Status: Final Date: December 15, 2005 Report No. LOXA05-004 Enhanced Water Quality Program Parameter Reduction Rationale Report No. LOXA05-004 Prepared by: Donatto Surratt, Ph.D. A.R.M. Loxahatchee National Wildlife Refuge December 2005 BACKGROUND In 2004 Congress authorized funding for an Enhanced Water Q uality Monitoring Program, which would aid in improving the scientific understanding of water quality issues in the Arthur R. Marshall Loxahatchee National Wildlife Refuge (Refuge). The 1991 Federal Consent Decree established a set of analytical parameters based on previous work carried out across the greater Everglades and 14 sampling location across the Refuge. These stations are also part of a larger effort conducted by the South Florida Water Management District to sample water quality in the Everglades Protection Area (EVPA). The Enhanced Water Quality Monitoring Program (LOXA), initiated in June 2004, consists of an additional 39 sampling locations that are monitored for 23 of the water quality parameters monitored at the 14-station Consent Decree network (Table 1). The LOXA Work Plan (Brandt et al. 2004) states that, “data will be reviewed… and the list of parameters reduced as appropriate” (p. 12) with the objective to optimize the marsh monitoring network. Further, the reduction of parameters has the potential to reduce analytical cost and these reclaimed funds can be applied to extending the duration of the LOXA monitoring effort. This document highlights the procedures and rationale applied in selecting the analytical parameters for reduction in the Refuge’s Enhanced Water Quality Monitoring Program. The analysis presented here are specific to the Refuge and the objectives of our study. Parameter elimination recommendations presented here may not be appropriate for other studies in the Everglades (e.g., EVPA). PARAMETER REDUCTION PROCEDURES In the effort to review and determine whether analytical parameters could be eliminated from the LOXA program several approaches were involved: (1) collaboration Status: Final Date: December 15, 2005 Report No. LOXA05-004 with members of the South Florida Water Management District (District) and experts in the Department of Interior, (2) times-series characterization and correlation analysis, and (3) consideration of rationale for parameter reduction by the District for the Storm Water Treatment Areas (STA) in the non-ECP EFA Permit Modification Request of February 2005. Resource requirements for analyzing a particular constituent (e.g., field or laboratory effort, cost to process) were not considered in this exercise, however the recovered cost resulting from elimination of the selected parameters is presented in Table 2. (1) Collaboration between the Refuge (Michael Waldon, Ph.D.; Matthew Harwell, Ph.D. ; Laura Brandt, Ph.D. ; and Donatto Surratt, Ph.D. ), the District (Sue Newman, Ph.D. and Scot Hagerthey, Ph.D.) and the United States Geological Survey (Paul McCormick, Ph.D.) involved discussions of expert opinions on parameters that may be less important from a management and scientific perspective, including those that generally do not lead to increased understanding of water quality within the Refuge. Eleven parameters identified as worth initial consideration for elimination were: alkaline phosphatase (APA), color, dissolved organic carbon (DOC), hardness (HARD), magnesium (Mg), sodium (Na), ammonium (NH4), nitrite (NO3), silica (SiO 2), total dissolved solids (TDS), and turbidity (TURB). (2) The next approach involved time-series characterization, spatial characterization, and simple correlation analysis for all the parameters ove r the entire sampling period (June 2004 – May 2005). Time-series characterization of the data was applied to explore, and validate where possible, some of the expert opinions when selecting parameters to eliminate. Time-series characterization was an observation of change in a parameter over the year that data was collected. The approach for characterization was two fold in some cases including (1) an all marsh site analysis and where necessary (particularly when observation of quantitative variability in the form of a coefficient of variation was necessary) (2) an individual site analysis. If the parameter had low variability over the year it generally did not yield quantitative information about localized changes to the ecosystem, thus the parameter could be consider for elimination. If the parameter demonstrated high variability then further analysis was necessary to determine if it could be eliminated. There was no consistent pattern in variability for the Status: Final Date: December 15, 2005 Report No. LOXA05-004 parameters in the marsh or in the canal. Another characterization approach was the spatial analysis approach which was performed from canal to the interior of the marsh on several parameters. In most cases this analysis was performed to determine if there was an observable pattern for the parameter demonstrating some dependence on canal water penetration into the interior of the Refuge. Simple correlation analysis was applied to determine if there were parameters that produced redundant information. The analysis included both canal and interior sites, which allowed the spectrum of rain-driven soft water and hard canal water to be analyzed collectively. Correlation values, reported as r, were used to assess how well parameters associated. Parameters with r greater than or equal to 0.95 (Table 3) were considered well correlated and in some cases resulted in one or more parameters being added to the list of parameters for elimination. (3) The PRO ECP Non-ECP EFA Permit Modification Request of February 2005 drafted by the District and sent to the Florida Department of Environmental Protection (FDEP) was reviewed for rationales for parameter reduction used by the District. This letter provided rationale for eliminating seven parameters also monitored in LOXA: ammonia, chloride, turbidity, total dis solved phosphorus, ortho-phosphate, total dissolved nitrogen, and total dissolved solids. FDEP approved elimination from Non-ECP EFA sampling of all except turbidity. Personal emails between Dr. Surratt and David Struve, Ph.D. provided further insight on the procedures for parameter reduction by the District. Dr. Struve was on the review panel for parameter reduction for the PRO ECP Non-ECP EFA Permit Modification Request of February 2005. Dr. Struve related some of the considerations used to generate the modification request and some of these considerations were used to address the parameter elimination issues in the present report. Two considerations applied for our purposes were: (1) is there any significant variability in the given parameters over time and (2) are there related parameters being measured that give the same information (example: plot chloride vs. conductance and see if the data are a good fit. If so, conductance values may be good enough to infer the chloride values). A final consideration was the comparison of the parameter time-series data with the Class III criteria defined in F.A.C. 62-302.530 and applied in the 2004 Everglades Consolidated Report (SFWMD, 2004). If the data were observed to extend beyond the Class III Status: Final Date: December 15, 2005 Report No. LOXA05-004 criteria over the monitored period then it was determined that the parameter should continue to be monitored. Some parameters only had