\nCN values are derived based on the\u00a0spatial intersection of\u00a0LU and HSG layers.<\/span><\/p>\n
The\u00a0tool operates in\u00a0three main\u00a0steps. Within\u00a0an area of\u00a0up to 20 km\u00b2, it automatically downloads and processes LU layers using a\u00a0uniform priority hierarchy and classification scheme, and vectorises the\u00a0downloaded HSG raster. It then performs geometric intersection of\u00a0these layers and assigns unique CN2 values to each combination, also deriving CN3 values. For the\u00a0specified design rainfall (for selected return periods or user-defined events), it calculates the\u00a0depths and volumes of\u00a0direct runoff. When using design rainfall from the\u00a0rain.fsv.cvut.cz service, the\u00a0tool weights the\u00a0resulting volumes across six synthetic hyetographs and the\u00a0probability of\u00a0antecedent saturation (based on API), in\u00a0accordance with the\u00a0latest design rainfall characteristics.<\/span><\/p>\n The\u00a0tool is published as open-source software, and its development is documented on GitHub platform. It is intended both for professional practice and for educational purposes. The\u00a0tool does not include the\u00a0calculation of\u00a0peak discharges or hyetograph shapes, as deriving these requires additional expertise in\u00a0hydrology, hydraulics, and GIS.<\/span><\/p>\n The\u00a0estimation of\u00a0direct runoff using the\u00a0SCS\u2013CN method\u00a0[1] is widely applied in\u00a0the\u00a0Czech Republic, particularly for determining hydrological response in\u00a0small catchments, ungauged profiles, and in\u00a0the\u00a0design of\u00a0water management measures in\u00a0the\u00a0landscape. The\u00a0use of\u00a0this method is accepted, for example, in\u00a0the\u00a0design of\u00a0infiltration strips, swales, ditches, and other shared facilities within\u00a0land consolidation projects, as well as in\u00a0the\u00a0preparation of\u00a0spatial studies. The\u00a0implementation of\u00a0this method under Czech conditions has been presented in\u00a0a\u00a0number of\u00a0standards and methodological guidelines prior to this publication. In\u00a0particular, this includes the\u00a0most recent methodology Soil Erosion Protection\u00a0[2], preceded in\u00a0the\u00a0field of\u00a0hydrology by the\u00a0methodology Short-term Rainfall for Hydrological Modelling and the\u00a0Design of\u00a0Small Water Management Structures in\u00a0the\u00a0Landscape\u00a0[3].<\/p>\n The\u00a0conceptual approach of\u00a0the\u00a0SCS\u2013CN method, which is widely used for its simplicity and clarity of\u00a0methodological procedures, nevertheless has its limitations. The\u00a0method itself was derived and developed for small catchments, typically a\u00a0few km\u00b2 in\u00a0size. Its application to larger areas, additionally affected by uniform rainfall, is therefore debatable. The\u00a0SCS\u2013CN method is used to derive runoff depth, i.e. runoff volume, and does not account for changes in\u00a0runoff conditions under different rainfall intensities. To obtain\u00a0peak (design) discharges, it is necessary to combine this method with another approach (e.g. a\u00a0unit hydrograph). A\u00a0major limitation is the\u00a0classification of\u00a0soils into only four hydrological soil groups. This limitation can be overcome by physically based methods. An example is the\u00a0Czech model SMODERP\u00a0[4], which can operate with inputs similar to those used by the\u00a0tool described here. However, due to their application demands, physically based models are currently used only to a\u00a0limited extent and primarily in\u00a0specific cases.<\/p>\n The\u00a0SCS\u2013CN method is implemented in\u00a0a\u00a0range of\u00a0proprietary and open-source modelling tools, for example in\u00a0the\u00a0freely available HEC-HMS\u00a0[5], the\u00a0Czech-localised Atlas HYDROLOGY\u00a0[6], and the\u00a0HydroRAIN web service\u00a0[7]. The\u00a0tool described here is partially based on the\u00a0latter. The\u00a0SCS\u2013CN method is also used by the\u00a0Czech Hydrometeorological Institute (CHMI) as one of\u00a0the\u00a0approaches for deriving hydrological characteristics of\u00a0surface waters in\u00a0ungauged profiles (\u010cSN 75\u20091400).