The River Macrophytes Database (RMD) is a Microsoft Access database constructed to house data on the plant communities of rivers in Great Britain and Northern Ireland. It includes data from over 7000 survey sites and is the most comprehensive database of its kind. Data have been collected from all over the UK between 1977 and the present day, following the methods of Holmes et al. (1999). The data held in the RMD are the result of collaborative work across all four statutory nature conservation bodies: Scottish Natural Heritage (SNH), Natural England (NE), Natural Resources Wales (NRW, formerly CCW) and the Northern Ireland Environment Agency (NIEA). The River Macrophytes Database can be downloaded from the JNCC website: https://hub.jncc.gov.uk/assets/0a26368d-400c-44e1-beaf-d4b89b7badcd#extent-detail

GROD v1.1 (filename: GROD_v1.1.csv), or Global River Obstruction Database version 1.1, contains 30549 manually identified human-made structures that obstructing river longitudinal flow. Obstructions have been identified on Google Earth Engine satellite map for all rivers mapped in the Global River Widths from Landsat (GRWL) database. Each obstruction has assigned one of the six types—Dam, Lock, Low head dam, Channel dam, Partial dam 1, Partial dam 2. Details of the mapping process and data quality can be found in the following publication:

Yang, X., Pavelsky, T.M., Ross, M.R.V., Januchowski-Hartley, S.R., Dolan,W., Altenau, E.H., Belanger, M., Byron, D.K., Durand, M.T., Dusen, I.V., Galit, H., Jorissen, M., Langhorst, T., Lawton, E., Lynch, R., Mcquillan, K.A., Pawar, S., Whittemore, A., in revision. Mapping ow-obstructing structures on global rivers. Water Resources Research.

The single csv file contain the version 1 of GROD that accompanying the above-mentioned publication. It contains 7 columns:

grod_id: unique identifier (character)

type: obstruction type (character)

lon: longitude in decimal degrees (float)

lat: latitude in decimal degrees (float)

sword_reach_id¹: nearest sword reach id (character)

distance_to_sword: distance to the nearest sword reach (float)

Note:

1. sword, or SWOT River Database (https://zenodo.org/record/3898570#.YU0urWZKhGo), is an improved version of GRWL with better topology. Majority of the GROD obstructions (N=30502) has been matched with the closest SWORD reach. The remaining 16 obstructions were further than 10km away from any SWORD reach and were not matched to SWORD.

2. GROD, along with many other large scale river obstructions databases, will be hosted on Global Dam Watch website (http://globaldamwatch.org/).

American Rivers’ Dam Removal Database includes all dam removals in the United States (of which we have been made aware) in which a significant portion of the dam has been removed for the full height of the dam, such that ecological function, natural river flow and fish passage can be restored at the site. This database is revised and updated annually with information provided by contributors across the country. The database may be used by anyone provided that citation is given to American Rivers and the DOI link is included.

Corresponding peer-reviewed publication

This dataset corresponds to all input and output files that were used in the study reported in:

Wade, J., David, C.H., Collins, E.L., Altenau, E.H., Coss, S., Cerbelaud, A., Tom, M., Durand, M., Pavelsky T.M. (In Review), Bidirectional Translations Between Observational and Topography-based Hydrographic Datasets: MERIT-Basins and the SWOT River Database (SWORD).

When making use of any of the files in this dataset, please cite both the aforementioned article and the dataset herein.

Summary

The MERIT-SWORD data product reconciles critical differences between the SWOT River Database (SWORD; Altenau et al., 2021), the hydrography dataset used to aggregate observations from the Surface Water and Ocean Topography (SWOT) Mission, and MERIT-Basins (MB; Lin et al., 2019; Yang et al., 2021), an elevation-derived vector hydrography dataset commonly used by global river routing models (Collins et al., 2024). The SWORD and MERIT-Basins river networks differ considerably in their representation of the location and extent of global river reaches, complicating potential synergistic data transfer between SWOT observations and existing hydrologic models.

MERIT-SWORD aims to:

Generate bidirectional, one-to-many links (i.e. translations) between river reaches in SWORD and MERIT-Basins (ms_translate files).

Provide a reach-specific evaluation of the quality of translations (ms_diagnostic files).