narrative criteria associated with them and the narrative criteria were not assessed in this study. Because of this limitation, Class III criteria comparison was performed on only a portion of the analytes, particularly alkalinity, dissolved oxygen, potassium, pH, and turbidity. Further, because of the short time-series (12 months), less weight was given to the time-series – Class III criteria comparison when determining whether or not to eliminate the parameter. PARAMETER REDUCTION RATIONALE There are 29 parameters used in the initial sample design. All the parameters were reviewed for potential to eliminate them from the water quality analytical suite. The following is a breakdown of all the parameters with rationales for elimination based on the three approaches presented above. Correlation analysis with coefficients greater than or equal to 0.95 is presented in Table 3. Because, in some case, we are eliminating parameters based on the ability to reproduce that parameter from another analytical parameter, it was necessary to choose an r cut-off limit that was high enough to show exceptionally strong proportional relationships between the two parameters. Appendix A presents the full correlation table for all parameters. Appendix B provides time-series (June 2004 – May 2005) graphs of the parameters for elimination and cross-plots of correlated parameters identified in Table 3. ALKA: Alkalinity correlated well with four parameters – calcium (r = 0.98), hardness (r = 0.98), specific conductance (r = 0.96), and total dissolved solids (r = 0.95). The high correlation between ALKA, calcium, and hardness is most likely an artifact of their dependence on the concentration of CaCO3. Although each of these parameters share this dependency they are still determined through unique analytical procedures and explain different conditions of the analyzed water parcels and as such yield different values and patterns through time, which is why they are not 100% correlated. Further, these correlations allow the observer to quickly qualify data points as outliers (prior to a rigorous quantitative assessment) with respect to specific conductance, which is Status: Final Date: December 15, 2005 Report No. LOXA05-004 exceptionally important for tracking water movement from the canal to the interior marsh. Alkalinity has a Class III criterion of = 20 mg L-1. In October and November 2004 five sampled sites had values below the criteria. This criterion is inappropriate for rainwater dominated wetlands where alkalinity is naturally below 20 mg L-1 (Weaver, 2005). A water quality goal of the Refuge is to maintain this low alkalinity condition. Because ALKA provides the ability to quickly determine specific conductance outliers; because there is an established numeric criterion pursuant to Section 62-302.530 F.A.C.; and because ALKA is important to the ecology of softwater marshes, ALKA will not be eliminated from the analytical suite. Alkalinity remains a monitored parameter. APA: There were no strong correlations associated with alkaline phosphatase activity (APA). APA can indicate phosphorus limitation in wetland ecosystems, as inducible phosphatase enzymes are produced when the system becomes deplete in phosphorus (Newman, et al., 2002). Spatial observation of APA for the marsh indicates that APA values increase towards the interior of the marsh (>1.5 km interior of the canals), which coincides with the reported phosphorus limitation for the Refuge (Vymazal, and Richardson, 1995). APA has also been applied as an early warning indicator of wetland eutrophication. The technique applied for analyzing APA for LOXA is not suitable for using APA as an ecological warning indicator. We only measure standing water APA, while the protocol requires the direct measurement of periphyton APA and the subtraction of standing water APA to determine periphyton response to nutrient enrichment (Newman et al., 2002). Because the only other applicable use of APA for the marsh found in this research effort is the determination of phosphorus limitation, which can be assessed with phosphorus and other parameters on the analytical list, APA can be removed from the list of analytical parameters. APA is on the list of parameters to be eliminated. Ca: Calcium correlated well with hardness (r = 0.99), specific conductance (r = 0.95), and total dissolved solids (r = 0.95). Temporal observation of Ca can indicate shifts between minerotrophic (flow [stream, canal, groundwater] driven) and ombortrophic Status: Final Date: December 15, 2005 Report No. LOXA05-004 (precipitation driven) wetland systems (Kadlec and Knight, 1996). Further, alterations to Ca levels impacts plant growth rates, which can alter community structures and reduce biodiversity. Calcium remains a monitored parameter. Cl: Chloride had strong correlations with sodium (r = 0.99) and specific conductance (r = 0.95). Elevated Cl has been associated with fertilizers and other contaminants that may reach the Refuge from Everglades Agricultural Area (EAA) operations during by-passes of untreated water (McPherson and Halley, 1996; Orem, 2004). Chloride concentrations can be used as a tracer of flow rates in wetland systems because of it low biological demand. Chloride remains a monitored parameter. COLOR: There were no strong correlations associated with color. Color was initially considered for elimination because it was expected to have a small range of values over time, ultimately providing little information about water quality. Based on a station by station time-series characterization the coefficient of variation for color was ~ 24%. Color is often used in assessing raw drinking water sources for treatability. It is often found to be correlated with the presence of organic materials including lignin and tanic acids. Color is a semi-qualitative parameter and did peak strongly when there were heavy precipitation events (i.e., hurricanes) in September and October 2004. Regardless of this observation color is being eliminated as a monitored parameter, because of its qualitative nature and the fact that there are other parameters that provide more quantitative information about hurricane and other events (i.e., SO4, TPO4, TDPO4, SiO 2, etc.). Color is on the list of parameters to be eliminated. DO: There were no strong correlations observed between dissolved oxygen and other analytes. Dissolved oxygen (DO) has an alternative Class III criterion called the site- specific alternative criteria (SSAC). The SSAC is a sinusoidal diel cycling algorithm (Weaver et al., 2001). The algorithm was adjusted, finalized, and presented in the Everglades Marsh Dissolved Oxygen Site Specific Alternative Criterion Technical Support Document (Weaver, 2004). Between June 2004 and May 2005 approximately 79% of the marsh samples were below the modeled criterion, 42% of the canal samples Status: Final Date: December 15, 2005 Report No. LOXA05-004 were below the criterion, and 72% of all the samples from the Refuge were below the determined criterion. DO in marsh systems gives a measure of oxidation potentials in the water column. When DO drops below 2 mg L-1 plant mortality significantly increases - (Kadlec and Knight, 1996). None of the calculated SSAC values were below the 2 mg L1, 43% of marsh samples, 18% of canal samples, and 39% of all the Refuge samples were below the 2 mg L-1. DO remains a monitored parameter. DOC: Dissolved