<\/p>\n Despite the\u00a0simplicity of\u00a0the\u00a0method itself, its practical application is constrained by (1) the\u00a0use of\u00a0different data sources and their varying quality, (2) inconsistent assignment of\u00a0CN values to land use (LU) classes, and (3) the\u00a0need to correctly handle design rainfall and antecedent moisture conditions.<\/p>\n The\u00a0aim of\u00a0the\u00a0tool developed and described here is to partially consolidate input data and derived outputs in\u00a0the\u00a0form of\u00a0a\u00a0plug-in\u00a0(hereafter referred to as the\u00a0\u201cplugin\u201d) for the\u00a0open-source geographic information system software QGIS\u00a0[8]. The\u00a0tool standardises and automates individual steps in\u00a0the\u00a0processing of\u00a0geospatial data using open data sources in\u00a0the\u00a0Czech Republic, and provides direct outputs both as geospatial layers (CN layers, direct runoff volume layers) and as tabular outputs within\u00a0the\u00a0QGIS environment.<\/p>\n The\u00a0tool has been designed to be:<\/p>\n The\u00a0SCS\u2013CN method is described in\u00a0a\u00a0number of\u00a0available publications [1\u20133]. Only a\u00a0very brief outline of\u00a0the\u00a0method is presented here. The\u00a0direct runoff depth H\u2080 from total rainfall Hs<\/sub> is determined by the\u00a0maximum potential retention\u00a0A and the\u00a0initial abstraction Ia<\/sub>\u00a0=\u00a0\u03bbA (with \u03bb\u00a0=\u00a00.2 by default), according to the\u00a0following equation:<\/p>\n The\u00a0retention is determined by the\u00a0curve number (CN) according to the\u00a0following equation:<\/span><\/p>\n The\u00a0CN value itself is tabulated for combinations of\u00a0land use (LU) and hydrological soil group (HSG) for three fixed antecedent moisture conditions, based on the\u00a0antecedent precipitation index. In\u00a0Czech practice, the\u00a0scenarios of\u00a0antecedent moisture CN2 (average conditions) and CN3 (increased antecedent moisture) are used, with CN3 derived from CN2 using a\u00a0standard transformation. CN1 (dry conditions) is of\u00a0marginal relevance in\u00a0the\u00a0Czech Republic and is practically not encountered. The\u00a0method was originally derived empirically for the\u00a0entire USA, including arid regions.<\/span><\/p>\n The\u00a0antecedent precipitation index is applied in\u00a0a\u00a0five-day form (API5<\/span>), determined according to the\u00a0following equation\u00a0[3]:<\/span><\/p>\n where:<\/p>\n Rn<\/sub>\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 denotes OGC web services for the\u00a0automated acquisition of\u00a0input vector data from open data sources use the\u00a0OGC Web Feature Service (WFS) specification, which enables distributed access to spatial features, including their geometric and attribute characteristics. Within\u00a0the\u00a0developed tool, datasets provided under the\u00a0open data framework in\u00a0the\u00a0Czech Republic (ZABAGED and LPIS) are retrieved using standard WFS operations. These two datasets form the\u00a0primary information input for subsequent land use classification. HSG and design rainfall data are obtained via the\u00a0OGC Web Processing Service (WPS) provided by the\u00a0rain.fsv.cvut.cz project. To ensure smooth operation of\u00a0the\u00a0plugin\u00a0user interface, communication with remote servers is handled asynchronously. This approach enables continuous updating of\u00a0input data, a\u00a0high degree of\u00a0automation, and full reproducibility of\u00a0the\u00a0computational workflow.