Data sources

The following sources were used to produce files in this dataset:

MERIT-Basins (version 1.0) derived from MERIT-Hydro (version 0.7) available under a CC BY-NC-SA 4.0 license. https://www.reachhydro.org/home/params/merit-basins

SWOT River Database (SWORD) (version 16) available under a CC BY 4.0. https://zenodo.org/records/10013982. DOI: 10.5281/zenodo.10013982

Mean Discharge Runoff and Storage (MeanDRS) dataset (version v0.4) available under a CC BY-NC-SA 4.0 license. https://zenodo.org/records/10013744. DOI: 10.5281/zenodo.10013744; 10.1038/s41561-024-01421-5

Software

The software that was used to produce files in this dataset are available at https://github.com/jswade/merit-sword.

Primary Data Products

The following files represent the primary data products of the MERIT-SWORD dataset. Each file class generally has 61 files, corresponding to the 61 global hydrologic regions (region ii). For typical use of this dataset, download the 3 following zip folders listed below. The ms_translate.zip and ms_diagnostic.zip NetCDF files are best suited for scripting applications, while the ms_translate_shp.zip shapefiles are best suited for GIS applications.

The MERIT-SWORD translation tables (.nc) establish links between corresponding river reaches in MERIT-Basins and SWORD in both directions. The mb_to_sword translations relate the COMID values of all MERIT-Basins reaches in region ii (as defined by MERIT-Basins) to corresponding SWORD reach_id values, which are ranked by their degree of overlap and stored in columns sword_1 – sword_40. The partial intersecting lengths (m) of SWORD reaches within related MERIT-Basins unit catchments are stored in columns part_len_1 – part_len_40 and can be used to weight data transfers from more than one SWORD reach. The sword_to_mb translations relate the reach_id values of all SWORD reaches in region ii (as defined by SWORD) to corresponding MERIT-Basins COMID values, which are ranked by their degree of overlap and stored in columns mb_1 – mb_40. The partial intersecting lengths (m) of SWORD reaches within related MERIT-Basins unit catchments are again stored in columns part_len_1 – part_len_40.

ms _translate.zip

mb_to_sword: mb_to_sword_pfaf_ii_translate.nc

sword_to_mb: sword_to_mb_pfaf_ii_translate.nc

The MERIT-SWORD diagnostic tables (.nc) contain evaluations of the quality of translations between MERIT-Basins and SWORD reaches, stored in column flag. The mb_to_sword diagnostic files contain integer quality flags for each MERIT-Basins reach translation in region ii. The sword_to_mb diagnostic files contain integer quality flags for each SWORD reach translation in region ii. The quality flags are as follows:

0 = Valid translation.

1 = Translated reaches are not topologically connected to each other.

2 = Reach does not have a corresponding reach in the other dataset (absent translation).

21 = Reach does not have a corresponding reach in the other dataset due to flow accumulation mismatches.

22 = Reach does not have a corresponding reach in the other dataset because it is located in what the other dataset defines as the ocean.

ms_diagnostic.zip

mb_to_sword: mb_to_sword_pfaf_ii_diagnostic.nc

sword_to_mb: sword_to_mb_pfaf_ii_diagnostic.nc

For GIS applications, the translations and diagnostic tables are also available in shapefile format, joined to their respective MERIT-Basins and SWORD river vector shapefiles. The MERIT-Basins and SWORD shapefiles retain their original attribute tables, in additional to the added translation and diagnostic columns.

ms _translate_shp.zip

mb: riv_pfaf_ii_MERIT_Hydro_v07_Basins_v01_translate.shp

sword: jj_sword_reaches_hbii_v16_translate.shp

Example Applications Data Products

The following files are example use cases of transferring data between MERIT-Basins and SWORD. They are not required for typical use of the MERIT-SWORD dataset.

The MeanDRS-to-SWORD application example files demonstrate how the MERIT-SWORD translation tables can be used to transfer discharge simulations along MERIT-Basins reaches (i.e. MeanDRS; https://zenodo.org/records/8264511) to corresponding SWORD reaches in region ii and continent xx. MeanDRS discharge simulations (m3 s-1) are transferred to SWORD reaches based on a weighted average translation of corresponding reaches and stored in the column meanDRS_Q.

app_meandrs_to_sword.zip: xx_sword_reaches_hbii_v16_meandrs.shp

The SWORD-to-MERIT-Basins application example files demonstrate how the MERIT-SWORD translation tables can be used to transfer variables of interest (in this case, river width) from SWORD reaches to corresponding MERIT-Basins reaches in region ii. SWORD width estimates (m) are transferred to MERIT-Basins reaches based on a weighted average translation of corresponding reaches and stored in the column sword_wid.

app_sword_to_mb.zip: riv_pfaf_ii_MERIT_Hydro_v07_Basins_v01_sword.shp

Intermediate Data Products

The following files are intermediates used in generating the primary data. They are not required for typical use of the MERIT-SWORD dataset.