organic carbon has been linked to microbial respiration and fulvic acid regulated mercury methylation in the southern Everglades (Reddy and Aiken, 2001). One of the experts in preliminary lab experiments revealed microbial respiration to be strongly carbon limited instead of phosphorus or nitrogen limited as for most system. Additional influxes of labile carbon can affect marsh soil processes. Reddy and Aiken (2001) demonstrate strong correlations between Hg bioaccumulation and DOC concentrations. Correlation analysis for DOC did not reveal significant correlation with other parameters. Because neither Hg nor microbial respiration are monitored in the Refuge ; because of the potential of DOC concentration changes to cause Hg bioaccumulation rates to change; and because no other parameters can serve as surrogates of DOC, monitoring DOC is recommended. DOC remains a monitored parameter. HARD: Hardness is measured as an equivalent concentration of calcium carbonate (CaCO3) and was expected to correlate well with alkalinity (which is also measured as an equivale nt concentration of CaCO3). Hardness correlated well with ALKA (r = 0.98), Ca (r = 0.99), magnesium (r = 0.97), specific conductance (r = 0.97), and TDS (r = 0.97). Although these strong correlations are observed the Florida Class III water quality criteria for various metals (Cd, Cr, Cu, Pb, Ni, Zn) require hardness in order to calculate the metal criterion (Bechtel, 2000). These metals have never been measured as a part of the LOXA project and where discontinued from the EVPA project in 2000. From the perspective of tracking softwater movement interior of the canals ALKA and specific conductance appears to provide the necessary information, thus the elimination of the hardness appears to be warranted. Hardness is on the list of parameters to eliminate. Status: Final 8 Date: December 15, 2005 Report No. LOXA05-004 K: Potassium did not correlate well with any other parameters and does not have a specific Class III criterion. Although there were no strong correlations associated with K, a station by station time-series characterization (not including August and September of 2004) showed low variability with a coefficient of variation of 19% and the pattern for the entire data set matched well with the TDS pattern. K in the interior of the Refuge did not drop below 0.64 mg L-1 and averaged 4.2±2.8 mg L-1, which is above the reported 2 mg L-1 K water column limitation value (Demaneche et al., 2001; Spijkerman and Coesel, 1998; Palmen et al., 1994). Based on the low variability in the K data set, the matching patterns between K and TDS, and because the K concentration was above the limitation concentration, K will be eliminated as a monitored water quality parameter. Potassium is on the list of parameters to be eliminated. Mg: Magnesium is expected to contribute to hardness. Mg had a strong correlation with hardness (r = 0.97), specific conductance (r = 0.97), and TDS (r = 0.97), thus specific conductance and TDS can serve as a proxy for Mg in future analysis. Because hardness was dropped and there was no other indication that Mg contributed to an increased understanding of water quality magnesium was added to the elimination list. Magnesium is on the list of parameters to be eliminated. Na: Sodium is expected to contribute to observed alkalinity values. Na had a strong correlation with specific conductanc e (r = 0.96) and TDS (r = 0.95), thus specific conductance and TDS can serve as a proxy for Na in future analysis. Na can serve as a conservative tracer for calculating dilution and concentration and for tracking groundwater discharges from wetlands (Kadlec and Knight, 1996). Based on the analysis Na provides redundant information with respect to water quality and SPCOND or TDS may be applied to replace sodium as a tracer. Sodium is on the list of parameters to be eliminated. NH4: There were no strong correlations associated with ammonia-nitrogen. Ammonium is on the list of parameters to be dropped although it may provide useful information about the seasonal processes of plant growth and decomposition in marsh systems Status: Final Date: December 15, 2005 Report No. LOXA05-004 (Kadlec and Knight, 1996). NH4 is a plant nutrient contributing to total Kjeldahl nitrogen. It may impact dissolved oxygen through oxidation to nitrate and may also cause aquatic toxicity. NH4 is generally quite low in Everglades Protection Area (EPA) waters and generally constitutes only a small fraction of EPA total nitrogen. The rationale for eliminating NH4 is that total nitrogen encompasses NH4 and because nitrogen is not considered a limiting nutrient in Refuge wetlands (Vymazal, and Richardson, 1995), the specific details of alteration to NH4 are of little consequence. Ammonium is on the list of parameters to be eliminated. NO3: There were no strong correlations associated with nitrate. Nitrate is generally a significant portion of nitrate-nitrite (NOX) measurement and in combination with the other components of total nitrogen provides information about nutrient lability. Because NO2 is generally only a small fraction of the NOX, it was expected that NOX values can be used to estimate the NO3 concentration. This rationale holds true for station LZ40 in Lake Okeechobee such that NO2 was ~11% and NO3 was ~90% of the NOX concentration. However, this pattern was not observed in the Refuge marsh. Observation of the data showed NO2 at ~43% and NO3 at ~58% of the NOX concentrations. Thirty- one percent of the NO3 values for the Refuge was below detection limits, which may have complicated this assessment and potentially confound any trend analysis attempted for the Refuge. Independently, NO3 is not particularly useful and the va lue is maintained in NOX. Nitrate is on the list of parameters to be eliminated. NO2: There were no strong correlations associated with nitrite. Nitrite is on the elimination list because it was thought to be short lived and represent only a small fraction of the nitrogen pool due to its rapid conversion to nitrate. Generally, NO2 found in wetlands above detection limits is indicative of an anthropogenic source of nitrogen (Kadlec and Knight, 1996). Inspection of Refuge data demonstrates that NO2 can contribute up to 43% of the inorganic nitrogen pool, which does not support the assumption that NO2 is a small portion of NOX. Again this assessment may be skewed as ~25% of the NO2 values were below detectable limits (0.004 mg L-1). Conversely, the observed imbalance between NO2 and NO3 with respect to NOX may reflect loading of Status: Final Date: December 15, 2005 Report No. LOXA05-004 anthropogenically generated nutrients in the Refuge. Regardless, nitrite will be incorporated in the analysis of nitrate-nitrite. Nitrite is on the list of parameters to be eliminated. NOX: There were no strong correlations associated with nitrate-nitrite. NOx is the sum of NO2 and NO3. NOX is a fraction of the inorganic nitrogen pool and in a balanced wetland can serve as a proxy for the essential plant nutrient, NO3. The sum of NOX and total Kjeldahl nitrogen approximates total nitrogen. Total nitrogen can be employed to determine the lability of food sources for marsh plant communities (Thomann, 1972). Total nitrogen is also a concern for water downstream of the Refuge. NOx remains a monitored parameter. OPO4: Ortho-phosphate had a strong correlation (r = 0.99) with total dissolved phosphate