Open Riverine Ecosystem Synthesis (OpenRES):

A QGIS plugin for automated extraction of hydrogeomorphic features to support functional process zone classification of river networks

General Information

OpenRES enables QGIS users to extract up to fifteen physical and environmental features along river segments (typically 5--10 km) to support classification of river networks into Functional Process Zones (FPZs).

Functional Process Zone (FPZ) classification is a method used to divide a river network into river valley scale (5-10 km) segments (or "zones") that share similar physical, hydrological, and geomorphic characteristics. Rather than treating a river as a continuous longitudinal gradient of changing physical conditions, FPZ classification recognizes that rivers are composed of a discontinuous set of hydrogeomorphic patches, each shaped by different landscape and hydrologic processes (Hestir 2007). These zones reflect how the river behaves in a given segment, including how it flows, how it transports sediment, how it interacts with its floodplain, and what types of habitats it supports.

After classifying a river network in FPZs, research questions posed by the tenets of the Riverine Ecosystem Synthesis hypothesis (Thorp et al. 2006, Thorp et al. 2023) can be explored.

---

Installation

Note: OpenRES requires QGIS version >=3.28.

Installation from QGIS:

• Open QGIS -> Plugins -> Manage and Install Plugins... -> select All tab -> search for OpenRES --> select and install plugin

Offline installation from .zip file :

• Go to releases of this repository -> select desired version -> download the .zip file (Note: use OpenRES.zip, not the source .tar).

• Open QGIS -> Plugins -> Manage and Install Plugins... -> install from ZIP tab --> select the downloaded zip --> install plugin (ignore warnings, if any).

---

Data Prerequisites

Users should prepare the following six datasets for your watershed of interest. Use consistent CRS and units (projected meters recommended, e.g., UTM) for all layers.

Dataset	Format	Description
Mean Annual Precipitation Layer	.tif	Raster of mean annual precipitation for the watershed.
Digital Elevation Model (DEM) Layer	.tif	Elevation raster used for slopes, valley floors, and longitudinal gradients.
Simplified Geology Layer	.shp	Polygon layer with generalized classes (e.g., alluvial, mixed, bedrock). Typically a simplified version of a detailed map.
Geomorphically Corrected Stream Network Layer	.shp	Stream lines generated from the DEM and manually corrected to follow observed channel positions in imagery for the analysis period (Whitebox Workflows recommended here).
Valley-Boundary Line Layer	.shp	Lines delineating both the valley floor boundary and valley-edge boundary. Suggested workflow: (1) delineate valley floor, (2) edit to remove holes and unrealistic extents, (3) derive 1-2 km² microsheds/isobasins from the DEM (Whitebox Workflows recommended here), (4) apply intersection/difference/polygon-to-line to extract combined boundaries as a line feature.
Channel Belt Layer (Optional)	.shp	Lines delineating the channel belt (active/recent fluvial influence, including channel and depositional features). Suggested workflow: (1) OpenRES → Geomorphology Tools → Generate Channel Belt, (2) manual refinement to match meanders and depositional forms visible in imagery.

Some of these layers may not be very common for most river systems (especially the Valley-Boundary Line Layer and Channel Belt Layer). For these less common datasets, OpenRES provides several geomorphology utility tools to allow users to create these datasets for their watershed of interest.

Data quality tips

• Ensure all inputs share the same projected CRS and unit (meters)

• Snap and clean linework to avoid sliver gaps that can break intersections

---

Core Functionality

OpenRES provides geomorphology utilities to prepare key boundary layers and a set of sequential extraction tools that produce a standard suite of 15 hydrogeomorphic attributes for FPZ classification.

Geomorphology Tools

Use these tools to produce your Valley-Boundary Line Layer and Channel Belt .

