Glossary

Overview

A Free-flowing Rivers Assessment (FRA) is an assessment to determine the connectivity status of rivers by taking into account both natural connectivity as well as fragmentation from infrastructure, such as dams, roads, urban areas, water use.

Figure 1 introduces important concepts, such as the elements of the hydrographic framework, the dimensions of river connectivity and the determination of the river status. The main result of a FRA is a connectivity status index (CSI) representing how well river stretches are connected in the up- and downstream, as well as other directions, given existing infrastructure. The CSI ranges from 0 to 100% on a sliding scale. The index is used in a subsequent step to classify rivers as free-flowing or not.

Principal Steps

In the global assessment, we gathered a team of scientists and practitioners, and first developed an integrated definition of free-flowing rivers (FFR) (Figure 2; step 1) according to multiple aspects of connectivity. Next, we identified five major pressure factors (step 2) that influence river connectivity according to an extensive literature review, and collated data for each factor. These five pressure factors include: (a) river fragmentation; (b) flow regulation; (c) sediment trapping; (d) water consumption; (e) road construction; and (f) urbanization.

We calculated proxy indicators (see Figure 3) for each factor using data from available global remote sensing products, other data compilations, or numerical model outputs such as discharge simulations. We specifically chose indicators that we expect to have substantial influence on connectivity and can be generated using robust global data sets of sufficient quality and consistency between countries and regions. All indicators were calculated for every river reach of the global river network (step 3).

Guided by literature reviews and expert judgement, we iteratively adjusted the weighting of each pressure indicator in a set of scenarios and tested different thresholds to yield a best match between the resulting FFRs and a benchmarking dataset of reported FFRs compiled from literature resources and expert input.

The final selection of weights was applied to a multi-criteria average calculation (step 4) to derive the Connectivity Status Index (CSI) for every river reach (step 5). The CSI ranges from 0% to 100%, the latter indicating full connectivity. Only river reaches with a CSI of >95% were considered as having ‘good connectivity status’ while river reaches below 95% were classified as impacted (step 6). Finally, river reaches were aggregated into rivers, i.e., contiguous flow paths from the source to the river outlet. If a river is above the CSI threshold of 95% over its entire length it is declared to be a FFR. Otherwise, the river as a whole is declared not free-flowing, yet it can maintain a mix of stretches with ‘good connectivity status’ and stretches that are impacted.

Figure 4 further illustrates the different concepts used in an FRA. The baseline river network consists of individual ‘river reaches’ (1–32 in a), defined as line segments separated by confluences (black dots). River reaches can be aggregated into ‘rivers’ according to a ‘backbone’ ordering system, which classifies river reaches as the mainstem or a tributary of various higher orders (b). Following this system, the river network can be distinguished into distinct rivers (1–16 in c), defined as contiguous stretches of river reaches from source to outlet on the mainstem or from source to confluence with the next-order river. CSI values for individual river reaches, as calculated with our model (d). If a value is at or above the CSI threshold (95%), the river reach is declared to have good connectivity status; if it is below the threshold, it is declared to be impacted. If an entire river (as defined in c) has good connectivity status )see panel e), it is defined to be an FFR (blue). A river can be partly above the CSI threshold, and thus contiguous stretches can have good connectivity status (green).

More info

For more detailed information on the FRA methodology, please also refer to:

Grill, G., B. Lehner, M. Thieme, B. Geenen, D. Tickner, F. Antonelli, S. Babu, P. Borrelli, L. Cheng, H. Crochetiere, H. Ehalt Macedo, R. Filgueiras, M. Goichot, J. Higgins, Z. Hogan, B. Lip, M. E. McClain, J. Meng, M. Mulligan, C. Nilsson, J. D. Olden, J. J. Opperman, P. Petry, C. Reidy Liermann, L. Saenz, S. Salinas-Rodriguez, P. Schelle, R. J. P. Schmitt, J. Snider, F. Tan, K. Tockner, P. H. Valdujo, A. van Soesbergen and C. Zarfl (2019). "Mapping the world's free-flowing rivers." Nature 569(7755): 215-221.

Welcome!

This website will provide materials, tools and guidance related to conducting a free-flowing river analysis (FRA). We will first introduce the FFR methodology more generally to provide a large-scale overview of the framework.

We will then focus our attention on a case study for the lower Mekong river basin. We will provide updated data and tools to conduct a FRA provided detailed training material, including video instructions. A web mapping tool is available to showcase the latest results of a tailor made assessment in the Lower Mekong River Basin. With this, conducting an FRA becomes accessible to a wide range of researchers and practitioners.

We furthermore provide guidance on how FRA's can be replicated in other parts of the world, and what steps are necessary to adapt the FRA to local needs.

What is a free-flowing river?

According to our definition, a free-flowing river (FFR) is a river where ecosystem functions and services are largely unaffected by changes to the fluvial connectivity allowing an unobstructed exchange of material, species and energy within the river system and surrounding landscapes (Grill et al., 2019).

FFRs are the freshwater equivalent of wilderness areas and they support many of the most diverse, complex and dynamic ecosystems globally, providing important societal and economic services. As FFRs are under increasing threat from infrastructure development by humans, leading to continued losses of biodiversity and ecosystem functions, we envision a world where the most critical FFRs are valued and protected for the enduring benefit of people, wildlife, and nature. Organisations such as World Wildlife Fund for Nature (WWF) is actively working to advocate for and communicate the importance of FFRs around the world with a vision to safeguard and sustain them into the future (https://www.worldwildlife.org/pages/free-flowing-rivers).

A part of this effort is to continuously monitor and assess the status of Free-flowing rivers worldwide. Under the term "Free-flowing River Assessment (FRA)" our team developed a framework and tool to assess river connectivity and the free-flowing status globally and regionally.

Free Flowing River Assessment (FRA)

A Free-flowing Rivers Assessment (FRA) is an assessment to determine the connectivity status of rivers by taking into consideration both natural connectivity as well as fragmentation from infrastructure, such as dams, roads, urban areas, and water use.

The main result of a FRA is a connectivity index representing how well river stretches are still connected in the lateral, and in the upstream and downstream direction given existing infrastructure. The index is termed Connectivity Status Index (CSI). As such the FRA provides a layer of information that is strictly focused on connectivity, and is therefore not a complete assessment of river health. However the results are meant to be combined and supplemented with other layers, such as species information, water quality or fluvio-geomorphological information to, for example, further assess and identify high-value conservation areas.

A second important result of this assessment is the classification of rivers into either “Free-flowing”, having a “Good Connectivity” status, or as being “Impacted”. This second layer of information helps addressing river connectivity from a "whole river" perspective. We dedicated a complete section to explain how to conduct a FRA for the lower Mekong river.

However, before deep-diving into the procedures and application of our tools in the Lower Mekong, we suggest to read more about the underlying research. A FRA includes many concepts and methods, that interact and play together, producing a rich set of results and statistics. It will help tremendously to first take some time to understand these key concepts.

4.1 Tool Parameters

The tool requires inputs for 17 parameters, with four optional parameters. The parameters are separated into seven categories. Each parameter will be described in further detail below. Additionally, if you are using the example data provided with the toolbox, for each parameter a suggested input will be recommended to provide a tutorial for using the tool.

4.1.1 Configuration File (optional)

This parameter allows users to quickly populate the parameters for the tool using a “config.xls” file that was created in the results folder of a previous execution of the tool (Figure 4.1). If an excel file is input into the parameter, the file is parsed to extract the values for the parameters of the tool. If there are any issues in reading the parameters from the excel file, an error message is displayed. Parameters that are unaffected by the error will be populated, but the parameters that encountered the error will need to be specified manually or corrected in the excel file and loaded again. If you are using the example data, navigate to the data folder provided in the extracted zip folder and select the “tutorial_config.xls” file. Alternatively, the parameters of the tool can be specified manually following the instructions provided with each parameter below. If you are using the “tutorial_config.xls”, the path settings (the file paths to the data) will need to be specified manually, all other settings will be populated automatically.