and appears to provide redundant information, thus OPO4 is on the list for parameter elimination. Although OPO4 is on the parameter elimination list it was argued that the presence of OPO4 was a good indicator of phosphorus enrichment in Water Conservation Area 2 (WCA2), and may play a similar role within the Refuge. Also, eleven percent of the reported OPO4 values were below detection limits (0.004 mg L-1). At greater than 1 km interior of the canals surrounding the Refuge, OPO4 values do not exceed 0.03 mg L-1, which suggest that OPO4 does not provide useful information about Refuge water quality dynamics. Ortho-phosphate is on the list of parameters to be eliminated. pH: There were no strong correlations associated with pH. pH has a Class III criterion that suggests the pH should remain between 6 and 8.5. The maximum limit of this criterion was not exceeded during the stud y period. In November 2004 the sample values fail below a pH of 6 in the northeastern region of the Refuge. pH remains a monitored parameter. SiO2: There were no strong correlations associated with silica. Silica shows a seasonal pattern for the interior sites of the Refuge. Silica is higher in the wet-warm season (late Status: Final Date: December 15, 2005 Report No. LOXA05-004 April – early November) and lower during the dry-cool season (November to early April). Independent research has demonstrated that silica concentrations decrease during the cooler season when algae populations (dominated by diatoms) are dying off and increase during the warmer season when these populations grow in again. Changes in these patterns can serve as a good indication of eutrophication and even biota shifts (Biggs, 1990). These assertions are generally for flowing waters (i.e., rivers) and historically appear to be of less import for the Refuge wetland ecosystem which ha s a low diatom abundance associated with the periphyton population dominating as the primary producer of the Refuge (Kadlec and Knight, 1996). However, with STA-1E becoming operational in September 2005, there will be larger volumes of higher mineral content water discharge into the L-40 Canal. Canal water penetration into the marsh has been documented (Harwell et al., 2005). The introduction of higher mineral content water into a soft-water ecosystem is expected to impact the biotic community. The combination of temperature, nitrogen to phosphorus, and silica to phosphorus ratios has been shown to relate to algae species distributions (Adamus et al., 2001). This suite of indicators can theoretically be applied as an indicator of changes in periphyton community dynamics as the high mineral content water impacts the Refuge. Presently, WCA-2 has a higher diatom abundance associated with the periphyton communities (relative to the Refuge) and the Refuge perimeter canals are major sources of water for WCA-2. Because of the potential of silica concentration change to cause biotic shifts, and because the canals of the Refuge provide water to other areas of the Everglades, silica is not on the elimination list. Silica remains a monitored parameter. SO4: There were no strong correlations associated with sulfate. Sulfate mediates methyl mercury production (Axelrad et al., 2005). Mercury contamination has been a concern for areas of the Everglades. As sulfate increases sulfide formation increases, which increases the potential for methyl mercury formation. At toxic levels sulfate can induce plant mortality, because of the reduction of sulfate to hydrogen sulfide (Armstrong et al., 1996). Increasing sulfate loads to the marsh can alter biogeochemical cycling (e.g., Fe and P) and result in decreased biodiversity for the ecosystem. Because of the significant role sulfate plays in the biogeochemical cycling and the generation of toxic forms of Status: Final Date: December 15, 2005 Report No. LOXA05-004 mercury sulfate is not being considered for elimination from the monitoring program. Sulfate remains a monitored parameter. SPCOND: Specific conductance is a conservative parameter that can be used to track canal water penetration into the marsh. SPCOND can also be employed to develop hydrological and chemical budgets for the Refuge. SPCOND had strong correlations with ALKA (r = 0.96), Ca (r = 0.95), Cl (r = 0.95), HARD (r = 0.97), Mg (r = 0.97), Na (r = 0.96), and TDS (r = 0.98). The Class III criterion of = 1250 µS/cm for SPCOND is well in excess of values experienced in the marsh and as such the criterion is meaningless when compared to values observed in the marsh. Specific conductance remains a monitored parameter. TDKN: There were no strong correlations associated with total dissolved Kjeldahl nitrogen (TDKN). TDKN is the fraction of the total Kjeldahl nitrogen (TKN) that passes through a 0.45 micron filter. The sum of organic nitrogen (in the trinegative oxidation state) and ammonia make up Kjeldahl Nitrogen (Clesceri et al., 1998). Sources of Kjeldahl nitrogen include the decay of organic material and urban and industrial organic waste. Large amounts of ammonia and organic nitrogen are applied to cropland as fertilizer. Both ammonia and organic nitrogen are relatively immobile in soils and ground water because of adsorption on soil surfaces and particulate filtering. These nitrogen constituents are susceptible to nitrification under aerobic conditions (Kadlec and Knight, 1996). The coefficient of variations for TDKN in the Refuge interior and canals were ~21% and ~20%, respectively. TDKN had a range of 0.65 and 2.57 mg L-1. At greater than 3 km interior of the canals impounding the Refuge (with the exception of 2 outliers in August and September 2004) TDKN values drop below 1.3 mg L-1, which suggest marsh TDKN maybe pulse driven by canal water penetration. TKN shows a similar pattern with a range of 0.63 to 4.11 mg L-1 (not including outliers) and dropping below 2 mg L-1 at greater than 3 km interior of the marsh canals. SPCOND serves as a more sensitive tracer of canal water penetration as the variable shows a finer gradient than TDKN. Because of the relatively low variability and because TKN will continue to Status: Final Date: December 15, 2005 Report No. LOXA05-004 be monitored TDKN can be removed from the suite of analytical water quality parameters. TDKN is on the list of parameters to be eliminated. TDPO4: Total dissolved phosphorus (not phosphate) had a strong correlation with OPO4 (0.99). TDPO4 provides a quick estimation of biologically available phosphorus. TDPO4 is subtracted from total phosphorus to yield total particulate phosphorus. The total particulate phosphorus value provides a good estimate of total phosphorus that is unavailable for biological uptake at the time of the measurement as well how much may become biologically available in the future. Total dissolve phosphate remains a monitored parameter. TDS: Total dissolved solids was suggested as a parameter to be eliminated. TDS and conductance are highly correlated (r = 0.98). TDS, as well as ALKA, are used to determine if outliers or invalid measurement values exist in SPCOND data. Further the ratio between Cl and TDS is expected to indicate sources of water (i.e., canal, marsh, rain, groundwater, etc.) for the Refuge (Waldon, 2005). Because of these applications, the conservative nature of TDS, and the observed variability in the time-series data for TDS it was suggested that the parameter not be dropped. Total dissolved solids