CUAHIS-HIS
RSS

Navigation





Quick Search
»
Advanced Search »

PoweredBy

Use Cases

(With contributions from Tim Whiteaker, Dan Ames, Jon Goodall, and Jeff Horsburgh)

Base Use Case: Data Discovery, Download, Visualization, Export (TW)

The user wants to search for hydrologic observations time series that meet certain criteria, extract sites/variable descriptions/time series of interest, and enable the result to be imported or used in a variety of analysis packages. Potential filter and search criteria include:

  • Geographic region or feature of interest
  • Time period of interest
  • Specific data sources, e.g., USGS NWIS, my local data, all data registered with HIS Central
  • Specific variables, e.g., pH, or categories of variables, e.g., Nutrients
  • Data themes, e.g., Drought, Flooding, Total Maximum Daily Load
  • Minimum number of data values per site location per variable
  • Methods used for sampling the data

Constraints:
  • The user knows what a time series is, and has a very basic understanding of hydrology, as well as a basic understanding of how to use a computer
  • The user may not have an advanced understanding of specific hydrologic variables.
  • The user may not know what a web service is or how to access one.
  • The system should operate quickly and intuitively.
  • The learning curve should be as negligible as possible.
  • The system should be free of charge, and work on the most common operating systems.
  • The system should record "state," so that user a can come back later and continue a given search for data, or refresh a given search in case any new data are now available.

Use Case 1: Basic Data Search, Download, Visualization, and Export (DA)

The user has a study area defined by an area of interest (AOI) as either a user supplied polygon, or previously defined HUC, or other hydrologic or political boundary (county, state). The AOI can be as large as a state or as small as a waste management facility. The user wants to find all data of parameter X that is within the bounds or within a buffer of interest near this AOI. For example, all streamflow data.
  • The user supplies the AOI, and parameter and sees a map showing all stations/sites that meet the criteria regardless of which HIS Server the data is coming from.
  • The user can see a layer of all stations OR can select specific "layers" from a legend that characterize the points based on the data available at that station. This gives the effect of a "streamflow layer" or a "phosphorus layer" that can be turned on or off.
  • The points/sites are symbolized based on count.
  • The user can then plot the data from any station
  • The user can view the description and metadata of every station
  • The user can view summary statistics of all the data
  • The user can export the time series data into one of several file formats

Constraints
  • The user does not know what an HIS server is, or a WSDL.
  • They have no idea which servers are out there, and which ones have data in them near his or her AOI.

Wishes
  • The user has an AOI, but wants more than one variable.
  • The user wants to pick a subset of the sites that are returned before they decide to do the data download.

DA. Is a layer just variable-location, or could it be location-location attribute(station type) (use case 7). How is layer information created/maintained



Use Case 2: Time Series Data Support for Modeling (JG)

The user wants to do analysis or modeling that requires hydrologic observations time series as input. Some of these time series are available from HIS Servers, but others are not. The time series not available from HIS Servers might be available from other data distribution systems or collected by the user, so the user is able to obtain these data. To meet this need, the system should provide a mechanism by which the user can easily combine "personal" data along with data stored in HIS Servers in a seamless way for modeling and analysis tasks.

Constraints:
  • The user is not interested in loading his/her data into a HIS Server.
  • The user has experience working hydrologic models and common data formats.
  • The user does not have programming experience, particularly related to web services and XML.

Can we get an specific use case for use case 2. With what data is needed for a model, and where we would expect it to be found. A case with local data that never shared is needed.

Use Case 3: Dynamic Data Update via Subscription (JH)

A user has assembled a number of datasets on their desktop machine using HIS Desktop. Some of those datasets are dynamic (i.e., realtime streamflow gages). The user wants to “subscribe” to the data services to make sure that every time they do an analysis on their desktop they have the most recent data in their local repository.
  • The user finds the data using HIS Desktop
  • The user subscribes to the dataset using HIS Desktop
  • The user gets an initial copy of the data using HIS Desktop
  • HIS Desktop periodically checks for updates and notifies user when new data are available
  • The user can manage their subscription (i.e., get new data, delete the whole thing, reload the whole thing).