OpenRES includes geomorphology utility tools to help users prepare the valley boundaries and channel belt layers required for subsequent feature extraction \autoref{tab:geomorph-tools}:

Tool	Purpose	Output	GIS Data Type
Generate Channel Belt	Creates lateral offsets from the stream network to approximate the channel belt extent; intended for manual refinement.	Channel belt layer	Vector (line)
Valley Floor Delineation – Sechu	Identifies low-relief valley floor areas from a DEM using a slope-based cost accumulation method [@sechu_2021].	Valley floor layer	Vector (polygon)
Generate Microsheds	Generates microsheds using a threshold-based watershed approach (1–3 km² typical) to capture valley tops in confining valleys.	Microshed layer	Vector (polygon)
Create Valley Boundary	Applies a difference operation between valley floor and microsheds and converts the result to a line layer representing valley boundaries.	Valley boundary layer	Vector (line)

Data Extraction Tools

Run these tools in order. Together, they generate all attributes required for FPZ classification.

Step / Tool	Features	Description	Required
[1] Generate Transects	t_ID	Generates perpendicular transects from river centerlines to valley boundaries, ensuring consistent sampling and linking outputs via a transect ID.	Yes
[2] Extract ELE, PRE, and GEO	ELE, PRE, GEO	Samples elevation, precipitation, and geologic class from user-provided datasets.	Yes
[3] Extract VW, VFW, and RAT	VW, VFW, RAT	Measures valley width and valley floor width from transects and computes their ratio.	Yes
[4] Extract LVS, RVS, and MVS	LVS, RVS, MVS	Computes left, right, and mean valley slopes from elevation differences along transects.	Yes
[5] Extract DVS and SIN	DVS, SIN	Calculates down-valley slope and river sinuosity from segment geometry.	Yes
[6] Extract CBW	CBW	Measures channel belt width from transect intersections with the channel belt layer.	Optional
[7] Extract LCS, RCS, and CBS	LCS, RCS, CBS	Quantifies within-belt channel sinuosity on each side and summarizes with a mean value.	Optional

---

Example usage: Eerste River catchment, South Africa

Note: All the following processing steps should be done in a sequential manner, following the instructions below. Sample data for hydrogeomorphic feature extraction is provided in sample_data folder.

To demonstrate the use of OpenRES, we have provided a dataset from the Eerste River catchment, a small watershed located in the Greater Cape Floristic Region of South Africa.

::: {align="center"}
_{Eerste River catchment, South Africa} :::

The Eerste River originates in the Jonkershoek Mountains, part of the Hottentots-Holland mountain range, and flows westward through the Stellenbosch area before reaching the False Bay coast near Strand. It drains a catchment area of approximately 390 km². Dominated by fynbos vegetation, the area hosts numerous endemic plant species and is under increasing pressure from urban development, invasive species, and agricultural runoff.

---

Prerequisites

Before starting the OpenRES workflow:

• Ensure all input data are properly prepared:
  • Mean annual precipitation (raster)
  • Digital Elevation Model (<= 30 m DEM, raster)
  • Simplified geologic class layer (i.e., alluvial, non-alluvial, bedrock; vector - polygon).
  • Stream network (vector - line)
  • Valley-boundary layer (vector - line)
  • (Optional) Channel belt layer (vector - line)
• The OpenRES plugin is installed and enabled in QGIS.
• The Processing Toolbox is open (via Processing > Toolbox).

::: {align="center"}
_{The OpenRES Processing Toolbox} :::

---

Using the Geomorphology Tool to Prepare Input Data

Generate Channel Belt Layer

Use "Generate Channel Belt"\

Location: Processing Toolbox > OpenRES > Geomorphology

::: {align="center"} :::

Inputs

• River Network Layer (polyline)

Outputs

• Channel Belt Layer

Notes

• Offsets each input stream segment to LEFT (+) and RIGHT (-) by the given distance.
• Offsets use layers CRS; use a projected CRS (e.g., meters).
• LEFT/RIGHT are relative to the digitized direction of each line
• Use Round join style for smooth banks; Miter for sharp corners (user will have to tune miter limit).
• Copies t_ID from input if present, otherwise creates sequential t_ID.
• Adds fields t_ID (int), side {'LEFT'|'RIGHT'}, offset (double).

