Periphyton‐based transfer functions to assess ecological imbalance and management of a subtropical ombrotrophic peatland

Abstract

Summary1. To assess the biological status and response of aquatic resources to management actions, managers and decision‐makers require accurate and precise metrics. This is especially true for some parts of the Florida Everglades where multiple stressors (e.g. hydrologic alterations and eutrophication) have resulted in a highly degraded and fragmented ecosystem. Biological assessments are required that directly allow for the evaluation of historical and current status and responses to implantation of large‐scale restoration projects.2. Utilising periphyton composition and water‐quality data obtained from long‐term (15 years) monitoring programmes, we developed calibrated and verified periphyton‐based numerical models (transfer functions) that could be used to simultaneously assess multiple stressors affecting the Everglades peatland (e.g. salinity, nitrogen and phosphorus). Periphyton is an ideal indicator because responses to stressors are rapid and predictable and possess valued ecological attributes.3. Weighted averaging partial least squares regression was used to develop models to infer water‐quality concentrations from 456 samples comprising 319 periphyton taxa. Measured versus periphyton‐inferred concentrations were strongly related for log‐transformed salinity ( = 0.81; RMSEP 0.15 mg L−1) and log‐transformed total phosphorus (TP; = 0.70; RMSEP 0.18 mg L−1), but weakly related for total Kjeldahl nitrogen (TKN) ( = 0.46; RMSEP 0.12 mg L−1). Validation results using an independent 455 sample data set were similar (log(salinity) r2 = 0.78, log(TP) r2 = 0.65 and log(TKN) r2 = 0.38).4. Water Conservation Area 1 (WCA‐1), a large ombrotrophic subtropical peatland impacted by multiple water‐quality stressors that has undergone major changes in water management, was used as a case study. The models were applied to a long‐term periphyton data set to reconstruct water‐quality trends in relation to restoration efforts to reduce nutrient loading to the Everglades. The combination of biologically inferred TP and salinity was used to identify the ecological status of periphyton assemblages. Periphyton assemblages were ecologically imbalanced with respect to salinity and TP. Salinity imbalance varied spatially and temporally, whereas TP was spatially restricted. Imbalances caused by water management were owing to salinity more so than to TP.5. The transfer functions developed for the Everglades are trait‐based quantitative numerical methods and are ideal because the abundances of species are modelled numerically in relation to a stressor. The resulting inferred value is a numerical representation of the stressor’s effect on biological condition that can be compared against the management of the stressor independent of other factors. The benefits are that biological lags or hysteresis events can easily be identified and environmental conditions can be estimated when measurements are lacking. Reporting biological assessments in terms of well‐defined water‐quality metrics (e.g. numeric criterion) increased the communicative ability of the assessment. The use of multiple metrics to assess ecological imbalance increased the ability to identify probable causes.