The MERIT-SWORD river trace files represent our first approximation of MERIT-Basins reaches that correspond to SWORD reaches in region ii, prior to the manual removal of mistakenly included reaches. The river trace files are only used to generate the final river network files and are not used elsewhere in the dataset.

ms_riv_trace.zip: meritsword_pfaf_ii_trace.shp

The MERIT-SWORD river network shapefiles contain the MERIT-Basins reaches that in aggregate best correspond to the location and extent of the SWORD river network for each of the Pfafstetter level 2 regions as defined by SWORD v16 (i.e. the 61 values of ii). The MERIT-SWORD river networks serve as an intermediary data product to enable reliable translations.

ms_riv_network.zip: meritsword_pfaf_ii_network.shp

The MERIT-SWORD transpose files are used to confirm that the translation tables in one direction can recreated in their entirety using only data from the translation tables in the other direction, ensuring ~3,500 less data transfer. These files are exact copies of the files contained in ms_translate.zip.

ms _transpose.zip

mb_transposed: mb_to_sword_pfaf_ii_transpose.nc

sword_transposed: sword_to_mb_pfaf_ii_transpose.nc

The MERIT-SWORD translation catchment files contain the MERIT-Basins unit catchments corresponding to each reach used in generating the mb_to_sword and sword_to_mb translations for each region ii. The files are used internally during the translation process and not required for typical dataset use.

ms_translate_cat.zip

mb_to_sword: mb_to_sword_pfaf_ii_translate_cat.nc

sword_to_mb: sword_to_mb_pfaf_ii_translate.cat.nc

The hydrologic regions as defined by MERIT-Basins and SWORD are not identical and overlap in many cases, complicating translations. The region overlap files provide bidirectional mappings between region identifiers in both datasets. The files are used in most dataset scripts to determine the regional files from each dataset that need to be loaded.

ms_region_overlap.zip: sword_to_mb_reg_overlap.csv, sword_to_mb_reg_overlap.csv

The MERIT-SWORD river edit files contain ~3,500 MERIT-Basins river reaches that were mistakenly included during river network generation and do not correspond to any SWORD reaches. These reaches are removed from the river trace files to generate the final MERIT-SWORD river network data product.

ms_riv_edit.zip: meritsword_edits.csv

Near the antimeridian, MERIT-Basins and SWORD shapefiles differ in their longitude convention. Additionally, the SWORD dataset lacks a shapefile for region 54, which does not have any SWORD reaches. The SWORD edit files contain copies of SWORD files, altered to match the longitude convention of MERIT-Basins and including a dummy shapefile for region 54.

sword_edit.zip: xx_sword_reaches_hbii_v16.shp

Known bugs in this dataset or the associated manuscript

No bugs have been identified at this time.

References

Altenau, E. H., Pavelsky, T. M., Durand, M. T., Yang, X., Frasson, R. P. de M., & Bendezu, L. (2021). The Surface Water and Ocean Topography (SWOT) Mission River Database (SWORD): A Global River Network for Satellite Data Products. Water Resources Research, 57(7), e2021WR030054. https://doi.org/10.1029/2021WR030054

Collins, E. L., David, C. H., Riggs, R., Allen, G. H., Pavelsky, T. M., Lin, P., Pan, M., Yamazaki,

The description for this record is not currently available.

The GEMS-GLORI register, circulated by UNEP for review in 1996, lists 555 world major rivers discharging to oceans (Q > 10 km**3/year, or A > 10 000 km**2, or sediment discharge > 5Mt/year, or basin population >5M people). Up to 48 river attributes are listed, including major ions and nutrients (C, N, P) in both dissolved, particulate, organic and inorganic forms. For many rivers, two or three sets of data are provided with relevant periods of records and references. Although half of the selected rivers are not yet documented for water quality, most of the first 40 rivers are well described (Irrawady, Zambezi, Ogooue, Magdalena, are noted exceptions). Altogether about 10 000 individual data from 500 references are listed. The global coverage in terms of river discharge and/or drainage area ranges from 40 to 67% for most major water quality attributes but drops to 25% for some organic and/or particulate forms of N and P. Planned development of the register includes collection of information on particulate chemistry and data on endorheic rivers and selected tributaries.

The River Macrophytes Database (RMD) is a Microsoft Access database constructed to house data on the plant communities of rivers in Great Britain and Northern Ireland. It includes data from over 7000 survey sites and is the most comprehensive database of its kind. Data have been collected from all over the UK between 1977 and the present day, following the methods of Holmes et al. (1999).

