SWOT River Database (SWORD)

** IMPORTANT UPDATE: ** Until now, the project and public versions of SWORD have been kept separate while algorithms were being developed in preparation for SWOT launch. Now that the SWOT mission is here, we have decided to publish the project version of SWORD which is why the version numbers jump after v2. The primary difference between the project and public versions of SWORD are extra "filler" variables in the NetCDF format that will be used for calculating discharge. Everything else, reach definition, attribute values, etc. are the same between the two versions. For details on the filler variables please reference the Product Description Document provided with the downloads. If you use the SWORD Database in your work, please cite: Altenau et al., (2021) The Surface Water and Ocean Topography (SWOT) Mission River Database (SWORD): A Global River Network for Satellite Data Products. Water Resources Research. https://doi.org/10.1029/2021WR030054 You can also visit www.swordexplorer.com to explore the current version of SWORD before downloading. 1. Summary: The upcoming Surface Water and Ocean Topography (SWOT) satellite mission, planned to launch in 2022, will vastly expand observations of river water surface elevation (WSE), width, and slope. In order to facilitate a wide range of new analyses with flexibility, the SWOT mission will provide a range of relevant data products. One product the SWOT mission will provide are river vector products stored in shapefile format for each SWOT overpass (JPL Internal Document, 2020b). The SWOT vector data products will be most broadly useful if they allow multitemporal analysis of river nodes and reaches covering the same river areas. Doing so requires defining SWOT reaches and nodes a priori, so that SWOT data can be assigned to them. The SWOt River Database (SWORD) combines multiple global river- and satellite-related datasets to define the nodes and reaches that will constitute SWOT river vector data products. SWORD provides high-resolution river nodes (200 m) and reaches (~10 km) in shapefile and netCDF formats with attached hydrologic variables (WSE, width, slope, etc.) as well as a consistent topological system for global rivers 30 m wide and greater. 2. Data Formats: The SWORD database is provided in netCDF, geopackage, and shapefile formats. All files start with a two-digit continent identifier ("af" – Africa, "as" – Asia / Siberia, "eu" – Europe / Middle East, "na" – North America, "oc" – Oceania, "sa" – South America). File syntax denotes the regional information for each file and varies slightly between netCDF and shapefile formats. NetCDF files are structured in 3 groups: centerlines, nodes, and reaches. The centerline group contains location information and associated reach and node ids along the original GRWL 30 m centerlines (Allen and Pavelsky, 2018). Node and reach groups contain hydrologic attributes at the ~200 m node and ~10 km reach locations (see description of attributes below). NetCDFs are distributed at continental scales with a filename convention as follows: [continent]_sword_v16.nc (i.e. na_sword_v16.nc). SWORD shapefiles consist of four main files (.dbf, .prj, .shp, .shx). There are separate shapefiles for nodes and reaches, where nodes are represented as ~200 m spaced points and reaches are represented as polylines. All shapefiles are in geographic (latitude/longitude) projection, referenced to datum WGS84. Shapefiles are split into HydroBASINS (Lehner and Grill, 2013) Pfafstetter level 2 basins (hbXX) for each continent with a naming convention as follows: [continent]_sword_[nodes/reaches]_hb[XX]_v16.shp (i.e. na_sword_nodes_hb74_v16.shp; na_sword_reaches_hb74_v16.shp). SWORD geopackage files are split into two files for nodes and reaches per continental region, where nodes are represented as 200 m spaced points and reaches are represented as polylines. All geopackage files are in geographic (latitude/longitude) projection, referenced to datum WGS84. Geopackage file names are distributed at continental scales and are defined by a two-digit identifier (Table 2): [continent]_sword_[nodes/reaches]_v16.gpkg (i.e. na_sword_nodes_v16.gpkg; na_sword_reaches_v16.gpkg). 3. Attribute Description: This list contains the primary attributes contained in the SWORD netCDFs and shapefiles. x: Longitude of the node or reach ranging from 180°E to 180°W (units: decimal degrees). y: Latitude of the node or reach ranging from 90°S to 90°N (units: decimal degrees). node_id: ID of each node. The format of the id is as follows: CBBBBBRRRRNNNT where C = Continent (the first number of the Pfafstetter basin code), B = Remaining Pfafstetter basin code up to level 6, R = Reach number (assigned sequentially within a level 6 basin starting at the downstream end working upstream), N = Node number (assigned sequentially within a reach starting at the downstream end working upstream), T = Type (1 – river, 3 – lake on river, 4 – dam or waterfall, 5 – unreliable topology, 6 – ghost node). node_length (node files only): Node length measured along the GRWL centerline points (units: meters). reach_id: ID of each reach. The format of the id is as follows: CBBBBBRRRRT where C = Continent (the first number of the Pfafstetter basin code), B = Remaining Pfafstetter basin codes up to level 6, R = Reach number (assigned sequentially within a level 6 basin starting at the downstream end working upstream, T = Type (1 – river, 3 – lake on river, 4 – dam or waterfall, 5 – unreliable topology, 6 – ghost reach). reach_length (reach files only): Reach length measured along the GRWL centerline