Frequently Asked Questions
Below is a list of common questions from StreamStats users. If you have any further question, please contact us.
How is the information from StreamStats used?
The primary products delivered by StreamStats are streamflow statistics and basin characteristics. Examples of streamflow statistics include the 100-year flood, the mean annual flow, and the 7-day, 10-year low flow. Examples of basin characteristics include the drainage area, stream slope, mean annual precipitation and percentage of forested area. Basin characteristics are the physical factors that control delivery of water to a point on a stream. This information is used to protect people and property from floods and droughts, and to manage land, water, and biological resources. StreamStats also provides descriptive information for data-collection stations, such as station name, identification number, latitude, longitude, and station type.
Engineers, land managers, biologists, and many others use streamflow statistics to help guide decisions in their everyday work. For example, engineers use streamflow statistics to map flood plains for cities and towns, and to design roads, bridges, and culverts. Scientists use streamflow statistics and basin characteristics to model the effects of changes in land use on streamflow; information that is valuable for planning purposes. Land and water-resource managers use streamflow statistics for design and management of water supplies, and for waste-water discharge permitting. Biologists use the information to determine flows needed to protect endangered aquatic animals and their habitat. Corporations use streamflow statistics to design and operate hydroelectric facilities and factories that use water or discharge waste into streams or water bodies.
How do I use StreamStats and why are there two versions available?
StreamStats operates through your Web browser window. You do not need to download anything to use it. Currently, two versions are available; version 3 and beta version 4. Version 3 should be used any official purposes, such as for permitting decisions. Beta version 4 is being made available on a trial basis to obtain user feedback and to provide access to some tools that are not available in version 3. See the StreamStats home page for a description of the differences between the tools that are available between version 3 and beta version 4. The Version 3 User Instructions and the Version 4 User Instructions pages provide detailed instructions on use of the applications and example outputs, which you should read before attempting to use StreamStats. Also, both StreamStats user interfaces contain a Help button near the upper right corner that gives access to the user instructions and additional information.
How are streamflow statistics determined?
The USGS currently operates a nation-wide network of about 8,000 streamgaging stations, where streamflow is monitored continuously. Historically, the USGS also has operated more than 12,000 streamgaging stations that are not currently operated. In addition, the USGS operates thousands of partial-record stations where streamflow is measured periodically. Data collected at the continuous streamgaging stations and partial-record stations are used to calculate the streamflow statistics for the stations. These data may be accessed through the NWIS-Web database. The statistics are calculated by following standard procedures prescribed by the USGS Office of Surface Water.
A process known as regionalization is used to develop equations that can be used to estimate streamflow statistics for ungaged sites. Regionalization involves use of regression analysis to relate streamflow statistics computed for a group of selected streamgages to basin characteristics measured for the streamgages. Basin characteristics measured for ungaged sites can be entered into the resulting equations to obtain estimates of the streamflow statistics. Regionalization studies most commonly have been done for individual states through a cooperative funding mechanism whereby local (usually state) agencies partner with the USGS to fund at least half of the cost for the study.
Each state has an introductory page that identifies the streamflow statistics that can be estimated by use of regression equations, provides citations for the reports that contain the regression equations and special notes, if any, about use of StreamStats for that state, and identities of the agencies that cooperated with the USGS to implement the state. In version 3, the introductory pages also provide a link to the version 3 application for the state. Users should read the introductory page and the cited reports before using StreamStats for a state.
Accessing State Introductory pages:
Version 3 – From the StreamStats home page, click on the link to the State Applications page, then select an implemented state or river basin from the provided pull-down lists or select a state from the status map on that page, and the introductory page for the selected state or river basin will appear. Alternately, the clicking on the Application Information link at the bottom of the version 3 user interface will open a web browser window on the State Applications page.
Beta version 4 – After first selecting the state of interest in the user interface, a black circle with an "i" will appear beside the name of the state. Click on the circle to see the introductory page for that state. The introductory pages also can be accessed in version 4 by clicking on the About link above the map and then on the page that will subsequently appear, click on the State/Regional Info tab.
How does StreamStats work?
