Regional Runoff Risk Tools for Nutrient Reduction in Great Lakes States

One method to reduce the impacts of excess nutrients leaving agricultural fields and degrading water quality across the Nation is to ensure nutrients are not applied right before a runoff event could occur.  Generally nutrient management approaches, including the 4-Rs (“right” timing, rate, placement, and source), include some discussion about the “right time” for nutrient applications, however that information is static guidance usually centered on the timing of crop needs.  What has been missing, and what will be discussed in this talk, will be the development and introduction to runoff risk decision support tools focused on providing farmers and producers real-time guidance on when to not apply nutrients in the next week to 10 days due to the risk of runoff capable of transporting those nutrients off their fields.  The voluntary adoption and use of runoff risk in short-term field management decisions could provide both environmental and economic benefits.

In response to the need for real-time nutrient application guidance and a request from states in the Great Lakes region, the National Weather Service (NWS) North Central River Forecast Center (NCRFC) has helped develop these runoff risk tools in collaboration with multiple state agencies and universities and with support from the Great Lakes Restoration Initiative (GLRI).  There are currently four active runoff risk tools in the Great Lakes region: Michigan, Minnesota, Ohio, and Wisconsin.  It is possible to develop similar tools for Illinois, Indiana, and New York if willing state partners are identified.  

What did we do?

Studies have shown that a few large runoff events per year contribute a majority of the annual load leaving fields.  In addition applications generally occur during the riskiest times of year for runoff (fall through spring) when fields experience the least vegetative cover and soils are vulnerable.  Knowing this information, real-time NWS weather and hydrologic models were evaluated to identify conditions that correlated with runoff observed at edge-of-field (EOF) locations.  The runoff risk algorithm identifies daily runoff events and stratifies the events by magnitude respective to each grid cell’s historical behavior.  The events are then classified into risk categories for the farmers and producers. In general, high risk events are larger magnitude events that don’t happen as often and also have a higher accuracy rate.  On the other end, low risk events are smaller magnitude events that have a higher chance of being a false alarm yet are also less likely to be associated with significant nutrient loss.

NWS models are run twice daily and simulate soil temperature, soil moisture, runoff, and snowpack conditions continuously.  The runoff risk algorithm is applied against the model output to produce runoff risk guidance which is sent to the state partners.  Each state has a working group and a lead agency or organization that manages the effort to produce and maintain the runoff risk websites as well as promote the tools and educate the users on how to interpret and use the guidance.  

What have we learned?

At this point there are four regional runoff risk tools available.  Response has been positive from both state agencies and when farming groups are asked about the runoff risk concept during post-presentation surveys and small focus groups.  There is a strong desire from the farming community to make the best decision during stressful times of the year when farming schedules and the weather are often in conflict.  

At this point, it is universally accepted among the runoff risk collaborators that there is a need to provide free, easily obtainable forecast guidance to the farming community so they can make the best nutrient application decisions for their operations and the environment.

Runoff risk tools are strictly for decision support and not meant to be a regulatory tool in nature.  This is due to the limitations in hydrologic models, weather forecasting, spatial scale issues, and that the tools have no way of incorporating farmer specific practices into the risk calculations.  Although model improvements will occur in the future, ensuring users understand the limitations but also the benefits they can provide are important components in the States’ outreach and education functions.  

Future Plans

Based on feedback from the states employing runoff tools, there is a second round of enhancement planned for the runoff risk algorithm in the summer of 2019.  Other improvements from the states’ perspective deal with updating webpages and building on and enhancing push notification capabilities such as text message and email alerts.

The next major step forward begins in spring 2019 with the start of version 3 runoff risk.  This 2-year development will transition runoff risk guidance from the current model over to the new NWS National Water Model (NWM).  The NWM framework will allow finer resolution guidance (1km or smaller) for numerous models runs per day all with full operational support.  Moving to the NWM also allows continuous improvement and future collaboration opportunities with universities to improve the underlying WRF-Hydro model as well as runoff risk and other derived decision support guidance.


Dustin Goering, Senior Hydrologist, North Central River Forecast Center, National Weather Service
Andrea Thorstensen, Hydrologist, North Central River Forecast Center, National Weather Service

Corresponding Author email

Additional Information

For further information on runoff risk background please visit this page:  (Still under construction)


To visit the state tools see the following links:







There are many individuals across a wide spectrum of agencies, industry, and universities that have been instrumental in the development of runoff risk to this point.

Support for the development of runoff risk across the Great Lakes and the upcoming version 3 runoff risk from the National Water Model has been provided by multi-year grants from the Great Lakes Restoration Initiative.



The authors are solely responsible for the content of these proceedings. The technical information does not necessarily reflect the official position of the sponsoring agencies or institutions represented by planning committee members, and inclusion and distribution herein does not constitute an endorsement of views expressed by the same. Printed materials included herein are not refereed publications. Citations should appear as follows. EXAMPLE: Authors. 2019. Title of presentation. Waste to Worth. Minneapolis, MN. April 22-26, 2019. URL of this page. Accessed on: today’s date.