The data held in the RMD are the result of collaborative work across all four statutory nature conservation bodies: Scottish Natural Heritage (SNH), Natural England (NE), Natural Resources Wales (NRW, formerly CCW) and the Northern Ireland Environment Agency (NIEA).

The River Macrophytes Database can be downloaded from the JNCC website: https://hub.jncc.gov.uk/assets/0a26368d-400c-44e1-beaf-d4b89b7badcd#extent-detail

These macro-invertebrate data incorporate the results from the national river water quality network (NRWQN) from 66 sites throughout New Zealand for the purpose of monitoring long-term trends. Data included: 1990 to 2008. The NRWQN was funded by the Foundation for Research, Science, & Technology through NIWA's Nationally Significant Database: Water Resources & Climate programme. Current funding (from July 2011) comes from the NIWA Environmental Information/Monitoring programme core funding. The data are collected annually in summer, and data collection was initiated in January 1989.

This data set contains small-scale base GIS data layers compiled by the National Park Service Servicewide Inventory and Monitoring Program and Water Resources Division for use in a Baseline Water Quality Data Inventory and Analysis Report that was prepared for the park. The report presents the results of surface water quality data retrievals for the park from six of the United States Environmental Protection Agency's (EPA) national databases: (1) Storage and Retrieval (STORET) water quality database management system; (2) River Reach File (RF3) Hydrography; (3) Industrial Facilities Discharges; (4) Drinking Water Supplies; (5) Water Gages; and (6) Water Impoundments. The small-scale GIS data layers were used to prepare the maps included in the report that depict the locations of water quality monitoring stations, industrial discharges, drinking intakes, water gages, and water impoundments. The data layers included in the maps (and this dataset) vary depending on availability, but generally include roads, hydrography, political boundaries, USGS 7.5' minute quadrangle outlines, hydrologic units, trails, and others as appropriate. The scales of each layer vary depending on data source but are generally 1:100,000.

This database gathers diatom taxa (to the lowest taxonomic level) counts (relative abundances) and main limnological variables georeferred to UTM coordinates for the rivers of the Duero Basin (NW Spain) More information on this dataset can be found in the Freshwater Metadatabase - BFE_3 (http://www.freshwatermetadata.eu/metadb/bf_mdb_view.php?entryID=BFE_3).

The SWOT Level 2 River Single-Pass Vector Data Product from the Surface Water Ocean Topography (SWOT) mission provides water surface elevation, slope, width, and discharge derived from the high rate (HR) data stream from the Ka-band Radar Interferometer (KaRIn). SWOT launched on December 16, 2022 from Vandenberg Air Force Base in California into a 1-day repeat orbit for the "calibration" or "fast-sampling" phase of the mission, which completed in early July 2023. After the calibration phase, SWOT entered a 21-day repeat orbit in August 2023 to start the "science" phase of the mission, which is expected to continue through 2025. Water surface elevation, slope, width, and discharge are provided for river reaches (approximately 10 km long) and nodes (approximately 200 m spacing) identified in the prior river database, and distributed as feature datasets covering the full swath for each continent-pass. These data are generally produced for inland and coastal hydrology surfaces, as controlled by the reloadable KaRIn HR mask. The dataset is distributed in ESRI Shapefile format. This dataset is the parent collection to the following sub-collections: https://podaac.jpl.nasa.gov/dataset/SWOT_L2_HR_RiverSP_node_2.0 https://podaac.jpl.nasa.gov/dataset/SWOT_L2_HR_RiverSP_reach_2.0

RiverSR database (River Surface Reflectance) v1.1.0

This database contains Landsat 5, 7, and 8 Level 1 Collection 1 surface reflectance from all rivers in the contiguous USA that are ~60 meters wide or greater. The surface reflectance values across bands (red, green, blue, nir, swir1, swir1) represent the median reflectance of pixels detected as water within each Landsat scene that are within the boundaries of each reach represented by NHDPlusV2 centerlines. Surface reflectance is therefore geo-referenced to river center lines with network topology (NHDPlusV2) for quick geospatial analysis.

Files:

1) Metadata (riverSR_v1.1_metadata.docx): Description of all data files associated with this repository.

2) Surface reflectance database (riverSR_usa_v1.1.feather). Feather files are text files readable in R and python with the feather package and this table is joinable to nhdplusv2_modified_v1.0.shp based on the "ID" column and to the original NHDplusV2 flowlines with the "COMID" column.