4.1.2 Load default settings (optional)

This button allows users to apply default settings to all the tool’s parameters. This is the fastest option for running the tool. The example data provided with the tool will be used as the default data. Note that for the default settings to load the required Path Settings, the folder structure will need to be consistent with the structure provided in the zip folder (see ZIP Folder Contents). The user can modify any of the loaded parameter settings if they would like to make adjusts before running the analysis.

4.1.3 Run Options

This parameter allows users to select whether to run a full set of calculations or a subset of the available calculations (Figure 4.2). Click the buttons next to DOF, DOR, SED, or CSI to select the desired options. For the tool to run, at least one run option must be selected. If you are using the example data, select all four options.

4.1.4 Path Settings

Five file paths are required for the tool to run (Figure 4.3). For each required path, navigate to the corresponding data through the browse folder button and select it. If you are using the example data, select the browse folder button and navigate to the folder where the example data was extracted from the zip folder.

4.1.5 General Settings

General settings include parameters that affect multiple calculations (Figure 4.4). For the barrier/dam inclusion field, use the drop down to select the field that is used to indicate which dams are included in the assessment from the “Barrier feature class” that was specified in the “Path Settings”. Values for this field should be 0 or 1. A value of 1 or 0 indicates to use or not to use the barrier, respectively. If you are using the example data, select the field: INC.

Update/Overwrite stream table parameter specifies if the newly calculated values for DOF, DOR, and SED should replace the current values in the corresponding fields (the fields specified in the Degree of Fragmentation Settings, Degree of Regulation Settings, and Sediment Trapping Index Settings). If you are using the example data, select “YES”.

Minimum length threshold is used to excluded small backbone rivers from the calculations. The minimum length threshold is in units of kilometers. Rivers with a length less than or equal to the specified threshold value are excluded, while rivers above the threshold value are included. If you are using the example data, input 100.

4.1.6 Degree of Fragmentation Settings

There are six Degree of Fragmentation Settings, but only four need to be specified (Figure 4.5). The degree of Fragmentation field is used for storing the calculated degree of fragmentation values. Use the drop down to select the desired field from the “Streams feature class” that was specified in the path settings. If you are using the example data, select the field: DOF.

The “use dam level discharge factors” parameter is used to select whether static or dam specific discharge range factors are used in the DOF calculations. From the dropdown, select “YES” or “NO”. The four parameters below “use dam level discharge factors” are disabled until an option is selected, after which the required parameters associated with the parameter selection will become enabled. If “NO” is selected, the algorithm applies static upstream and downstream discharge range factor values to all dams included in the calculations. The “Upstream Discharge Range Factor” and “Downstream Discharge Range Factor” parameters become enabled and are required if “NO” is selected. If “YES” is selected, the algorithm uses the values from the fields specified in the “Dam Level Upstream Discharge Factor Field” and “Dam Level Downstream Discharge Factor Field” parameters to calculate dam-specific DOF results. The “Dam Level Upstream Discharge Factor Field” and “Dam Level Downstream Discharge Factor Field” become enabled and required if “YES” is selected. If you are using the example data, select "NO".

For the “Upstream Discharge Range Factor” and “Downstream Discharge Range Factor” parameters, the value is pre-populated as 5. You can either leave the value as the default of 5 or input a new number to be used for the discharge range factor value. The number must be greater than or equal to 1. A discharge range factor of 1 signifies that there is no effect. There is no upper limit, but a value larger than 10 is not recommended. The standard value is 5, which means that the DOF effect occurs in river reaches with up to five times larger or smaller from the location of the dam. Please review the methodology section of the research article (Grill et al., 2018) for further details. If you are using the example data, leave the value as the default value of 5.

For the “Dam Level Upstream Discharge Factor Field” and “Dam Level Downstream Discharge Factor Field”, select the desired fields from the dropdowns containing the discharge range factor values for each dam. The fields provided in the dropdown are the fields available in the “Barrier Feature Class” that was specified in the Path Settings. If using the Dam level Discharge Range Factors with the example data, two example fields are provided: drf_up and drf_dwn, respectively. The values provided in these fields were created as random integer values and are for demonstrative purposes only.

4.2.7 Degree of Regulation Settings

These two parameters are specific to the Degree of Regulation calculations (Figure 4.6). The Degree of Regulation Field is used for storing the calculated degree of regulation values. Use the dropdown to select the desired field from the “Streams feature class” that was specified in the Path Settings. If you are using the example data, select the field: DOR.

For the “Source Field for Reservoir Storage Volume”, use the dropdown to select the desired field from the “Barriers feature class” that was specified in the Path Settings. The values of this field indicate the storage volume for each dam (Note: units must be in million cubic meters). Please review the methodology section of the research article (Grill et al., 2018) for further details on how reservoir storage volume is incorporated into the degree of regulation calculations. If you are using the example data, select the field: STOR_MCM.

4.1.8 Sediment Trapping Index Settings

The Sediment Trapping Index Field is used for storing the calculated sediment trapping index values (Figure 4.7). Use the dropdown to select the desired field from the “Streams feature class” that was specified in the Path Settings. If you are using the example data, select the field: SED.

4.1.9 Scenario Settings

The scenario settings allow users to define individual scenarios for calculating CSI (Figure 4.8). If the “tutorial_config.xls” file was used in the “Auto-populate parameters using config excel sheet (optional)” parameter, an example scenario called “tutorial” is provided. Typing a name into the parameter “Scenario Name” adds a new scenario to the table. The name provided must be under 10 characters, cannot have spaces or special characters, and cannot start with a number or underscore. It is recommended to use simple names, something like “CSI01”.

Once a scenario has been added to the table, the remaining columns of the scenario must be completed. For the indicator columns, select a field from the dropdown that contains values for the corresponding indicator. The fields provided in the dropdown are from the fields in the “Streams feature class” that was specified in the Path Settings.

The weight columns are used to apply a weighting to the corresponding indicator. Input values ranging from 0 – 100. However, the sum of the six weights must equal 100. The CSI threshold is used for determining the Free-Flowing river status. Values for CSI threshold must be greater than zero and less than or equal to 100%. The “Flood Plain Damp” parameter (fld_damp) determines the strength of the floodplain weighting, which interacts with the RDD and URB pressure indicators. See the methodology section of the research article (Grill et al., 2018) for further details. Values for Flood Plain Damp can range from 0 to 100. The “Filter Threshold” scenario setting is used for filtering extreme outliers. Values for “Filter Threshold” can range from 0 to 100. Filter Threshold is considered an advanced parameter. It is recommended to set the filter threshold to 0.1. The “To Process” scenario setting determines whether the scenario will be processed when the tool is executed. From the drop down, select “Yes” if the scenario should be processed when the tool is run or “NO” to skip the scenario when the tool is run. The “To Export” scenario setting determines if a results table, feature class, and ArcMap document will be created for the scenario in the output folder. From the dropdown, select “YES” to create the files and select “NO” to skip creating the files. If you are using the example data, select the options displayed in Figure 4.8.

4.2 Starting the model

When all the parameters have been populated, click the “OK” button at the bottom of the tool window (Figure 4.9a). If no errors are encountered with the parameters, the tool will start conducting the analysis and the tool dialog window will appear (Figure 4.9b). It is recommended to have the “close this dialog when completed successfully” unchecked so that the messages in the dialog window can be reviewed after the tool has finished running. The dialog window provides messages during the execution of the tool regarding the progression of the tool and information about the analysis (Figure 4.9c). When finished the dialog window will say “succeeded” followed by the date and the elapsed time for the tool to run (Figure 4.9d). The results excel file and an ArcMap document displaying the spatial results will automatically open.

