream channel can be composed of an unknown number of features, due to the presence of inner or false nodes mostly generated during the manual digitization of drainage network. The solution implemented in the BGeomorphic analysis^ tool is to automatically assign a code number to group the multiple features that hydraulically represent a single stream segment. Namely, starting from the outlet stream selected by the user, the algorithm ideally travels upstream along the drainage network and analyses the connections between the different features. If two features of the same Strahler order share a point where no other connections are found, a code is automatically generated and assigned to both features. The procedure is iterated for all the nodes of the stream network layer. All the features having the same code number are considered as a single stream segment in further calculations (Fig. 7-8). Once the stream segments have been grouped following the above described procedure, the number of stream segments for each Strahler order is derived. Stream segments of

Parameters reported in the anomalous_inflow_indices table

Parameter

Unit

Na

nd

Description imax −2

number of hierarchical anomaly: N a ¼ ∑

imax

∑ Nai;r ⋅f i;r with fi , r = 2r − 2 − 2i − 1 for i ≤ r − 2

i¼1 r¼iþ2

where Nai,r are the numbers of stream segments of ith order anomalously draining into segments of rth order; Ia

nd

index of hierarchical anomaly: I a a ¼ NN a1 where N1 is the number of 1st order streams

Da

km−2

density of hierarchical anomaly: Da a ¼ NAa

Earth Sci Inform Table 5

Parameters reported in the drainage_density table

Parameter

Unit

Description

Ltot Fs

km km−2

total length of the drainage network, given by the sum of all the stream segments length stream frequency: F s s ¼ NA , being N = number of stream segments

−1

Dd

km

drainage density: Dd d ¼ LAtot

SQRT_Dd

nd

Bmodified^ drainage density: SQRT Dd ¼ LptotffiffiAffi

order i directly flowing into segments of order (i + 1) are counted, along with Banomalous^ segments flowing into segments of order greater than (i + 1). Finally, the geomorphic parameters are derived and written in tables that are automatically added to the map (Figs. 9 and 10). Overall, 7 tables are generated in the QMorphoStream environment. A list and a short description of computed parameters, in addition to those already described in Table 1, are presented in Tables 2, 3, 4, 5 and 6. Another table reports, for each pair of Strahler order numbers i and r, the number Nair of anomalous stream segments of i order draining into r order stream segments (r ≥ i + 2). Most parameters are non-dimensional (Bnd^). For those having dimensions, the units of measurement are shown.

Stream slope analysis The Bstream slope analysis^ processing tool calculates the slope of each stream segment and the mean slope of all segments of the same order. Input data are the DEM and the stream network layer processed by the BGeomorphic analysis^ tool. For each hydraulic stream segment, comprised either of a single feature or of a Bgroup^ of features, the processing tool performs the following steps: 1. Calculation of the stream length (Ls);

Table 6

2. Rasterization of the stream segment features using the GRASS tool Bv.to.rast.value^; 3. Creation of a raster layer using the SAGA tool Braster calculator^, interpolating the rasterized stream segment and the DEM. The following formula is used: Bifelse(eq(A,0/0),0/0,B)^, being A = rasterized stream segment, and B = DEM; 4. Computation of the minimum (zsmin) and maximum (zsmax) elevation values in the raster layer created at previous step, using the GRASS tool Br.info^; Finally, the stream slope Ss is derived, given by the ratio: Ss ¼

zsmax −zsmin Ls

For each Strahler order i, the mean value of the stream slope for all the segments of ith order is finally calculated and written in the STREAM_SLOPE table, that is added to the map (Fig. 11).

Discussion and conclusions The new toolset QMorphoStream was developed as a plugin to QGIS, in order to provide a complete, user-friendly integrated set of processing tools aimed at automatically and sequentially performing the geomorphic analysis of watersheds and drainage networks.

Parameters reported in the morphometric_indices table. Circularity and elongation ratios are both equal to 1 for an ideal circular basin

Parameter

Unit

Description

Hf

nd

Fournier’s orographic coefficient: Hf = Hmean ⋅ tan α where tanα ¼ H mean A

Hy Rc

nd nd

hypsometric integral of the basin circularity ratio, i.e. the ratio of the basin area to the area of the circle with circumference length equal to the perimeter of the basin

Rh

nd

relief ratio: Rh ¼ ΔH Le

Re

nd

elongation ratio: Re ¼ DLec , where Dc is the diameter of the circle having the same area of the basin

Earth Sci Inform

Fig. 11 Output of the Bstream slope analysis^ tool. For each Strahler order number, the total number of segments and the mean values of stream slope are reported in the table

