Florida Geolcgical Survey icr any anomalous lemperature decreases. Temperatures generally ranged between 80 and 112 degrees F al depths greater than 1.G00 feel and generally increased with deplh; therefore, no cyclic or convective circulation is suspected in the study area. Temperature decreases have been documenled in south Florida where the Strails of Florida are adjacent lo the Florida Platform, The occurrence of convective cilrculalion should be invesligated when injection wells are proposed for Ihese areas. GROUND-WATER CHEMISTRY ANALYSIS Introduction Water-quality data, confining zone, and inieclion zone information for seven wells in Brevard County and one well in Indian River County {Hercutes Corporation) were analyzed for this investigation. Three wells, the Merritt Island, Soulh Beaches, and D. B. Lee injection wells, were chosen for detailed study because o obvious trends observed in water-quality data. For these wells, confining, injection, and monitor zone llhologies were examined lo detewrnme any physical properties that might help explain observed monilor zone contamination. Determining the mechanical iniegity oP the irijeclion wells was beyond the scope of ihis study. The injection inlervats of all eight wells occur in the Boulder Zone of the lower Floridan aqufer syste1 (Figure 36), in the Oidsma.r Formalion (with ihe possible exception of tMe Port Malabar injection interval where the Boulder Zone is riot well developed; see Plate 3). This zone is generally highly fractured and cavernous, with transmissivity values ranging up lo 21 million gpdift (Haberfeld, 1991)- The high transmissivities in the injection zone, and pumping rales which can be tens-of-millions of gallons per day, result in only minimal increases in wellhead pressure in most werls (Haberfeld, 1991 ), This implies the possibility that ihe injected waters are circulaling freely. Fractures, discontinuities, and -cavities in tJhe designated confining zones of 'he wells could prov.de conduits !or the ciroua ng water. The inlected fluids are generally low.sa'inily, treated municipal waste waler. Irdustrial waste waters are inlected at he Harris Corp. and Hercules, Inc. sites. Injecled waters ar less dense than formation water, and sii'ce fluids in the injection and lower monitor zones are high:y saline, "contaminatdon" irOTi ir'jecied fli.ids will be seen as Ireshening trends in mon, or well data. For example, such [rends show Lip as a decrease in tolal dissolved solids (TDS) and,"cr chloride concentration. Occasionally. maKed increases in these parainelers are observed, and this is attribuled to deeper saline waters being displaced upward by injected Iluicis (J. Haberfeld, DEIR, personal cortimuni cation, 1991). Nitrogen conlenil is monilored oecaise treated waste water will generally have higrier niogen concentrations 1han ambient formation water. It is measured as tolal kjeldahl nitrcge ':TKNI':, which is organic nitrogen plus ammonia. Another important measurement is the depth al which ihe TDS value exceeds 10,000 mgL. Ur-iled Slates Envirorrnental Protection Agency guidelines slate that the TDS value of formalion wale.s in an injection zone mus" exceedJ 10.000 n)'L, -so consultants note 1he depth al wh.ch the transilion occurs. For 1he transition deplhs, only preinection values ate availabe. The TDS, chloride, and TKN are among the parameters racked in "he ,aiiO.mo nilor zones, arid tnese three were chosen fc.r close anvestigalion because time series data on, them are available for the rmroriilor zones of the injeclion wellsThe three primary wells are the Merrilt Island, the South Beaches, and ilie D. B, Lee ijeclion weils. These are discussed first, and are followed by summaries of dala lor the Harris Corp., Grant Streel, Port Malabar, West Melbourne, and Hercules, Inc. injeclion wells.