Modeling phosphorus retention at low concentrations in Florida Everglades mesocosms

• A phosphorus model was built based on a mesocosm study of phosphorus retention.
• Outflow total phosphorus was modeled with relative errors <35.0 for all treatments.
• Simulations predicted performance an additional year beyond the experiment's end.

Reducing phosphorus (P) concentration in surface water is a primary component of the ongoing effort to restore the Florida Everglades. Engineered wetlands are currently being used to retain P from stormwater inflows but are not consistently achieving outflow P concentration goals. A three-year mesocosm study was performed investigating the effects of different plant communities on P retention within engineered wetlands. A dynamic model was constructed in the high-level simulation software STELLA, using water, soil, weather, and plant data from this mesocosm study. The model consists of three interconnected submodels: plant growth, hydrology, and P dynamics. The model simulates processes in water and soil related to all four forms of P: dissolved organic, dissolved inorganic, particulate organic, and particulate inorganic. Model verification and subsequent calibration was performed using biweekly outflow water quality data from a mesocosm containing a submerged aquatic vegetation (SAV) community consisting of Najas guadalupensis and the algae Chara sp. Model validation was then conducted using data from separate mesocosms with three different plant communities: monocultures of Typha domingensis or Cladium jamaicense, and a combination of Nymphaea odorata and SAV. The model was able to simulate outflow concentrations of total phosphorus from all four plant communities with average relative errors of less than 35%. A sensitivity analysis revealed the relative importance of the various processes involved in the retention of all P forms and the effects of different vegetation communities on these processes. Further simulations were run to predict the outflow total P concentrations for an additional year beyond the end of the mesocosm study.