6.1 Customizing Template Symbology

The Template ArcMap document accompanying the tool (i.e., the template.mxd document in the zip folder) is used to apply the symbology to the output maps discussed in the ArcMap Document section of Results & Interpretation. User’s can alter the symbology (e.g., size and colour) of the layers in the template document. Any changes will be applied to new output map results created from running the tool. The structure of the layers and how they are symbolized by default in the template.mxd document is described below.

Other than the “Benchmark Rivers” layer, all the layers contained in the template document are group layers, with sub-layers. These sub-layers make use of definition queries that are programmatically generated by the tool during execution. This allows the output documents to display the data selected by the user for each corresponding layer. For instance, the Barriers Feature class contains two sub-layers: “Included Barriers” and “Excluded Barriers”. Each of these layers has a definition query that is programmatically set based on the “barrier/dam inclusion field”, described in the “General Settings” section. The definition query for the “Included Barriers” layer is set as “[user selected barrier/dam inclusion field] = 1”. Setting the definition query only displays the features in the layer that satisfy the query, in this case only the included barriers will be displayed in this layer.

Many of the stream map layers (i.e., MAP CSI, MAP DOM – IMP, MAP DOM – ALL, MAP DOF, MAP DOR, MAP SED, MAP USE, MAP URB, and MAP RDD) have sub-layers labeled with numbers from 1 – 10, which corresponds to river order. These sub-layers allow for variable line feature width for rivers of different orders, while applying a graduated colour scheme to display the values of the different indices. In this case the streams with a river order of 1 are given a greater width which decreases as river order increases. It is recommended to keep the graduated colour scheme consistent across the sub-layers.

MAP FFR STATUS is slightly different in that it is first broken into three sub-groups: IMP, GOOD, FREE. Each of these sub-groups contains four layers: very long, long, medium, and short. These layer groups display the line features using graduated symbols based on river order, with lower river orders having a greater line width. The colours are then varied between the layer groups with very long having the darkest colour and short having the lightest colour.

6.2 Customizing pre-determined definition queries

Some of the layers make use of definition queries that are not programmatically set and are instead hard-coded into the tool. For instance in the MAP FFR STATUS layer, the definition queries are set to ensure that only rivers with the matching FFR status are displayed in the corresponding layer group. To illustrate, for rivers in the group layers imp, good, and free to be displayed they must have a FFR value of 3, 2, and 1 respectively. Additionally, each layer corresponds to rivers of a certain length with very long, long, medium, and short rivers corresponding to backbone rivers with lengths greater than 1000 km, between 1000 and 500 km, between 500 and 100 km, and less than 100 km, respectively.

In order, to edit these definition queries, the code of the tool will need to be edited. This will require some coding experience and opening the tool in a script editing software, such as idle, pycharm, or visual studio code. Simply right-click the toolbox in the catalog panel of ArcMap and select edit, which will open the tool in the default script tool editor specified under the geoprocessing options of ArcMap. To navigate to the area of code to edit the definition queries, you can search for “Customizing The Map Template”, which will bring you to a variable in the code called “template_data_list”. If you look for the entries corresponding to the “MAP FFR STATUS” layer, you can edit the text string at the end of each line to adjust the definition queries for these layers (Figure 6.1). Once done, save the code and refresh the tool in ArcMap. The next time the tool produces output map results they will be generated using the new definition queries specified in the code.

1.1 Introduction to Lower Mekong FRA

1.2 Overview of the Lower Mekong Free-flowing Rivers Assessment (FRA)

The Lower Mekong Free-flowing River assessment has been conducted between January and September 2022 by WWF-US, WWF-Greater Mekong and Confluvio. The following describes the general steps of the assessment and the specific objectives and outcomes

1.2.1 Lower Mekong FRA - Steps

1. Preparation and data review: The first step in conducting a Free-flowing river assessment (FRA) entails the review of existing (global) data and methodology, as well as a discussion with local groups, experts, and stakeholders of possible methods refinements and enhancements to better represent the local context.

2. Data processing: Simultaneously, a review of available data layers that could replace global data layers should be conducted with the help of regional experts. The data review is followed by data collection and subsequent data processing using Geographic Information Systems (GIS). The result of the data processing are various layers representing pressures on natural connectivity (dams, roads, urban areas, etc.). Next, the data sets collected and processed are translated to the river reach scale using the necessary data processing steps.

3. Model: Once the relevant datasets and indicators are in place (at the river reach scale), the FRA model, a set of python tools bundled in a Python Toolbox is used to calculate:

   • the Connectivity Status Index (CSI) of the river reaches, by combining the pressure variables,

   • the Free-flowing status (free-flowing or not), based on the CSI values and a threshold,

   • benchmarking statistics, which determine if the settings used to create the outcome is reasonable based on pre-defined benchmarking rivers,

   • an ArcMap document providing production-ready maps for various data layers

   • a series of statistics as Excel output

4. Refinements: The production of summary statistics and maps is typically followed by several cycles of refinement, based on stakeholder discussion, including local river experts.

1.2.2 Lower Mekong FRA - Specific Outcomes

A free-flowing river assessment should engage a variety of stakeholders into a discussion of the importance of free-flowing rivers and river connectivity, their benefit and their vulnerabilities to local pressures, such as fragmentation. Aside from this, the specific products of the assessment include:

• High-resolution hydrographic framework including river networks and network metrics; flow direction maps; hydrologic connectivity information and attribute information. This framework can be reused for other assessments, for example for strategic environmental assessment of hydropower development options.

• Free-flowing rivers map for the study area, including maps of underlying metrics (i.e., degree of fragmentation, degree of regulation, sediment loss, road and urban development, water abstraction from rivers, and connectivity status index).

• Free-flowing status report, including statistical analysis of extent of free-flowing rivers (number, length, connectivity to ocean)

• Adjusted Python tool: Although extensive modifications to the main toolbox are rarely required, some modification may be necessary in the template and configuration files to tailor the analysis for specific needs. In the case of the Lower Mekong basin assessment, the analysis was conducted at the scale of the entire Mekong Basin, and a few simple modifications of the source code were necessary to focus the maps and statistical analysis to only the area of interest - the lower Mekong basin. See this section for tool downloads.

1.2.3 Lower Mekong FRA - Training and training tools

To aid current and future FFR assessments, comprehensive documentation, guidance materials and training tools for understanding the FFR methodology as well as for implementing the approach at basin, national or regional scales will be developed and/or updated.

The training tools and materials will be developed in conjunction with applying the FFR methods in the Lower Mekong Basin in Vietnam and Cambodia as an additional case study. To allow current and future users to learn more about completed assessments, selected examples will be described and added as case studies to the training materials, more clearly explaining how to produce and use maps, results, and figures, among others.

Training tools

The results of this assessment were translated to an online interactive tool to explore the results in detail. The Lower Mekong Free-flowing River Atlas App provides data layers summarizing the six individual pressure indices, as well as the combined indices, including the Connectivity Status Index (CSI), the Free-flowing River status (FFR) and the Dominant pressure factors (DOM). The user can interact with the map interface by zooming and panning, by filtering layers based on their attributes, and by clicking on river reaches and dams to receive key statistics. The tool is accessible through this page or directly at https://free-flowing-lower-mekong.web.app/

Workshop

Our team participated in a Regional Joint Workshop termed 'Rivers of the Lower Mekong Region', supported and organized by USAID, Stockholm Environment Institute, World Wildlife Fund for Nature (WWF), and the Asian Disaster Preparedness Center (ADPC).