remains a monitored parameter. TEMP: There were no strong correlations associated with temperature. Temperature is an integral parameter that changes seasonally. It is used to adjust conductance, dissolved oxygen measurements, determine DO saturation, and the DO-SSAC. There were no strong correlations associated with temperature. Temperature remains a monitored parameter. TKN: There were no strong correlations associated with total Kjeldahl nitrogen. TKN combined with NOX provides a value for total nitrogen, which is important in understand water quality and is of particular concern downstream of the Refuge. There were no strong correlations associated with TKN. TKN remains a monitored parameter. Status: Final Date: December 15, 2005 Report No. LOXA05-004 TOC: There were no strong correlations associated with TOC. From the time-series characterization TOC demonstrated a consistent range and did not appear to indicate any anthropogenic or natural events that may have impacted the marsh. One caveat for TOC was presented by a recent USGS study that demonstrated problems in obtaining consistent TOC results (Aiken et al. 2001). The difference between TOC and DOC yields particulate organic carbon (POC). POC has been linked to mercury methylation in WCA2 (Reddy and Aiken, 2001). Because TOC is necessary for determining POC and because there are no surrogates for TOC, TOC will remain a monitored parameter. TOC remains a monitored parameter. TPO4: Total phosphorus (not total phosphate). Phosphorus is a nutrient required for plant growth and in most freshwater systems phosphorus is considered a limiting nutrient. Historic phosphorus loads for the Everglades were generally low and derived from atmospheric deposition. Recent (last half century) increases in phosphorus loading are associated, in part, with run-off from agricultural sources upstream of the Everglades in the EAA. Regardless of this relatively new source of nutrients for the Everglades, total phosphorus has been identified as the key limiting nutrient (McCormick et al., 2002). With the development of agricultural in regions north of the Refuge, nutrient run-off towards the Refuge has been an increasing concern. The Consent Decree issued in 1991 mandates the monitoring of phosphorus in the Refuge. Total phosphate remains a monitored parameter. TSS: There were no strong correlations associated with total suspended solids. Total suspended solids are generally a concern for potable waters. There are conditions when water is pumped out of the Refuge as water supply for neighboring residential areas (e.g., Lake Worth Drainage District) and TSS serves as an indicator water clarity and hence the suitability of using the water for drinking and hygienic purposes. Further, high TSS has been associated with marsh disturbance and can help indicate sample sites that may have been agitated during sample collection. TSS remains a monitored parameter Status: Final Date: December 15, 2005 Report No. LOXA05-004 TURB: There were no strong correlations associated with turbidity. Turbidity has a Class III criteria of less than or equal to 29 NTU above natural background. Natural background was defined for the STAs and cannot be applied for the marshes. Regardless, turbidity was suggested as a parameter to be dropped because it was expected to be correlated to TSS. Inversely, the coefficient of determination (r = 0.55) was less than expected. A station by station time-series varia tion analysis shows that turbidity had a coefficient of variation of ~65% compared to the even higher TSS coefficient of variation of ~128%. The observed high variability in turbidity suggest that the parameter is sensitive to small scale changes to water conditions and further analysis of the parameter may lead to increased understanding of temporal and spatial hydrologic dynamics. Turbidit y is a proxy for the condition and productivity of natural bodies of water as it rela ys information about the clarity of the water parcel. Turbidity is a measure of light scatter and absorbance and can reveal a level of detail about light penetration to the marsh peat and algal communities (Clesceri et al., 1998). Also, high turbidity has been associated with marsh disturbance and can help indicate sample sites that may have been agitated during sample collection. Turbidity remains a monitored parameter. FINAL LIST OF PARAMETERS FOR ELIMINATION As a result of this exercise 11 monitoring parameters are being proposed for elimination: · Alkaline phosphatase, · Ammonium, · Color, · Hardness, · Magnesium, · Nitrate, · Nitrite, · Ortho-phosphate. · Potassium, · Sodium, · Total dissolved Kjeldahl nitrogen, and Status: Final Date: December 15, 2005 Report No. LOXA05-004 The work presented in this document was an important exercise for the LOXA program. Part of the recognition that these parameters may be acceptable to eliminate is that they are presently monitored at EVPA sites. Therefore, parameter reduction recommendations for the LOXA program may not be appropriate for the EVPA program, or other water quality studies in the Everglades. Status: Final Date: December 15, 2005 Report No. LOXA05-004 REFERENCES Adamus, P., Danielson, T.J., Gonyaw, A., 2001. Indicators for Monitoring Biological Integrity of Inland, Freshwater Wetlands A Survey of North American Technical Literature (1990-2000). Environmental Protection Agency, EPA843- R-01. Aiken, G., Kaplan, L.A., Weishaarc, J., 2001. Assessment of relative accuracy in the determination of organic matter concentrations in aquatic systems. Journal of Environmental Monitoring v4, p70–74. Armstrong, J., F. Afreen-Zobayed, and W. Armstrong. 1996. Phragmites die-back: Sulphide and acetic acid-induced bud and root death, lignifications, and blockages within aeration and vascular systems. New Phytologist, v134, p601-614. Axelrad, D., Atkeson, T., Pollman, C., Lange, T., 2005. Chapter 2B: Mercury monitoring research and environmental assessment in South Florida. G. Redfield, ed. In: 2005 South Florida Environmental Report. West Palm Beach, FL. Bechtel, T., Hill, S., Iricanin, N., Mo, C., Van Horn, S., 2000. Chapter 4: Status of water quality criteria compliance in the Everglades Protection Area and at Non-ECP Structures. G. Redfield, ed. In: 2000 Everglades Consolidated Report. West Palm Beach, FL. Biggs, B. J., 1990. Periphyton communities and their environments in New Zealand Rivers. New Zealand Journal of Marine and Freshwater Research, v24, p367386. Brandt, L.A., Harwell, M., Waldon, M. 2004. Work Plan: Water Quality Monitoring and Modeling for the A.R.M. Loxahatchee National Wildlife Refuge: 2004-2006. Prepared for the A.R.M. Loxahatchee National Wildlife Refuge. April, 2004. 