In this section
The amount of water that flows from a spring depends on several factors, including rainfall, size of the caverns within the rocks, the water pressure in the aquifer, and the size of spring recharge basin. Human activities can also influence the volume of water that discharges from a spring; for example, increased groundwater withdrawals can reduce the hydraulic pressure in an aquifer, causing water levels to decline and spring flows to decrease.
Historical spring flow data has been collected by the U.S. Geological Survey since 1929 and by the St. Johns River Water Management District since 1984. The amount of discharge data collected at each spring varies from only one measurement in several of the smaller springs to several hundred measurements taken at the first magnitude springs. Descriptive statistics for the discharge from each spring include the number of flow measurements, period of record, and the minimum, mean, median, and maximum values. Spring discharge through time is graphed where sufficient measurement data are available.
For the springs with discharge data graphs, locally weighted scatterplot smoothing (LOESS) was utilized to depict spring discharge over time. LOESS smooth lines emphasize the shapes of the data sets and can be used to observe possible patterns over time. The smoothing is done by selection of a smoothing factor (f) and a polynomial fit (λ). The f controls the window width over which a smooth is calculated for each data point, and λ is either a locally linear or locally quadratic fit (Helsel and Hirsch, 1992). The selection of f and λ is determined subjectively according to the purpose for which the smooth is to be utilized. In this report, the intent is to observe the changes in discharge over time and an f value of 0.25 and a locally quadratic fit was utilized.
Posted on 9-16-2013