INTRODUCTION

The U.S. Geological Survey (USGS), in cooperation with State, County, and other Federal agencies, obtains a large amount of data pertaining to the water resources of Florida each water year. These data, accumulated during many water years, constitute a valuable data base for developing an improved understanding of the water resources of the state. To make these data readily available to interested parties outside the USGS, the data are published annually in this report series entitled "Water Resources Data - Florida, Volume 2A: South Florida Surface Water and Volume 2B: South Florida Ground Water".

This report series includes records of stage, discharge, and water quality for streams; stage, contents, and water quality for lakes; and ground-water levels, contents, and water quality of ground-water wells. The data for South Florida include continuous or daily discharge for 86 streams, continuous or daily stage for 54 streams (including stage published at discharge and stage only sites), continuous elevations for 1 lake, continuous ground-water levels for 257 wells, periodic ground-water levels for 226 wells, and quality-of-water data for 39 surface-water sites and 149 wells.

Publication of this series of annual reports for Florida began with the 1961 water year, with a report that contained only data relating to the quantities of surface water. For the 1964 water year, a similar report was introduced that contained only data relating to water quality. For the 1975 water year, the report format was modified to one volume presenting data on quantities of surface water, quality of surface and ground water, and ground-water levels. For the 1977 water year, the report format was modified to a two volume set: one volume presenting data on quantity as well as quality of surface water and one volume presenting data on water levels along with quality of ground water.

Prior to introduction of this series and for several concurrent water years, water-resources data for Florida were published in USGS Water-Supply Papers. Data on stream discharge and stage and on lake or reservoir contents and stage through September 1960 were published annually under the title "Surface-Water Supply of the United States". For the 1961 through 1970 water years, the data were published in two 5-year reports. Data on chemical quality, temperature, and suspended sediment for the 1941 through 1970 water years were published annually under the title "Quality of Surface Waters of the United States", and water levels for the 1935 through 1974 water years were published under the title "Ground-Water Levels in the United States". The aforementioned Water-Supply Papers may be consulted in the federal repository libraries of the principal cities of the United States and may be purchased from the U.S. Geological Survey, Branch of Information Services, Box 25286, Federal Center, Denver, CO 80115 (telephone: 888-ASK-USGS).

Similar reports are published annually by the USGS for all of the United States. These official USGS reports have an identification number consisting of the two-letter State abbreviation, the last two digits of the water year, and the volume number. For example, this volume is identified as "U.S. Geological Survey Water-Data Report FL-xx-2B," where xx represents the current water year. For archiving and general distribution, reports for the 1971-74 water years also are identified as water-data reports. These water-data reports are for sale in paper copy or microfiche by the National Technical Information Service, U.S. Department of Commerce, Springfield, VA 22161. Additional information, including current prices, for ordering specific reports may be obtained from the Office Chief at the address given on the back of the title page or by telephone (305) 717-5800.

COOPERATION

The USGS and various Federal, State, and local organizations have had cooperative agreements for the collection of water-resource records since 1930. Organizations that assisted in collecting the data presented in this report through cooperative agreement with the USGS are:

Broward County City of Boca Raton City of Cape Coral City of Ft. Lauderdale City of Hallandale Beach City of Hollywood Everglades National Park Florida Keys Aqueduct Authority

Lee County Miami-Dade County Department of Environmental Resource Management Seminole Tribe of Florida South Florida Water Management District St. Lucie County U.S. Army Corps of Engineers U.S. Fish and Wildlife Service

Organizations that provided data are acknowledged in station manuscripts.

SUMMARY OF HYDROLOGIC CONDITIONS

This section summarizes important hydrologic events that occurred during the 2004 water year (October 1, 2003 to September 30, 2004) as well as significant natural and water-management responses to these events. Figure 2 provides a frame of reference for some of the major land areas of hydrologic significance mentioned in the summary.