Use Case 4: Data Combination and Transformation (JH)

A user has assembled a number of datasets on their desktop machine using HIS Desktop. Some of those datasets have different units from their native sources. The user wants to combine data from multiple sources in common analyses using desktop analysis software (e.g., Excel, MATLAB, R, etc.). The user wants to make sure to keep information about how the data were transformed so they can be tracked back to their individual sources.
  • The user has found the data and downloaded it using HIS Desktop
  • The user converts units of data they have downloaded
  • The user exports data in common formats that can be easily used in analysis software (Excel, MATLAB, R, etc.)
  • HIS Desktop maintains the provenance and metadata for the datasets

Where is the provenance for processing stored, and what core information is it associated with (data value, site, or variable)

Use Case 5: Data Search for Spatially/Temporally Contemporaneous Data (JH)

A user has variable(s)/topic(s)/process(es) of interest. For example, I am studying primary productivity in streams and rivers. I want to use high frequency, diurnal dissolved oxygen, temperature, specific conductance, and pH to extract estimates for photosynthesis, respiration, and reaeration rates using R. I am interested in doing this at many different sites across the country so I want to use HIS Desktop to search for and download data from locations that have corresponding data for all of these variables. I only want sites where the period of record overlaps for all of the variables and where the data have the same or similar spacing (i.e., hourly data or more frequent). The data retrieved using HIS Desktop must be easily imported to R.
  • The user supplies keywords for searching across all published data networks
  • The user supplies complex query criteria (number of observations, date range, presence of other variables, data spacing, etc.)
  • The user sees a GIS map of all sites that meet the search criteria
  • The user can get different types of data (observations versus model simulation results)
  • The user can visualize and screen data prior to download (extended metadata view, plots, and summary statistics)
  • The user can select a subset of the search results for download for local analysis

Is this a request for a bi-directional link to systems, or download and export?

Use Case 6: Complex Queries and Subsetting (JH)

With the WaterOneFlow Web Services, users are very limited in the set of methods available for querying and retrieving subsets of data. Using HIS Desktop, a user has assembled a number of different datasets in a data repository on their desktop machine. The user wants to manipulate the data and do complex queries for subsetting and combining datasets (i.e., generate a time-matched data table with DateTime, Temperature, Dissolved Oxygen Concentration as columns for export).
  • The user can connect to the local HIS Desktop data repository directly from Excel, MATLAB, or R
  • The user can run standard TSQL queries on the data to retrieve data directly into the analysis software (i.e., "return the number of days in a data series that dissolved oxygen dipped below 6 mg/L", or "give me daily average discharge derived from 15-minute data for the month of June")
  • The user can export query results to a number of formats compatible with analysis software (text files, Excel files)

Use Case 7: Combine Spatial and Temporal Data (JH)

A user is studying the effect of land use on water quality in the Little Bear River watershed. The user wants to get locations of water quality monitoring sites and plot them on a map that shows streams, the underlying terrain, and land use data. The user then wants to delineate the catchment areas contributing to each monitoring site and then summarize the land use within each site's catchment. The user does the following in HIS Desktop. Spatial data downloads would optimally be done automatically by HIS Desktop.
  • The user gets the locations of water quality monitoring sites in the Little Bear River and plots them on a map
  • The user gets National Hydrography Dataset streams for the Little Bear River watershed and adds them to the map
  • The user gets the National Elevation Dataset elevation data for the Little Bear River watershed and adds it to the map
  • The user gets the National Land Cover Dataset land cover data for the Little Bear River watershed and adds it to the map
  • The user delineates catchments for monitoring sites and then summarizes the land cover data within the catchment polygons to determine land cover areas contributing to each site
  • The user then exports the derived areas in a format that can be accessed using R for analyses
  • The user gets specific water quality data for monitoring sites in the Little Bear River from published data services and imports them to R for analyses
    Good detail. What is the definition of "Water quality monitioring site?"