Note: You must edit this layer to capture the channel belt appropriately. This means delineating areas of active/recent fluvial influence, including channel and depositional features AND removing areas with no visible channel belt.

Delineate Valley Floor

Use "Valley Floor Delineation - Sechu"

Location: Processing Toolbox > OpenRES > Geomorphology

::: {align="center"} :::

Inputs

• River Network Layer (polyline)
• Elevation Raster

Outputs

• Valley Floor Layer (MultiPolygon)

Notes

• Delineates valley bottom by building slope from DEM, using slope as cost surface in GRASS r.cost from a stream network, taking an initial (max) cost threshold, computing mean cost inside that belt, re-thresholding with that mean, and cleaning, smoothing and filling skinny gaps.
• Initial cost distance threshold: [500*(resolution/10m)] is a good starting point

Generate Microsheds

Use "Generate Microsheds"

Location: Processing Toolbox > OpenRES > Geomorphology

::: {align="center"} :::

Inputs

• Input DEM

Parameters

• Threshold: minimum exterior basin size in cells
• GRASS memory: memory allocated to GRASS r.watershed, in MB

Outputs

• Output basins raster
• Microsheds polygon layer

Notes

• Runs GRASS r.watershed to generate uniquely labeled watershed basins from an input DEM.
• The threshold controls the minimum exterior basin size in raster cells.
• The polygon output is created by polygonizing the labeled basin raster using basin_id as the output field.
• A threshold corresponding to approximately 2–3 km² is a useful starting point, depending on DEM resolution.

Create Valley Boundary

Use "Create Valley Boundary"

Location: Processing Toolbox > OpenRES > Geomorphology

::: {align="center"} :::

Inputs

• Microsheds polygons
• Valley floor polygons

Parameters

• Smoothing iterations
• Smoothing offset
• Maximum node angle

Outputs

• Valley boundary lines

Notes

• Selects microshed polygons that intersect the valley floor, subtracts the valley floor polygons from those microsheds, then converts the remaining polygon boundaries to lines.
• The resulting linework is lightly smoothed and dissolved to create a continuous valley boundary layer.
• The output should delineate the valley floor boundary and the confining valley margins.
• Lower smoothing values preserve sharper geomorphic breaks; higher values produce cleaner but more generalized boundaries.

---

Overview of Extracted Features

The following table summarizes the 15 geomorphic and environmental features that will be automatically derived across the Eerste River catchment using the OpenRES tool suite. Each transect, generated perpendicular to the stream network, will be assigned a unique identifier t_ID, and the attributes listed below will be extracted or calculated at the transect or segment level. Transects, segment centers, and the river network segments are all linked together by the t_ID field, enabling subsequent FPZ classification methods to link using joins and relates to the stream network, river segment centers, or transects as desired for visualization purposes.

Feature	Name	Hydrogeomorphic role
ELE	Elevation	Longitudinal position and energy gradient
PRE	Precipitation	Hydroclimatic setting
GEO	Geologic class	Substrate and structural control
VW	Valley Width	Lateral accommodation space at the valley scale
VFW	Valley Floor Width	Floodplain/low relief valley floor extent
RAT	VW:VFW Ratio	Relative valley confinement
LVS	Left Valley Slope	Left side valley confinement
RVS	Right Valley Slope	Right side valley confinement
MVS	Mean Valley Slope	Overall valley confinement
DVS	Down Valley Slope	Longitudinal channel gradient
SIN	River Sinuosity	Planform complexity of the river
CBW	Channel Belt Width	Active channel belt extent
LCS	Left Channel Sinuosity	Left side within-belt planform curvature
RCS	Right Channel Sinuosity	Right side within-belt planform curvature
CBS	Channel Belt Sinuosity	Mean within-belt planform curvature

---

Step 1: Generate Transects

Use: "[1] Generate Transects"
Location: Processing Toolbox > OpenRES > Feature Extraction

::: {align="center"} :::

Inputs

• River Network Layer (polyline)
• Valley Lines Layer (line)
• Extension Increment (optional, default = 250m)
• Max Length (optional, default = 50000m)

Outputs

• Transects -- Multiline layer across the valley
• Segment Centers -- Points at the center of each transect

Notes

• Each river segment gets a unique t_ID.
• Transects are generated perpendicular to the river segment direction.
• Verify that transects intersect both valley lines properly; occasional mismatches may occur due to geometry errors.
• Stream network will also be updated to include the t_ID field.