points (units: meters). wse: Average water surface elevation (WSE) value for a node or reach. WSEs are extracted from the MERIT Hydro dataset (Yamazaki et al., 2019) and referenced to the EGM96 geoid (units: meters). wse_var: WSE variance along the GRWL centerline points used to calculate the average WSE for each node or reach (units: square meters). width: Average width for a node or reach (units: meters). width_var: Width variance along the GRWL centerline points used to calculate the average width for each node or reach (units: square meters). max_width: Maximum width value across the channel for each node or reach that includes island and bar areas (units: meters). facc: Maximum flow accumulation value for a node or reach. Flow accumulation values are extracted from the MERIT Hydro dataset (Yamazaki et al., 2019) (units: square kilometers). n_chan_max: Maximum number of channels for each node or reach. n_chan_mod: Mode of the number of channels for each node or reach. obstr_type: Type of obstruction for each node or reach based on the Globale Obstruction Database (GROD, Whittemore et al., 2020) and HydroFALLS data (http://wp.geog.mcgill.ca/hydrolab/hydrofalls). Obstr_type values: 0 - No Dam, 1 - Dam, 2 - Channel Dam, 3 - Lock, 4 - Low Permeable Dam, 5 - Waterfall. grod_id: The unique GROD ID for each node or reach with obstr_type values 1-4. hfalls_id: The unique HydroFALLS ID for each node or reach with obstr_type value 5. dist_out: Distance from the river outlet for each node or reach (units: meters). type: Type identifier for a node or reach: 1 – river, 2 – lake off river, 3 – lake on river, 4 – dam or waterfall, 5 – unreliable topology, 6 – ghost reach/node. lakeflag: GRWL water body identifier for each reach: 0 – river, 1 – lake/reservoir, 2 – canal, 3 – tidally influenced river. manual_add (node files only): Binary flag indicating whether the node was manually added to the public GRWL centerlines (Allen and Pavelsky, 2018). These nodes were originally given a width = 1, but have since been updated to have the reach width values. meand_len (node files only): Length of the meander that a node belongs to, measured from beginning of the meander to its end in meters. For nodes longer than one meander, the meander length will represent the average length of all meanders belonging to the node (units: meters). sinuosity (node files only): The total reach length the node belongs to divided by the Euclidean distance between the reach end points. slope (reach files only): Reach average slope calculated along the GRWL centerline points. Slopes are calculated using a linear regression (units: meters/kilometer). n_nodes (reach files only): Number of nodes associated with each reach. n_rch_up (reach files only): Number of upstream reaches for each reach. n_rch_down (reach files only): Number of downstream reaches for each reach. rch_id_up (reach files only): Reach IDs of the upstream neighboring reaches. rch_id_dn (reach files only): Reach IDs of the downstream neighboring reaches. swot_obs (reach files only): The maximum number of SWOT passes to intersect each reach during the 21 day orbit cycle. swot_orbits (reach files only): A list of the SWOT orbit tracks that intersect each reach during the 21 day orbit cycle. river_name: All river names associated with a node or reach. If there are multiple names for a node or reach they are listed in alphabetical order and separated by a semicolon. edit_flag: Numerical flag indicating the type of update applied to SWORD nodes or reaches from the previous version. Flag descriptions are listed in the Product Description Documentation included with the file downloads. trib_flag: Binary flag indicating if a large tributary not represented in SWORD is entering a node or reach. 0 - no tributary, 1 - tributary. 4. References: Allen, G. H., & Pavelsky, T. M. (2018). Global extent of rivers and streams. Science, 361(6402), 585-588. Altenau, E. H., Pavelsky, T. M., Durand, M. T., Yang X., Frasson, R. P. d. 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. Biancamaria, S., Lettenmaier, D. P., & Pavelsky, T. M. (2016). The SWOT mission and its capabilities for land hydrology. In Remote Sensing and Water Resources (pp. 117-147). Springer, Cham. JPL Internal Document (2020b). Surface Water and Ocean Topography Mission Level 2 KaRIn high rate river single pass vector product, JPL D-56413, Rev. A, https://podaac-tools.jpl.nasa.gov/drive/files/misc/web/misc/swot_mission_docs/pdd/D-56413_SWOT_Product_Description_L2_HR_RiverSP_20200825a.pdf 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(15): 2171–2186. Data is available at www.hydrosheds.org. Tessler, Z. D., Vörösmarty, C. J., Grossberg, M., Gladkova, I., Aizenman, H., Syvitski, J. P. M., & Foufoula-Georgiou, E. (2015). Profiling risk and sustainability in coastal deltas of the world. Science, 349(6248), 638-643. Whittemore, A., Ross, M. R., Dolan, W., Langhorst, T., Yang, X., Pawar, S., Jorissen, M., Lawton, E., Januchowski-Hartley, S., & Pavelsky, T. (2020). A Participatory Science Approach to Expanding Instream Infrastructure Inventories. Earth's Future, 8(11), e2020EF001558. Yamazaki, D., Ikeshima, D., Sosa, J., Bates, P. D., Allen, G., & Pavelsky, T. (2019). MERIT Hydro: A high-resolution global hydrography map based on latest topography datasets. Water Resources Research. https://doi.org/10.1029/2019WR024873. Yang, X., Pavelsky, T. M., Allen, G. H. (2019). The past and future of global river ice. Nature. SWOT Orbits: https://www.aviso.altimetry.fr/en/missions/future-missions/swot/orbit.html HydroFALLS: http://wp.geog.mcgill.ca/hydrolab/hydrofalls/