The goal of this workshop is to bring together stakeholders from across the region to build knowledge and technical capacities, aid the development of new networks and partnerships that support the sustainable development and use of the rivers, and support the inclusion of those most impacted and least consulted when it comes to the development of the Lower Mekong Region. Topics covered include learning about the importance of river connectivity and its threats, tools that can be used for river planning, guidance to address governance and transparency issues, and potential entry points for inclusion in decision-making and advocacy efforts for CSOs, INGOs, and research institutions.

At the workshop we introduce the Free-flowing rivers Initiative, provide an overview of the methodology, conduct a technical session to introduce a Toolbox for Geographic Information Systems (GIS).

1.2.3 USAID - Wonders of the Mekong

This projects was realized with the generous support of the USAID Wonders of the Mekong project. The Wonders of the Mekong project conducts applied research, builds capacity, and develops outreach and communications products to highlight the economic, ecological, and cultural values of biodiversity and ecosystem services associated with the Lower Mekong River. The outputs and resulting products, developed as an integrated package, will lead to better protection of a vibrant and healthy Lower Mekong system.

References

Alcamo, J., Döll, P., Henrichs, T., Kaspar, F., Lehner, B., Rösch, T., Siebert, S. (2003) Development and testing of the WaterGAP 2 global model of water use and availability. Hydrological Sciences Journal 48, 317-337.

Borelli, P., Robinson, D. A., Fleischer, L. R., Lugato, E., Ballabio, C., Alewell, C., Meusburger, K., Modugno, S., Schütt, B., Ferro, V., Bagarello, V., Van Oost, K., Montanarella, L., Panagos, P. (2017) An assessment of the global impact of 21st century land use change on soil erosion. Nature Communications 8, 2013 (2017).

Döll, P., Kaspar, F., Lehner, B. (2003) A global hydrological model for deriving water availability indicators: model tuning and validation. Journal of Hydrology 270, 105-134.

Grill, G., Lehner, B., Lumsdon, A.E., MacDonald, G.K., Zarfl, C., Liermann, C.R. (2015) An index-based framework for assessing patterns and trends in river fragmentation and flow regulation by global dams at multiple scales. Environmental Research Letters 10, 015001.

Grill, G., Lehner, B., Thieme, M., Geenen, B., Tickner, D., Antonelli, F., Babu, S., Cheng, L., Crochetiere, H., Filgueiras, R., Goichot, M., Higgins, J., Hogan, Z., Lip, B., McClain, M., Meng, J.-H., Mulligan, M., Nilsson, C., Olden, J.D., Opperman, J., Petry, P., Reidy Liermann, C., Saenz, L., Salinas-Rodríguez, S., Schelle, P., Snider, J., Tockner, K., Valdujo, P.H., van Soesbergen, A., Zarfl, C. (2018) Assessing global river connectivity to map the world’s remaining free-flowing rivers. in review.

Lehner, B., Grill, G. (2013) Global river hydrography and network routing: baseline data and new approaches to study the world's large river systems. Hydrological Processes 27, 2171-2186.

Lehner, B., Verdin, K., Jarvis, A. (2008) New global hydrography derived from spaceborne elevation data. EOS, Transactions of the American Geophysical Union 89, 93.

Myung Sik Cho, Jiaguo Qi, Quantifying spatiotemporal impacts of hydro-dams on land use/land cover changes in the Lower Mekong River Basin, Applied Geography, Volume 136, 2021, 102588, ISSN 0143-6228, https://doi.org/10.1016/j.apgeog.2021.102588.

The Free-flowing River Toolbox for ArcGIS Desktop is a tool to conduct a Free-flowing river analysis (FFR) - replicating the methodology described in Grill et al. (2019): Mapping the world's free-flowing rivers. The study can be downloaded at this link:

https://hydrolab.io/wom/Grill et al 2019-Mapping the world's free-flowing _Author copy.pdf

The original source code that describes the methodology, as well as global data is available at this link. However, the user is encouraged to use the newer, more flexible FFR Toolbox that can be launched from a Geographic Information System.

Tutorial

A tutorial, including video guidance, is provided to guide the user through the installation and the use of the tool at this page.

Download the tool

There are two different versions:

The original version covers the entire Mekong River basin. It can be downloaded from this link:

An alternative version has been developed to focus the analysis to the Lower Mekong river basin. The link to download this version is here:

Join us on September 29th for an interactive training session on using the Free-flowing Rivers toolbox.

* Workshop materials will be distributed here after the session *

5.1 Results Folder

A results folder is created in the output folder specified in the “Path Settings”. The time stamp for when the tool was run is appended to the results folder to differentiate the results of separate runs of the tool. The file structure of the results folder is described below and shown in Figure 5.1.

• STAT: This folder contains the tabular results of the analysis in an excel workbook. See the Tabular Data section below for further details regarding these results.

• STATS_CSI: This folder contains a stats_csi.dbf file. This table contains descriptive statistics of the CSI values produced from the different scenarios. These statistics can be used to conduct a sensitivity analysis. See the methodology section of the research article (Grill et al., 2018) for further details.

• CSI.gdb: This file geodatabase contains the spatial data results of the analysis. See the Spatial Data section below for further details regarding the spatial data results.

• ArcMap Document (.mxd): For each scenario conducted in the analysis, an ArcMap document is created in the results folder to display the results visually. See the ArcMap Document section below for further details.

• config_[stamp].xls: This excel file is the config file created from running the analysis. The config file is stored with the results to document the files and parameter settings used to create the results. The date and time stamp for when the tool was run is appended to the file name to associate the config file with the results. Additionally, the config file can be used in the “Auto-populate parameters using config excel sheet (optional)” parameter to auto-populate the tool’s parameters to rerun the analysis or make minor adjustments and run a new analysis with modifications.

5.2 Tabular Data

In the STAT subfolder of the results, an excel workbook file is created to store the tabular data from the analysis. Each sheet of the workbook provides different statics and insight into the analysis. Each sheet and the values presented are described in further detail in this section.

Global_stats: This sheet provides statistics regarding the river reaches included in the analysis. Each row of the sheet corresponds to one scenario. Each column of the Global_stats sheet is explained below.

• Scenario columns: The first 18 columns of the sheet are a duplicate of the scenario parameters used when running the tool. This is to serve as documentation for the scenario settings.

• Sce_name: The name of the scenario that the results are associated with.

• Count_reaches: The total number of river reaches considered in the analysis.

• Count_reaches_affected: The number of river reaches affected by a loss of connectivity.

• Perc_reaches_affected: The percent of affected river reaches out of all reaches considered in the analysis.

• Mean_CSI: The mean connectivity status index.

• Count_nff: The number of river reaches that were not free flowing.

• Perc_nff: The percent of non-free-flowing river reaches out of the total number of river reaches in the analysis.

• Bench_match: The number of benchmark rivers that the model determined to be free-flowing.

Global_dom: This sheet provides the number of river reaches in the entire input streams dataset impacted by the different pressure indicators. Each row corresponds to a dominant pressure indicator in one scenario. Each column of the Global_dom sheet is explained below.

• Stamp: The date and time stamp of when the tool was initialized.

• Sce_name: The name of the scenario that the results are associated with.

• Pressure: Identifies the dominant pressure.

• Number of reaches: The number of impacted reaches where the associated pressure is the dominant pressure.

Bench_dom: This sheet reports on the model’s performance with regards to the benchmark rivers. If the model produces results indicating a benchmark river is not free flowing, information for the benchmark river will be provided on this sheet. Each row of the sheet corresponds to a pressure indicator impacting the reaches of a benchmark river. Each column of the Bench_dom sheet is explained below.

• Stamp: The date and time stamp of when the tool was initialized.

• Sce_name: The name of the scenario that the results are associated with.

• FFR_ID: The benchmark river identifier.

