Evaluating the relation of phosphorus to turbidity during high flow events in western Iowa rivers to improve phosphorus load estimates
Issue
The presence of phosphorus (P) in waterways is a major concern in the state of Iowa. Although this nutrient is essential for biological processes, excessive quantities sourced back to anthropogenic origins disrupt the ecological balance of both fresh and marine aquatic systems. Human activity has historically elevated the amount of P discharged into Iowa’s waterbodies. The state has set targets to lower this P load via the Iowa Nutrient Reduction Strategy (INRS), and a key benchmark of the INRS is the reduction of annual P load exported by Iowa’s major rivers.
Quantifying P loads in rivers remains a challenge, with estimations subject to several potential sources of error and uncertainty. Turbidity has shown promise as a surrogate for particulate phosphorus (Part P) concentrations in rivers. While monthly sampled datasets are large, peak flow conditions have not been not well represented, and this lack of data at higher flows is a shortcoming in applying turbidity as a surrogate for Part P. Additional data are needed above to better characterize the surrogacy relationship during high flows, which is essential for quantifying annual P export.
A previous, related project proved highly successful, with measurements from nine sites and over 100 new samples collected. Incorporating these new data alongside the existing monthly samples both improved the surrogacy models and validated their use during peak flows in eastern Iowa. While this project made great strides in quantifying P export, the scope was limited to eastern Iowa. The western Iowa watersheds possess a separate set of geologic and land use conditions. Similar methods will be applied to seven sites in western Iowa to expand our approach and address the estimation of Part P concentrations in these watersheds during high flows.
Objective
This project’s primary goal is to gather new data supporting the improvement of surrogacy models that predict particulate P concentrations using turbidity. The data will be collected during high flow conditions at several locations on four western Iowa rivers adjacent to existing IDNR monitoring sites and USGS stream gauges, where historical data has been lacking. Specific objectives are to:
- collect water samples at western Iowa’s seven terminal monitoring sites when flows exceed the predetermined thresholds;
- measure the Part P and turbidity levels within these samples; and
- compare these new data points to the established surrogacy models at each site.
Approach
Researchers plan to replicate methods that yielded success in a related, INRC-funded project in eastern Iowa. The equipment purchased as part of the previous project can be redeployed in western Iowa. Since travel times between the University of Iowa and the western watersheds are significant, local collaborators will assist in sample collection. Each sample will be analyzed by the Iowa State Hygienic Laboratory to determine its turbidity, total phosphorus (TP), and orthophosphate (OP) levels. TP and OP concentrations are needed to calculate Part P, which is the difference between TP and OP.
The resulting models will enable a scientifically defensible method for estimating Part P loads exported from the seven terminal sites in western Iowa.
Project Updates
Note: Project reports published on the INRC website are often revised from researchers' original reports to increase consistency.
April 2025
FINAL REPORT
This project aimed to validate the relationship between turbidity and particulate phosphorus (Part P) during high-flow conditions in western Iowa rivers. Turbidity has demonstrated great effectiveness as a surrogate for predicting Part P in numerous Midwest rivers and streams. Since turbidity can be continuously measured in situ using turbidimeters, its monitoring can considerably improve estimates of a stream P by producing high-frequency predictions of Part P concentrations. These models have applications for determining watersheds’ long-term P export (including calculating the statewide P load), as well as describing the short-term dynamics of P transport.
While several such models have been constructed for Iowa rivers using historical water quality samples, existing datasets contain few samples collected during high flow periods (i.e., flows above the 90th percentile). Thus, the performance of these surrogacy models during wet weather events was largely unknown. This project collected new water quality samples, where turbidity and Part P were measured, in seven western Iowa rivers. These samples were strategically collected when streamflow levels exceeded predefined thresholds.
This project was a continuation of a previous INRC project (2020-05), which sampled nine eastern Iowa rivers. Following the successful completion of the eastern Iowa effort, this project was undertaken to complete data collection in Iowa’s western rivers. Attachment 1 displays the sites where sampling occurred and lists additional information for each site. Attachment 2 contains a graphical abstract describing this project’s general workflow and provides several pictures of the western Iowa rivers and relevant sampling equipment. While this project was initially scheduled to conclude in 2023, periods of drought necessitated an extension to allow for sampling under wetter conditions.
Findings:
The newly collected data largely validated and improved the existing models linking turbidity and Part P. From 2022 – 2024, 41 wet weather events were monitored, with 177 total samples collected (see Attachment 2). Turbidity measurements ranged from 70 – 6400 NTU, and Part P concentrations ranged from 0.14 – 8.06 mg/L. Attachment 3 contains scatterplots for each river overlaying the new samples on the original water quality data. These plots also display the regression lines used to convert turbidity measurements to Part P predictions. R2 values for the seven rivers ranged from 0.83 – 0.93.