**重要更新：** 此前，在为SWOT（Surface Water and Ocean Topography）任务发射开发算法期间，SWORD的项目版与公共版一直保持分离。如今SWOT任务已正式启动，我们决定发布SWORD的项目版，这也是v2版本后版本号出现跳跃的原因。SWORD项目版与公共版的核心差异仅在于NetCDF格式中用于径流计算的额外“填充”变量。其余内容，包括河段定义、属性值等，两个版本完全一致。有关填充变量的详细信息，请参阅下载包附带的产品说明文档。若您在研究中使用了SWORD数据库，请引用如下文献：Altenau等人（2021）《地表水与海洋地形（SWOT）任务河流数据库（SWORD）：面向卫星数据产品的全球河网》，《水资源研究》。https://doi.org/10.1029/2021WR030054。您也可以在下载前访问www.swordexplorer.com浏览当前版本的SWORD。

1. 摘要：计划于2022年发射的地表水与海洋地形（SWOT）卫星任务，将大幅拓展河流水面高程（Water Surface Elevation, WSE）、宽度与坡度的观测能力。为灵活支持各类新型分析研究，SWOT任务将提供一系列相关数据产品。其中一类产品为针对每次SWOT过境场景的矢量河流数据，以shapefile格式存储（JPL内部文档，2020b）。若要实现对同一河流区域的河道节点与河段进行多时间序列分析，SWOT矢量数据产品将具备最广泛的应用价值，而这需要先验定义SWOT河段与节点，以便将SWOT观测数据匹配至这些要素中。SWOT河流数据库（SWORD）整合了多套全球河流与卫星相关数据集，用于定义构成SWOT矢量河流数据产品的节点与河段。SWORD以shapefile与netCDF格式提供高分辨率的河流节点（200米）与河段（约10千米），附带水文变量（包括WSE、宽度、坡度等），并为宽度≥30米的全球河流构建了统一的拓扑系统。