33 pp. http://sofia.usgs.gov/lox_monitor_model/workplans/2004- 2006_workplan.html Clesceri, L.S., Greenberg, A.E., Eaton, A.D., 1998. Standard Methods for the Examination of Water and Wastewater, 20th edition. Published by: American Public Health Association, American Water Works Association, and Water Environment Federation. Demaneche, S., Kay, E., Gourbiere, F., and Simonet, P., 2001. Natural transformation of Pseudomonas fluorescens and Agrobacterium tumefaciens in soil. Applied Environmental Microbiology, v67, i6, p2617-2621. Status: Final Date: December 15, 2005 Report No. LOXA05-004 Harwell, M, Surratt, D, Waldon, M, Walker, B, Laura, B. 2005. A.R.M Loxahatchee National Wildlife Refuge Enhanced Water Quality Monitoring and Modeling Interim Report Kadlec, R. H., Knight, R. L., 1996. Treatment of Wetlands. Lewis Publishers, CRC Press LLC. McCormick, P.V., Newman, S., Miao, S., Gawlik, D.E., Marley, D., Reddy, K.R., Fontaine, T.D., 2002. Effects of anthropogenic phosphorus inputs on the Everglades. J.W. Porter and K.G. Porter, ed. In: The Everglades, Florida Bay, and Coral Reefs of the Florida Keys – An ecosystem sourcebook. CRC Press LLC. McPherson, B., Halley, R., 1996. The South Florida Environment: a region under stress. USGS, Circular no.1134. United States Government Printing Office. Newman, S., McCormick, P.V., Backus, J.G., 2002. Phosphatase activity as an early waring indicator of wetland eutrophication: problems and prospects. Journal of Applied Phycology v15,i1, p45-59. Non-ECP PM, 2005. Non-ECP EFA Permit Modification Request, February 2005 Orem, W., 2004. Impacts of sulfate contamination on the Florida Everglades. USGS, Fact Sheet, FS 109-03. Palmen, R., Buijsman, P., Hellingwerf, K.J., 1994. Physiological regulation of competence induction for natural transformation in Acinetobacter calcoaceticus. Archives of Microbiology, v162, p344–351. Reddy, M.M., Aiken, G.R., 2001. Fulvic acid-sulfide ion competition for mercury ion binding in the Florida Everglades. Water, Air, and Soil Pollution, v132, p89-104. SFWMD, 2004. Everglades Consolidated Report 2004. South Florid a Water Management District, West Palm Beach, Florida. Spijkerman, E., Coesel, P.F.M., 1998. Different response mechanisms of two planktonic desmid species (Chlorophyceae) to a single, saturating addition of phosphate. Journal of Phycology v34, p 438-445. Thomann, RV, 1972. System analysis and water quality measurement. Environmental Research and Application, Inc., New York. Vymazal, J., Richardson, C.J., 1995. Species composition, biomass, and nutrient content of periphyton in the Florida Everglades. Journal of Phycology v31, p343-354. Status: Final Date: December 15, 2005 Report No. LOXA05-004 Waldon, M., 2005. Verbal communication on September 19. Environmental Program Team, ARM Loxahatchee Wildlife Refuge.. Weaver, K., Bennett, T., Payne, G., Germain, G., Hill, S., Iricanin, N., 2001. Chapter 4: Status of Water Quality Criteria Compliance in the Everglades Protection Area. G. Redfield ed. In: 2001 Everglades Consolidate Report. South Florida Water Management District. Weaver, K., 2004. Everglades Marsh Dissolved Oxygen Site Specific Alternative Criterion Technical Support Document. Water Quality Standards and Special Projects Program Division of Water Resource Management Florida Department of Environmental Protection Tallahassee, FL. http://www.dep.state.fl.us/water/wqssp/everglades/dossac.htm Weaver, K., Payne, G., 2005. Chapter 2A: Status of Water Quality in the Everglades Protection Area. G. Redfield, ed. In: 2005 South Florida Environmental Report. South Florida Water Management District, West Palm Beach, FL. EXPERTS CONSULTED A.R.M. Loxahatchee National Wildlife Refuge Michael Waldon, Ph.D., Senior Hydrologist Matthew Harwell, Ph.D., Senior Ecologist Laura Brandt, Ph.D., Senior Wildlife Biologist South Florida Water Management District Sue Newman, Ph.D., Senior Supervising Environmental Scientist Scot Hagerthey, Ph.D., Senior Environmental Scientist David Struve, Ph.D., Water Quality Division Director United States Geological Survey Paul McCormick, Ph.D., Ecologist Status: Final Date: December 15, 2005 Report No. LOXA05-004 Table 1. Analytical water quality monitoring parameters for the A.R.M. Loxahatchee National Wildlife Refuge - Enhanced Water Quality Project. Parameter descriptions and IDs are as listed in the SFWMD DBHYDRO database. PARAMETER ID UNITS ALKALINITY TOTAL as CaCO3 ALKA mg L-1 COLOR COLOR PCU TURBIDITY TURB NTU ALKALINE PHOSPATASE APA nM/minmL CALCIUM Ca mg L-1 CHLORIDE Cl mg L-1 DISSOLVED OXYGEN DO mg L-1 DISSOLVED ORGANIC CARBON DOC mg L-1 HARDNESS as CaCO3 HARD mg L-1 POTASSIUM K mg L-1 MAGNESIUM Mg mg L-1 SODIUM Na mg L-1 AMMONIUM NH4 mg L-1 NITRATE NO3 mg L-1 NITRITE NO2 mg L-1 NITRATES and NITRITES as N NOX mg L-1 PHOSPHATE, ORTHO as P OPO4 mg L-1 pH pH mg L-1 SILICA SiO2 mg L-1 SULFATE SO4 mg L-1 SP CONDUCTANCE SpCOND uS/cm KJELDAHL NITROGEN, DISSOLVED TDKN mg L-1 PHOSPHATE, DISSOLVED as P TDPO4 mg L-1 TOTAL DISSOLVED SOLIDS TDS mg L-1 TEMPERATURE TEMP Deg. C KJELDAHL NITROGEN, TOTAL TKN mg L-1 CARBON, TOTAL ORGANIC TOC mg L-1 PHOSPHATE, TOTAL as P TPO4 mg L-1 TOTAL SUSPENDED SOLIDS TSS mg L-1 Status: Final Date: December 15, 2005 Report No. LOXA05-004 Table 2. Cost recovery resulting from parameter elimination. The reported cost of analysis from SFWMD for the second quarter of 2004 was projected over all four quarters of year 2 for the before parameter elimination cost assessment and only in the first quarter of year 2 for the after parameter elimination cost assessment. YR2 TOTAL ESTIMATED COST BEFORE PARAMETER ELIMINATION YR2 TOTAL ESTIMATED COST AFTER PARAMETER ELIMINATION COST DIFFERENCE TEST CODE PRICE COUNT COST COUNT COST COST ALKA $6.53 488 $3,186.64 488 $3,186.64 $0.00 APA $17.21 488 $8,398.48 122 $2,099.62 $6,298.86 CA $8.13 488 $3,967.44 488 $3,967.44 $0.00 CL $8.07 512 $4,131.84 512 $4,131.84 $0.00 COLOR $6.30 488 $3,074.40 122 $768.60 $2,305.80 DOC $12.20 488 $5,953.60 122 $5,953.60 $0.00 K $8.13 488 $3,967.44 122 $991.86 $2,975.58 MG $8.13 488 $3,967.44 122 $991.86 $2,975.58 NA $8.13 488 $3,967.44 122 $991.86 $2,975.58 NH4 $6.53 492 $3,212.76 123 $803.19 $2,409.57 NO2 $6.53 488 $3,186.64 122 $796.66 $2,389.98 NOX $6.53 504 $3,291.12 504 $3,291.12 $0.00 OPO4 $6.53 492 $3,212.76 123 $803.19 $2,409.57 SIO2 $8.25 488 $4,026.00 488 $4,026.00 $0.00 SO4 $8.07 512 $4,131.84 512 $4,131.84 $0.00 TDKN $12.26 496 $6,080.96 124 $1,520.24 $4,560.72 TDPO4 $11.42 496 $5,664.32 496 $5,664.32 $0.00 TDS $12.20 488 $5,953.60 488 $5,953.60 $0.00 TKN $12.26 492 $6,031.92 492 $6,031.92 $0.00 TOC $12.20 492 $6,002.40 123 $6,002.40 $0.00 TPO4 $9.14 520 $4,752.80 520 $4,752.80 $0.00 TSS $7.77 488 $3,791.76 488 $3,791.76 $0.00 TURB $6.53 488 $3,186.64 488 $3,186.64 $0.00 TOTAL 11352 $103,140.24 7311 $64,872.00 $38,268.24 Status: Final Date: December 15, 2005 Report No. LOXA05-004 Table 3. Correlation parameter with correlation coefficient (r) greater than or equal to 0.95. CORRELATED PARAMETERS r VALUE ALKA Ca 0.98 ALKA HARD 0.98 ALKA SPCOND 0.96 ALKA TDS 0.95 Ca HARD 0.99 Ca SPCOND 0.95 Ca TDS 0.95 Cl Na 0.997 Cl SPCOND 0.95 HARD Mg 0.97 HARD SPCOND 0.97 HARD TDS 0.97 Mg Na 0.95 Mg SPCOND 0.97 Mg TDS 0.97 Na SPCOND 0.96 Na TDS 0.95 OPO4 TDPO4 0.997 SPCOND TDS 0.98 Status: Final Date: December 15, 2005 Report No. LOXA05-004 Table 4. Finalized list of water quality parameters to be eliminated from the Enhanced Water Quality Monitoring Program. Parameter Primary Justification for Elimination APA Non-stand alone parameter requiring further data to be useful COLOR Low value with respect to water quality in the marsh HARD Redundant to alkalinity K Redundant to TDS; low value as WQ indicator for the marsh Mg Redundant to hardness and thus alkalinity Na Redundant to alkalinity NH4 Encompassed in total nitrogen measurement NO3 Captured in NOX measurement NO2 Short lived and captured in NOX measurement OPO4 Redundant to TDPO4 TDKN Captured in total nitrogen measurement Status: Final Date: December 15, 2005 Report No. LOXA05-004 Appendix A Table A.1. Pearson product moment correlation analysis for the suite of 29 parameters considered for elimination