In every case, these were improvements upon models constructed solely using the original data. Differences in parameters (e.g., slope and intercept) between the updated regression equations and those built using the original data were not significantly significant (0.05). With the validation of these models under high flow conditions, they can now be confidently implemented under a full range of hydrologic conditions. This enables real-time predictions of Part P levels, improves Iowa’s statewide P load assessments, and increases the capacity for Part P transport modeling. These results also suggest that western Iowa’s existing Part P data is more robust than previously thought. All collected data for this project (for the eastern and western rivers) has been included with Attachment 3.
Following acceptance of a peer-reviewed manuscript, this project’s data will be shared with state and federal agencies, including the Iowa Nutrient Research Center and the Iowa Department of Natural Resources. New P loads will be calculated using the newly validated models, updating previous values to include the 2024 calendar year. The next steps for this work include implementing real-time P estimates for major Iowa rivers that will be made available for public use via the Iowa Water Quality Information System web platform.
This work completed the initiative that set out to model Iowa’s statewide P export using surrogacy models. The nature of this project also provided insights into best practices for event-based sampling at remote, rural sites. These lessons learned will likely be helpful for future projects that involve sampling during high flows. Our sampling in western Iowa involved working alongside volunteers at Iowa Lakeside Labs and Golden Hills RC&D, thus bolstering public involvement in issues surrounding nutrients in Iowa and increasing stakeholder collaboration.
Related accomplishments and activities
5 presentations.
Publications related to this project:
- Anderson, E. S., Schilling, K. E., & Weber, L. J. (in preparation). Validation of phosphorus surrogacy models under high-flow conditions. Hydrology.
- “Monitoring Iowa’s Phosphorus using Turbidity” article submitted for the 2022 edition of Iowa State publication Getting into Soil & Water
- “Monitoring Iowa’s Phosphorus using Turbidity” article submitted for the 2021-22 edition of IGS Geode (activities of the Iowa Geological Survey)
January 2025
All fieldwork and water quality sampling for this project was completed. The multiple heavy rainfall events that occurred in western Iowa from April to July 2024 allowed ample data collection at high flows for the seven rivers covered in this project. Following sampling efforts, all field equipment was retrieved. We also completed all data analysis and model validation required for this project.
The phosphorus data collected during high flows for this project largely aligned with previous surrogacy models. This process thus validated (and improved) current models that related in-situ turbidity observations to corresponding particulate phosphorus concentrations. The results of this project allow us to fully implement the surrogacy models (which relate turbidity to particulate phosphorus) on a statewide level. These models help estimate statewide phosphorus loads and improve our understanding of suspended phosphorus dynamics. Previously, we were unsure how these models performed during high flows, but the data collected by this project has validated their usage under all hydrologic conditions.
We are currently preparing this project's final impact report.
Related accomplishments and activities
1 project presentation is planned: to IDNR, Jan 14, 2025, Des Moines, Iowa.
December 2023
Project activities consisted of continued efforts to collect water samples during high-flow events in seven western Iowa rivers. We maintained four automated sampling devices deployed along our waterbodies of interest. We collaborated with local volunteers (Iowa Lakeside Laboratory and Golden Hills RC&D) whose staff assisted with grab sample collection.
Unfortunately, drought conditions persisted across Iowa, and very few wet-weather events occurred. While we have at least one event at all seven sites, we would like to obtain samples at two more events in five of the rivers.
Initial results suggest that our new data collected validates our models. All equipment has been retrieved for the winter, and sampling efforts will resume in March 2024.
Related accomplishments and activities
1 Presentation.
Our collaboration with local volunteers provided an opportunity to informally share information about nutrient issues in Iowa and the work being conducted by the Iowa Nutrient Research Center.
June 2023
Over the past six months, our activities have almost entirely consisted of sample collection in seven western Iowa rivers during high-flow events. Several ISCO devices were deployed, and we have worked extensively with local volunteers who have assisted in sample collection (namely Lakeside Laboratory and Golden Hills RC&D). We have successfully sampled events at each site, but recent drought conditions have resulted in few flow events in western Iowa. We continue monitoring river levels and are prepared to collect the necessary final data points during future high-flow conditions.
Other Activities
- 10 field days
- 1 presentation
Other accomplishments:
We have built partnerships with local volunteers in western Iowa assisting us with sample collection.