2. 数据格式：SWORD数据库提供netCDF、geopackage与shapefile三种格式。所有文件均以两位大陆标识符开头（"af"对应非洲，"as"对应亚洲/西伯利亚，"eu"对应欧洲/中东，"na"对应北美洲，"oc"对应大洋洲，"sa"对应南美洲）。文件命名规则用于标识各文件的区域信息，netCDF与shapefile格式的命名规则略有差异。NetCDF文件分为3个组：中心线组、节点组与河段组。中心线组包含原始GRWL 30米中心线的位置信息，以及关联的河段与节点ID（Allen与Pavelsky，2018）。节点组与河段组则包含~200米节点与~10千米河段位置处的水文属性（详见下文属性说明）。NetCDF文件按大陆尺度分发，文件名约定格式如下：[continent]_sword_v16.nc（例如na_sword_v16.nc）。SWORD的shapefile包含四个核心文件（.dbf、.prj、.shp、.shx），节点与河段分别拥有独立的shapefile：节点以~200米间隔的点要素表示，河段以多段线要素表示。所有shapefile均采用地理（纬度/经度）投影，基准为WGS84大地基准。针对各大陆，shapefile按HydroBASINS（Lehner与Grill，2013）的Pfafstetter二级流域（hbXX）进行拆分，文件名约定格式如下：[continent]_sword_[nodes/reaches]_hb[XX]_v16.shp（例如na_sword_nodes_hb74_v16.shp；na_sword_reaches_hb74_v16.shp）。SWORD的geopackage文件按大陆区域拆分为节点与河段两个文件，节点以200米间隔的点要素表示，河段以多段线要素表示。所有geopackage文件均采用地理（纬度/经度）投影，基准为WGS84大地基准。Geopackage文件按大陆尺度分发，文件名由两位标识符定义（见表2），格式如下：[continent]_sword_[nodes/reaches]_v16.gpkg（例如na_sword_nodes_v16.gpkg；na_sword_reaches_v16.gpkg）。

3. 属性说明：以下为SWORD的NetCDF与shapefile文件中包含的核心属性。
x：节点或河段的经度，范围为东经180°至西经180°，单位：十进制度。
y：节点或河段的纬度，范围为南纬90°至北纬90°，单位：十进制度。
node_id：每个节点的ID，格式为CBBBBBRRRRNNNT，其中：C为大陆编码（Pfafstetter流域编码的首位数字）；B为剩余的Pfafstetter流域编码（最高至6级）；R为河段编号（在6级流域内从下游向上游依次分配）；N为节点编号（在河段内从下游向上游依次分配）；T为类型（1-河道，3-河道上的湖泊，4-大坝或瀑布，5-不可靠拓扑，6-伪节点）。
node_length（仅节点文件包含）：沿GRWL中心线点测量的节点长度，单位：米。
reach_id：每个河段的ID，格式为CBBBBBRRRRT，其中：C为大陆编码（Pfafstetter流域编码的首位数字）；B为剩余的Pfafstetter流域编码（最高至6级）；R为河段编号（在6级流域内从下游向上游依次分配）；T为类型（1-河道，3-河道上的湖泊，4-大坝或瀑布，5-不可靠拓扑，6-伪河段）。
reach_length（仅河段文件包含）：沿GRWL中心线点测量的河段长度，单位：米。
wse：节点或河段的平均水面高程（WSE）值。WSE数据提取自MERIT Hydro数据集（Yamazaki等人，2019），并以EGM96大地水准面为基准，单位：米。
wse_var：用于计算每个节点或河段平均WSE的GRWL中心线点上的WSE方差，单位：平方米。
width：节点或河段的平均宽度，单位：米。
width_var：用于计算每个节点或河段平均宽度的GRWL中心线点上的宽度方差，单位：平方米。
max_width：每个节点或河段的河道最大宽度，包含岛屿与沙洲区域，单位：米。
facc：每个节点或河段的最大汇流值。汇流值提取自MERIT Hydro数据集（Yamazaki等人，2019），单位：平方千米。
n_chan_max：每个节点或河段的最大河道数量。
n_chan_mod：每个节点或河段的河道数量众数。
obstr_type：基于全球障碍物数据库（GROD，Whittemore等人，2020）与HydroFALLS数据（http://wp.geog.mcgill.ca/hydrolab/hydrofalls）定义的节点或河段障碍物类型。obstr_type取值如下：0-无坝，1-大坝，2-河道坝，3-船闸，4-低渗透性坝，5-瀑布。
grod_id：obstr_type取值为1-4的节点或河段对应的唯一GROD ID。
hfalls_id：obstr_type取值为5的节点或河段对应的唯一HydroFALLS ID。
dist_out：节点或河段至河流出口的距离，单位：米。
type：节点或河段的类型标识符：1-河道，2-河道外湖泊，3-河道上的湖泊，4-大坝或瀑布，5-不可靠拓扑，6-伪河段/节点。
lakeflag：每个河段的GRWL水体标识符：0-河道，1-湖泊/水库，2-运河，3-受潮汐影响的河道。
manual_add（仅节点文件包含）：二进制标记，用于标识该节点是否被手动添加至公共GRWL中心线（Allen与Pavelsky，2018）。此类节点最初被赋值width=1，后续已更新为对应河段的宽度值。
meand_len（仅节点文件包含）：节点所属曲流的长度，从曲流起点至终点的测量值，单位：米。若一个节点包含多个曲流，则meand_len为该节点所有曲流的平均长度。
sinuosity（仅节点文件包含）：节点所属河段的总长度除以河段端点间的欧氏距离。
slope（仅河段文件包含）：沿GRWL中心线点计算的河段平均坡度，采用线性回归方法计算，单位：米/千米。
n_nodes（仅河段文件包含）：每个河段关联的节点数量。
n_rch_up（仅河段文件包含）：每个河段的上游河段数量。
n_rch_down（仅河段文件包含）：每个河段的下游河段数量。
rch_id_up（仅河段文件包含）：上游相邻河段的ID。
rch_id_dn（仅河段文件包含）：下游相邻河段的ID。
swot_obs（仅河段文件包含）：在21天轨道周期内，可与该河段相交的最大SWOT过境次数。
swot_orbits（仅河段文件包含）：在21天轨道周期内，可与该河段相交的所有SWOT轨道轨迹列表。
river_name：与节点或河段关联的所有河流名称。若一个节点或河段存在多个名称，则按字母顺序排列，以分号分隔。
edit_flag：数值标记，用于标识SWORD节点或河段相较于上一版本的更新类型，标记的详细说明请参阅下载包附带的产品说明文档。
trib_flag：二进制标记，用于标识是否有未被SWORD收录的大型支流汇入该节点或河段。0-无支流汇入，1-有支流汇入。