• Rivername: The name of the benchmark river.

• Pressure: The dominant pressure for 1 or more impacted reaches of the benchmark river.

• Number of reaches: The number of impacted reaches of the benchmark river.

• Source: The source that identified the benchmark river as a free-flowing river (e.g., an expert or research article).

River_stats_1: This sheet provides the results for rivers. The rivers are grouped based on river length categories and free-flowing status. A variety of statistics are provided for the rivers in each river length category. Each column of the River_stats_1 sheet is explained below.

• CON_ID: An integer identifying the continent where the rivers are located.

• LCAT: The length category of the rivers (10 = rivers between 10 km – 100 km; 100 = rivers between 100 km – 500 km; 500 = rivers between 500 km – 1000 km; 1000 = rivers longer than 1000 km).

• CAT_FFR: The free-flowing river category (1 = free-flowing ; 3 = non-free-flowing).

• SCE: The name of the scenario that the results are associated with.

• VOLUME_TCM: The total volume of the rivers measured in thousand cubic metres.

• LENGTH_KM: The total length of rivers.

• NUM: The number of rivers.

• BB_OCEAN: The number of rivers that are connected to the ocean.

River_stats_2: This sheet is very similar to River_stats_1, however, an additional free-flowing river category has been added: good status. This additional category was added to identify rivers that are no longer free-flowing but have some long stretches that maintain good connectivity. Each row corresponds to a river length category. Each column of the River_stats_2 sheet is explained below.

• CON_ID: An integer identifying the continent where the rivers are located.

• LCAT: The length category of the rivers (10 = rivers between 10 km – 100 km; 100 = rivers between 100 km – 500 km; 500 = rivers between 500 km – 1000 km; 1000 = rivers longer than 1000 km).

• CAT_FFR: The free-flowing river category (1 = free-flowing ; 2 = good connectivity status; 3 = non-free-flowing).

• SCE: The name of the scenario that the results are associated with.

• VOLUME_TCM: The total volume of the rivers measured in thousand cubic metres.

• LENGTH_KM: The total length of rivers.

• NUM: The number of rivers.

• BB_OCEAN: The number of rivers that are connected to the ocean.

River_stats_good: This sheet provides results describing the rivers with good connectivity status (i.e., rivers that are considered non-free-flowing, but with long stretches that maintain good connectivity). The rivers are grouped into length categories and information regarding each length category is provided. Each row corresponds to a river length category. Each column is described below.

• CON_ID: An integer identifying the continent where the rivers are located.

• LCAT: The length category (10 = rivers between 10 km – 100 km; 100 = rivers between 100 km – 500 km; 500 = rivers between 500 km – 1000 km; 1000 = rivers longer than 1000 km).

• SCE: The scenario name.

• VOLUME_TCM: The volume of rivers measured in thousand cubic metres.

• LENGTH_KM: The length of rivers measured in kilometres.

• NUM: The number of rivers.

List_of_FFRs_###: This sheet provides a list of free-flowing rivers and some additional information about the rivers. Only free-flowing rivers that are longer than the min_length parameter specified in the General Settings are reported on this sheet. The min_length value is appended to the end of the sheet name (i.e., the min length value replaces the “###” in the sheet name “List_of_FFRs_###”) to indicate what min_length value was specified as a cut off. Each row represents a river. Each column is described below.

• SCE: The name of the scenario that the results are associated with.

• CON_ID: An integer identifying the continent where the rivers are located.

• BAS_NAME: The basin name where the river is located.

• BB_ID: An integer value identifying the river.

• BB_NAME: The name of the river.

• LENGTH_KM: The length of the river in kilometres.

• RIV_ORD: The river order.

• BB_OCEAN: A binary field indicating whether river is connected to the ocean (value = 1) or not connected to the ocean (value = 2).

5.3 Spatial Data

The results are also created in the form of a line feature class, stored in the CSI.gdb. For each scenario a separate line feature class is created in the geodata base. These feature classes contain all the fields from the input streams feature class specified in the Path Settings as well as six additional fields that contain the results of the CSI analysis. The fields containing the results of the CSI are described below.

• CSI: The connectivity status index values calculated for the scenario the data is associated with. Values can range from 0 – 100.

• CSI_D: The dominant pressure index if the river reaches connectivity is impacted (i.e., CSI less than 100).

• CSI_FF: A binary field indicating if the river reach is free-flowing (value = 1) or not free-flowing (value = 0).

• CSI_FF1: Indicates if the river is free-flowing (value = 1) or not free-flowing (value = 3).

• CSI_FF2: Indicates if the river is free-flowing (value = 1), has good connectivity status (value = 2), or is not free-flowing (value = 3). Good connectivity status indicates that a river is not free flowing but has large segments that are considered free-flowing.

• CSI_FFID: The free-flowing river identifier.

5.4 ArcMap Document

Lastly, for each scenario an ArcMap document (.mxd) is created in the results folder using the scenario name and time stamp. The ArcMap documents contain the input and output geodata used in the analysis with symbology applied to intuitively display the results graphically. Each layer of the ArcMap document is described in further detail below.

5.4.1 Map CSI

The MAP CSI layer displays the Connectivity Status Index values for the river reaches included in the analysis (Figure 5.2). The symbology values assume a CSI threshold of 95%, but the symbology can be adjusted by the user if desired. The MAP CSI layer is a layer group containing 10 layers, each corresponding to the river order. The colour symbology is applied using the same categories across the 10 layers, but the width of the line segments is greater for larger river reaches.

5.4.2 Barriers

The Barriers layer is a layer group containing two layers: Included Barriers and Excluded Barriers (Figure 5.3). These layers apply the same size symbology with larger symbols representing barriers with a greater storage volume (field: STOR_MCM). However, different colours are applied to the barriers that were included in the analysis (orange) and barriers that were excluded from the analysis (grey) based on the barrier inclusion field specified in the General Settings when the tool was run.

5.4.3 MAP DOM

There are two MAP DOM layer groups: MAP DOM – IMP and MAP DOM – ALL. These layers groups display the dominant pressure indicators for the river reaches (Figure 5.4). MAP DOM – IMP displays only the impacted rivers reaches (rivers with a CSI below the CSI threshold) and their dominant pressure indicator. MAP DOM – ALL displays all the river reaches and the dominant pressure indicator. The symbology is applied the same across the MAP DOM – IMP and MAP DOM – ALL, with different colours corresponding to the dominant pressure indicators and the width of the line segments representing the size of the river reaches.

5.4.4 MAP FFR STATUS

MAP FFR STATUS is a layer group that indicates river reaches that are classified as free flowing, having good connectivity status, or impacted (Figure 5.5). The colour symbology of this layer varies based on connectivity category and length of the river. Impacted reaches are in red, good connectivity status reaches are in green, and free-flowing reaches are in blue. The saturation of the colour decreases as river length decreases. The width of the line segments is greater for larger river reaches.

5.4.5 Benchmark Rivers

The Benchmark Rivers layer displays the benchmark rivers feature class that was provided in the “Path Settings” (Figure 5.6).

5.4.6 Pressure Indicators

The pressure indicator layers (i.e., MAP DOF, MAP DOR, MAP SED, MAP USE, MAP URB, and MAP RRD) display the values for each indicator that was specified for use in the CSI calculations (Figure 5.7).

USAID Wonders of the Mekong: Lower Mekong Free-flowing River App

The purpose of this tool is explore the results of Free-flowing river assessment for the Lower Mekong Basin in 2022 conducted by WWF-US, WWF-Greater Mekong, Dr. Doan Van Binh, Vietnamese-German University, Vietnam, and Confluvio. This project was realized with the generous support of the USAID Wonders of the Mekong project. The Wonders of the Mekong project conducts applied research, builds capacity, and develops outreach and communications products to highlight the economic, ecological, and cultural values of biodiversity and ecosystem services associated with the Lower Mekong River.