StreamStats was initially developed through a cooperative effort of the USGS and ESRI, Inc1. It is an integrated GIS application that is partly based on a combination of ArcGIS Server technology and the ArcHydro Tools,as well as on Python scripts. Most StreamStats functionality is available as web services. StreamStats incorporates a map-based user interface for site selection; a database that contains information for data-collection stations; a GIS program that delineates drainage-basin boundaries and measures physical and climatic characteristics of the drainage basins; and a GIS database that contains data needed for delineating basins, for measuring drainage-basin characteristics, and for locating sites of interest in the user interface. GIS data needed for delineations generally were developed by integrating elevation data from the USGS 3D Elevation Program, streams from the National Hydrography Dataset, and previously determined watershed boundaries from the Watershed Boundary Dataset. After StreamStats measures the drainage-basin characteristics for a selected site, the values are input to a separate program named the USGS National Streamflow Statistics (NSS) program, which is a Microsoft Windows program that contains all of the USGS-developed equations for estimating flood-frequency statistics in the Nation. NSS has been modified for StreamStats so that it can also contain equations for estimating other types of streamflow statistics. All of the equations in NSS are documented through links to each individual State from the NSS Web site. Output from NSS and from the Access database is presented to StreamStats users in a pop-up window on their computer desktop.
What streamflow statistics and basin characteristics does StreamStats provide?
The information provided varies by State, as programming for each State is separate in StreamStats. This allows the application to be customized for the needs of individual States. All streamflow statistics and basin characteristics provided for USGS data-collection stations were published previously in reports or on the Web, and a reference is provided for each item of information. The database contains more than 2,500 different types of streamflow statistics and basin characteristics, but only a small number of them will commonly be available for individual data-collection stations or ungaged sites. The Basin Characteristic Definitions and Streamflow Statistic Definitions pages provide tables showing the types of basin characteristics and streamflow statistics that may be available, respectively, and their definitions. The basin characteristics that are provided for ungaged sites will include all characteristics that are used in the regression equations that are applicable to the location and may include some additional basin characteristics.
What do the error terms mean in the outputs, such as Prediction Error and Prediction Interval?
Please refer to the descriptions of these terms on the Ungaged Sites page at http://streamstats.usgs.gov/ungaged.html. The section on Streamflow Statistics Tables describes the various error indicators.
Some of the statistics in the StreamStats report for a streamgage that I selected are more than 10 years old. Are there any newer values available?
The information that StreamStats provides in reports for data-collection stations is maintained in a database that must be updated by scientists who work in our USGS state offices. These offices generally rely on partnering with local agencies to cover at least half of the cost of the needed database work. Often, the database is updated as part of the process of initially implementing StreamStats, but additional updates subsequent to implementation and updates for states that are not implemented are difficult to achieve. As a result, a lot of the data in our database is out of date. The StreamStats development team plans to reduce this problem during 2016 by beginning a process of annually updating a set of about 100 non-interpretive statistics, which can be computed automatically by a computer without human intervention. These non-interpretive statistics will include means, medians, minimums, maximums, and standard deviations of annual, monthly, and daily flow time series for streamgages, as well as harmonic means and flow-duration percentiles computed from the daily flows. However, low-flow and peak-flow frequency statistics, such as the 7-day, 10-year low flow and the 100-year (1-percent probability) peak flow, are considered interpretive, meaning that a scientist needs to use professional judgment when computing those statistics. As a result, it is not possible to automate the updating of those types of statistics. The most recent interpretive statistics that are available for streamgages within a state generally are in reports that are cited on the StreamStats state introductory page for that state, if the state is implemented. A more comprehensive list of reports that contain regression equations and computed interpretive streamflow statistics for the streamgages used to develop the equations is at http://water.usgs.gov/osw/programs/nss/NSSpubs_Rural.html. Also, the USGS office for each state maintains a web page at http://xx.water.usgs.gov, where xx is the 2-letter postal cost for the state. The web pages normally provide links to publications that have produced by scientists in that state. A search of these publications may turn up additional reports that contain streamflow statistics.
If you need newer statistics for streamgages and can't wait for the UGGS to make them available, then you may be able to compute them yourself. You can download time series of daily flows needed to compute most flow statistics and annual series of instantaneous peak flows needed to compute flood-frequency statistics from our NWIS-Web site at http://waterdata.usgs.gov/nwis. The USGS's PeakFQ program is available for computing peak-flow frequency statistics. It can be downloaded at http://water.usgs.gov/software/PeakFQ/. This software uses procedures prescribed by the Interagency Advisory Committee on Water Data in Bulletin 17B. Users should note that a new version of PeakFQ is expected to be released with the release of Bulletin 17C during 2016. The new version of PeakFQ will incorporate the new methods for computing peak-flow statistics that will be prescribed in Bulletin 17C.