4. 参考文献：
Allen, G. H. 与 Pavelsky, T. M. (2018). 全球河流与溪流分布范围. 《科学》, 361(6402), 585-588.
Altenau, E. H., Pavelsky, T. M., Durand, M. T., Yang, X., Frasson, R. P. d. M. 与 Bendezu, L. (2021). 《地表水与海洋地形（SWOT）任务河流数据库（SWORD）：面向卫星数据产品的全球河网》. 《水资源研究》.
Biancamaria, S., Lettenmaier, D. P. 与 Pavelsky, T. M. (2016). SWOT任务及其在陆地水文学中的应用能力. 收录于《遥感与水资源》(pp. 117-147). 瑞士尚贝里：Springer.
JPL内部文档（2020b）. 地表水与海洋地形任务二级KaRIn高速河流单次过境矢量产品, JPL D-56413, 修订版A. https://podaac-tools.jpl.nasa.gov/drive/files/misc/web/misc/swot_mission_docs/pdd/D-56413_SWOT_Product_Description_L2_HR_RiverSP_20200825a.pdf
Lehner, B., Grill G. (2013). 全球河流水文与网络汇流：研究世界大型河系的基准数据与新方法. 《水文过程》, 27(15): 2171–2186. 数据可从www.hydrosheds.org获取.
Tessler, Z. D., Vörösmarty, C. J., Grossberg, M., Gladkova, I., Aizenman, H., Syvitski, J. P. M. 与 Foufoula-Georgiou, E. (2015). 全球沿海三角洲的风险与可持续性分析. 《科学》, 349(6248), 638-643.
Whittemore, A., Ross, M. R., Dolan, W., Langhorst, T., Yang, X., Pawar, S., Jorissen, M., Lawton, E., Januchowski-Hartley, S. 与 Pavelsky, T. (2020). 采用参与式科学方法拓展河道基础设施清单. 《地球的未来》, 8(11), e2020EF001558.
Yamazaki, D., Ikeshima, D., Sosa, J., Bates, P. D., Allen, G. 与 Pavelsky, T. (2019). MERIT Hydro：基于最新地形数据集的高分辨率全球水文地图. 《水资源研究》. https://doi.org/10.1029/2019WR024873.
Yang, X., Pavelsky, T. M., Allen, G. H. (2019). 全球河流冰的过去与未来. 《自然》.
SWOT轨道信息：https://www.aviso.altimetry.fr/en/missions/future-missions/swot/orbit.html
HydroFALLS：http://wp.geog.mcgill.ca/hydrolab/hydrofalls/

搜集汇总

背景与挑战

背景概述

SWOT River Database (SWORD) 是一个全球河流网络数据库，旨在为SWOT卫星任务提供先验定义的河流节点和河段，以支持多时相水文分析。该数据集提供高分辨率（节点200米、河段约10公里）的河流数据，包含丰富的水文属性（如水位高程、宽度、坡度），覆盖全球宽度30米及以上的河流，并以netCDF、shapefile和geopackage多种格式发布。