Abstract Introduction Study Area Material & Methods >Results Discussion Acknowledgements Literature Cited Figures Tables

The distribution of porosity within the HFCs (Fig. 2) in the study area indicates that the most connectively porous rock is found throughout the Q5 HFC and near the base of the Q3A HFC. The least porous matrix is a semiconfining layer that spans nearly the entire thickness of the Q4 HFC and the upper part of the Q3B HFC. Well data show that the top of the semiconfining layer ranges 3.0 to 3.7 m below ground surface in the study area. Although the rock matrix of this layer is characteristically dense, it appears to contain many semivertical solution cavities.

The total number of HFCs sampled varied among wells. Q5 and Q3A HFCs were sampled in most wells given the high porosity and thickness of these layers. Wells on different canals reached to different HFCs (Figs. 1, 2): Q3A for canal L-31N; Q4 for the western section of canal C-4; Q5 for the eastern section of canal C-4; Q4 on Bird Road: Q3A for canal C-1W. Ground water levels at the 15 monitoring wells varied as a result of rainfall patterns and flood control operations on each canal (data for L-31 canal, are presented in Figure 3).

Ground water temperature varied according to seasonal variations in air temperature, with higher values at the end of summer (Table 1). During summer, higher ground water temperatures were recorded near the surface. Ground water temperature was directly correlated with groundwater depth (Table 1).

[larger table]

Table 1. Pearson correlation coefficient (r) and its significance level (p) for correlations between groundwater physical-chemical variables (temperature, conductivity, oxygen concentration), groundwater level from surface (GW depth), abundance of individuals (N. ind.), and number of species (N. sp.). Go to Table 1

Figure 3. (above) Average ground water level along canal L-31 measured as average of the values recorded at all the wells located on the canal (cm from surface, right = R), L-31 canal stage (cm, left = L), and average rainfall measured at ENP monitoring station NE1 in Northeast Shark River Slough (cm, left), from June 2000 to May 2001. [larger image]

Figure 4. (above) Total number of individuals (adults + copepodids, open symbols) and number of species (closed symbols) collected each month, over all wells, from June 2000 to May 2001. Bars represent standard errors. [larger image]

Figure 5. (above) Total number of individuals (adults + copepodids) collected at each well, plotted against the depth of the closest HFC. [larger image]

Conductivity was correlated with water depth (Table 1), and values varied between a maximum of 807 µS at A1-3m in June, and a minimum of 141 µS at the same well and depth in September (Table 1). In summer, low-conductivity values were recorded near the ground water surface, and the range of values was more variable than in winter.

Oxygen concentrations were usually very low (Table 1), < 1 mg/L; they varied from a maximum of 3 mg/L, recorded in February at A1-3m, to a minimum value of < 0.01 mg/L, recorded several times during the sampling period. Oxygen concentration was significantly correlated with ground water depth and conductivity (Table 1).

The abundance of individuals and the number of species were positively correlated (Table 1), and number of species was positively correlated with temperature (Table 1). Species richness (Fig. 4) was highest in summer and in late spring, at the end of the dry season. Specimens collected for all taxa were male and female adults or larval stages; fertilized females were never recorded. We collected a total of 4,660 copepods, both adults and copepodids, 4,428 of which were identified. The remaining 232 were early stage copepodids, which were not identified. A list of the species collected for this study, and total number of individuals for each species, as well as their ecological classification in ENP (Bruno et al. 2003), are listed in Table 3. The collection included 2 species of calanoids (Arctodiaptomus floridanus and Osphranticum labronectum), 19 species of cyclopoids, and 1 harpacticoid (Phyllognathopus viguieri). All the calanoids and harpacticoids were stygoxenes (i.e., organisms having no affinity with ground water systems, are usually foreign to that habitat, and can influence ground water ecosystems by functioning as either predators or prey (Gibert et al. 1994a)), and 2 cyclopoids were stygophiles (i.e.,organisms that actively exploit resources in the ground water system, as well as seek protection from unfavorable situations in the surface environment (Gibert et al. 1994a)).

Most of the wells had low numbers of individuals, except well B1, located on the C-4 canal along levee L-29 (Fig. 1), which had 852 individuals per sampled depth (Table 4). Samples from a group of four wells, A2 and A3 located along L-31N, A4 on Krome Avenue, and C2 on Bird Road canal, had values between 114 and 176 ind/depth (Table 4). At most sites, the majority of specimens were collected in the top stratum, Q5. However, at well B1, most specimens were collected at depths below the top of the Fort Thompson Formation in HFCs Q3A and Q3B (Fig. 2, Table 4), and in deeper samples in the test well A3 (Fig. 2, Table 4). In well A3, the maximum number of individuals was collected within the Q2 HFC of the Fort Thomson Formation, but high numbers were collected also from the Q3A and Q3B HFCs.