from the A.R.M. Loxahatchee National Wildlife Refuge - Enhanced Water Quality Monitoring Program. ALKA APA CA CL COLOR DEPTH DISS_O2 DOC HARD K MG NA NH4 NO2 NOX OPO4 PH_FIELD SIO2 SO4 SpCond TDKN TDPO4 TDS TEMP TKN TOC TPO4 TSS TURB TN TPP ALKA 1.000 APA -0.622 1.000 CA 0.981 -0.615 1.000 CL 0.883 -0.528 0.857 1.000 COLOR 0.385 -0.292 0.419 0.199 1.000 DEPTH 0.536 -0.359 0.561 0.556 -0.072 1.000 DISS_O2 -0.022 0.095 -0.030 0.131 -0.463 0.331 1.000 DOC 0.705 -0.391 0.683 0.772 0.594 0.137 -0.106 1.000 HARD 0.984 -0.610 0.993 0.897 0.395 0.553 -0.012 0.720 1.000 K 0.916 -0.597 0.916 0.860 0.515 0.473 -0.081 0.776 0.929 1.000 MG 0.943 -0.570 0.931 0.937 0.324 0.511 0.027 0.762 0.968 0.911 1.000 NA 0.894 -0.539 0.868 0.997 0.217 0.553 0.110 0.770 0.909 0.867 0.950 1.000 NH4 0.387 -0.224 0.414 0.320 0.251 0.505 -0.087 0.249 0.389 0.393 0.320 0.326 1.000 NO2 0.429 -0.242 0.450 0.392 0.264 0.498 0.089 0.318 0.445 0.434 0.415 0.396 0.491 1.000 NOX 0.382 -0.202 0.386 0.365 -0.024 0.536 0.297 0.158 0.390 0.341 0.381 0.367 0.339 0.775 1.000 OPO4 0.258 -0.219 0.322 0.100 0.370 0.323 -0.164 0.062 0.264 0.362 0.131 0.103 0.514 0.286 0.161 1.000 PH_FIELD 0.774 -0.556 0.770 0.797 0.061 0.683 0.327 0.484 0.786 0.743 0.783 0.793 0.320 0.415 0.431 0.209 1.000 SIO2 0.641 -0.403 0.620 0.613 0.521 0.191 -0.295 0.609 0.659 0.690 0.709 0.632 0.114 0.215 0.065 0.033 0.354 1.000 SO4 0.842 -0.482 0.865 0.835 0.344 0.524 -0.003 0.656 0.902 0.847 0.937 0.854 0.343 0.427 0.368 0.217 0.731 0.676 1.000 SpCond 0.958 -0.590 0.950 0.954 0.330 0.552 0.022 0.759 0.972 0.917 0.972 0.962 0.403 0.429 0.381 0.199 0.796 0.659 0.895 1.000 TDKN 0.741 -0.371 0.752 0.753 0.602 0.283 -0.098 0.907 0.769 0.815 0.769 0.752 0.432 0.420 0.255 0.269 0.560 0.548 0.728 0.782 1.000 TDPO4 0.294 -0.241 0.358 0.136 0.365 0.361 -0.145 0.078 0.301 0.395 0.165 0.138 0.526 0.301 0.182 0.997 0.249 0.040 0.248 0.236 0.290 1.000 TDS 0.953 -0.574 0.950 0.941 0.415 0.525 -0.020 0.794 0.972 0.938 0.974 0.949 0.377 0.430 0.342 0.214 0.765 0.721 0.910 0.979 0.816 0.247 1.000 TEMP 0.218 -0.119 0.233 0.203 0.218 0.215 -0.170 0.090 0.244 0.293 0.255 0.215 0.189 0.134 -0.039 0.222 0.236 0.412 0.381 0.260 0.189 0.226 0.289 1.000 TKN -0.021 0.138 -0.015 0.032 0.285 -0.073 -0.054 0.227 -0.011 0.045 -0.001 0.021 0.018 0.048 -0.014 0.002 -0.050 0.089 -0.006 0.010 0.232 -0.003 0.037 -0.332 1.000 TOC 0.675 -0.425 0.666 0.673 0.491 0.144 -0.134 0.811 0.690 0.727 0.708 0.679 0.287 0.258 0.117 0.117 0.463 0.557 0.635 0.700 0.769 0.137 0.730 0.305 -0.280 1.000 TPO4 0.023 0.157 0.049 0.008 0.254 0.042 -0.081 0.109 0.033 0.087 -0.002 0.002 0.138 0.106 0.038 0.275 -0.015 0.020 0.029 0.024 0.133 0.274 0.044 -0.276 0.728 -0.227 1.000 TSS 0.021 0.010 0.027 0.059 -0.211 0.038 0.157 -0.001 0.030 -0.023 0.035 0.047 0.114 0.024 0.054 0.002 0.127 -0.124 0.017 0.035 0.078 0.009 0.015 0.024 0.139 0.017 0.056 1.000 TURB 0.431 -0.218 0.445 0.490 -0.188 0.587 0.341 0.149 0.452 0.363 0.446 0.478 0.321 0.384 0.489 0.150 0.584 0.076 0.443 0.467 0.268 0.184 0.427 0.130 0.053 0.124 0.107 0.547 1.000 TN -0.015 0.135 -0.009 0.038 0.284 -0.065 -0.050 0.230 -0.005 0.050 0.005 0.027 0.023 0.060 0.002 0.004 -0.043 0.090 0.000 0.016 0.236 0.000 0.042 -0.333 1.000 -0.279 0.729 0.140 0.061 1.000 TPP -0.061 0.232 -0.053 -0.034 0.163 -0.062 -0.045 0.091 -0.053 -0.021 -0.051 -0.040 -0.007 0.028 -0.012 0.004 -0.090 0.010 -0.041 -0.045 0.053 0.003 -0.026 -0.354 0.751 -0.274 0.962 0.017 0.044 0.751 1.000 Status: Final Date: December 15, 2005 Report No. LOXA05-004 Appendix B Temporal graphs of parameters considered for elimination by the expert scientists. Graphs with stars (*) in the upper left corner are also on the final list of parameters to be dropped. Figure 1. *Alkaline Phosphatase for the period June 2004 to May 2004. Figure 2. *Color for the period June 2004 to May 2004. Figure 3. Dissolved organic carbon for the period June 2004 to May 2004. Figure 4. *Hardness for the period June 2004 to May 2004. Figure 5. *Potassium for the period June 2004 to May 2004. Figure 6. *Magnesium for the period June 2004 to May 2004. Figure 7. Total organic carbon for the period June 2004 to May 2004. Figure 8. *Sodium for the period June 2004 to May 2004. Figure 9. *Ammonium for the period June 2004 to May 2004. Figure 10. *Nitrite for the period June 2004 to May 2004. Figure 11. *Total dissolved Kjeldahl nitrogen for the period June 2004 to May 2004. Figure 12. *Ortho-phosphate for the period June 2004 to May 2004. Figure 13. Silica for the period June 2004 to May 2004. Figure 14. Total dissolved solids for the period June 2004 to May 2004. Figure 15. Turbidity for the period June 2004 to May 2004. Figure 16. Total organic carbon for the period June 2004 to May 2004. Correlation graphs for parameters with R values greater than or equal to 0.95. Figure 17. Correlation graphs for (a) Alkalinity verses Calcium, (b) Alkalinity verses Hardness, (c) Alkalinity verses Specific conductance, and (d) Alkalinity verses Total dissolved solids. Figure 18. Correlation graphs for (a) Calcium verses Hardness, (b) Calcium verses Specific conductance, (c) Calcium verses Total dissolved solids, and (d) Chloride verses Sodium. Figure 19. Correlation graphs for (a) Chloride verses Specific Conductance, (b) Magnesium verses Specific Conductance, (c) Hardness verses Specific Conductance, and (d) Hardness verses Total dissolved solids. Figure 20. Correlation graphs for (a) Alkalinity Magnesium verses Specific Conductance, (b) Magnesium verses Total Dissolved Solids, (c) Sodium verses Specific Conductance, and (d) Sodium verses Total Dissolved Solids. Figure 21. Correlation graphs for (a) Ortho-phosphate verses Total Dissolved Phosphorus and (b) Specific Conductance verses Total Dissolved Solids. Status: Final 26 Date: December 15, 2005 Report No. LOXA05-004 Figure 1. Alkaline Phosphatase for the period June 2004 to May 2004. * COLOR * LOXA 200 300 180 160 250 200 150 100 50 0 23-Apr-04 22-Jun-04 21-Aug-04 20-Oct-04 19-Dec-04 17-Feb-05 18-Apr-05 COLOR_COLOR_DATE Figure 2. Color for the period June 2004 to May 2004. COLOR (PCU) Status: Final 27 Date: December 15, 2005 Report No. LOXA05-004 DISSOLVED ORGANIC CARBON LOXA 50 45 40 35 30 25 20 15 10 5 DOC_INTERIOR DOC_CANAL 0 23-Apr-04 22-Jun-04 21-Aug-04 20-Oct-04 19-Dec-04 17-Feb-05 18-Apr-05 17-Jun-05 DATE Figure 3. Dissolved organic carbon for the period June 2004 to May 2004. DOC (mg/L) Status: Final Date: December 15, 2005 Report No. LOXA05-004 HARD (mg/L) * HARDNESS as CaCO3 350 300 250 200 150 100 50 0 23-Apr-04 22-Jun-04 21-Aug-04 20-Oct-04 19-Dec-04 17-Feb-05 18-Apr-05 17-Jun-05 LOXA HARD_INTERIOR HARD_CANAL DATE Figure 4. Hardness for the period June 2004 to May 2004. Status: Final Date: December 15, 2005 Report No. LOXA05-004 K (mg/L) POTASSIUM * LOXA 14 12 10 8 6 4 2 0 K_INTERIOR K_CANAL 23-Apr-04 22-Jun-04 21-Aug-04 20-Oct-04 19-Dec-04 17-Feb-05 18-Apr-05 17-Jun-05 DATE Figure 5. Potassium for the period June 2004 to May 2004. Status: Final Date: December 15, 2005 Report No. LOXA05-004 MG (mg/L) MAGNESIUM * ALL LOXA 30 25 20 15 10 5 0 23-Apr-04 22-Jun-04 21-Aug-04 20-Oct-04 19-Dec-04 17-Feb-05 18-Apr-05 17-Jun-05 MG_INTERIOR MG_CANAL DATE Figure 6. Magnesium for the period June 2004 to May 2004. Status: Final Date: December 15, 2005 Report No. LOXA05-004 TOTAL ORGANIC CARBON LOXA 60 TOC_INTERIOR TOC_CANAL 50 40 30 20 10 0 23-Apr-04 22-Jun-04 21-Aug-04 20-Oct-04 19-Dec-04 17-Feb-05 18-Apr-05 17-Jun-05 DATE