This tool is currently under development. Users exploring the current draft may experience slow load times or possible bugs. We please ask for your patience if you encounter any issues, thank you.

Link to map tool

Functionality of Tool

The following describes the main functions and components of the tool.

Theme panel

The data in the prioritization tool is organized into two themes and each theme contains multiple data layers. The main theme is the "FFR" theme, showing the layers that resulted from the assessment. The second theme, "Atlas", contains additional data layers that are of interest in the region. We plan on adding additional layers, as the tool develops.

Layer Panel

The layer panel provides the individual data layers viewable in the tool. The top layer displays the Barriers data used in this assessment which can be toggled on or off independently from the river reach layers.

River reach layers can be turned on or off one at a time in the map display by toggling the eye symbol associated with each layer. The map display can then be used to visually explore the layer by zooming and panning in the map interface.

Filter function

With the filter function, the user can identify river reaches that meet specific criteria based on attribute values of one or multiple layers. Using the button, the user can adjust the range filter to only display map results in the output that meet the criteria specified in the range filter, other river reaches become grey.

Range filters from multiple layers can be applied simultaneously. For example, the user may filter river reaches with a discharge above 100 cubic meters per second in the "Natural Discharge" layer of the "Atlas" theme, and then additionally filter the CSI layers to highlight only reaches above 95%. This selection reveals medium-to-large rivers with good connectivity.

Note that specific values can be entered manually in the 'min' and 'max' field of the layer. When a range filter is applied to a layer, the layer will be highlighted in yellow. The user also has the option of resetting the filter to the default range by selecting the circular arrow symbol next to the range filter.

Map panel

The map panel provides a 'slippy map' interface and allows the user to zoom and pan the map content. With a mouse-click on a river feature or barrier, a pop-up provides key information of the clicked feature underneath the cursor.

The map contains a satellite imagery basemap with thematic features that was specifically created for this project. It emphasizes water and conservation features and subdues roads and cities.

On the top left, a legend is provided with a short description of the layer, as well as the value categories for the specific layer.

The top-right side of the map panel provides functions to zoom, maximize the map interface to the full screen, and to return to the default zoom showing the Lower Mekong Region.

With a mouse-click on a river feature or barrier, a pop-up provides key information of the clicked feature.

Contact

For questions or comments, please get in touch with the developer of this tool at 'info at confluvio.com'. Thank you!

Four feature classes are provided with the toolbox as examples of the required data for running the tool. These files are stored in the example_data.gdb of the data folder. Each file is described in further detail below and additional tables are provided in the annex to describe the fields contained in the different feature classes.

3.1 Stream network

The streams feature class is a stream network. Each row of the database represents a river reach. In this context, we define a river reach as a line segment between two confluences; a river stretch as two or more contiguous reaches but not a full river; and a river as an aggregation of river reaches that form a single-threaded, contiguous flowpath from headwater source to river outlet. The river outlet can represent either the river mouth at the ocean; a terminal inland depression; or the confluence with a larger river.

To create the streams feature class, a stream network delineation has been extracted from World Wildlife Fund’s HydroSHEDS database (Lehner et al. 2008) at a grid resolution of 15 arc-seconds (approx. 500 m at the equator); for more information please refer to the Technical Documentation at http://www.hydrosheds.org. In HydroSHEDS, rivers were defined to start at all pixels where the accumulated upstream catchment area exceeds 10 km2, or where the long-term average natural discharge exceeds 100 liters per second, resulting in a total global river length of 35.9 million kilometers (excluding Antarctica). Rivers are broken into reaches at all confluences, creating 8,477,883 million individual river reaches with an average reach length of 4.2 km. Each river reach is linked to a polygon of its contributing hydrological sub-catchment, with an average area of ~12 km2.

There are many fields in the streams feature class, but not all of them are actually required for the analysis. The required fields in the streams feature class are necessary to preform a number of critical aspects in conducting the FRA. A series of fields are used to conduct the routing of the stream network. Examples of these fields include the NOID, NUOID, and NDOID, which correspond to the Network Object ID, the Upstream Network ID, and Downstream Network ID, respectively. Additionally, some fields in the streams feature class are required for conducting calculations in the FRA. For instance, the streams feature class contains fields to store indicator values for the degree of regulation (DOR), degree of fragmentation (DOF), sediment trapping index (SED), urban areas (URB), road density (RDD), and water consumption (USE). Lastly, the output values of theses calculations are stored in user specified fields of the streams feature class, which are used to display some of the output spatial data results discussed in the Results & Interpretation section. The table in the annex here lists fields in the streams feature class and provides a description of the field values.

3.2 Barriers

The barriers feature class is a point feature class representing the locations of barriers to water flow along the stream network. The FFR barrier database (Grill et al.2019) was used as a blueprint for this analysis, as it contains the necessary fields required to run the FRA. The global data contains only larger projects that existed approximately at or before 2018. Updating the barrier database was necessary, and the main sources of dam and reservoir data in this project were the following:

• Stimson Tracker Database: Very good and complete database including reservoirs.

• Mekong River Commission. Technical reference database with detailed technical information

• GeoDAR: Georeferenced global dam and reservoir dataset for bridging attributes and geolocations

Other sources used for validation purposes:

1. CSIS Reconnecting Asia Project Database. Reconnecting Asia’s database tracks seven types of infrastructure projects–power plants, roads, rails, ports, intermodal, transmission, and pipelines–active across the Eurasian supercontinent since 2006. See https://reconasia.csis.org/reconnecting-asia-map/

2. WRI Power Database: Download here. Citation: Global Energy Observatory, Google, KTH Royal Institute of Technology in Stockholm, Enipedia, World Resources Institute. 2018. Global Power Plant Database.

3. Water, Land and Ecosystems (WLE) 2016: Greater Mekong Dam observatory. https://wle-mekong.cgiar.org/changes/our-research/greater-mekong-dams-observatory/ (CGIAR)

4. Open Street Maps (OSM) and Google Maps: Both provide a way to verify existence and location of dams. Tip: use this extension to easily switch from ArcMAP to Google Maps or Bing maps.

The barrier database has many fields, but not all of them are actually required for the analysis. The table in the annex here lists fields in the barriers feature class. A few of the important fields are BAS_ID, GOID, STOR_MCM, PCAP_MW and INC. Each of these are described below.

BAS_ID: This field identifies which hydroBASIN the barrier is in.

GOID: GOID stands for “Global Object Identifier”. Each dam has a value in the field called GOID. The value corresponds to the GOID of the stream nearby. For the model to recognize on which river the dam is located, each dam must have the correct GOID value (i.e., the one that is corresponding to the GOID of the river reach). Note: The position of the dam feature doesn't need to be close to the river reach. The value in this field is used to identify the position of the dam in the stream network.

STOR_MCM: This field contains the storage volume, if available. The units are “million cubic meters”. Note: If you don’t have storage capacity, you could use the power production capacity to derive a rough estimate (Get in touch for details). If the storage capacity is zero, the DOR will not be calculated for the dam, and only the fragmentation will be calculated.

PCAP_MW: This field contains the power capacity of the barrier, if available. The units are “megawatts”.

INC: This field is a binary field (values of 0 or 1) indicating whether to include the barrier (value = 1) or exclude the barrier (value = 0) from the analysis. Variations of this field are included in the annex table here, which can be used to run different scenarios. For instance, variations of this field might indicate to include only existing dams, dams under construction, or planned dams as well.

3.3 Benchmark Rivers

The benchmark rivers feature class is composed of line features and represents rivers that were positively identified as free-flowing. The benchmarking dataset of reported FFRs was compiled from literature resources and expert input.