Use Case 8: Flood Inundation Mapping (Nick Arnold)

In an effort to provide more informative flood data to the citizens of Iowa City, a user wants to develop semi-real time flood inundation mapping system relating with Iowa River discharges released from the Coralville Dam extending downstream to the southern corporate limit of Iowa City. The user also wants to archive flood inundation maps for further water management purposes.

  • The user will use HEC-RAS model to simulate a range of flow conditions in the Iowa River.
  • The user desires to use the HEC-RAS model in OpenMI working in HIS Desktop coupled with varying flow conditions using HydroLink.
  • The user wants to dynamically get discharge data from Coralville Dam (US Army Corps) as an initial condition of HEC-RAS and run    the HEC-RAS model based on a given flow condition.
  • The user already has pre-processed topologic data for the HEC-RAS, and imports them into HIS Desktop.
  • The HEC-RAS model is calibrated and validated against existing information (e.g., USGS gage height) by adjusting a series of parameters.
  • The user wants to get a time series of polygons of flood inundation maps from the HEC-RAS simulated results.
  • The user wants to export the developed and archived Inundation maps to Google Earth format for public dissemination.

Use Case 9: Effects of Land-use/Practice Changes on Water Quantity and Quality in Streams (Nick Arnold)

Changes of the crop type (driven by economic or other incentives) trigger adjustments of land practices that further modify infiltration rates and erosion patterns on the lots subjected to land change. Multiple stakeholders want to determine (if possible in near-real time) the implications of adopting alternative land use scenarios on the individual agricultural lots and decide collectively on the optimum land use solution using a transparent and easily accessible communication platform. The user wants to run SWAT model to obtain water quantity and quality indicators that are compared with their threshold values. The process is iteratively repeated to guide the decision making.

  • User has pre-processed input files for SWAT via AGWA/AVSWAT
  • User wants an interface for editing SWAT input files related with land use such as crop rotation, land use managements and  others to reflect alternative land use scenario.
  • User wants to prepare weather data dynamically obtained from data sources such as NCDC NWS, and Hydro-Nexrad, and converts them SWAT input file format.
  • User executes SWAT and visualizes results.
  • User wants to calibrate SWAT modeling using existing flow (USGS gage station) and water quality data (EPA Storet).
  • User wants to archive the results of SWAT modeling based on land use scenarios.
    Can we get a document on the SWAT format, and a mapping from values to SWAT


Use Case 10: Scenarios contributed by Paul Montagna, Harte Research Institute for Gulf of Mexico Studies

I can imagine at least 3 different scenarios: 1) primarily a time-series question, 2) primarily a spatial variability question, and 3) a experimental question about space and time. The following all assume a map-based user interface (like DASH).

  1. Time series scenario: User wants to locate an area where there is the highest (or lowest) probability of low dissolved oxygen or high ammonium concentrations and plot a time series to identify when it is most likely to occur. There are really several compound problems here. First, which station? There needs to be a way to quickly identify the right station. Second, which variables are available? Third, finally graph out the time series of selected data. The main metadata here is xyz, that is latitude, longitude, and depth.
  2. Spatial scenario: User wants to analyze how a variable behaves spatially. Here the problem is to select all the stations in an area. Imagine being able to draw a box on a map and then all those stations are selected. The next problem is how to aggregate the data when there are multiple time measurements? There might need to be a seasonal weighting. For example, suppose 5 measurements are made in summer and only 1 is made in winter, then you first calculate the means by month so the number of samples doesn’t drive the answer, and then take the means over the year of just the monthly data. Finally, might want to interpolate the data over space. Again, have to sort out depth issues.
  3. Experimental studies: A typical experimental study will have spatial treatments and repeated measures over time. The resulting data is a matrix where rows are samples at different times, and columns are different locations in space. The problem is that each cell might have replicates or multiple measurements as well. How does one select that data set where specific space and time combinations are wanted? The user would want means, variances, and sample sizes.

ScrewTurn Wiki version 3.0.1.400. Some of the icons created by FamFamFam.