3) Shapefile of river centerlines to which the reflectance data can be attached (nhdplusv2_modified_v1.0.shp).

4) Shapefile of the reach polygons associated with each nhdplusv2_modified reach. (nhdplusv2_polygons.shp).

5) The reach IDs of original and new NHDplusV2 centerlines. (COMID_ID.csv).

The NRRSS (National River Restoration Science Systhensis) Summary Database is a representative database of stream restoration projects throughout the country, with focus on seven geographic regions--California, Chesapeake Bay, Central US, Pacific Northwest, Southeast, Southwest, and Upper Midwest. These data were collected from a variety of electronic, paper, and human sources from 2002 to 2004. Information from this database has been summarized in Bernhardt et al 2005 (link) as well as published and forthcoming regional summaries. As of December 15, 2005, the information in this database is now freely searchable, downloadable, and usable by members of the public.

The USGS’s FORE-SCE model was used to produce a long-term landscape dataset for the Delaware River Basin (DRB). Using historical landscape reconstruction and scenario-based future projections, the data provided land-use and land-cover (LULC) data for the DRB from year 1680 through 2100, with future projections from 2020-2100 modeled for 7 different socioeconomic-based scenarios, and 3 climate realizations for each socioeconomic scenario (21 scenario combinations in total). The projections are characterized by 1) high spatial resolution (30-meter cells), 2) high thematic resolution (20 land use and land cover classes), 3) broad spatial extent (covering the entirety of the Delaware River basin, corresponding to USGS HUC codes 020401 and 020402), 4) use of real land ownership boundaries to ensure realistic representation of landscape patterns, and 5) representation of both anthropogenic land use and natural vegetation change that respond to projected climate change. Data are provided in 10-year time steps from 1680 through 2100 (43 individual dates). Historical landscape data is provided in one downloadable zip file, containing 34 individual land cover datasets for 10-year intervals from 1680 through 2010. “Current” (2020) and “future” (2030 through 2100) data are provided at 10-year time steps in files corresponding to the 21 different scenario combinations. The following provides a brief summary of the 7 major land-use scenarios. 1) Business-as-usual - Based on an extrapolation of recent land-cover trends as derived from remote-sensing data. Overall trends were provided by 2001 to 2016 change in the National Land Cover Database, while change in crop types were extrapolated from 2008 to 2018 change in the Cropland Data Layer. 2) Billion Ton Update (BTU) scenario ($40 farmgate price) - This scenario is based on US Department of Energy biofuel scenarios from the BTU. The $40 scenario represents likely agricultural conditions under an assumed farmgate price of $40 per dry ton of biomass (for the production of biofuel). All three BTU scenarios include the representation of a “perennial grass” class (class #20) that represents grass crops such as miscanthus, switchgrass, or prairie grasses grown for production of cellulosic biofuel. 3) BTU scenario ($60 farmgate price) - This scenario is based on US Department of Energy biofuel scenarios from the BTU. The $60 scenario represents likely agricultural conditions under an assumed farmgate price of $60 per dry ton of biomass (for the production of biofuel). 4) BTU scenario ($80 farmgate price) - This scenario is based on US Department of Energy biofuel scenarios from the BTU. The $80 scenario represents likely agricultural conditions under an assumed farmgate price of $80 per dry ton of biomass (for the production of biofuel). 5) Global Change Analysis Model (GCAM) Reference scenario - Based on global-scale scenarios from the GCAM model, the "reference" scenario provides a likely landscape under a world without specific carbon or climate mitigation efforts. As such, it's another form of a "business-as-usual" scenario. 6) GCAM 2.6 scenario - Based on global-scale scenarios from the GCAM model, the GCAM 2.6 model represents a very aggressive climate mitigation scenario, where carbon payments and other mitigation efforts result in a net radiative forcing of only ~2.6 W/m2 by 2100. 7) GCAM 4.5 scenario - Based on global-scale scenarios from the GCAM model, the GCAM 4.5 model represents a mid-level climate mitigation scenario, where carbon payments and other mitigation efforts result in a net radiative forcing of ~4.5 W/m2 by 2100. For each of the 7 land-use scenarios, three alternative climate / vegetation scenarios were modeled, resulting in 21 unique scenario combinations. The alternative vegetation scenarios represent the potential changes in quantity and distribution of the major vegetation classes that were modeled (grassland, shrubland, deciduous forest, mixed forest, and evergreen forest), as a response to potential future climate conditions. The three alternative vegetation scenarios correspond to climate conditions consistent with 1) The Intergovernmental Panel on Climate Change (IPCC's) Representative Concentration Pathway (RCP) 8.5 scenario (a scenario of high climate change), 2) the RCP 4.5 scenario (a mid-level climate change scenario), and 3) a mid-point climate that averages RCP4.5 and RCP8.5 conditions Data are provided here as compressed ZIP files for 1) the historical landscape reconstruction time frame (1680 through 2010), and 2) for each of the 21 future scenario combinations, including the starting 2020 year and extending through 2100 (thus 22 downloadable ZIP files). The “attributes” section of the metadata provides a key for identifying file names associated with each of the scenario combinations and historical period.