In the global study, we created 100 scenarios where we manipulated the individual weights within the plausible ranges and compared the results of each scenario to the set of benchmark rivers reported to be free-flowing. For the final CSI application, we selected the weights of the scenario that best reproduced the FFR status of the benchmark rivers. The same procedure was applied in this case study.

The Benchmark Rivers feature class contains 5 fields described below.

• GOID: GOID stands for “Global Object Identifier”. The value in this field is derived from the streams feature class to ensure the GOID of the benchmark river matches the corresponding GOID of the river reach in the streams feature class. For the model to recognize which river is the benchmark river, each benchmark river must have the correct GOID (i.e., the GOID that corresponds to the GOID of the river reach).

• BB_ID: BB_ID stands for "backbone identifier". The value in this field is a integer identifying the backbone river the river reach corresponds to. The BB_ID is derived from the streams feature class and the values for the same river reach must match between the two datasets.

• FFRID: FFRID stands for "free-flowing river identifier". This field contains a integer identifier for the benchmark river. Each benchmark river must have a unique FFRID value.

• Name_Expert: Provides the name of the benchmark river. The name must be consistent across the same FFRID.

• BENCH_SRC: Provides the source used to designate the river as a benchmark river. This can be the name of a researcher or a publication, a category such as "Expert" or "Publication" etc. The field is used to analyze the results of the benchmarking separately by source name or category.

A benchmark river can be a full river, i.e. all reaches with the same BB_ID are included or part of a river. Only the segments that are benchmark rivers are included in this feature class, the rest are deleted.

3.4 Lakes

The lakes feature class is a point feature class representing the location of lakes through out the study area. The fields contained in the lakes feature class are mainly used for establishing positional relationships of the lake features within the stream network and calculating the Sediment Trapping Index (SED).

The lakes feature class is based off of HydroLAKES, which contains many of the fields required in the lakes feature class for running the FRA. A few new fields have been added to the HydroLAKES point features for use in the FRA analysis. These fields include:

• GOOD: This is a binary field indicating if the lake is a reservoir identified in the global georeferenced database of dams (GOODD) database.

• GOID: GOID stands for “Global Object Identifier”. Each lake feature has a value in the field called GOID. The value corresponds to the GOID of the stream nearby. For the model to recognize on which river the lake is located, each lake must have the correct GOID value, the one that is corresponding to the GOID of the river reach. Note: You don't have to move the lake close to the river reach. You only need to set the value in this field correctly.

• SED_ACC: This field provides the sediment accumulation, measured in tons per year. The values in this field come from Borrelli et al., (2017).

• IN_STREAM: This is a binary field that indicates if the lake is connected to the stream network.

• IN_CATCH: This a binary field indicating if the lake is in a catchment of the stream network.

A brief description of the fields in the lakes feature class is provided in an annex table here.

Key References

GeoDAR: Wang, J., Walter, B.A., Yao, F., Song, C., Ding, M., Maroof, A.S., Zhu, J., Fan, C., Xin, A., McAlister, J.M., Sikder, S., Sheng, Y., Allen, G.H., Crétaux, J.F., Wada, Y. (2021) GeoDAR: Georeferenced global dam and reservoir dataset for bridging attributes and geolocations. Earth Syst. Sci. Data Discuss. 2021, 1-52.

2.1 Requirements

The components to set up the analysis are as follows:

• the data and code, which can be downloaded in this section

• the software ArcGIS 10.6 or higher, preferably ArcGIS 10.8.2

• a program that can read .xlsx files, for example Microsoft Excel, or LibreOffice.

• Optional: the ArcGIS 64-bit geoprocessing module if you are processing larger regions. This module is part of the ArcGIS installation package and contains the 64-bit python geoprocessing scripts.

The following steps are required to install the toolbox:

1. Extract the zip folder to a directory on your computer (Figure 2.1a).

2. Open ArcMap and access the ArcToolbox window (Figure 2.1b).

3. Right-click in the ArcToolbox window and select “Add Toolbox” (Figure 2.1c).

4. In the “Add Toolbox” window, navigate to the file directory where the zip folder was extracted and select the FFRA_v001 toolbox (Figure 2.1d).

2.3 ZIP Folder Contents

The contents of the zip file provided through the dropbox link are shown in Figure 2.1a and are briefly described below.

Data Folder: Contains the data used in the tutorial for running the FRA. The data folder contains:

• example_data.gdb: A file geodatabase containing four feature classes: barriers, benchmark_rivers, lakes, and streams. Each of these feature classes is explained in further detail in the next section, 3 Input Data.

• tutorial_config.xls: Provides an example config that can be used to auto-populate the tool’s parameters for the tutorial.

output folder: This folder is provided as an empty folder and is intended to be set as the output location of the tool’s results. A different folder can be specified in the “Path Settings” as the output folder if desired.

FFRA_v001.pyt: Contains the Free_Flowing_Rivers_Analysis tool.

• Free_Flowing_Rivers_Analysis: The tool used to conduct the free-flowing river analysis.

input_data.mxd: An ArcMap document that can be used as a working document for conducting analyses. To make getting started a bit easier, the document has the example data added to the Table of Contents and the FFRA_v001 tool box has been added to the ArcToolbox menu.

template.mxd: An ArcMap document file that contains the symbology to be applied for output ArcMap documents displaying the spatial data outputs of the tool. It is recommended not to remove this document from the source directory (the folder containing FFRA_v001.pyt) as changes may affect the tool’s ability to display the spatial data outputs. User’s can customize the output map results by altering the template document to their liking. Editing the template document is covered in the Tool customization section.

7.1 Examples of previous applications of the FFR methodology

Aside from the global study, the FFR methodology has been applied in a variety of different ways. The FFR methodology can a) be "downscaled", meaning that the global data sources are replaced with local data sources, and that new benchmark rivers are defined, and weighting may be updated based on a sensitivity analysis or b) the results of the FFR analysis (global or local) could be combined with other data layers, supporting new types of analysis.

Examples:

1) Applying the CSI for monitoring connectivity in protected areas in Canada.

The CSI, calculated at the river reach scale, can be aggregated to larger spatial scales. For this assessment, we aggregated the CSI to protected area, by calculated a weighted average of all river reaches inside a protected area.

2) High-Conservation Value River Assessment in Nepal (USAID - PAANI project)

In this application we combined the free-flowing status of rivers with an assessment of freshwater-related ecosystem values, and a water quality pressure assessment. The map below shows area of high/low connectivity, high/low water quality pressures, and the color intensity indicates increasing importance of freshwater values.

3) Identification of Water Resource Protection Areas in Zambia

4) Hydropower development scenarios in the Maranon river basin, Peru

Here, the CSI and the FFR status is used as indicators of impact as part of a system-scale hydropower planning model. The goal of this assessment was to identify future hydropower portfolios that minimize the impact of hydropower development on the Maranon river in Peru.

4) Linking FFR with freshwater connectivity corridors in the Amazon river basin

For this research on important migratory corridors in the Amazon Basin, we combined the FFR status of rivers with the mapping of migratory corridors of long-distance migratory fishes, dolphins and turtles to identify important freshwater connectivity corridors

7.2 General steps in conducting a regional FFR application

Preparation and data review: review of existing (global) data and methodology; discussion with local groups, experts, and stakeholders of possible methods refinements and enhancements to better represent the local context; review of available data layers that could replace global data layers should be conducted with the help of regional experts

Data processing: The data review is followed by data collection and subsequent data processing using Geographic Information Systems (GIS). The result of the data processing are various layers representing pressures on natural connectivity (dams, roads, urban areas, etc.) at river reach scale.