GROUND-WATER MONITORING NETWORK

During the 2004 water year (October 1, 2003 to September 30, 2004), the Florida Integrated Science Center - Water and Restoration Studies (FISC-WRS) monitored 511 wells in southern Florida to assess regional ground-water conditions. In southeastern Florida, the principal aquifers monitored are the Biscayne aquifer in Miami-Dade and Broward Counties (209 wells) and the surficial aquifer system in Palm Beach, St. Lucie, and Martin Counties (52 wells). In southwestern Florida, the principal aquifers are the water-table aquifer (86 wells), lower Tamiami aquifer (49 wells), sandstone aquifer (44 wells), mid-Hawthorn aquifer (39 wells), lower Hawthorn aquifer (or lower Hawthorn producing zone) (25 wells), and the Floridan aquifer system (7 wells). The generalized geology and hydrogeology of southern Florida is shown in figure 3.

DATA FROM SELECTED WELLS

The most extensive data are provided by 262 monitoring wells equipped with data recorders that measure hourly water levels. The daily maximum water-level elevations presented in this report are derived from these hourly measurements. Seven recorder-equipped wells have been selected as index stations to depict ground-water conditions for the 2004 water year. These wells were selected to show changes in ground-water levels that occurred in seven municipal water-supply aquifers used in southern Florida. Relations shown for these index wells can only indicate changes occurring in the aquifers in the vicinity of each well.

A detailed assessment of hydrologic conditions throughout each aquifer would require similar statistical evaluations of data from many additional wells. For detailed assessment of aquifer conditions, the USGS has performed statistical analyses of data from additional network wells, and results are available at the U.S. Geological Survey (USGS) Current Water-level Conditions website http://www.sflorida.er.usgs.gov/ddn_data/index.html.

Two hydrographs are shown for each of the seven selected stations (figs. 4-10). The first hydrograph compares the water levels for the 2004 water year to historical water-level data. Daily maximum water levels from the 2004 water year are compared to: (1) normal monthly means of the daily maximum water levels, computed using all data available for each month of the period October 1979 to September 2004; (2) highest and lowest daily maximum water levels for the period of record; and (3) the monthly standard deviation of water levels above and below the normal monthly mean. Relations depicted in this first graph could be skewed by long-term water-level trends.

The second hydrograph shows the annual mean of daily maximum water levels and statistical data obtained from the Seasonal Kendall Trend Test (SKTT), depicting long-term trends. The SKTT is a nonparametric test for a monotonic trend in daily values. Two results of this test are the p-value and the Seasonal Kendall Slope Estimator (SKSE). The p-value indicates whether the trend determined by the SKTT is statistically significant. The null hypothesis for this test assumes that the random variable (water level) has not changed over time. The test makes pairwise comparisons of data values from the same seasons to eliminate seasonal variability. If the null hypothesis is disproven (p-values less than 0.05), there is a statistically significant trend in the data. The SKSE is a positive or negative slope representing a trend in water levels that is either increasing or decreasing. The SKSE is expressed as the change in water level in feet per year. The SKTT is determined using monthly mean values.

The statistical analyses used to evaluate the water-level data can be affected by missing data. Almost all index stations selected have periods of missing data. Cooperative support for ground-water monitoring at some locations has fluctuated. As a result, monitoring at some wells was terminated, but later resumed. Additionally, some stations experienced mechanical problems with the float system or water-level recorder. One example is L-2434 (fig. 10), which is one of the only recorder-equipped monitoring stations in the lower Hawthorn producing zone in southwestern Florida. Water levels at this station fluctuate rapidly and extensively, The mechanical systems used to monitor the well used to slip or become entangled. This necessitated the deletion of some segments of erroneous data until a submersible pressure transducer was installed during the 1997 water year. Pressure transducers at stations with rapidly fluctuating water levels have provided a more complete record than previously possible using mechanical devices.

The data analyzed were not censored for missing record because certain stations, such as L-2434, would have insufficient data for analysis if all partial monthly or annual records were removed. Despite the missing data from well L-2434, there is clearly a significant trend toward decreased water levels that is apparent in the record available. This trend must be considered in order to understand why the current water-level data remain below the normal monthly mean water levels from this well.