The RivFISH database aggregates the available data on freshwater-dependent fish presence in Europe, validated at the river basin level and considering taxonomical synonyms for species names, thus allowing for a maximization of data usage and robustness. This database also promotes interoperability with other datasets, including the IUCN Red List of Threatened Species, FishBase and the Catchment Characterisation and Modelling (CCM2) – River and Catchment Database v2.1. It is, as far as the authors know, the most up-to-date and comprehensive database on the presence of freshwater-dependent fish species for European river basins. The structure of the database is also prepared to deal with future alterations in species taxonomy, as well as new records of species occurrence in river basins.

This site contains a compilation of monthly mean river discharge data for over 3500 sites worldwide. The data sources are RivDis2.0, the United States Geological Survey, Brazilian National Department of Water and Electrical Energy, and HYDAT-Environment Canada. The period of record for each station is variable, from 3 years to greater than 100. All data is in m3/s.

To access the data click on the map below to zoom in to the desired stations and data. Alternatively, the data can be accessed by using a key word search or by entering the river ID number if that is known. The data is provided in a tab-delimited format compatible with most spreadsheet programs.

We are continually looking to improve this data set and welcome any additional data and comments. Base map image courtesy of NASA. The image came from a single remote sensing device - NASA's MODIS (Moderate Resolution Imaging Spectroradiometer) onboard the satellite Terra.

Contact: Md Safat Sikder (mssikder@illinois.edu), Jida Wang (jidaw@illinois.edu)

Citation

Sikder, M. S., Wang, J., Allen, G. H., Sheng, Y., Yamazaki, D., Crétaux, J.-F., and Pavelsky, T. M., 2024. HarP: Harmonized Prior river-lake database. Zenodo, https://doi.org/10.5281/zenodo.14205131.

If you only use the PLD-TopoCat dataset, please cite the following paper:

Sikder, M. S., Wang, J., Allen, G. H., Sheng, Y., Yamazaki, D., Song, C., Ding, M., Crétaux, J.-F., and Pavelsky, T. M., 2023. Lake-TopoCat: A global lake drainage topology and catchment dataset. Earth System Science Data, 15, 3483-3511, https://doi.org/10.5194/essd-15-3483-2023.

Data description and components

The Harmonized Prior river-lake database (HarP) for SWOT integrated the SWOT River Database (SWORD) (Altenau et al., 2021) and the SWOT Prior Lake Database (PLD) (Wang et al., 2023) into a geometrically (lake/river) explicit but topologically harmonized vector database to allow for coupled fluvial-lacustrine applications, including a synergistic use of both river and lake products from SWOT.

In addition to the input river network (SWORD v16) and lake database (PLD v106), we used the MERIT Hydro v1.0.1 (Yamazaki et al., 2019), a high-resolution (~90 m) global hydrography dataset, to develop this database.

The SWORD-PLD harmonization process involves three major steps, with Step 3 being divided into three sub-steps. The processing chain is illustrated in the attached Figure "SWORD-PLD_harmonization_steps.jpg", as well as in Section 2 of the product description document. The HarP database consists of the outputs from each of the steps. For convenience, the global landmass (excluding Antarctica) was partitioned to 68 Pfafstetter Level-2 basins/regions, with their IDs shown in Figure "Pfaf2_basins.jpg" attached.

The HarP database consists of five datasets or components (outputs from each step), each with multiple features. The five datasets are described below, and more details are elaborated in the product description document.

1. Harmonized SWORD-PLD (file name "Harmonized_SWORD_PLD"): This is the fully harmonized SWORD-PLD dataset, the primary product of HarP (i.e., output of Step 3.3 in Figure "SWORD-PLD_harmonization_steps.jpg"). This dataset couples SWORD and PLD into a geometrically segmented but topologically integrated dataset at the node, reach, and catchment scales (stored by three feature layers, respectively):

(a) Harmonized feature nodes: Harmonized_feature_nodes_pfaf_xx
(b) Harmonized river network: Harmonized_river_network_pfaf_xx
(c) Harmonized feature catchments: Harmonized_feature_catchments_pfaf_xx
Note: ''pfaf_xx'' indicates the Pfafstetter Level-2 basin ID (shown in Fig. 'Pfaf2_basins.jpg').

