1. Key Considerations
   • a. Data and Modeling Tools
     • i. Data on streamflow and stage from available gauges.
       1. U.S. Geological Survey Water Data for the Nation: http://waterdata.usgs.gov/nwis/
       2. Environment Canada Real Time Hydrometric Data: http://scitech.pyr.ec.gc.ca/waterweb/main.asp?lang=0
       3. Advisory Committee on Water Information, organized through the U.S. Department of the Interior and U.S. Geological Survey, provides a variety of water resources: http://acwi.gov/
       4. Other information collected by the States/Provinces.
     • ii. Hydrologic models simulate streamflow hydrographs for ungauged sites. To understand existing cumulative impacts, flows need to be estimated for reference conditions and current conditions. Examples:
       1. Massachusetts' Sustainable-Yield Estimator (SYE) uses a "modified QPPQ" regression approach for estimating daily streamflow at ungauged sites. USGS is currently extending this approach to Pennsylvania and the entire Connecticut River basin.
       2. Colorado Stream Simulation Model (StateMod) is a water balance model that simulates daily and monthly water allocation. It compares water management scenarios in large river basins and is part of the Colorado Water Conservation Board's Colorado Decision Support System. It may be accessed at: http://cdss.state.co.us/DNN/default.aspx
       3. Texas Water Availability Model (WAM) is a water balance model that simulates water management and allocation throughout a State. It may be accessed at: www.tceq.state.tx.us/permitting/water_supply/water_rights/wam.html
       4. AFINCH (Analysis of Flows in Networks of Channels-- a product of the National Assessment of Water Availability and Use Program) calculates monthly flow duration curves which may be used to generate flow statistics. AFINCH cannot estimate streamflow hydrographs, so it cannot estimate interannual flow variation, nor can it be used to analyze the extent to which natural flows have been altered. The Great Lakes Basin Pilot, a model of AFINCH, was performed in southwest Michigan and may be extended to the entire Basin as part of the Great Lakes Restoration Initiative. It may be accessed at: http://mi.water.usgs.gov/projects/GreatLakesWaterAvailability/surface.html
       5. In Quebec, a semi-distributed hydrological model named Hydrotel is in use for a large number of watersheds. The model algorithms are derived as much as possible from physical processes, together with more conceptual or empirical approaches. Computer code may be accessed at: http://hydrotel.codeplex.com/
       6. Other models that can be used to estimate streamflows are listed at:
          http://conserveonline.org/workspaces/eloha/documents/tools-for-building-a-hydrologic-foundation. Information includes capabilities, limitations and skills required to use the models. Approaches include watershed (rainfall-runoff) models such as SWAT, HSPF, PRMS and others. Some models have been used for Total Maximum Daily Load and other programs in various parts of the Great Lakes Basin.
          
          The Environment Canada Water Modeling Site (http://www.ec.gc.ca/water/en/manage/model/e_model.htm) describes water management models that have been applied in Canada and provides brief descriptions of specific applications.
       7. A general compilation of groundwater and surface water modeling software is available as part of the U.S. Geological Survey Water Resources Applications Software. It may be accessed at: http://water.usgs.gov/software/
2. Water Use. Cumulative impact analysis must consider the impacts of current and future water uses including return flows. A decision support model can be used to apply water use data to streamflow conditions and arrive at current hydrologic conditions for ungauged sites.
   • a. Key Considerations.
     • i. Diffuse (irrigation) vs. Discrete (wells or water intakes) water uses.
     • ii. How is use managed or modeled?
   • b. Key Data Sources.
     • i. Information collected by States and Provinces including remotely gathered (aerial photos and satellite images).
     • ii. Great Lakes Regional Water Use Database. It may be accessed at: http://www.glc.org/wateruse/database/

II. Biological data.

1. Inventory of important water-dependent biological resources.
   • a. Key Considerations.
     • i. Availability of data--fish have more data than other taxonomic groups.
     • ii. Sensitivity of taxa to altered flows.
     • iii. Sociopolitical connection with the biota (perceived "value").
     • iv. Ease and cost of monitoring.
   • b. Key Data Sources
     • i. State and Provincial water quality regulatory programs.
     • ii. State and Provincial fish and game programs.
     • iii. Federal biological monitoring programs.
     1. U.S. Geological Survey National Water-Quality Assessment Program: http://water.usgs.gov/nawqa/
     2. U.S. Environmental Protection Agency-Regional Environmental Monitoring and Assessment Program: http://www.epa.gov/emap/remap/index.html
     3. National Park Service-Inventory and Monitoring Program: http://science.nature.nps.gov/im/index.cfm
     • iv. Others including universities, museums or scientific literature.

III. Stream/River Classes or Types.

Of course, different regions have different climates, geology, soils, vegetation and topography creating differences in daily, seasonal and annual hydrology and biological systems. Organizing streams into ecohydrologic classes can assist in understanding and accounting for variability among streams. This is one means to help manage flows and better understand statistical relationships between flow alteration and changes in river conditions. Further information including examples for classifying rivers may be accessed at: http://conserveonline.org/workspaces/eloha/documents/river-types-0

Ecological Response Curves.
Modeled relationships between flow alteration and ecological response allow for impact estimation. A combination of models may be used:

1. Conceptual models (narrative, schematic or combination)
   • a. Consensus of interdisciplinary experts. Example: Verde River.
   • b. Literature review and analysis. Example: Colorado.
2. Statistical models (various analytical approaches)
   • a. Indirect (link to biota is indirectly established through key habitat variables, e.g., IFIM)
   • b. Empirical (use existing flow and biological data to develop statistical relationship) Example: Michigan.
   • c. Experimental (Manipulate key flow parameters and monitor biological response).

An overview and examples of these approaches using different biological indicators may be accessed at: http://conserveonline.org/workspaces/eloha/documents/flow-ecology-relationships-0. Information about specific modeling software may be accessed at: http://ecobas.org/www-server/index.html

V. Determining Instream Flow Needs and Assessing Cumulative Impacts of Withdrawals on Instream Flows.
The Instream Flow Council, comprising Provincial and State Instream Flow Program managers, provides guidance on maintaining environmental flows. It may be accessed at: http://www.instreamflowcouncil.org/

Models for determining instream flows include the following:
1. River-by-river basis (This is accomplished through field work and analyses of river segments.)
   • a. Southwest Florida Water Management District.
   • Washington and Texas are attempting to set river-by-river flow targets: http://www.texaswatermatters.org/flows.htm.
2. Determine flow targets regionally (This is accomplished without studying each individual water body.)
   • a. Ecological Limits of Hydrologic Alteration (ELOHA) combines hydrologic, water use and biological information to determine ecological response curves for different river types. It may be accessed at: http://conserveonline.org/workspaces/eloha/.
Tools for assessing cumulative impacts of water withdrawals on instream flows and their impacts on biological integrity include the following:
3. Michigan’s Water Withdrawal Assessment Tool links ecological conditions to a hydrologic model. This tool allows prospective water users to determine whether a new water use, combined with upstream uses, would adversely impact biological resources. It may be accessed at: http://www.miwwat.org/
4. Colorado’s Watershed Flow Evaluation Tool requires a less intensive data collection than the Michigan tool. This tool allows scientists to develop ecological response curves relating to the degree of high- and low-flow alteration. It may be accessed at: http://www.cdss.state.co.us/DNN/default.aspx.