A validation test has been conducted on sample watersheds and drainage networks of limited areal extent for the purpose of verifying the accuracy of algorithms and related computation performance. Sample stream networks were composed of 500 to 1000 features and were derived as sub-sets of larger Italian watersheds. For each watershed, the geomorphic analysis was first performed running the QMorphoStream processing tools. Then, the stream ordering was manually performed analysing and counting the stream segments, paying attention to identification of Bfalse nodes^ and detection of direct and anomalous confluences. Results of both manual and automatic procedures were the same. Moreover, the watershed aerial and relief parameters were semi-automatically computed using geoprocessing and analysis tools available in QGIS, and results were consistent with those provided by QMorphoStream. Despite only a limited number of small watersheds was used, the above described testing procedure allowed to verify that QMorphoStream algorithms are conceptually correct. Further tests are planned on larger watersheds that will undergo geomorphic analysis performed through different software tools. The extensive use of the QMorphoStream tools could also suggest further implementations aiming, for instance, at extending the set of calculated parameters by introducing new ones, and improving the efficiency of the algorithms. The present work was developed in the framework of a wide research program aiming at developing tools for the

integrated study of the sediment cycle in the watershedcoastal zone system. In particular, the application of geomorphic analysis to a number of Italian river basins is planned, to improve existing methods for predicting the sediment yield variations under hypothetical climate changes and estimating the possible effects of sea-level changes.

References Abera W, Antonello A, Franceschi S, Formetta G, Rigon R (2014) The uDig spatial toolbox for hydro-geomorphic analysis. Geomorphological Techniques 2:1–19 Anderson HW (1957) Relating sediment yield to watershed variables. Trans Am Geophys Union 38:921–924 Avena GC, Giuliano G, Lupia Palmieri E (1967) Sulla valutazione quantitativa della gerarchizzazione ed evoluzione dei reticoli fluviali (Quantitative assessment of fluvial networks hierarchy and evolution). Boll Soc Geol Ital 86:781–796 Blasi L, Grauso S, Marzocchi S (1990) Automazione delle procedure di analisi geomorfica quantitativa mediante utilizzo di un sistema informativo geografico. Technical Report RT/AMB/90/31, Italian National Committee for Research and Development of Nuclear Energy and Alternative Energies (ENEA), Roma Ciccacci S, Fredi P, Lupia Palmieri E, Pugliese F (1980) Contributo dell’analisi geomorfica quantitativa alla valutazione dell’entità dell’erosione nei bacini fluviali. Boll Soc Geol Ital 99:455–516 Ciccacci S, Fredi P, Lupia Palmieri E, Pugliese F (1987) Indirect evaluation of erosion entity in drainage basins through geomorphic,

Earth Sci Inform climatic and hydrological parameters. In: Gardiner V (ed) International Geomorphology 1986. John Wiley & Sons, Chichester, UK: 33–48 Conrad O, Bechtel B, Bock M, Dietrich H, Fischer E, Gerlitz L, Wehberg J, Wichmann V, Böhner J (2015) System for automated geoscientific analyses (SAGA) v. 2.1.4. Geosci Model Dev 8:1991–2007 De Bonis P, Fattoruso G, Grauso S, Peloso A, Regina P (2002) Computation of geomorphic parameters via G.I.S.-based algorithms: a support tool in river basin management. In: Proceedings of International Conference BNew Trends in Water and Environmental Engineering for Safety and Life. Eco-compatible Solutions for Aquatic Environments^. Capri, Italy Delahaye A (2016) A Strahler plugin for QGIS. https://github. com/ArMoraer/QGISStrahler Della Seta M, Del Monte M, Pascoli A (2005) Quantitative geomorphic analysis to evaluate flood hazards. Geografia Fisica e Dinamica Quaternaria 28:117–124 Dorillo I (2010) r.geomorphic.py: un algoritmo per il calcolo dei parametri dell’analisi geomorfica quantitativa. Un’applicazione al torrente Fersinone (Bacino del F. Tevere, Umbria centrale. r. geomorphic.py, a GIS algorithm for calculating quantitative geomorphic analysis parameters. An application to Fersinone creek (central Umbria). Dissertation, Università degli Studi di Perugia ESRI. GIS mapping software, solutions, services, map apps, and data (2016). http://www.esri.com/ Genchi SA, Vitale AJ, Perillo GME, Piccolo MC (2016) Geomorphometric assessment of drainage systems in a semi-arid region of Argentina using geospatial tools and multivariate statistics. Earth Science Informatics online first articles Gigli C, Miniero O, Grauso S, Polizzano C, Aurelio V, Oliveri F, Marzocchi S (1991) ELISA e GEOMORF. Un sistema informativo territoriale ambientale e un applicativo idrografico come sua evoluzione. Technical Report RT/AMB/90/42, Italian National Committee for Research and Development of Nuclear Energy and Alternative Energies (ENEA), Roma Grauso S (1986) Analisi geomorfica quantitativa: un contributo alla descrizione e qualificazione dell’ambiente. Technical Report RT/ PAS/86/6,Italian National Committee for Research and Development of Nuclear Energy and Alternative Energies (ENEA), Roma Grauso S, Pagano A, Fattoruso G, De Bonis P, Onori F, Regina P, Tebano C (2008) Relation between climatic-geomorphical parameters an sediment yield in a mediterranean semi-arid area (Sicily, southern Italy). Environ Geol 54:219–234 Harlin JM (1978) Statistical moments of the hypsometric curve and its density function. Math Geol 10:59–72 Horton RE (1945) Erosional development of streams and their drainage basins; hydrophysical approach to quantitative analysis. Bull Geol Soc Am 56:275–370 Jasiewicz J, Metz M (2011) A new GRASS GIS toolkit for Hortonian analysis of drainage networks. Comput Geosci 37:1162–1173 Karymbalis E, Chalkias C, Ferentinou M (2011) Flood hazard evaluation in small catchments based on quantitative geomorphology and GIS modeling: The case of Diakoniaris torrent (W. Peloponnese, Greece). In: Advances in the Research of Aquatic Environment. Springer, Heidelberg, Germany; Dordrecht, The Netherlands; London, UK; New York, USA:137–145 Langen G, Griffith J (2012) Alternative method for automatic coding of stream order. In: ESRI International User Conference. San Diego, CA, USA