Figure "HarP_example.jpg", attached to this database, is an example of the fully harmonized SWORD-PLD dataset for the Ohio River Basin. The example shows three main features of the dataset: feature nodes (i.e., reach downstream ends, lake inlets, and lake outlets; see Fig. 3 in the product description document for definitions), river reaches (i.e., reaches characterized by SWORD alone, characterized by TopoCat alone, and shared by both SWORD and TopoCat), and catchments segmented by each of the feature nodes.

2. Intersected SWORD-PLD drainage configuration (file name "Intersected_SWORD_PLD"): This dataset is the intersected SWORD-PLD (prior river-lake) features (i.e., output of Step 2 in Figure "SWORD-PLD_harmonization_steps.jpg"). This dataset was constructed independently from Step 1 and Step 3. In this dataset, the original geometries of SWORD and PLD are not altered, but instead, their geometric and drainage topological relationships are configured in the attribute tables. This dataset consists of three features:

(a) Intersected reaches: Intersected_SWORD_reaches_pfaf_xx
(b) Intersected nodes: Intersected_SWORD_nodes_pfaf_xx
(c) Intersected lakes: Intersected_PLD_lakes_pfaf_xx

3. PLD-TopoCat (file name "PLD_TopoCat"): This dataset is the lake drainage topology and catchments (TopoCat) for PLD lakes (i.e., output of Step 1 in Figure "SWORD-PLD_harmonization_steps.jpg"). PLD-TopoCat was developed to generate detailed lake drainage topology and connecting paths, which were later used to configure the off-SWORD-network PLD lakes into the tributaries that drain to SWORD. PLD-TopoCat was generated from PLD v106 and MERIT Hydro. Details of the developiong process and algorithm for TopoCat can be found at Sikder at al., (2023). PLD-TopoCat dataset contains six features:

(a) Lake original polygon: PLD_lakes_pfaf_xx
(b) Lake raster polygon: Lake_raster_polygons_pfaf_xx
(c) Lake outlets: Lake_outlets_pfaf_xx
(d) Lake catchments: Lake_catchments_pfaf_xx
(e) Inter-lake reaches: Inter_lake_reaches_pfaf_xx
(f) Lake-network basins: Lake_network_basins_pfaf_xx
Note: full version of the PLD-TopoCat is available here.

4. SWORD-mirror network (file name "SWORD_mirror"): The SWORD-mirror network was constructed to facilitate the SWORD-TopoCat network merging process (i.e., output of Step 3.1 in Figure "SWORD-PLD_harmonization_steps.jpg"). It is essentially a replica of SWORD except that the original SWORD reaches are geometrically modified to be aligned with the topological/hydrographic information depicted in MERIT Hydro. The SWORD-mirror network consists of four features:

(a) SWORD-original reaches: SWORD_original_reaches_pfaf_xx
(b) SWORD-mirror prelim. reaches: SWORD_mirror_prelim_reaches_pfaf_xx
(c) SWORD-mirror reaches: SWORD_mirror_reaches_pfaf_xx
(d) SWORD-mirror reach catchments: SWORD_mirror_reach_catchments_pfaf_xx

5. Merged SWORD-mirror – TopoCat network (file name "SWORD_TopoCat_merged"): This dataset is the output of Step 3.2 in Figure "SWORD-PLD_harmonization_steps.jpg". It is essentially the merged product of the inter-lake reaches (from Step 2) and SWORD-mirror reaches (from Step 3.1). The merged SWORD-mirror – TopoCat network consists of three features:

(a) Merged SWORD-TopoCat reaches: SWORD_TopoCat_merged_reaches_pfaf_xx
(b) SWORD nodes at SWORD-TopoCat confluence: SWORD_TopoCat_confluence_nodes_pfaf_xx
(c) Reach catchments for merged network: SWORD_TopoCat_reach_catchments_pfaf_xx

The attribute tables for each of the feature components are explained in Section 4 of the product description document. All files of HarP are available in both shapefile and geodatabase formats.

Disclaimer
Authors of this dataset claim no responsibility or liability for any consequences related to the use, citation, or dissemination of HarP. For any quesitons, please contact Safat Sikder and Jida Wang.