What are the blue pixelated lines? Are they real streams? Why are there so many/few of them?
We call these the “stream grid” or “synthetic stream grid”, which are used mainly as a guide for delineation. If you click off of these lines, your delineation may only include a few grid cells near the point you clicked. The grid lines are not visible in version 3 until you have zoomed into a scale of at least 1:18,056, or in beta version 4 until you have reached zoom level 15.
The grids are derived from digital elevation models (DEMs) of the terrain, which have been processed to agree with the digital representations of the stream lines from either 1:100,000- or 1:24,000-scale USGS topographic maps. They aren't necessarily true streams, especially in headwaters areas. For most states, they represent 10x10-meter grid cells having at least 900 grid cells upslope from them, based on the DEMs used to do basin delineations. In states using a 30x30-meter grid cell, including California, Oregon, and Washington, they represent grid cells having at least 100 grid cells upslope from them. Outside of headwaters areas, the stream grids usually will agree fairly well with the streams that are shown on the USGS topographic base map in the user interface. When they disagree, it usually will be because either there is a difference in scale between the source data used to generate the grids and the source material used to create the base maps, or the source materials are of a different age and changes were made in the interim.
In more weathered terrain, with more humid climates, the stream grids tend to correspond to drainage channels, and will often coincide with streams. However, they are not intended to mean there is a stream in that place; only that if there were overland flow that these are the areas where that flow would tend to accumulate, given the terrain represented in the DEM.
It's not really possible to say, in general terms, that the synthetic stream grid shown in StreamStats represents the "natural stream path." What it represents is the path water would take given the digital representation of the terrain from the DEM. There really are too many variables to make any blanket statements regarding what the natural stream path would be. The DEMs generally, (but not always) give us a pretty good idea of the size of the drainage basin and other basin characteristics, and from those we can estimate the streamflow volumes we would expect in typical situations, using regression equations. The characteristics of a particular site, however, always could include some unique circumstances that could change things, and site-specific information should be preferred over StreamStats.
What is the smallest drainage area that StreamStats can delineate accurately, and can I rely on flow estimates for small areas?
The accuracy of drainage-area delineations for small areas depends on accuracy and scale of the elevation data used in StreamStats for the delineations. The elevation data that StreamStats uses for most areas are from the USGS 3D Elevation Program dataset, which is a seamless national elevation dataset that is refreshed regularly as new data are made available. When implementing StreamStats, we take a snapshot of the elevation data that is available at the time for the area to be implemented, and the area may include data of varying ages and accuracy. As a result, it is difficult to say without detailed investigation what the accuracy of the elevation is at a particular location.
Generally, StreamStats is able to delineate basins with reasonable accuracy down to around 0.05 square miles (32 acres) in terrain with moderate relief. Delineations for flat areas will have lower accuracy. Although, StreamStats can delineate such small basins with reasonable accuracy, the errors associated with estimates of streamflow statistics for such small area often is unknown, and may be very large. This is because the USGS operates very few streamgages in such small basins that can be used in regression analyses to develop equations for estimating the streamflow statistics. In outputs for user-selected sites, StreamStats indicates when the drainage area, or any other basin characteristic used as an explanatory variable in the regression equations, is outside of the ranges of the values computed for the streamgages that were used to develop the equations. When that happens, the errors associated with the estimated streamflow statistics are unknown.
What does the cross-hatching on the map mean?
StreamStats has limited functionality in some areas. These are indicated on the map with a cross-hatch pattern: These are areas where StreamStats either can't delineate the drainage area, or there is some other limitation, such as not being able to solve regression equations there. If you try to do a delineation in those areas, you will see a message explaining what the situation is in that case. Some of the areas allow some functionality, but others don't allow you do delineations at all. It just depends on the circumstances. The way to find out is just to attempt a delineation there.
In a series of estimated flow statistics for an ungaged site, such as peak-flows at various recurrence intervals or flow-duration statistics, why do I sometimes get estimates for lower flows that are higher than the estimates for higher flows?
