The comid field of these data can be used to join to the National Hydrography Dataset Plus V2.1 (NHDPlusV2) flowline comid or catchment featureid attributes. The included attributes follow the same data model as the NHDPlusV2 but include numerous updates and improvements to network connectivity. All attributes that depend on network connectivity have been recalculated. These attributes are based on the NHDPlusV2 network geometry and modifications retrieved from the National Water Model V2.1 (NWMv2.1) and "E2NHDPlusV2_us: Database of Ancillary Hydrologic Attributes and Modified Routing for NHDPlus Version 2.1 Flowlines" (E2NHDPlusV2) datasets. The attributes included are: comid, tocomid, fcode, nameID, lengthkm, reachcode, frommeas, tomeas, arbolate_sum, terminalpa ,hydroseq, levelpathi, pathlength, dnhydroseq, areasqkm, totdasqkm, terminalfl, dnlevelpat. These attributes are available in three formats: csv, fst, and parquet. "fst" is a high-performance format for use with the R programming language "fst" package. "parquet" is a high-performance format for use with multiple programming languages (including python) that support the Apache Arrow Parquet format.

Large-scale, accurate and fully attributed digital river centreline covering England and Wales. The dataset has full-feature network geometry cross-referenced with OS MasterMap following Digital National Framework principles. The dataset has full-feature network geometry cross-referenced with OS MasterMap following Digital National Framework. It is made of the three following layers: - Links: lines representing the river network. It is a river centreline dataset, based on OS MasterMap for surface features and Environment Agency culvert surveys for underground features (where available). There are many attributes associated with this dataset to enable it to be used for many different business purposes. It is topologically correct to allow it's use in network tracing tasks. - Offline Drainage: lines representing the sections of river and drains that do not obviously connect to the main online drainage network represented by the DRN. Sections with uncertain flow direction and connectivity are presented here, although in reality some may connect to the main DRN, and be added to it as more information becomes available. - Nodes: points representing the junctions between discrete stretches of the online DRN. It is used to assist in connectivity and flow direction, as every DRN stretch is attributed with the 'from' and 'to' nodes. Nodes are also included where line features cross, but do not intersect, such as an aqueduct passing over a river. Nodes have types to determine whether they are at for example junction or at a change in river type.

Digital line dataset of River Network of Kathmandu Valley, Nepal. This dataset is created using Topographic sheet data at scale of 25,000/50,000, of 1995 acquired from National Geographic Information Infrastructure Project (NGIIP), Dept. of Survey Kathmandu, Nepal.

Ordnance Survey (OS) make an open dataset of rivers available as OS Open Rivers vector polylines under an Open Government Licence, derived from the detailed OS MasterMap Water Layer. Although widely used by the practitioner and the academic community it is not fully topologically connected, limiting its suitability for several uses and research applications such as linear referencing and reach analysis, hydro-ecological analysis, water quality monitoring, restoration and remediation prioritisation, connectivity planning and integration in decision support tools. The entire river network for Great Britain was corrected for topological errors and attributed with additional data. The network consisted of 183,349 polylines representing 147,387 kilometres.To ensure the network was a topologically correct river network, canals and channels that broke the dendritic connectivity of the river network were removed. The network was further simplified by removing loops. Small unconnected sections that were within the great catchment they sat within were also deleted out.With the topological errors removed, the network was passed through the river network processing tool RivEX to create added-value attribution. Encoding the network with these pre-computed values allows for rapidly analyse of the network alongside site data (points snapped to the network).Further details on edits made to the network and the attribute fields added are recorded in the lineage and fields section of the metadata.For further info see https://openrivers.net/

Digital line dataset of River Network of Lamjung district, Nepal. This dataset is created using Topographic sheet data at scale of 50,000 acquired from Department of Survey, Kathmandu, Nepal.

River network in Glasgow showing the river and coast lines. To view or use these files, a compression software and GIS software like ESRI ArcGIS or QGIS is needed. Data extracted 2013-10-15T14:30:45 Contains Ordnance Survey data (c) Crown Copyright 2013. Licence: None

Digital polyline data of River Systems of Nepal. This dataset is created using Topographic Zonal Map of 250000 scale published by Department of Survey Nepal in 1988 and is in geographic coordinates.