In the second hydrograph presented for each index station, dashed lines indicate periods for which the computed annual means included one or more months that were missing more than 15 days. The SKSE provided for these trends should also be considered approximate. Within the applicable graphs, the word "approximately" is used to indicate uncertainty for those stations that have excessive periods of missing record.

LONG-TERM TRENDS

The potential effect of long-term trends must be considered before current ground-water conditions can be reasonably evaluated. Hydrographs showing daily ground-water levels, means, and SKTT statistics (figs. 4-10) show statistically significant (p-value less than 0.05) long-term trends in water-levels at all index wells, except for well PB-561 (fig. 5, surficial aquifer system). Trends were evaluated for the 1980-2004 water-year period for all index stations, except L-2434; which only had record available since 1981.

Well S-196A (fig. 4), completed in the Biscayne aquifer in Miami-Dade County indicates a water-level increase of about 0.01 ft/yr (foot per year). Well C-496 (fig. 6), completed in the water-table aquifer in Collier County indicates a water-level increase of 0.02 ft/yr.

Wells completed in the lower Tamiami aquifer (fig. 7, L-2194), sandstone aquifer (fig. 8, L-729), mid-Hawthorn aquifer (fig. 9, L-1993), and lower Hawthorn producing zone (fig. 10, L-2434) in Lee County, all indicate long-term water-level declines. A 0.10 ft/yr downward trend in water-level data from well L-2194 was determined. This represents an approximate 2-ft decrease in average water levels for the 1980 to 2004 period (fig. 7). Water-level data from well L-729 indicated a 0.22 ft/yr downward trend that represents an approximate 6-ft decline in average water levels for the same period (fig. 8). Water level data from well L-1993 indicated a 0.71 ft/yr downward trend that represents an approximate 18-ft decline in average water levels for the 1980 to 2004 period (fig. 9). Water-level data from well L-2434 indicated an overall downward trend of 1.49 ft/yr, representing an approximate 36-ft decline in water levels for the 1981 to 2004 period (fig. 10). The trend in water levels in well L-2434 has resulted in period-of-record extreme minimum water levels that are progressively lower each year during the 1999 to 2004 period. Wells L-729, L-1993, and L-2194 are all completed in confined aquifers and located in areas affected by municipal water-supply withdrawals.

Because of the effect of substantial long-term declines in water level at wells L-729 (fig. 8, sandstone aquifer), L-1993 (fig. 9, mid-Hawthorn aquifer), and L-2434 (fig. 10, lower Hawthorn producing zone), recent water levels are, on average, below the monthly means of historic water levels (figs. 7-10). These monthly means are influenced by water-levels that had been much higher in the past. The effect of the declining water levels on monthly means of historic water levels is most obvious in the 2004 water year hydrograph from well L-1993 (fig. 9). Therefore, comparative analysis of recent and historical water levels is required to develop a comprehensive understanding of long-term water-level trends.

RAINFALL

The water-table aquifer in southwestern Florida and the Biscayne aquifer in Broward and Miami-Dade counties are locally named aquifers within the surficial aquifer system (fig. 3). Water levels in the surficial aquifer system respond readily to precipitation. The wells completed in these aquifers usually indicate water-level increases in response to rainfall events. The same is true for most wells completed in the lower Tamiami aquifer because a portion of this aquifer in southern Florida is unconfined. The response of water levels to precipitation in the deep confined aquifers is less pronounced. This is because the recharge areas of the deep confined aquifers are in the central Florida. In the deeper confined aquifers, the effects of municipal water-supply withdrawals on aquifer water levels are often more evident than the water-level variations caused by rainfall events. But all wells in all aquifers monitored indicate a seasonal variation in water levels that corresponds to the seasonal variation in rainfall.