The regression equations that are used to estimate flow statistics for ungaged sites were developed using computed streamflow statistics and measured basin characteristics for a selected group of streamgages. The regression equations have errors associated with them that can result in reversals of magnitude in a series of flow estimates. These errors occur because the stations that are used in the regression analysis are only a sample of the streams within the region, and the sample does not capture all of the variability in the streamflow statistics and basin characteristics that would be obtained if all streams within the region were gaged. In addition, flow statistics and basin characteristics can’t be measured with 100 percent accuracy, and it is not possible to include in the regression equations all possible basin characteristics or other factors that cause flows to vary from site to site within the region. StreamStats provides one or more indicators of the accuracy of the estimates produced from regression equations. Often, differences in the magnitudes of flow estimates within a series may be small (for example, the difference between the 85- and 90-percent flow), and the errors associated with the estimates may be relatively large. As a result, it is possible that an equation for the 90-percent flow will give an estimate that is larger than the estimate from an equation for the 85-percent flow, even though the 85-percent flow is a larger flow than the 90-percent flow. Usually in these cases, if prediction intervals are available for the estimates, the prediction intervals for the estimates will overlap each other.
If I use StreamStats at an existing gage location, will I get the same results as if I computed statistics using the gage records?
It depends. If you use the Query Streamgage tool to get the statistics, then the statistics should be the same as what you computed if the same period of record was used, and if the statistics are interpretive (peak- and low-flow frequency statistics), the same interpretation of the data was made. If you obtained estimates using the Estimate Flows Using Regression Equations tool in version 3 or the Regression-Based Scenarios tool in beta version 4, then the estimates will be different, as they will be computed from regression equations rather than from the actual record at the streamgage. The regression equations provide estimates that represent somewhat of an area-based average of the peak-flow statistics for the streamgages in the region in which your site of interest is located. Flows at the streamgages used to develop the equations generally are minimally affected by human alterations. As a result, the equations provide estimates of flows that would be expected under essentially natural flow conditions.
What happened to the version 2 that allowed estimating flows based on the flows at an nearby streamgage and the other tools that allowed stream network navigation?
StreamStats version 2 had to be shut down on July 15, 2015 because the servers on which it ran used an obsolete operating system that was considered a security risk by the Department of Interior, of which the USGS is a part. Version 3 was placed in use before it was possible to develop code for all of the tools that were in version 3. Rather than continue to develop version 3, it was decided to put priority on replacing it with version 4, which feedback has indicated has a much improved user interface. After the initial release of beta version 4, emphasis is being placed on redeveloping the tools that were lost from version 2. It is planned to have all of the tools that were in version 2 operating in version 4 during 2016.
StreamStats is not available where I need information. Are there any alternatives?
Here are some alternatives to StreamStats, in case we have not yet implemented your area of interest:
If you want to get only a basin delineation, then you can use ESRI’s Watershed Explorer tool, which can be found at http://www.arcgis.com/home/item.html?id=bbd10d91da4f4ffbbdfd8db1d677053c. Use of this tool requires an ArcGIS Online account. Alternately, if you have a basin that is less than about 1.5 square miles, then you can use the USGS National Map Viewer to hand delineate the boundary and get the area. To do that, go to http://viewer.nationalmap.gov/viewer/, zoom into your location of interest, click on the Advanced tab above the map and then on the Area button, and then begin your delineation. Double-click to end it. The resulting area will be shown.
If you want a watershed polygon, that is quite a bit more complicated. There is a web service from EPA here: http://www.epa.gov/waters/geoservices/docs/navigation_delineation_service.html
You need to be a web developer, though, to figure out how to use that.
If you need to get the polygon, and you have access to ArcGIS 9.3.1. You can try the NHDPlus Basin Delineator Tool. You can get it here: http://www.horizon-systems.com/NHDPlus/NHDPlusV1_tools.php
There is newer data (NHDPlus Version 2), and of course newer ArcGIS software (10.1), but the tools haven't been updated to work on either of those yet.
What can I do to help get StreamStats operational for my State?
The USGS Water Science Center Director for your State.
How should StreamStats be cited?
If you want to cite the program in general, use:
U.S. Geological Survey, 2012, The StreamStats program, online at http://streamstats.usgs.gov.
If you want to cite a specific state application, for example, use:
U.S. Geological Survey, 2012, The StreamStats program for Colorado, online at http://water.usgs.gov/osw/streamstats/colorado.html.