River Water Segmentation Dataset (RIWA)

New Version 2: It is the largest high quality (min size of 400x400) dataset as far as we know (01/2023).

The dataset called RIWA represents a pixel-wise binary river water segmentation. It consist of manually labelled smartphone, drone and DSLR images of rivers as well as suiting images of the Water Segmentation Dataset and high quality AED20K images. The COCO dataset was withdrawn since the segmentation quality is extremely poor.

Version hisoty:

Version 2: (declared as Version 4 by kaggle) - contains 1142 training, 167 validation and 323 test images. - Min size: 400 x 400 (h x w) - High quality segmentations. If you find an error, please message us.

Version 1: - contains 789 training, 228 validation and 111 test images. - Min size: 174 x 200 (hxw) - Some segmentations are not perfect.

Citation

If you use this dataset, please cite as:

 @misc{RIWA_Dataset,
  title={River Water Segmentation Dataset (RIWA)},
  url={https://www.kaggle.com/dsv/4901781},
  DOI={10.34740/KAGGLE/DSV/4901781},
  publisher={Kaggle},
  author={Xabier Blanch and Franz Wagner and Anette Eltner},
  year={2023}
}

Contact: - Xabier Blanch, TU Dresden see at SCIENTIFIC STAFF - Franz Wagner, TU Dresden - Anette Eltner, TU Dresden

CNN comparison

In 2023, we carried out a comparison to find the best CNN on this domain. If you are interested, please see our paper: River water segmentation in surveillance camera images: A comparative study of offline and online augmentation using 32 CNNs.

We conducted the tests using the AiSeg GitLab repository. It is capable of interactively train 2D and 3D CNNs, augmenting data with offline and online augmentation, analyzing single networks, comparing multiple networks, and applying trained CNNs to new data. The RIWA dataset can be used directly.

Background:

The handling of natural disasters, especially heavy rainfall and corresponding floods, requires special demands on emergency services. The need to obtain a quick, efficient and real-time estimation of the water level is critical for monitoring a flood event. This is a challenging task and usually requires specially prepared river sections. In addition, in heavy flood events, some classical observation methods may be compromised.

With the technological advances derived from image-based observation methods and segmentation algorithms based on neural networks (NN), it is possible to generate real-time, low-cost monitoring systems. This new approach makes it possible to densify the observation network, improving flood warning and management. In addition, images can be obtained by remotely positioned cameras, preventing data loss during a major event.

The workflow we have developed for real-time monitoring consists of the integration of 3 different techniques. The first step consists of a topographic survey using Structure from Motion (SfM) strategies. In this stage, images of the area of interest are obtained using both terrestrial cameras and UAV images. The survey is completed by obtaining ground control point coordinates with multi-band GNSS equipment. The result is a 3D SfM model georeferenced to centimetre accuracy that allows us to reconstruct not only the river environment but also the riverbed.

The second step consists of segmenting the images obtained with a surveillance camera installed ad hoc to monitor the river. This segmentation is achieved with the use of convolutional neural networks (CNN). The aim is to automatically segment the time-lapse images obtained every 15 minutes. We have carried out this research by testing different CNN to choose the most suitable structure for river segmentation, adapted to each study area and at each time of the day (day and night).

The third step is based on the integration between the automatically segmented images and the 3D model acquired. The CNN-segmented river boundary is projected into the 3D SfM model to obtain a metric result of the water level based on the point of the 3D model closest to the image ray.

The possibility of automating the segmentation and reprojection in the 3D model will allow the generation of a robust centimetre-accurate workflow, capable of estimating the water level in near real time both day and night. This strategy represents the basis for a better understanding of river flo...

The River Obstacles dataset is an inventory of weirs, waterfalls, sluices, dams, culverts, fords and flap gates, compiled initially from digital Ordnance Survey maps and the Environment Agency's Detailed River Network AfA036 (DRN) and improved and extended using information submitted by users via the River Obstacles App.
The information collected using the River Obstacles App is quality checked and verified before being added to the dataset.