Rainfall data collected and evaluated by the South Florida Water Management District (SFWMD) during the 2004 water year, provide a framework for understanding monthly water level variations (South Florida Water Management District, 2005). The rainfall data provided by the SFWMD for southern Florida are subdivided into 16 geographic areas. Monthly rainfall totals from individual stations within each area were averaged and compared to the historical total monthly rainfall averages. The percentage of average monthly rainfall is computed for each of the 16 geographic areas. This percentage is used throughout the discussion of ground-water conditions for the 2004 water year.

Weekly precipitation anomalies (percentage of 1971 to 2000 normal), are provided by the National Climatic Data Center, National Oceanic and Atmospheric Administration (NCDC-NOAA) (2005), for the years 2003 and 2004. The weekly precipitation anomalies were computed for the entire United States and include two geographic areas in southern Florida. The first geographic area is southeastern Florida from Key Largo to St. Lucie County. The second geographic area includes the rest of southern Florida, south of the northern extent of Lake Okeechobee. Unlike the percentages of average monthly rainfall provided by SFWMD, which are exact values, the weekly precipitation anomalies provided by NCDC-NOAA are divided into seven categories or ranges: very dry (less than 30 percent of average), severely dry (30 to 50 percent of average), moderately dry (50 to 70 percent of average), mid range (70 to 140 percent of average), moderately moist (140 to 200 percent of average), very moist (200 to 330 percent of average), and extremely moist (greater than 330 percent of average). Weekly and monthly precipitation anomalies were used in conjunction with water-level data to describe hydrologic conditions in aquifers during the 2004 water year.

GROUND-WATER CONDITIONS DURING THE 2004 WATER YEAR

Water levels in well C-496 (fig. 6, water-table aquifer) were above normal monthly mean water levels for about 9 months during the 2004 water year. This may be explained, in part, by the 0.02 ft/yr long-term water-level increase observed in this well. Conversely water levels in wells PB-561 (fig. 5, surficial aquifer system), L-729 (fig. 8, sandstone aquifer), L-1993 (fig. 9, mid-Hawthorn aquifer), and L-2434 (fig. 10, lower Hawthorn producing zone) were below normal monthly mean water levels for most of the year. This is consistent with the -0.22, -0.71, and -1.49 ft/yr water-level trends observed, respectively, in wells L-729 (fig. 8, sandstone aquifer), L-1993 (fig. 9, mid-Hawthorn aquifer), and L-2434 (fig. 10, lower Hawthorn producing zone). However, there was no statistically significant long-term water-level trend at well PB-561 (fig. 5, surficial aquifer system). Annual means of water-level data from this well suggest a general decline since about 1999.

During October, rainfall was substantially lower than normal (8 to 39 percent of average) throughout southern Florida. Water levels in all wells declined during this month. Water levels in wells S-196A (fig. 4, Biscayne aquifer), C-496 (fig. 6, water-table aquifer), L-2194 (fig. 7, lower Tamiami aquifer), and L-729 (fig. 8, sandstone aquifer) all declined by about one to two standard deviations. Water levels in well S-196A (fig. 4, Biscayne aquifer) and L-729 (fig. 8, sandstone aquifer) declined from about one standard deviation above the mean to about half of one standard deviation below the mean. Water levels in well L-2194 (fig. 7, lower Tamiami aquifer) decreased from about one standard deviation above the mean in the beginning of the month to near normal monthly mean water level by the end of the month. Water-level declines were slight in wells L-1993 and PB-561.

Rainfall during the first week of November was 200 to 330 percent of average. As a result, most of the index wells indicated increases of one to two standard deviations in water levels during the first week of the month. Water levels in wells S-196A (fig. 4, Biscayne aquifer) and C-496 (fig. 6, water-table aquifer) increased to about two standard deviations above the normal monthly mean. Water levels in well L-2194 (fig. 7, lower Tamiami aquifer) increased to about one standard deviation above the normal monthly mean. Water levels at wells PB-561 (fig. 5, surficial aquifer system) and L-729 (fig. 8, sandstone aquifer) increased from about one standard deviation below the normal monthly mean to levels slightly above the normal monthly mean. Rainfall during the remaining 3 weeks of November was substantially lower than normal (less then 30 percent of average). Water levels in all of the index wells declined during this period. Rainfall for the month was 86 to 114 percent of average in southwestern Florida, 98 to 152 percent of average in southeastern Florida, and 59 to 87 percent of average in south-central Florida.