Distribution of copepods was analyzed with regard to depth by plotting the total number of individuals collected over all months at each depth in each well against the depth from ground level (Fig. 5). Depth below ground level was measured as the depth of the nearest high porosity layer to the pumping depth. Most of the samples had low numbers of individuals (< 100), regardless of depth. Samples with 100 to 200 individuals were mostly collected in Q4, Q3A, and Q3B, as well as the few samples with more than 500 individuals (Table 4). Most of the samples (32%), were from the Q5 HFC, above the semiconfining layer that marks the top of Q4. Of the total specimens that were collected, 62% were from Q4 to Q3A HFCs (depths of 3.2 to 6 m).

[larger table]

Table 2. Mean ± standard error, maximum and minimum value, for conductivity, oxygen concentration, and temperature, from June 2000 to May 2001. Go to Table 2

[larger table]

Table 3. Total number of individuals collected (adults + copepodids) for each species, and their ecological classification in ENP (from Bruno et al., 2003). Go to Table 3

[larger table]

Table 4. Mean number of individuals (adults + copepodids) shown in three different manners. Go to Table 4

[larger table]

Table 5. Comparison of mean number of individuals (adults + copepodids) collected across time. Go to Table 5

The overall abundance and vertical distribution of copepods varied over time (Table 5). More individuals were collected in the dry season than in the wet season (dry: 2,762, wet: 1,898), and mean monthly abundances were significantly greater during the dry season (Kruskal- Wallis test p = 0.01). High numbers of individuals were recorded in June (15 ind/sample), October (20 ind/sample), March (44 ind/sample), and May (14 ind/sample) (Table 5). The mean numbers of individuals per HFC were high for Q5 HFC in June and October, for Q3A and Q3B HFCs in March, and Q4 HFC in May (Table 5, Fig. 6). More than 20 ind/sample were collected from deeper samples, Q2 and Q1 HFCs only in March.

The ordination diagram obtained with detrended correspondence analysis for the first and second axis (eigenvalue axis 1: 0.741, axis 2: 0.144) (Fig. 7), groups the wells that were along the same canal, or along canals connected to each other, on the basis of a similar species composition and relative species abundance. Wells A2, A3, A5, A6, and A7 on L-31N canal, and A1 on C-1W canal at its confluence with L-31N, formed one group (group B, Fig. 7). A second group was made of wells B1, B2, B3, B4, and B5 on L-29 canal, test well A9 on L-30 canal, and test wells C2 and C3 on Bird Road canal (group A, Fig. 7). The only well not directly located along a canal, A4 (group C, Fig. 7), stands separately, and appears to have almost no species in common with the remaining wells (Fig. 7). Two species Mesocyclops reidae and Metacyclops cushae, were collected only in this well, in October and August respectively, and Microcyclops varicans and Mesocyclops americanus had high numbers only here.

Figure 6. (above) Percentage of total individuals (adults + copepodids) that were collected at each HFC for each sample date, from June 2000 to May 2001. The averaged number of individuals collcted from each HFC for all dates combined is shown on right as Total. [larger image]

Figure 7. (above) Detrended Correspondence Analysis, ordination diagram for axes 1 and 2. [larger image]

Figure 8. Relative abundance of copepods (adults + copepodids) on a geometric scale with X2 size grouping (equivalent to log2 scale). [larger image]

Species were ranked on the basis of the total number of individuals (adults+copepodids) collected over the entire sampling period (Table 3). The assemblage of species was heterogeneous. Only the five most numerous species had more than 100 individuals (Table 3, Fig. 8). Most of the species had about 100 individuals, and several species were rare, with < 10 individuals (Fig. 8). The temporal and spatial distribution of the first four numerous species, which totaled 78% of all the identified individuals, varied. The first four species, Orthocyclops modestus, A. floridanus, Mesocyclops edax, Thermocyclops parvus, were the only species present in samples collected at the Q2 HFC, together with 1 specimen of Diacyclops nearcticus, Ectocyclops phaleratus, M. varicans, Microcyclops rubellus, and 3 specimens of M. reidae. In the only Q1 sample, O. modestus was the most numerous species (26 individuals over 33 collected). The dominant species, O. modestus, was collected in the spring months at all depths, and was extremely abundant in March at B1 (1,558 individuals) (Fig. 9). Mesocyclops edax was abundant in late spring and early summer, mainly in June from Q5 to Q2 (Fig. 9). This species was collected at all depths. Arctodiaptomus floridanus was almost four times more abundant than the other calanoid, O. labronectum, and it was quite abundant at the deepest depth (105 individuals total); this species was present and abundant in late spring (Fig. 9). Thermocyclops parvus, ranking third together with A. floridanus, was present nearly all year. It was abundant only in August and in October at the Q5 and Q3B HFCs (Fig. 9).

Almost all the species collected in the study area were epigean in ENP (Table 3) and are considered stygoxenes. In contrast, only two species were stygophiles, D. nearcticus and Diacyclops crassicaudis brachycercus (fifth and fifteenth most abundant numerically, respectively). No samples of stygobites (i.e. specialized and obligatory hypogean organisms that are adapted to ground water in their physiology, biology, etc. (Gibert et al. 1994a)) were collected.

Figure 9. Number of individuals (adults and copepodids) collected monthly (top) and total of individuals collected for all dates at each HFC (bottom) for the first four ranked species. [larger image]

< Material and Methods | Discussion >