This data set contains sediment size data collected at research sites using a Wolman Pebble Count method. This dataset is associated with the following publication: Collins , S., M. Thoms, and J. Flotemersch. Hydrogeomorphic zones characterize riverbed sediment patterns within a river network. River Systems. E. Schweizerbart'sche Verlagsbuchhandlung, Stuttgart, GERMANY, 21(4): 203-213, (2015).

The River Sediment Database (RivSed) database contains surface suspended sediment concentrations (SSC) derived from Landsat 5, 7, and 8 Level 1 Collection 1 surface reflectance from all rivers in the contiguous USA that are ~60 meters wide or greater. SSC represent spatially integrated "reach" median concentrations over the footprint of NHDPlusV2 centerlines where high quality river water pixels were detected within each Landsat image from 1984-2018. This is built in the River Surface Reflectance database (RiverSR) also in Zenodo (Gardner et al,. 2020 Geophysical Research Letters).

The paper associated with RivSed: Gardner, J., Pavelsky, T. M., Topp, S., Yang, X., Ross, M. R., & Cohen, S. (2023). Human activities change suspended sediment concentration along rivers. Environmental Research Letters. https://iopscience.iop.org/article/10.1088/1748-9326/acd8d8

Files:

1) Metadata (riverSed_v1.0_metadata.pdf): Description of all data files associated with this repository.

2) RiverSed (RiverSed_USA_v1.1.txt). Table of SSC and associated data that is joinable to nhdplusv2_modified_v1.0.shp based on the "ID" column and to the original NHDplusV2 flowlines with the "COMID" column.

3) Shapefile of river centerlines to which the reflectance data can be attached (nhdplusv2_modified_v1.0.shp).

4) Shapefile of the reach polygons associated with each nhdplusv2_modified reach. (nhdplusv2_polygons_v1.0.shp).

5) The look up table for reach IDs of original (COMID) and modified (ID) NHDplusV2 centerlines. (COMID_ID.csv). Short reaches were joined together to optimize for remote sensing data collection and make more consistent reach lengths.

6) SSC-Landsat matchup database with extended metadata on locations and in-situ data derived from Aquasat (Ross et al., 2019) (Aquasat_TSS_v1.1.csv)

7) The final training data used to build the xgboost machine learning model (train_clean_xgb_v1.1.csv)

8) The xgboost model that can make SSC predictions over inland waters in USA using Landsat bands/band combinations (finalmodel_xgb_v1.1.rds and .RData). The model can only be loaded in R for now.

A major problem related to large-scale water quality modeling has been the lack of available observation data with a good spatiotemporal coverage. This has affected the reproducibility of previous studies and the potential improvement of existing models. In addition to the observation data itself, insufficient or poor quality metadata has also discouraged researchers to integrate the already available datasets. Therefore, improving both the availability and quality of open water quality data woould increase the potential to implement predictive modeling on a global scale. We aim to address the aforementioned issues by presenting the new Global River Water Quality Archive (GRQA) by integrating data from five existing global and regional sources: Canadian Environmental Sustainability Indicators program (CESI), Global Freshwater Quality Database (GEMStat), GLObal RIver Chemistry database (GLORICH), European Environment Agency (Waterbase) and USGS Water Quality Portal (WQP). The resulting dataset covering the timeframe 1898 - 2020 contains a total of over 17 million observations for 42 different forms of some of the most important water quality parameters, focusing on nutrients, carbon, oxygen and sediments. Supplementary metadata and statistics are provided with the observation time series to improve the usability of the dataset.

Last update: 2022-03-11

GRQA_v1.2 contains three updated files compared to GRQA_v1.1:

• NH4N_GRQA.csv
• NO2N_GRQA.csv
• NO3N_GRQA.csv

The files were updated, because the assumed conversion constants used for the corresponding GLORICH observations were found to be incorrect. The corresponding files in GRQA_figures.zip and GRQA_meta.zip are yet to be updated, but will be in GRQA_v1.3.

The explanation for the updated conversion constants is given in this notebook:
https://nbviewer.org/github/LandscapeGeoinformatics/GRQA_src/blob/main/testing/glorich_conversion_test.ipynb

An overview of all the files in the dataset can be found in README_v1.2.txt.

Statistical overview of all 42 parameters is given in the data catalog file GRQA_data_catalog.pdf.

For more information about the development of this dataset look for Virro, H., Amatulli, G., Kmoch, A., Shen, L., and Uuemaa, E.: GRQA: Global River Water Quality Archive, Earth Syst. Sci. Data, 13, 5483–5507, https://doi.org/10.5194/essd-13-5483-2021, 2021.