Modeling: Once the relevant datasets and indicators are in place (at the river reach scale), the FRA model, a set of python tools bundled in a Python Toolbox is used to calculate or produce maps and statistical outcome

Refinements: The production of summary statistics and maps is typically followed by several cycles of refinement, based on stakeholder discussion, including local river experts.

7.3 Common data preparation tasks

7.3.1 Barriers

The first step, and perhaps the most important one is typically to update the barrier database. Find the most reliable data layers and combine them into a set of point features. Identify and remove duplicate barriers, but review the attribute information first. If the attribute information is different between duplicate projects it is worthwhile choosing the most reliable, or even merge the two duplicates and only maintain reliable attributes form each duplicate to be transferred to the new barrier.

The barrier feature class contains a set of barrier points. The reservoir of a dam is not required to run the FFR analysis, however some attributes of the reservoirs must be linked to the barrier point database, e.g., Storage capacity (STOR_MCM).

Often times, barrier databases do not have precise locations, which can create uncertainty about its location in the river network. Within the FFR framework, barrier points are mapped to corresponding river reaches. It is required to assign each barrier point a values that identifies the river reach (Field: GOID) at which the barrier is located. Since the point is mapped to the river reach as a whole, the barrier point does not have to be located with great geographical precision, nor must its location be snapped to a river reach using GIS. However, it is important to assign the correct river reach (Field: GOID). To ensure this, each barrier must be manually verified, using the available barrier attributes.

For example, if the barrier has a field indicating upstream catchment area, it can be used to determine the correct river reach, using the upstream catchment area attribute of the river network. Similarly, if the barrier database indicates the river name at which the barrier is located, the user can find the river using satellite images and approximate its location.

7.3.2 Updating Waterfalls attribute

The waterfalls database stems from Lehner et al, (2016), and aids in calculating the Degree of Fragmentation (DOF). Waterfalls naturally reduce connectivity in river networks by reducing or blocking species movement in river networks. Waterfalls should therefore be considered in the fragmentation analysis. They are blocking species movement upstream, while the downstream connectivity remains intact. As such waterfalls influence the degree of fragmentation (DOF).

Use any waterfall database you have available visualize the points together with the river network. The waterfalls database is not a direct source data layer to the model, but will be used to update the field "HYFALL" in the stream database. The attribute "HYFALL" in the streams feature class indicates the occurrence of at least one waterfall on the river reach (HYFALL = 1); otherwise 'HYFALL = 0". Only waterfalls that you consider to be a true barrier for species should be set to "1". Typically, rapids are included in waterfalls databases, but these are generally passable by most species, and should therefore be excluded as a waterfall.

• Review each waterfall and determine if it actually constitutes a barrier for species. Use satellite images and other sources for verification. If in any doubt, for example, the waterfall is not creating a break (e.g. rapids), exclude the waterfall from the assessment.

• Select all river reaches that have at least one waterfall, and set the HYFALL attribute to "1", otherwise to "0"

7.3.3 Updating urban areas and nightlights (URB)

Urban areas are an indicator for dense infrastructure development, which includes sealed surfaces, including parking lots, various types of buildings, as well as difficult to map features, such as weirs, locks channels etc., which may all impact adjacent rivers and streams by inhibiting lateral connectivity.

The indicator is a combination of urban areas and nightlights within urban areas. The original urban dataset used in the FFR study was based on Modis imagery (Schneider et al., 2009). However, new data became available globally developed by the Global Urban Footprint (GUF) project (Esch et al., 2017). These data provide superior urban footprints at an unprecedented spatial resolution of 0.4 arc seconds (~12 m; for more details see https://www.dlr.de/eoc/en/PortalData/60/Resources/dokumente/guf/GUF_Product_Specifications_GUF_DLR_v01.pdf). Urban areas represented by this dataset are far more precise than in previous global maps, resulting in less overall coverage of urban areas compared to the dataset of the global FFR analysis (Schneider et al., 2009) which showed a more blurred representation of urban areas. Also, smaller urban areas are now more clearly distinguished and are found widespread in the study area.

The nightlight dataset used in the global FFR analysis (Doll, 2008) is no longer recommended for further analysis. Instead, we suggest using the dataset by Elvidge et al., (2017) at higher spatial resolution (90 m versus 500 m previously) providing superior imaging accuracy. This new nightlight composite map uses different units and value ranges than the previous dataset. New values are given in nW/area unit reaching up to a maximum of 107,000 nW/area unit (in areas where gas flares are visible year-round). Due to these extreme values an upper cut-off was applied to avoid distortion in our data standardization process caused by outliers. Elvidge et al. (2017) suggested a value of 300 nW as maximum value for city light sources, depending on atmospheric conditions and type of light source. To simplify the standardization with the FFR assessment, we used a threshold where pixels with values larger than 100 were set to 100. This threshold produced well-defined gradients both within cities and between cities, whereas the dataset previously used in the global FFR analysis typically showed maximum light values for most cities across the world, from large to small ones.

Like in the global FFR analysis, the nightlight values should be clipped to urban areas and to a 1-km buffer around the river.

The above described datasets are fairly detailed, and can be used in a local context. However, similarly suitable local datasets can be used instead. The following steps serve a general guideline on calculating the URB attribute for the FFR analysis.

Steps for calculating urban density for each river catchment

• Identify and analyze different urban area extents, as well as nightlight intensity maps for the region and create a list of candidate layers.

• Choose the most adequate layers for the analysis, considering accuracy (check with satellite images), completeness, consistency and attribute availability.

• Reclassify (normalize) the nightlight intensity layer to a scale from 0 to 100.

• Overlay the urban areas and the nightlight intensity to nightlight intensity only within urban extents.

• Overlay the resulting raster with the buffer raster to eliminate urban areas outside the buffer area (map algebra)

• Calculate zonal statistics using the stream catchment to calculate the average urban density in each river reach catchment.

7.3.4 Roads (RDD)

A typical task for updating the road dataset is to use a local dataset that is more updated than the global one. In addition to replacing the road dataset with an updated version, we used new methodological elements to better capture the effects of roads on connectivity, especially for smaller roads. The global FFR analysis used the GRIP v3 dataset (Meijer and Klein Goldewijk, 2009), whereas for this study, we integrated used the polylines from OpenStreetMap as the basis for mapping.

While in the global FFR study the width for all road was kept constant at 50 meters regardless of type (highway, primary roads, etc.), the road width typically vary between 5 and 30 meters depending on road types (see table). Also, calculations of road density should be conducted at 5 m instead of 50 m resolution, which can contributed to a higher accuracy for estimating the road density in each river reach catchment.

7.3.5 Benchmark rivers

Benchmark rivers are rivers that were positively identified as free-flowing. The benchmarking dataset of reported FFRs was compiled from literature resources and expert input. For any local assessment, the research team should select their own benchmark rivers specific to the region. The benchmark rivers should be nominated and discussed by a group of experts, ideally as part of a workshop, webinar, or group call. Benchmark rivers can be named, drawn, marked up in existing maps, or provided digitally. The assessment team should then translate the information in to the provided template, using the following steps:

1. Use the streams feature class to select the benchmark river that should be add to the database.

2. Export only the selection and merge it with benchmark_rivers feature class. Using this technique, the required fields GOID and BB_ID are derived from the stream feature class, and will be present in the benchmark_rivers feature class.

3. Update the fields FFRID, BENCH_SRC, and Name_Expert.

The field "BENCH_SRC" indicates the source of the benchmark river. This can be the name of a researcher or a publication, a category such as "Expert" or "Publication" etc. The field is used to analyze the results of the benchmarking separately by source name or category.

A benchmark river can be a full river, i.e. all reaches with same BB_ID are included or part of a river. Only the segments that are benchmark rivers are included in this feature class, the rest are deleted. Each benchmark river must have a unique FFRID and the river name ("Name_Expert") must be consistent across the same FFRID.