---

Step 2: Extract Elevation, Precipitation, and Geology

Use: "[2] Extract Point Data"
Location: Processing Toolbox > OpenRES > Feature Extraction

::: {align="center"} :::

Inputs

• Segment Centers Layer (from Step 1)
• Elevation Raster
• Precipitation Raster
• Geology Polygon Layer
• Geology Field (attribute from polygon layer, e.g., GEO or LITH)

Output

• Segment Centers with ELE, PRE, GEO -- Updated point layer

Notes

• Elevation and precipitation are sampled directly from rasters.
• Geology is assigned from intersecting polygon based on the selected field.
• Missing data (e.g., no polygon overlap or invalid raster) will be filled with fallback values (e.g., -9999 or "No Data").

---

Step 3: Extract Valley Width (VW), Valley Floor Width (VFW), and their ratio (RAT)

Use: "[3] Extract VW, VFW", and RAT
Location: Processing Toolbox > OpenRES > Feature Extraction

::: {align="center"} :::

Inputs

• Transects Layer (from Step 1)
• Segment Centers Layer (from Step 2)
• Valley Lines Layer
• Stream Network Layer

Outputs

• Left/Right VFW Reference Points
• Left/Right VW Reference Points
• Updated Segment Centers with:
  • VFW -- Valley floor width
  • VW -- Valley width

Notes

• Uses intersection logic to find points where transects intersect valley floor and valley edge.
• Then calculates distance between these intersections to compute VFW and VW.
• Reference points (left/right) are saved as point layers for inspection or QA/QC.

---

Step 4: Extract Side Slopes (LVS, RVS, and MVS)

Use: "[4] Extract LVS, RVS, and MVS"
Location: Processing Toolbox > OpenRES > Feature Extraction

::: {align="center"} :::

Inputs

• Segment Centers Layer (from Step 3)
• Left VW / VFW Reference Points
• Right VW / VFW Reference Points
• Elevation Raster

Output

• Segment Centers updated with:
  • LVS -- Left valley side slope (%)
  • RVS -- Right valley side slope (%)

Notes

• Slopes are computed as the elevation difference between VFW and VW reference points on each side divided by the horizontal distance.
• All calculations are in percent slope (rise/run * 100).
• Output replaces the segment center layer with updated slope fields.

---

Step 5: Extract Down-Valley Slope and Sinuosity (DVS and SIN)

Use: "[5] Extract DVS and SIN"
Location: Processing Toolbox > OpenRES > Feature Extraction

::: {align="center"} :::

Inputs

• Segment Centers Layer (from Step 4)
• Stream Network Layer
• Elevation Raster

Output

• Segment Centers updated with:
  • DVS -- Down-valley slope (%)
  • SIN -- Sinuosity (unitless)

Notes

• For each stream segment:
  • Elevation is sampled at start and end points.
  • DVS is calculated as (start - end) / length * 100.
  • SIN is the ratio of actual segment length to straight-line distance.
• Features with insufficient geometry or elevation data are skipped.

Step 6: Extract Channel Belt Width (CBW)

Use: "[6] Extract CBW"\ Location:Processing Toolbox > OpenRES > Feature Extraction`

::: {align="center"} :::

Inputs

• Transects Layer
• Segment Centers Layer (from Step 5)
• Channel Belt Layer (may be generated from Geomorphology > Generate Channel Belt
• River Network Layer

Output

• Left Channel Belt Width Reference
• Right Channel Belt Width Reference
• Segment Centers updated with: - Channel belt width

Notes

• Uses intersection logic to find points where transects intersect the channel belt right and left reference.
• Then calculates distance between these intersections to compute 'CBW'.
• Reference points (left/right) are saved as point layers for inspection or QA/QC.