During the first, second, and fourth weeks of December 2003, rainfall was generally substantially lower than normal (30 to 50 percent of average or less). Rainfall in the third week of the month was substantially greater than normal (greater than 330 percent of normal in southwestern and south-central Florida and 200 to 330 percent of normal in southeastern Florida). As a result, water levels generally declined throughout the month but increased during the third week of the month. During the third week of the month, water levels in wells S-196A (fig. 4, Biscayne aquifer) and C-496 (fig. 6, water-table aquifer) increased to levels slightly greater than one standard deviation above the normal monthly mean, and water levels in wells PB-561 (fig. 5, surficial aquifer system) and L-729 (fig. 8, sandstone aquifer) increased to levels slightly below the normal monthly mean by the middle of the month. Water levels at well L-2194 (fig. 7, lower Tamiami aquifer) increased to levels slightly above the normal monthly mean. Rainfall during the month varied spatially from 47 (eastern Miami-Dade geographic area) to 184 (east Caloosahatchee geographic area) percent of average.

During January 2004, rainfall generally was substantially lower than normal (less than 30 percent of average), except during the last week of the month. Rainfall during this week exceeded 330 percent of average. During most of the month, water levels for most of the index wells declined, except for monitoring well L-729 (fig. 8, sandstone aquifer) in Lee County where water levels increased from values lower than one standard deviation below the normal monthly mean in the mid-January to about normal monthly mean values at the end of the month. Water levels in most of the index wells increased at the end of the month in response to heavy rainfall in late January. This heavy rainfall late in the month throughout most of southern Florida also resulted in monthly rainfall totals that were slightly above normal despite the prevailing dry conditions that existed during the first 3 weeks of the month. Monthly rainfall totals varied spatially from 71 (eastern Palm Beach geographic area) to an estimated 190 (Everglades National Park geographic area) percent of average.

During the first 3 weeks of February, rainfall in southern Florida varied temporally from normal (70 to 140 percent of average) to well below normal (less than 30 percent of average). Rainfall during the last week of February was much higher than normal (between 200 and 330 percent of average in southeastern Florida and more than 330 percent of average in southwestern and south-central Florida). Monthly rainfall totals were about normal for southern Florida. Monthly rainfall varied spatially from 65 (eastern Broward geographic area) to 170 (Everglades National Park geographic area) percent of average. Water levels at most index stations increased twice: once at the beginning of the month as a result of heavy rainfall late in January and early in February, and again at the end of the month as a result of heavy rainfall in the last week of February. During the first week of the month, water levels in well S-196A (fig. 4, Biscayne aquifer) increased from about the normal monthly mean to slightly above one standard deviation above the normal monthly mean. At the end of the month water levels in S-196A again rose to about one standard deviation above the normal monthly mean. Water levels at well PB-561 (fig. 5, surficial aquifer system), increased from one standard deviation below the normal monthly mean at the beginning of the month to about one half of one standard deviation below the normal monthly mean.

During March, rainfall was substantially lower than average. Monthly rainfall ranged spatially from 3 to 37 percent of average throughout southern Florida. During this month, water levels for all the index wells declined. Water levels at well S-196A (fig. 4, Biscayne aquifer) declined from one standard deviation above normal monthly mean at the beginning of the month to about one standard deviation below normal monthly mean by the end of the month Water levels at well C-496 (fig. 6, water-table aquifer) were about one standard deviation above the normal monthly mean at the beginning of March and declined at about the normal rate during the month. Water levels at well L-2194 (fig. 7, lower Tamiami aquifer) and L-729 (fig. 8, sandstone aquifer) declined from about the normal monthly mean at the beginning of the month to about one standard deviation below the normal monthly mean by the end of the month. Water levels at well L-2434 (fig. 10, lower Hawthorn producing zone) declined to about two standard deviations below the normal monthly mean.

