Michigan Stream Evaluator

Background and Instructions

Background

Those interested in stream fish populations and habitats often have more questions than answers when they see survey data for a particular stream reach. They want to know what the data tells them about that reach. Are fish species that you would expect for this type of stream present and in typical abundances? Is growth of game fish above or below average? Is the habitat typical for this type of stream? Is there too much (or too little) sand or gravel on the channel bottom? Are there adequate amounts and types of woody habitat? Expectations for each type of stream are needed to answer these questions, and this requires data from many surveys done using standardized methods. The Michigan Stream Evaluator tool uses data from standardized surveys conducted throughout the state to allow users to explore such questions.

The Michigan DNR Fisheries Division initiated a statewide Stream Status and Trends Program¹ (SSTP) in 2002 to assess fish populations and aquatic habitat in Michigan's inland waters. This program initiated statewide use of standardized survey data collection and procedures as part of its Random Sites sampling component, which is intended to provide a comprehensive survey of all types of permanently-flowing (non-intermittent) streams in Michigan. (Note: Data from the Trends component can be accessed via the online 'Stream Fish Population Trend Viewer'.)

The SSTP Random Site protocols specify what data should be collected on field surveys and how data are to be collected. Data are collected on Random Site protocol surveys to characterize: fish community composition and abundance; age and growth of game fishes; stream channel attributes such as width, depth, bottom type, and occurrence of aquatic vegetation; stream bank and riparian conditions; amount of different types of woody habitat in the channel; stream discharge; and various observations and notes associated with the sample reach and survey.

In addition to MDNR Fisheries Division staff dedicating a portion of their annual survey time to SSTP Random Site surveys, Random Site protocols have been used on additional surveys by MDNR Fisheries Division and others due to the rigor of the information collected and the potential for comparing the survey data to other surveys done using Random Site protocols. Please read the freely-available report at this link https://doi.org/10.1002/fsh.10890 for further detail on MDNR Fisheries Division’s Stream Evaluator and Stream Fish Population Trend Viewer tools.

¹Wills, T. C., T. G. Zorn, and A. J. Nuhfer. 2006. Stream Status and Trends Program sampling protocols. Chapter 26 in Schneider, James C. (ed.) 2000. Manual of fisheries survey methods II: with periodic updates. Michigan Department of Natural Resources, Fisheries Special Report 25, Ann Arbor. Link to FishCat

Zorn, T., Wills, T., Hessenauer, J.-M., Lenz, J., Bissell, E., DePottey, A., Kilijanczyk, D.F. and Francis, A. (2023), Combining Statewide Surveys and Classification to Support Management of Streams. Fisheries, 48: 157-167. https://doi.org/10.1002/fsh.10890

Steps for using the Stream Evaluator

When you first look at the map you will see stream segments with surveys in red, segments currently selected for comparison in green, and all other segments in blue.

Zoom in to select an existing survey to evaluate. The segment you selected to evaluate will display in light blue on the map display. (You can skip this step if you only wish to see typical values for a certain type of stream segment.)
Select attributes of the type of streams you want to compare your survey to then hit the 'Submit' button to see the output. If there are adequate numbers of surveys (say, 8 or more), better comparisons will result from matching as many characteristics as possible between your survey and the set of stream surveys used for comparisons. You should at least match the size and July mean water temperature classifications, but may be less selective about attributes (such as geographic area, channel gradient, Great Lakes accessibility) if there are not many comparable stream reaches with surveys. Reaches used for comparisons will be highlighted in green on the map, and you can zoom out to see them on the map.
View the results. You will see the following:

Observed values of survey parameters for the stream reach you selected. Hover over a parameter to get a description of it.
Mean (or average) and standard deviation (a measure of variation in) values for the same parameters from the surveys you specified as benchmarks for comparison. In the table, 'Dev from mean' (relative difference) is defined as the observed value minus the benchmark mean, divided by the standard deviation² of the benchmark mean. This 'weighting' puts relative difference values for all parameters on the same numeric scale (i.e., numbers of standard deviations above or below the benchmark mean). The further the relative difference is from zero, the more conditions at the site differ from those at benchmark locations. For example, a 'Dev from mean' value of zero indicates no difference between observed and benchmark mean values. A value of 2 means that the observed value is considerably (2 standard deviations) higher than the benchmark mean, and if benchmark streams have a statistically 'normal' distribution, the observed value is higher than about 95% of benchmark locations.
Simple bar charts showing the 'Dev from mean' value (relative difference between the survey value and benchmark values) for each parameter. Scanning the graphs allows for quick identification of parameters that are well above or below benchmark values.

²For more info on standard deviation, see this link: https://en.wikipedia.org/wiki/Standard_deviation or other sources online.

For questions or concerns about this application, please contact: DNR-Fisheries@michigan.gov