not always agree with field measure- 

 ments (for example Figures 3B and C) , 

 it is apparent that averages of 

 observed and computed discharges 

 taken over several points are in good 

 agreement. A statistical analysis 

 revealed that when averaging over a 

 month the discharges computed from 

 the 6-min tide records yields a 

 maximum probable error of 0.075 

 m /sec at the 90 percent confidence 

 level (Daddio and van de Kreeke 

 1979). 



RESULTS 



Figure 4 is a plot of the monthly 

 rainfall recorded at the Rookery Bay 

 Marine Station and the net monthly 

 water discharge through the culverts 

 computed using Eq. 4. The net water 

 discharge is exclusively toward the 

 south for the entire recording 

 period. The discharge hydrograph 

 shows a definite seasonal trend with 

 the largest discharges occurring 

 during the wet season (here defined 

 as June through September) . The 

 maximum monthly discharge for one 

 year beginning June 1977 is 3.51 x 

 10 m /month in July 1977. The dry 

 season discharges are still su t>7 

 stantial and on the order of 10, 

 m„/month with a minimum of 0.27 x 10 

 m / month during the month of Nov- 

 ember 1977. For comparison the 

 subsurface ^run-off is estimated at 

 only 10 m /month based on figures 

 presented by Amy (1980). 



It is noteworthy that although 

 water flowing through the culverts 

 experiences tidal reversals for most 

 of the year relatively few reversals 

 occur through the wet season months. 

 This is exemplified by measured water 

 levels at the north and southside of 

 the culverts during April (dry 

 season) , and September (wet season) ; 



see Figure 5. During September the 

 water levels on the northside are 

 practically always higher than the 

 levels on the southside leading to 

 unidirectional flow. During April 

 the head difference changes sign. 

 Mean water levels during September 

 are about 0.20 m higher than during 

 April . 



The total rainfall for the 1- 

 year period beginning 7 June 1977 is 

 144 cm or 3.11 x 10 m for the 

 drainage basin of which 1.66 x 10 

 m was discharged through the cul- 

 verts. This yields a run-off ra- 

 tio of 0.53 for the drainage ba- 

 sin. Assuming no net water stor- 

 age over a period of a year, the 

 remaining 47 percent of rainfall 

 is lost largely by evapotranspi- 

 ration. This percentage is consid- 

 erably lower than the 76.2 per- 

 cent evapotranspiration reported 

 for the Big Cypress Swamp (Carter 

 et al. 1973) . This may be a result 

 of accelerated runoff associated with 

 channelization in this study area. 

 Also, there exists a possibility that 

 the freshwater lens on the golf 

 course (Amy 1980) forces ground water 

 derived from areas north of the 

 drainage basin to surface and be- 

 come part of the surface runoff. 



CONCLUSIONS 



Water export is determined 

 with a semi-empirical relationship 

 between discharge and water eleva- 

 tion differences across culverts 

 which convey surface run-off out 

 of the study area. Monthly water 

 export ranges from 0.27 x 10 m / 

 November 1977 to 3.51 x 

 for July 1977. The 

 total flow fox a one-year period 

 is 1.66 x 10 m representing 53 

 percent of the rainfall. 



mouth for 

 10 m /month 



239 



