Development of a total phosphorus concentration goal in the TMDL process for Lake Okeechobee, Florida (USA)

This paper describes the approach used to establish an in-lake concentration goal for total phosphorus in the total maximum daily load (TMDL) process carried out by the Florida Department of Environmental Protection (FDEP) for Lake Okeechobee. In order to specify the in-lake phosphorus goal, the first consideration was to identify the most suitable indicator of "imbalance in flora or fauna" pursuant to Subsection 62-302.530(48)(b) of the Florida Administrative Code (FAC). Blooms of cyanobacteria (blue-green algae) were previously identified as one of the most serious symptoms of cultural eutrophication in this lake, and there existed a large data set relating total phosphorus to chlorophyll a, which can be used to index the occurrence of blooms. We evaluated the occurrence of samples with chlorophyll a in excess of 40 mug.L-1 (moderate bloom) and 60 mug.L-1 (severe bloom) as a function of total phosphorus concentrations in order to specify the lake phosphorus goal. A cross-tabulation procedure was used to identify a range of total phosphorus corresponding to a rapid increase in bloom frequency. Near-shore and pelagic data sets were sorted by total phosphorus and each was sub-divided into ten intervals of approximately equal sample size. When total phosphorus averaged below 30 mug.L-1, the probability of moderate blooms was below 3% and the probability of severe blooms was near 1% in the near-shore region. When total phosphorus averaged between 35 and 45 mug.L-1, frequencies were between 15 and 35% for moderate blooms and between 2 and 5% for severe blooms, respectively. Pelagic bloom frequencies also increased with increasing phosphorus, but the response was considerably muted relative to that observed in the near-shore area. To ensure an acceptable level of risk in terms of algal bloom occurrence, a total phosphorus goal of 40 mug.L-1 was selected by the FDEP. Mass-balance modeling results (Walker 2000) indicate that an average external phosphorus load of 140 metric tons y(-1) (compared to a 1973 to 1999 mean of 498 metric tons y(-1)) would provide a long-term aver-age phosphorus concentration in the take's pelagic zone of 40 mug.L-1. Based on our empirical model relating bloom frequencies to total phosphorus, it is predicted that under these TMDL loading conditions, bloom frequencies in the near-shore region would be 2 to 9%, as compared to 5 to 33% under present conditions. Successful implementation of the TDML should significantly reduce near-shore bloom frequencies in Lake Okeechobee.