Rainfall during the month of April ranged spatially from about normal to a little below normal (71 to 124 percent of average). Rainfall was generally lowest in south-central Florida and highest in southwestern Florida. Abundant rainfall (200 to 330 percent of average) occurred in the third week of the month; however, rainfall during the rest of the month was less than 30 percent of average rainfall. As a result of rainfall in the third week, water levels observed at index wells rebounded. Water levels in index wells S-196A (fig. 4, Biscayne aquifer), PB-561 (fig. 5, surficial aquifer system), and L-2194 (fig. 7, lower Tamiami aquifer) were generally a little higher near the end of the month than at the beginning of the month. Water levels at these wells were generally between the normal monthly mean and one standard deviation below the normal monthly mean during April. Water levels in well C-496 (fig. 6, water-table aquifer) declined sharply during the month - from slightly below one standard deviation above the normal monthly mean to a little above the normal monthly mean.

May rainfall varied spatially and temporally. During the first week of May, rainfall was 50 to 70 percent of average in southeastern Florida and 70 to 140 percent of average in south-central and southwestern Florida. During the remaining weeks, very dry conditions prevailed (generally less than 30 percent of average.) Resulting monthly rainfall was substantially lower than normal. Rainfall for the month varied spatially from 16 to 50 percent of average. Water levels in all of the index wells declined during the month. Water levels in wells S-196A (fig. 4, Biscayne aquifer) and C-496 (fig. 6, water table aquifer) declined from a little above the normal monthly mean to about one standard deviation below the normal monthly mean. Water levels in well L-2194 (fig. 7, lower Tamiami aquifer) and L-729 (fig. 8, sandstone aquifer) declined to two standard deviations below the normal monthly mean.

On June 2, water levels in well L-2434 (fig. 7, lower Tamiami aquifer) declined to a new period of record lowest daily maximum water level (43.44 ft below the National Geodetic Vertical Datum of 1929). By the middle of the first week of June, water levels in well S-196A (fig. 4, Biscayne aquifer) had declined to about 1.5 standard deviations below the normal monthly mean. Rainfall was about normal (70 to 140 percent of average) during the first 2 weeks of the month, except in southeastern Florida where rainfall was slightly below normal (50 to 70 percent of average). For the rest of June, rainfall was lower than normal (30 to 50 percent of average or lower in southeastern Florida). June monthly rainfall totals were slightly below average in much of southern Florida but substantially lower than average (30 to 55 percent) in the geographic areas of; eastern Palm Beach County, eastern Broward County, Water Conservation Area 1, and eastern Miami-Dade. Water levels in wells S-196A (fig. 4, Biscayne aquifer) and C-496 (fig. 6, water-table aquifer) rebounded gradually in the first half of the month, but declined once again in the last half of the month. By the end of June, water levels in well S-196A (fig. 4, Biscayne aquifer) had declined to two standard deviations below the normal monthly mean. Water levels in well PB-561 (fig. 5, surficial aquifer system), where rainfall was lower than average, decreased from slightly above to slightly below one standard deviation below normal monthly mean. Water levels in wells L-2194 (fig. 7, lower Tamiami aquifer) and L-729 (fig. 8, sandstone aquifer) increased from levels considerably below one standard deviation below normal monthly mean to about one standard deviation below normal monthly mean by the end of the month. Water levels in well L-2434 (fig. 10, lower Hawthorn producing zone), where rainfall was slightly below average, continued to show values considerably lower than one standard deviation below normal monthly mean.