López-Vicente M, Álvarez S (2016) Effect of DEM resolution on simulated hydrological connectivity in a Mediterranean agro-ecosystem with different crops. Conference poster: 4th MC meeting & WG meetings 21-23rd of September 2016, Prague, Czech Republic Lupia Palmieri E (1983) Il problema della valutazione dell’entità dell’erosione nei bacini fluviali. In: Proceedings of 23rd Italian Geographic Congress. Italian Geographic Society, Catania, pp 143–176 Lupia Palmieri E, Ciccacci S, Civitelli G, Corda L, D’Alessandro L, Del Monte M, Fredi P, Pugliese F (1995) Geomorfologia quantitativa e morfodinamica: I – Il bacino idrografico del Fiume Sinello – Quantitative geomorphology and morphodinamics of the Abruzzo, Italy: I – The drainage basin of the River Sinello. Geografia Fisica e Dinamica Quaternaria 18:31–46 Mahmood SA, Gloaguen R (2012) Appraisal of active tectonics in Hindu Kush: insights from DEM derived geomorphic indices and drainage analysis. Geosci Front 3:407–428 Neteler M, Bowman MH, Landa M, Metz M (2012) GRASS GIS: a multi-purpose open source GIS. Environ Model Softw 31:124–130 Noti V (2014) GIS Open Source per geologia e ambiente. Dario Flaccovio Editore, Palermo, Italy Numpy and Scipy Documentation (2016). http://docs.scipy.org/doc/ Pérez-Peña JV, Azañon JM, Azor A (2009) CalHypso: an ArcGIS extension to calculate hypsometric curves and their statistical moments. Applications to drainage basin analysis in SE Spain. Comput Geosci 35:1214–1223 Pérez-Peña JV, Azor A, Azañon JM, Keller EA (2010) Active tectonics in the sierra Nevada (Betic cordillera, SE Spain): insights from geomorphic indexes and drainage pattern analysis. Geomorphology 119:74–87 QGIS. A Free and Open Source Geographic Information System (2016). http://www.qgis.org Sarangi A, Madramootoo CA, Enright P (2003) Development of User Interface in ArcGIS for Estimation of Watershed Geomorphology. Proceedings of CSAE/SGR 2003 Meeting, Canadian Society for Engineering in Agricultural, Food, and Biological Systems, Montreal, Canada Scheidegger AE (1968) Horton’s law of stream numbers. Water Resour Res 4:655–658 Singh O, Sarangi A, Sharma MC (2008) Hypsometric integral estimation methods and its relevance on erosion status of north-western lesser Himalayan watersheds. Water Resour Manag 22:1545–1560 Stall JB, Bartelli LJ (1959) Correlation of reservoir sedimentation and watershed factors. Springfield plain, Illinois. Illinois State Water Survey, Urbana, IL, USA Strahler AN (1952a) Dynamic basis of geomorphology. Bull Geol Soc Am 63:923–938 Strahler AN (1952b) Hypsometric (area-altitude) analysis of erosional topography. Bull Geol Soc Am 63:1117–1142 Strahler AN (1957) Quantitative analysis of watershed geomorphology. Transactions. Am Geophys Union 38:913–920 Tarboton DG, Bras RL, Rodriguez-Iturbe I (1991) On the extraction of channel networks from digital elevation data. Hydrol Process 5:81–100 uDig. User-friendly Desktop Internet GIS (2016). http://udig. refractions.net Wuttichaikitcharoen P, Babel MS (2014) Principal component and multiple regression analyses for the estimation of suspended sediment yield in ungauged basins of northern Thailand. Water 6:2412–2435