Step 7: Extract Left, Right and Center Channel Sinuosity

Use: "[7] Extract LCS, RCS, and CBS"
Location: Processing Toolbox > OpenRES > Feature Extraction

Inputs

• Transects Layer
• Segment Centers Layer (from Step 5)
• Channel Belt Layer (may be generated from Geomorphology > Generate Channel Belt)
• River Network Layer

Output

• Left Channel Sinuoisty (LCS)
• Right Channel Sinuosity (RCS)
• Channel Belt Sinuosity (CBS)
• Segment Centers updated with:
  • Left channel sinuoisty
  • Right channel sinuoisty
  • Channel belt sinuosity

---

Completion

At the end of Step 7, your segment center point layer will contain all 15 hydrogeomorphic attributes:

• t_ID, ELE, PRE, GEO, VFW, VW, RAT,LVS, RVS,MVS, DVS, SIN, CBW, LCS, RCS, CBS

::: {align="center"} :::

The extracted attributes from OpenRES can be joined to the river network output from [1] Generate Transects: by using t_ID as the joining feature, then exported to Python, R, or another software for hierarchical clustering analyses commonly used to delineate FPZs.

To assist users in this process, we developed a separate Shiny app in R, ShinyFPZ, which contains common methods for FPZ classification as well as visualization tools for OpenRES output data.

For users preferring to stay in the QGIS environment for this step, there are also QGIS plugins that contain the appropriate capabilities, such as the Attribute based clustering plugin.

---

Issues

1. Report issues or problems with the software here: https://github.com/jollygoodjacob/OpenRES/issues

2. For questions about the OpenRES plugin, contact: [email protected]{.email}

References

Elgueta, Anaysa, Martin C Thoms, Konrad Górski, Gustavo Díaz, and Evelyn M Habit. 2019. "Functional Process Zones and Their Fish Communities in Temperate Andean River Networks." River Research and Applications 35 (10): 1702--11.

Hestir, Erin L. 2007. "Functional Process Zones and the River Continuum Concept." Center for Watershed Sciences, University of California, Davis, Los Angeles, USA.

Maasri, Alain, James H Thorp, Jon K Gelhaus, Flavia Tromboni, Sudeep Chandra, and Scott J Kenner. 2019. "Communities Associated with the Functional Process Zone Scale: A Case Study of Stream Macroinvertebrates in Endorheic Drainages." Science of the Total Environment 677: 184--93.

Sechu, Gasper L., Bertel Nilsson, Bo V. Iversen, Mette B. Greve, Christen D. Børgesen, and Mogens H. Greve. 2021. "A Stepwise GIS Approach for the Delineation of River Valley Bottom within Drainage Basins Using a Cost Distance Accumulation Analysis." Water 13 (6). https://doi.org/10.3390/w13060827.

Thorp, James H, Martin C Thoms, and Michael D Delong. 2006. "The Riverine Ecosystem Synthesis: Biocomplexity in River Networks across Space and Time." River Research and Applications 22 (2): 123--47.

Thorp, James H, Martin C Thoms, and Michael D Delong. 2010. The Riverine Ecosystem Synthesis: Toward Conceptual Cohesiveness in River Science. Elsevier.

Thorp, James H, Martin C Thoms, Michael D Delong, and Alain Maasri. 2023. "The Ecological Nature of Whole River Macrosystems: New Perspectives from the Riverine Ecosystem Synthesis." Frontiers in Ecology and Evolution 11: 1184433.

Williams, Bradley S, Ellen D'Amico, Jude H Kastens, James H Thorp, Joseph E Flotemersch, and Martin C Thoms. 2013. "Automated Riverine Landscape Characterization: GIS-Based Tools for Watershed-Scale Research, Assessment, and Management." Environmental Monitoring and Assessment 185: 7485--99.