July monthly rainfall totals were generally about average for southern Florida. Rainfall for the month generally varied spatially from 75 to 127 percent of average. Geographic areas that received more than 90 percent of the average rainfall were the east and west Everglades Agricultural Area (EAA), Water Conservation Areas 1, 2 and 3, Big Cypress Preserve, and eastern Broward. Rainfall in the eastern Miami-Dade geographic area was slightly above average (127 percent of average) for the period. Rainfall in southeastern Florida was lower than normal (30 to 50 percent of average) in the first week of the month. Rainfall in southwestern and south-central Florida was lower than normal (30 to 50 percent of average) in the second week of the month. Rainfall throughout southern Florida was higher than normal (140 to 200 percent of average) in the last week of the month. During July, water levels in most index wells increased. Water levels in well S-196A (fig. 4, Biscayne aquifer) increased from about two standard deviations below the normal monthly mean at the beginning of the month to about the normal monthly mean by the end of the month. Water levels in well C-496 (fig. 6, water-table aquifer) varied between the normal monthly mean and one standard deviation below the normal monthly mean. By the end of the month, water levels in well C-496 (fig. 6, water-table aquifer) were slightly above the normal monthly mean. Water levels in well L-729 (fig. 8, sandstone aquifer) increased from one standard deviation below the normal monthly mean at the beginning of the month to about the normal monthly mean by the end of the month.

August monthly rainfall totals for southeastern Florida were slightly above normal (99 to 135 percent of average). Rainfall in southeastern Florida was considerably greater than normal (200 to 330 percent of average) in the first week of the month, below normal (50 to 70 percent of average) in the second and third weeks of the month, and about normal (70 to 140 percent of average) in the last week of the month. August monthly rainfall totals for parts of southwestern and south-central Florida were considerably above average and ranged from 139 percent of normal in the East EAA geographic area to 171 percent of average in the east Caloosahatchee geographic area. Rainfall was greatest in the second week of the month and that can be attributed to Hurricane Charley which brought more than 5 inches of rain to Florida on August 13-14, 2004. In southwestern Florida, where monthly rainfall was 162 percent of average, water levels in well L-2194 (fig. 7, lower Tamiami aquifer) increased from about one standard deviation below the normal monthly mean at the beginning of the month to about one standard deviation above the normal monthly mean by the last week of the month. Water levels in well L-729 (fig. 8, sandstone aquifer) increased from about the normal monthly mean at the beginning of the month to about one standard deviation above the mean by the end of the month. Despite rainfall amounts that were considerably above average for southwestern Florida, water levels in well L-2434 (fig. 10, lower Hawthorn producing zone) decreased from about one standard deviation below normal mean at the beginning of the month to about two standard deviations below the normal monthly mean by the end of the month. In the eastern Miami-Dade geographic area, where rainfall was 99 percent of average, water levels in well S-196A (fig. 4, Biscayne aquifer) decreased from about one standard deviation above the normal monthly mean to about one standard deviation below the normal monthly mean by the third week of the month.

September monthly rainfall totals varied spatially from 302 percent of average in the upper Kissimmee geographic area to 60 percent of average, in the Everglades National Park geographic area. Monthly rainfall totals were well above normal in the Martin (249 percent of average), eastern Palm Beach (208 percent of average) and lower Kissimmee (217 percent of average) geographic areas. This can be attributed to rainfall associated with Hurricanes Frances (September 5-6, 2004) and Jeanne (September 26-27, 2004), which produced 10 and 8 in., respectively, of rainfall in eastern, central, and northern Florida. Rainfall from Hurricane Frances caused water levels in well PB-561 (fig. 5, surficial aquifer system) to increase dramatically from about one standard deviation below the normal monthly mean to about one standard deviation above the normal monthly mean. Water levels increased again in response to rainfall from Hurricane Jeanne - from values slightly above the normal monthly mean on September 20, 2004, to values about two standard deviations above the normal monthly mean by September 26, 2004. Water levels in well S-196A (fig. 4, Biscayne aquifer) decreased from about the normal monthly mean to about one standard deviation below the normal monthly mean during the first 2 weeks of the month.