<\/span><\/p>\n The\u00a0computational workflow was implemented in\u00a0the\u00a0open-source QGIS environment using a\u00a0combination of\u00a0built-in\u00a0analytical tools and functionality extended through a\u00a0custom plugin. The\u00a0core principle of\u00a0the\u00a0processing is the\u00a0stepwise transformation of\u00a0input vector layers representing land use (LU), hydrological soil groups (HSG), and rainfall characteristics into derived thematic layers used as inputs to the\u00a0SCS\u2013CN model. This process employs spatial overlay operations, attribute- and geometry-based joins, buffer generation, and raster vectorisation. These computational operations are carried out using standard QGIS tools executed via the\u00a0Python interface PyQGIS, enabling their automated chaining and ensuring output consistency. The\u00a0entire workflow is complemented by a\u00a0system of\u00a0input data validation and ongoing checks of\u00a0computational steps, aimed at eliminating errors in\u00a0the\u00a0generation of\u00a0CN layers and the\u00a0subsequent calculation of\u00a0direct runoff volumes. The\u00a0objective of\u00a0the\u00a0technological solution is to ensure numerical stability and reproducibility of\u00a0results.<\/span><\/p>\n The\u00a0development of\u00a0the\u00a0plugin\u00a0was conceived as a\u00a0stepwise integration of\u00a0individual analytical steps into the\u00a0QGIS environment. The\u00a0implementation was carried out in\u00a0the\u00a0Python programming language using the\u00a0PyQGIS interface and the\u00a0PyQt library for the\u00a0development of\u00a0the\u00a0graphical user interface. Emphasis was placed on the\u00a0modularity of\u00a0the\u00a0solution, the\u00a0transparency of\u00a0individual computational steps, and the\u00a0possibility for user-defined adjustments of\u00a0input parameters and layer prioritisation via configuration files in\u00a0YAML format, thereby enhancing the\u00a0flexibility and practical applicability of\u00a0the\u00a0resulting plugin.<\/span><\/p>\n The plugin is developed as open-source software under the GNU GPL licence and is available on the GitHub platform. Installation instructions and the user manual are available on the rain.fsv.cvut.cz website.<\/p>\n The\u00a0workflow is divided into four consecutive steps. This allows the\u00a0user to\u00a0 verify individual steps and use their own or modified data, thereby enabling the\u00a0use of\u00a0only selected parts of\u00a0the\u00a0workflow.<\/p>\n The\u00a0tool validates input layers (specifically the\u00a0existence and types of\u00a0attributes defining land use (LU), hydrological soil groups (HSG), and CN2 values), parameter ranges (\u03bb in\u00a0the\u00a0interval 0.1\u20130.3), the\u00a0numerical format of\u00a0user-defined rainfall inputs, and the\u00a0structure of\u00a0configuration files and CSV tables. It also includes a\u00a0uniformly applied symbology for generated LU, CN, and direct runoff volume layers (using a\u00a0quantile classification), comprehensive user documentation (CZ\/EN, MkDocs), and a\u00a0suite of\u00a0automated tests implemented via GitHub Actions, verifying data retrieval, editing operations, CN value assignment, and the\u00a0generation of\u00a0direct runoff volume layers.<\/p>\n The\u00a0model operates with national open data and other available data sources.<\/span><\/p>\n The\u00a0derivation of\u00a0land use (LU) is based on a\u00a0combination of\u00a0selected ZABAGED and LPIS layers with a\u00a0defined priority hierarchy. Selected linear features are interpreted as polygon features using buffers defined by attribute values. The\u00a0result is a\u00a0topologically consistent layer without overlaps between input layers.<\/span><\/p>\n Hydrological soil groups (HSG) are obtained using the\u00a0integrated WPS service soil-texture-hsg<\/span> (https:\/\/rain.fsv.cvut.cz\/webapp2\/ogc-wps\/#1-wps), operated on the\u00a0rain.fsv.cvut.cz server and used to provide soil data. The\u00a0HSG layer was created using digital soil mapping in\u00a0combination with pedotransfer functions (PTFs) as part of\u00a0the\u00a0TA CR project No. TJ02000234. The\u00a0derivation is described in\u00a0two publications [9, 10]. The\u00a0data are provided in\u00a0raster format with a\u00a0spatial resolution of\u00a020\u00a0m. HSG are not derived in\u00a0military training areas, where soil mapping is not available, in\u00a0areas of\u00a0surface mining, in\u00a0the\u00a0centres of\u00a0large cities, or in\u00a0water bodies. The\u00a0raster data are subsequently vectorised without smoothing, and values are supplemented in\u00a0water bodies to enable the\u00a0assignment of\u00a0CN = 99 for these areas.