 |
|||||
| Ground-water conditions in southern Florida |
|
|||||
| Ground-water conditions in southern Florida |
This report has been reformatted for presentation on the World Wide Web. The official text of WRIR 01-4275 (6.3 MB download) is available in PDF format. The Adobe PDF Reader program is available, at no cost, from Adobe.
Water managers and others need readily accessible information from the real-time ground-water level monitoring network. Software programs have been created that automatically analyze data and portray the results on the real-time prototype website (http://www.sflorida.er.usgs.gov/ddn_data/index.html). Data from the real-time network is presented in graphs and maps that have been designed for easy access and interpretation while retaining the content necessary to support water-management decisions. As part of the website update process, the software developed for the site incorporates many of the statistical analyses used for the well network analysis. These analyses include: use of the Seasonal Kendall trend test, mathematical removal of long-term trends using linear or polynomial regressions, and duration analysis of data. Where needed, these analytical techniques have been modified to reflect the website focus on the relation of current data to historical data.
Although this website is specifically designed to present ground-water data from the real-time ground-water level monitoring network, a select group of surface-water monitoring stations can add valuable information to aid in analysis. Using an extensive canal network, water from Lake Okeechobee can be used to recharge surficial aquifers in many areas of southern Florida. When the water level in Lake Okeechobee falls below 12 ft; however, a minimal amount of water is available to provide this recharge. During past droughts, water levels in Lake Okeechobee remained below 12 ft for extended periods. To aid in assessing drought severity and the effects that these droughts may have on aquifers, water levels from 11 surface-water monitoring stations (fig. 33 (96K)) in southern Florida also are presented on this website.
When evaluating data for a given well, it is useful to compare current values to those that occurred in previous water years. These comparisons create a framework for understanding the data from the current year and aid in rapid assessment of the status of the aquifer near each indicator well. For this reason, the prototype website includes a page presenting background information and current water-level and salinity data where available for each network station.
The most basic graphic generated by the prototype website shows the long-term chloride data for a well (fig. 34 (25K)). Chloride water samples are only collected in a few of the real-time monitoring wells, but when these data are available, it can provide direct assessment of saltwater intrusion at a monitoring well. Chloride samples are not collected continuously at most sites, but one real-time monitoring well (L-4820), has been equipped with conductivity probes. The prototype website provides plots of the conductivity data.
Another basic method of depicting data is to use hydrographs that show changes in water level through time. The website software generates hydrographs for time periods of 30 and 90 days, current water year to date, and 25 years. The first three hydrographs shown on the webpage include water-level data from previous years and descriptive statistics (weekly maximum and minimums) for comparison with the most recent data from the current year. Representations of these types of hydrographs are shown in figure 35A-C (149K). Descriptive statistics are computed for the post-1974 period of record to the most recently published year. The current year water-level data shown in the 30- and 90-day graphs are hourly values, which are collected and updated every 4 hours. For these short-period hydrographs, hourly data are used to show the daily water-level cycles that occur at many ground-water monitoring wells. The current year water-level data shown in longer duration hydrographs are daily maximum water levels for ground-water monitoring wells and daily mean water levels for surface-water monitoring stations. For those wells where water levels are near the top of the aquifer, this aquifer elevation is also shown on the 25-year hydrographs.
Daily water levels from selected previous years are included on the first three hydrographs to provide a reference to historical water levels during significant events, such as droughts or unusually wet periods. The water-level data from a dry year (1989) and a wet year (1995) are used to provide historical reference. For those stations that do not have daily water-level data for previous years, instantaneous water-level measurements are used, if available. To aid in the analysis of the two monitoring stations on Lake Okeechobee, regulation schedules (U.S. Army Corps of Engineers, 1999) for the lake are also shown on these hydrographs.
Examining current water-level data in light of historical data provides a valuable tool for water managers. Yet, because of the presence of strong long term trends in water levels, these simple comparisons may not be the best tools for trying to assess the effect of drought conditions. Long-term trends can mask water-level changes that may occur during short-term conditions such as droughts or floods.
As an example, figure 35A-C (149K) shows that as of September 27, 2000, the water levels for the period shown at monitoring well L-581 were very low relative to the historical reference levels. However, water levels at this monitoring well have declined by about 20 ft over the last 25 years (fig. 36 (44K)). Therefore, water levels for the current year should be expected to be below many historical reference levels. Direct historical comparisons therefore, may not represent the full picture of current hydrologic conditions throughout the aquifer because of bias introduced by long-term effects. Furthermore, these plots show that the spread between historical water-level highs and lows can be increased by a long-term trend. Because of this, it is difficult to quantify the significance of departures from historical conditions based solely on direct historical comparison. Detailed assessment of ground-water conditions requires a quantifiable knowledge of long-term, as well as short-term, changes in water levels.
To address the potential effects of long-term trends, historical water-level data are reviewed annually for long-term trends using a modified Seasonal Kendall trend test. If a statistically significant long-term trend is identified, the website support software runs a series of linear and polynomial regressions to obtain a best-fit regression of water level against time. The regression characteristics and Kendall trend test results are then stored for further reference. The derived trend line is plotted on the 25-year hydrograph (fig. 36 (44K)). A zero-slope line representing the period of record mean water level is plotted on the 25-year hydrograph if a statistically significant trend was not identified.
For sites with identified long-term trends, regression residuals are used for performing a frequency analysis of the available data, grouped by the week of the year collected. If a statistically significant long-term trend was not identified, residuals from the period of record mean water level are used for the frequency analysis. The regression result (or mean) is combined with the results of the frequency analysis to generate expected weekly water-level frequencies for the current year. The most recent 365 days of water-level data are then plotted against this prediction (fig. 37 (56K)). The frequency prediction displays a tighter range in water-level variation for the current year than would have been obtained had long-term trends not been considered. Using this prediction, it is much easier to accurately identify variation in the current year's data that may be caused by such factors as meteorologic droughts.
Data from the regression and frequency analyses are used to classify water-level monitoring sites. Maps are produced for each county and aquifer in addition to a map depicting the study area. For each site, symbols and colors are used to show the results of a comparison of the 7-day average of daily water levels to the weekly frequency predictions. Figure 38 (74K) is a representation of the water-level comparison carried out for the real-time monitoring network in Lee County on September 27, 2000.
Return to Table of Contents || Real-Time Ground-Water Level Monitoring Network Design || Real-time prototype website || Selected References
Funding for the USGS to design and maintain this site has been provided through a cooperative agreement with the South Florida Water Management District (SFWMD). Water-level conditions are monitored by the USGS with support from Federal, State, and local cooperators.
| Real-time conditions (de-trended data) |
End-of-month conditions (de-trended data) |
FISC-WRS, home page |
| Accessibility FOIA Privacy Policies and Notices | |
Page Last Modified: Thursday, 27-May-2004 16:03:12 EDT |
  |
