Internal Draft for Management Review – Do Not Circulate or Cite

Stormwater Treatment Area 1-West Lessons Learned

Introduction

The 1994 Everglades Forever Act (EFA; ξ 373.4592, F.S.) required that all water released from the Everglades Agricultural Area into the Everglades Protection Area achieve compliance with applicable State and Federal water quality standards, including criteria for phosphorus (P). The use of large constructed wetlands, i.e., the Stormwater Treatment Areas (STAs), to reduce total P (TP) concentrations in agricultural runoff is a key component of the South Florida Water Management District’s Everglades Restoration Program. Six STAs with a combined treatment area of 16,154 ha (39,317 acres) have been constructed to date. Current plans call for adding approximately 8,000 ha (~19,700 acres) of new treatment area to the STAs in the near future.

The size of each STA was determined during design using a mass-balance model whose P removal coefficient was derived from water quality and sediment data collected along the nutrient gradient in Water Conservation Area 2A, a portion of the remnant Everglades impacted by agricultural stormwater runoff (Walker, 1995). The mean design hydraulic loading rate for the STAs was 2.5 cm d-1 (1.0 in d-1; Burns & McDonnell, 1994), but varied from 1.6 to 3.0 cm d-1 (0.6 to 1.2 in d-1) depending on estimates of annual basin runoff into each STA. The design minimum and maximum water depths for the STAs were 0.2 and 1.4 m (0.5 to 4.5 ft), respectively.

Design Envelope vs. Operational Data for STA-1W

STA-1W (Figure 1) was constructed in two phases and encompasses 2,699 ha (6,670 acres). The first phase, known as the Everglades Nutrient Removal Project (ENRP; 1,545 ha [3,818 acres]), comprised Cells 1, 2, 3 and 4 and become operational in August 1994. The second phase, which nearly doubled the original ENRP treatment area, added Cells 5A and 5B and began flow-through operations in July 2000. The revised design parameters for STA-1W are summarized in Table 1 and are discussed below in relation to the operational data.

Flow: STA-1W receives water from the S-5A Pump Station via G-302 (Fig. 1). Inflow is then distributed to the three flow-ways that comprise the wetland (i.e., Cells 1→3; Cells 2→4; Cells 5A→5B) and eventually discharged to Water Conservation Area 1A via the G-251 and G-310 Pump Stations. The
annual hydraulic load anticipated for STA-1W during design (i.e., design HL) was 1.97 x 108 m3 (159,985 acre-feet) (Table 1). This assumed that STA-1E was fully operational and receiving a portion of the S-5A basin runoff.

During the past year, the 365-day cumulative inflow volume to STA-1W generally exceeded the design HL. Additionally, the 30-day cumulative HL increased sharply beginning in August 2004. The HL in September 2004 alone was 1.36 x 108 m3 (109,912 acre-feet), which equated to 69% of the design HL and reflected the large volume of rainfall the drainage basin received during Hurricanes Frances and Jeanne.

Water depth: The long-term average water depth throughout STA-1W has been 0.58 m (1.9 ft) which is close to the design depth of 0.6 m (2.0 ft). However, water depth has reached 1.2 m (4 ft) during storm events. Because extended periods of deep water causes physiological stress on the emergent vegetation community and promotes floating cattail islands, the target water depth between storm events was reduced to 0.4 to 0.5 m (1.25 to 1.75 ft) in the various treatment cells of STA-1W.

Total phosphorus: The annual TP load to STA-1W used for design (i.e., design P load) was 27,430 kg (60,473 lbs), which is based on the design HL  a mean inflow TP concentration of 139 µg L-1 (= ppb) (Table 1). From May 1994 through September 2004, the flow-weighted mean inflow TP concentration at G-250 and G-302 was 151 µg L-1. Beginning in June 2004, weekly TP concentrations at G-302 often exceeded 150 µg L-1, with a peak monthly mean TP concentration during September 2004 of 296 µg L-1. Elevated inflow TP concentrations, combined with high inflow water volumes, resulted in 365-day cumulative TP loads that were often greater than the design P load. Conversely, the 30-day cumulative TP loads between November 2003 and late August 2004 were generally at or below the anticipated average monthly load, but experienced a sharp increase beginning September 2004 in response to increased runoff from Hurricanes Francis and Jeanne (Fig. 2).

Treatment performance

Data presented in the 2004 Everglades Consolidated Report indicated that treatment performance, i.e., TP removal, in STA-1W was related to inflow TP concentrations and the aerial TP loading rate. More recent data suggest that if TP loading is reduced, the TP removal performance of STA-1W (expressed as percent TP load reduction) will improve (Fig. 2); TP load reduction is moderately correlated with TP load. The monthly TP load into STA-1W has been at or below the design envelope since November 2003. Correspondingly, TP load reduction generally increased from December 2003 through September 2004. The decline in TP removal performance that began in September 2004 (and the corresponding increase in TP load) can be attributed to the effects of Hurricanes Francis and Jeanne. In addition, the high winds and inflows experienced during the hurricanes may have reduced plant coverage in some treatment cells; field surveys indicate that much submersed aquatic vegetation was lost from Cell 5B during these storms.

Cell 5B Berm

A berm was constructed across the width of Cell 5B in anticipation that it would improve cell hydraulics and thereby improve treatment performance. However, hydraulic modeling performed during design predicted at best only slight changes in flow patterns throughout the cell after the berm was installed compared to pre-berm conditions (Piccone 2003). In addition, a post-berm tracer study could not determine the exact contribution, if any, that the berm made to the hydraulic performance of this cell (DBEL 2004). A tracer study was not conducted prior to constructing the berm to establish the need to improve cell hydraulics.

Lessons Learned

Loading: The treatment performance of STA-1W is sensitive to both the hydraulic and P loads to this wetland. Prolonged loading above the design envelope can result in reduced TP removal as measured on a percent basis. When this STA has been operated for flood control, treatment performance has suffered. However, the data also indicate that the performance of STA-1W can recover when loading is subsequently reduced. Total P load reduction is moderately correlated with TP load.

Water depth: It is well known that the composition of the emergent vegetation community in wetlands (i.e., cattail) is influenced by water depth. Prolonged periods of deep water have been implicated in the formation of floating cattail islands in Cells 1 and 2 of STA-1W. The depth threshold for this process appears to be in the range of 60 to 90 cm. However, we suspect that soil conditions and high wind events also have a significant role in cattail island formation. The interaction of these factors complicates the development of a simple “maximum depth” rule to prevent cattail from floating in the future.

Cell 5B berm: Cell hydraulics should be fully characterized before installing additional berms in the STA treatment cells in the future.

References

Burns & McDonnell. 1994. Everglades Protection Project - Conceptual Design. 92-166-1-002. Report prepared for the South Florida Water Management District, West Palm Beach, FL.

DBEL. 2004. Evaluation of Full Scale Stormwater Treatment Area Enhancements: Tracer Project. Contract ML040332. DB Environmental Laboratories, Inc., 365 Gus Hipp Blvd., Rockledge, FL.

Piccone, T.T. 2003. Two-dimensional Hydraulic Analysis for Cell 5 of STA-1W. Prepared for DEP Agreement No. G0040 Subtask 1.1 Hydraulic Analysis. South Florida Water Management District, West Palm Beach, FL.

Walker, W.W. Jr., 1995. Design basis for Everglades stormwater treatment areas. Water Resources Bulletin 31, 671-685.

STA-1W Lessons Learned_V2.docPage 1 of 45/19/19