<\/span><\/p>\n The\u00a0plugin\u00a0primarily operates with six-hour design rainfall totals for return periods ranging from two to 100 years, which were derived for the\u00a0purposes of\u00a0hydrological modelling\u00a0[11]. In\u00a0addition to the\u00a0design rainfall depth itself, the\u00a0plugin, in\u00a0accordance with current methodologies [2, 3], incorporates the\u00a0probability distribution of\u00a0hyetograph shapes\u00a0[12] in\u00a0combination with the\u00a0probability of\u00a0elevated (abnormal) initial saturation conditions. The\u00a0user is also allowed to specify a\u00a0custom rainfall total.<\/span><\/p>\n <\/p>\n The\u00a0characteristics of\u00a0design rainfall (total depth, distribution of\u00a0hyetograph shapes, and probability of\u00a0initial saturation) for the\u00a0selected return periods are retrieved via an integrated WPS service, d-rain6h-timedist<\/span> (https:\/\/rain.fsv.cvut.cz\/ The assignment of CN values is carried out in two steps. In the first step, an overlay of the LU and HSP layers is performed. This allows the user to work directly with downloaded data, data refined to reflect actual conditions, user-defined datasets, or future scenarios\/designs. This is followed by the assignment of the CN value itself based on the code designation of LU and HSP. The plugin includes an integrated CN table derived from the original United States Department of Agriculture (USDA) methodologies [1] and their interpretation for Czech conditions [2], published in more detail at https:\/\/rain.fsv.cvut.cz\/scs-cn\/scs-cn-met\/ for average antecedent moisture conditions (CN2). For the\u00a0determination of\u00a0CN3, the\u00a0tool applies a\u00a0derivation from CN2 according to the\u00a0following equation:<\/span><\/p>\nINTRODUCTION<\/h2>\n
\n
MATERIALS AND METHODS<\/h2>\n
The\u00a0SCS\u2013CN method and selected parameters<\/h3>\n
![\"\"](\"https:\/\/www.vtei.cz\/wp-content\/uploads\/2026\/04\/Jehlicka-vzorec-1.jpg\")
<\/a>\n \u00a0 \u00a0 \u00a0 \u00a0 \u00a0<\/a>(1)<\/p>\n\u00a0\u00a0\u00a0\u00a0<\/a>(2)<\/p>\n
\nthe\u00a024-hour rainfall total for the\u00a0period beginning n\u00a0days prior to the\u00a0rainfall event. A\u00a030-day antecedent precipitation index (API30) is also used in\u00a0some cases.<\/p>\nTechnical solution<\/h3>\n
QGIS processing tools<\/h3>\n
Development of\u00a0the\u00a0custom plugin<\/h3>\n
DATA AND RESULTS<\/h2>\n
Description of\u00a0the\u00a0technical solution<\/h3>\n
Input data and their basic processing<\/h3>\n
<\/a>\nFig. 1. Plugin interface: a) data download, b) linking\u00a0LU and HSG layers, c) assignment of CN values, d) retrieving or\u00a0entering precipitation data and running the calculation of runoff depth and runoff volume<\/h6>\n
Land use<\/h3>\n
Hydrological soil groups<\/h3>\n
Design rainfall, probability distribution of\u00a0hyetograph shapes and initial conditions<\/h3>\n
<\/a>\nFig. 2. Input data processed with the plugin for T\u0159ebe\u0161ice and B\u00fdkovice (Bene\u0161ov District): a) LU layer derived from ZABAGED and LPIS data, b) HSG<\/h6>\n
\nwebapp2\/ogc-wps\/#1-wps), which is incorporated into the\u00a0plugin.<\/span><\/p>\n<\/a>\nFig. 3. Interface providing pre-prepared design rainfall data for fourth-order catchments (https:\/\/rain1.fsv.cvut.cz\/); the plugin retrieves the data for the user-specified area<\/h6>\n
Assignment of\u00a0CN values and calculation of\u00a0runoff depth and volume<\/h3>\n
\u00a0\u00a0\u00a0\u00a0<\/a>(3)<\/span><\/h3>\n<\/a><\/h6>\nFig. 4. Assigned CN values based on the intersection of LU and HSG layers<\/h6>\n
Calculation of\u00a0runoff depth and volume<\/h3>\n