This water flow network dataset is a route feature class rather than a simple polyline. The geometry is generated by merging the river lines of individual geometric network datasets. This layer contains an integrated flow network that includes known flow connections through rivers, lakes and groundwater aquifers. In places where the network is depicted flowing through lakes or through underground channels, the flow channels are schematic only, and do not represent the precise location of these flow channels. The appropriate Geological Survey Ireland data sets should be consulted where underground flows or connections are known or suspected.

Comprehensive index is an important index to measure the depth and breadth of regional land use. In 2020, the Yellow River basin 1km network land use degree data set is based on the 2020 annual change survey 30 meters land use grid data, land use is divided into unused land level, forest water level, agricultural land level and urban settlement with level 4, comprehensive calculation of regional land use by human intervention, quantitative reflect the strength of land use in the research area. This data set can be used for the research of territorial spatial planning, regional development and development evaluation, main functional area evaluation, and coupling analysis of human-land relationship in the Yellow River Basin.

This water flow network dataset is a route feature class rather than a simple polyline. The geometry is generated by merging the river lines of individual geometric network datasets. This layer contains an integrated flow network that includes known flow connections through rivers, lakes and groundwater aquifers. In places where the network is depicted flowing through lakes or through underground channels, the flow channels are schematic only, and do not represent the precise location of these flow channels. The appropriate Geological Survey Ireland data sets should be consulted where underground flows or connections are known or suspected.This dataset is provided by the Environmental Protection Agency (EPA). For more information please see https://gis.epa.ie/geonetwork/srv/eng/catalog.search#/metadata/c4043e19-38ec-4120-a588-8cd01ac94a9c

Digital line data of River Network of Ludhi Khola Watershed, Gorkha, Nepal. This dataset is created using Topographic sheet data at scale of 25,000/50,000, of 1995 acquired from National Geographic Information Infrastructure Project (NGIIP), Dept. of Survey, Kathmandu, Nepal. The data was prepared for the Reducing Emission from Deforestation and Forest Degradation (REDD) Pilot Project. This is a joint project of ICIMOD with Asian Network for Sustainable Agriculture and Bio-resources (ANSAB) and Federation of Community for Forest Users of Nepal (FECOFUN) funded by NORAD.

The National Water Quality Network (NWQN) for Rivers and Streams includes 113 surface-water river and stream sites monitored by the U.S. Geological Survey (USGS) National Water Quality Program (NWQP). The NWQN represents the consolidation of four historical national networks: the USGS National Water-Quality Assessment (NAWQA) Project, the USGS National Stream Quality Accounting Network (NASQAN), the National Monitoring Network (NMN), and the Hydrologic Benchmark Network (HBN). The NWQN includes 22 large river coastal sites, 41 large river inland sites, 30 wadeable stream reference sites, 10 wadeable stream urban sites, and 10 wadeable stream agricultural sites. In addition to the 113 NWQN sites, 3 large inland river monitoring sites from the USGS Cooperative Matching Funds (Co-op) program are also included in this annual water-quality reporting Web site to be consistent with previous USGS studies of nutrient transport in the Mississippi-Atchafalaya River Basin. This data release c ...

The GeoGraphNetworks data repository provides a comprehensive collection of networks, including the road and railway networks of the United States of America (USA) and the road and river networks of Great Britain (GB). This dataset offers nationwide coverage for both countries.The networks representing the road infrastructure of the USA include primary and secondary roads, covering all 50 states, 1 federal district (Washington, D.C.), and 5 territories, resulting in a total of 56 networks. The rail line infrastructure of the USA is represented as a single network that covers the entire country and includes connectivity to Canada. A total of 114 files , including 57 in Excel (.xlsx) format and 57 in JSON format, representing a total of 56 road networks and 1 rail network.Great Britain dataset comprises a comprehensive collection of 106 files, including 53 in Excel (.xlsx) format and 53 in JSON format, representing a total of 53 unique networks, 52 road networks and 1 river network. Each file is named with a unique code assigned by Ordnance Survey, which systematically divides Great Britain into 100 km by 100 km tiles (https://www.ordnancesurvey.co.uk/documents/product-support/user-guide/os-open-roads-overview.pdf).It allows researchers, urban planners, and anyone interested to explore how these networks connect people and places. With GeoGraphNetworks, you can easily analyze the travel routes and waterways that play a crucial role in the transportation systems of GB and the USA.The JSON files contain graph objects created using the widely used Python library NetworkX, allowing for immediate use without the need for pre-processing. Meanwhile, the Excel files are designed to support the construction of these graph networks across various platforms and programming languages, providing users with flexibility and ease of integration into their projects. This ensures that whether you're a developer, researcher, or data analyst, you can leverage this dataset effectively in your work.Visual representation of each network along with the code to use these networks (in Notebooks) are hosted on the Github Profile: https://github.com/Harsh9650/GeoGraphNetworks

Long-term monitoring data of geomorphic, hydrological, and biological characteristics of landscapes. This information provides an effective means of relating observed change to possible causes of the change. Identification of changes in basin characteristics, especially in arid areas where the response to altered climate or land use is generally rapid and readily apparent, might provide the initial direct indications that factors such as global warming and cultural impacts have affected the environment. The Vigil Network provides an opportunity for earth and life scientists to participate in a systematic monitoring effort to detect landscape changes over time, and to relate such changes to possible causes. This data release includes 70 sites and basins used to monitor landscape features. This data release includes information for Vigil Network sites monitored in the United States. The data and information in this data release are historical and were obtained from original documents.

Long-term monitoring data of geomorphic, hydrological, and biological characteristics of landscapes. This information provides an effective means of relating observed change to possible causes of the change. Identification of changes in basin characteristics, especially in arid areas where the response to altered climate or land use is generally rapid and readily apparent, might provide the initial direct indications that factors such as global warming and cultural impacts have affected the environment. The Vigil Network provides an opportunity for earth and life scientists to participate in a systematic monitoring effort to detect landscape changes over time, and to relate such changes to possible causes. This data release includes 70 sites and basins used to monitor landscape features. This data release includes information for Vigil Network sites monitored in the United States. The data and information in this data release are historical and were obtained from original documents.

The National Hydro Network (NHN) focuses on providing a quality geometric description and a set of basic attributes describing Canada's inland surface waters. It provides geospatial digital data compliant with the NHN Standard such as lakes, reservoirs, watercourses (rivers and streams), canals, islands, drainage linear network, toponyms or geographical names, constructions and obstacles related to surface waters, etc. The best available federal and provincial data are used for its production, which is done jointly by the federal and interested provincial and territorial partners. The NHN is created from existing data at the 1:50 000 scale or better. The NHN data have a great potential for analysis, cartographic representation and display and will serve as base data in many applications. The NHN Work Unit Limits were created based on Water Survey of Canada Sub-Sub-Drainage Area.

Nitrogen, phosphorus, and suspended-sediment loads, and changes in loads, in major rivers across the Chesapeake Bay watershed have been calculated using monitoring data from the Chesapeake Bay River Input Monitoring Network (RIM) stations for the period 1985 through 2017. Nutrient and suspended-sediment loads and changes in loads were determined by applying a weighted regression approach called WRTDS (Weighted Regression on Time, Discharge, and Season). The load results represent the total mass of nitrogen, phosphorus, and suspended sediment that was exported from each of the RIM watersheds.

A collection of analysis-ready datasets for the U.S. Geological Survey - Idaho National Laboratory (USGS-INL) groundwater and surface-water monitoring networks, administered by the USGS-INL Project Office in cooperation with the U.S. Department of Energy. The data collected from wells and surface-water stations at the Idaho National Laboratory and surrounding areas have been used to describe the effects of waste disposal on water contained in the eastern Snake River Plain aquifer, located in the southeastern part of Idaho, and the availability of water for long-term consumptive and industrial use. The datasets include long-term monitoring records dating back to measurements from 1922. Geospatial data describing the areas from which samples were collected or observations were made are also included.