Figure 7. Total organic carbon for the period June 2004 to May 2004. TOC (mg/L) Status: Final Date: December 15, 2005 Report No. LOXA05-004 NA (mg/L) * SODIUM LOXA 140 120 100 80 60 40 20 0 23-Apr-04 22-Jun-04 21-Aug-04 20-Oct-04 19-Dec-04 17-Feb-05 18-Apr-05 17-Jun-05 NA_INTERIOR NA_CANAL DATE Figure 8. Sodium for the period June 2004 to May 2004. Status: Final 33 Date: December 15, 2005 Report No. LOXA05-004 * AMMONIUM LOXA NH4 (mg/L) 0.4 0.35 0.3 0.25 0.2 0.15 0.1 0.05 0 23-Apr-04 12-Jun-04 1-Aug-04 20-Sep-04 9-Nov-04 29-Dec-04 17-Feb-05 8-Apr-05 28-May-05 17-Jul-05 NH4_INTERIOR NH4_CANAL DATE Figure 9. Ammonium for the period June 2004 to May 2004. Status: Final Date: December 15, 2005 Report No. LOXA05-004 NO2 (mg/L) * NITRITE LOXA 0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0 23-Apr-04 22-Jun-04 21-Aug-04 20-Oct-04 19-Dec-04 17-Feb-05 18-Apr-05 17-Jun-05 NO2_INTERIOR NO2_CANAL DATE Figure 10. Nitrite for the period June 2004 to May 2004. Status: Final Date: December 15, 2005 Report No. LOXA05-004 TDKN (mg/L) TOTAL DISSOLVE KJELDAHL NITROGEN * LOXA 3 2.5 2 1.5 1 0.5 0 23-Apr-04 22-Jun-04 21-Aug-04 20-Oct-04 19-Dec-04 17-Feb-05 18-Apr-05 17-Jun-05 DATE TDKN_INTERIOR TDKN_CANAL Figure 11. Total dissolved Kjeldahl nitrogen for the period June 2004 to May 2004. Status: Final 36 Date: December 15, 2005 Report No. LOXA05-004 ORTHO PHOSPHATE * LOXA OPO4 (mg/L) 0.6 0.5 0.4 0.3 0.2 0.1 0 23-Apr-04 22-Jun-04 21-Aug-04 20-Oct-04 19-Dec-04 17-Feb-05 18-Apr-05 17-Jun-05 OPO4_INTERIOR OPO4_CANAL DATE Figure 12. Ortho-phosphate for the period June 2004 to May 2004. Status: Final 37 Date: December 15, 2005 Report No. LOXA05-004 SILICA LOXA SIO2 (mg/L) 45 40 35 30 25 20 15 10 5 0 23-Apr-04 22-Jun-04 21-Aug-04 20-Oct-04 19-Dec-04 17-Feb-05 18-Apr-05 17-Jun-05 SIO2_INTERIOR SIO2_CANAL DATE Figure 13. Silica for the period June 2004 to May 2004. Status: Final 38 Date: December 15, 2005 Report No. LOXA05-004 TOTAL DISSOLVED SOLIDS LOXA TDS (mg/L) 800 700 600 500 400 300 200 100 0 TDS_INTERIOR TDS_CANAL 23-Apr-04 22-Jun-04 21-Aug-04 20-Oct-04 19-Dec-04 17-Feb-05 18-Apr-05 17-Jun-05 DATE Figure 14. Total dissolved solids for the period June 2004 to May 2004. Status: Final Date: December 15, 2005 Report No. LOXA05-004 TURBIDITY LOXA 25 TURB_INTERIOR TURB_CANAL 20 15 10 5 0 23-Apr-04 22-Jun-04 21-Aug-04 20-Oct-04 19-Dec-04 17-Feb-05 18-Apr-05 17-Jun-05 DATE Figure 15. Turbidity for the period June 2004 to May 2004. TURB (NTU) Status: Final Date: December 15, 2005 Report No. LOXA05-004 TOC (mg/L) TOTAL ORGANIC CARBON * LOXA 60 50 40 30 20 10 0 23-Apr-04 22-Jun-04 21-Aug-04 20-Oct-04 19-Dec-04 17-Feb-05 18-Apr-05 17-Jun-05 TOC_INTERIOR TOC_CANAL DATE Figure 16. Total organic carbon for the period June 2004 to May 2004. Status: Final Date: December 15, 2005 Report No. LOXA05-004 a 350 100 b 90 300 80 20 50 10 0 0 0 50 100 150 200 250 300 0 50 100 150 200 250 300 ALKALINITY (mg/L) ALKALINITY AS CaCO3 (mg/L) 1200 c 800 d 700 TOTAL DISSOLVED SOLIDS (mg/L) HARDNESS AS CaCO3 (mg/L) 250 70 60 SPECIFIC CONDUCTIVITY (mS/cm) CALCIUM (mg/L) 200 150 40 30 100 200 100 0 0 0 50 100 150 200 250 300 0 50 100 150 200 250 300 ALKALINITY AS CaCO3 (mg/L) ALKALINITY AS CaCO3 (mg/L) Figure 17. Correlation graphs for (a) Alkalinity verses Calcium, (b) Alkalinity verses Hardness, (c) Alkalinity verses Specific conductance, and (d) Alkalinity verses Total dissolved solids. 600 500 400 300 200 Status: Final 42 Date: December 15, 2005 Report No. LOXA05-004 350 a 1200 b 300 1000 TOTAL DISSOLVED SOLIDS (mg/L) HARDNESS AS CaCO3 (mg/L) SODIUM (mg/L) SPECIFIC CONDUCTIVITY ( mS/cm) 800 600 400 200 50 0 0 0102030405060708090100 0102030405060708090100 CALCIUM (mg/L) CALCIUM (mg/L) 800 140 c d 700 120 600 100 40 200 20 100 0 0 0 10 20 30 40 50 60 70 80 90 100 0 20 40 60 80 100120140160180200 CALCIUM (mg/L) CHLORIDE (mg/L) 80 60 Figure 18. Correlation graphs for (a) Calcium verses Hardness, (b) Calcium verses Specific conductance, (c) Calcium verses Total dissolved solids, and (d) Chloride verses Sodium. Status: Final 43 Date: December 15, 2005 Report No. LOXA05-004 1400 a 1200 b 1200 1000 TOTAL DISSOLVED SOLIDS (mg/L) SPECIFIC CONDUCTIIVITY (mS/cm) SPECIFIC CONDUCTIVITY ( mS/cm) SPECIFIC CONDUCTIVITY mS/cm 800 600 400 200 200 0 0 0 20 40 60 80 100120140160180200 0 5 10 15 20 25 30 CHLORIDE (mg/L) MAGNESIUM (mg/L) 1200 800 c d 700 200 100 0 0 0 50 100 150 200 250 300 350 0 50 100 150 200 250 300 350 HARDNESS (mg/L) HARDNESS (mg/L) Figure 19. Correlation graphs for (a) Chloride verses Specific Conductance, (b) Magnesium verses Specific Conductance, (c) Hardness verses Specific Conductance, and (d) Hardness verses Total dissolved solids . 600 500 400 300 200 Status: Final 44 Date: December 15, 2005 Report No. LOXA05-004 1200 a 800 b 700 SPECIFIC CONDUCTIVITY ( mS/cm) SPECIFIC CONDUCTIIVITY (mS/cm) TOTAL DISSOLVED SOLIDS (mg/L) TOTAL DISSOLVED SOLIDS (mg/L) 600 500 400 300 200 200 100 0 0 0 5 1015202530 0 5 10 15 2025 30 MAGNESIUM (mg/L) MAGNESIUM (mg/L) 1600 c 900 d 800 1400 200 100 0 0 0 20 40 60 80 100120140 0 20 40 60 80 100120140 SODIUM (mg/L) SODIUM (mg/L) Figure 20. Correlation graphs for (a) Alkalinity Magnesium verses Specific Conductance, (b) Magnesium verses Total Dissolved Solids , (c) Sodium verses Specific Conductance, and (d) Sodium verses Total Dissolved Solids . 700 600 500 400 300 200 Status: Final Date: December 15, 2005 Report No. LOXA05-004 800 700 b TOTAL DISSOLVED PHOSPHORUS (mg/L) a 0.1 100 0 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0 200 400 600 800 1000 1200 ORTHO-PHOSPHATE (mg/L) SPECIFIC CONDUCTIVITY ( mS/cm) TOTAL DISSOLVED SOLIDS (mg/L) 600 500 400 300 200 Figure 21. Correlation graphs for (a) Ortho-phosphate verses Total Dissolved Phosphorus and (b) Specific Conductance verses Total Dissolved Solids . Status: Final Date: December 15, 2005 Report No. LOXA05-004 10216 Lee Road, Boynton Beach, Florida 33437-4796 APPENDIX C Phone (561) 732-3684 Fax (561) 369-7190 January 17, 2006 To: TOC and interested parties From: Donatto Surratt, Ph.D. Matt Harwell, Ph.D. Subject: Response to written comments on the November 7, 2005 draft A.R.M. Loxahatchee National Wildlife Refuge Enhanced Water Quality Program – Parameter Reduction Rationale We appreciate the continued interest in our water quality monitoring program and the careful and detailed review that outside interested parties gave our draft report, “Enhanced Water Quality Program Parameter Reduction Rationale. Report No. LOXA05-004” by Surratt, D. This exercise to reduce unnecessary parameters from our monitoring program benefited from the constructive technical comments we received. Below we provide a written response to the comments we received, as well as identify how the final version of the report (available at: http://www.sofia.usgs.gov/lox_monitor_model) was modified. SFWMD Comments First bullet: “The objective for reducing parameters, as stated in the second paragraph on the first page, was to reduce analytical costs and reclaim funds to extend the duration of the LOXA monitoring effort. The rationale for selecting parameters to meet this objective did not appear to be consistently applied across parameters…” Response – As stated in the cover memo when the draft report was circulated for review and comment, the primary aim of the exercise was to eliminate redundant parameters and those not providing valuable information. Cost savings was intentionally not a driving factor in the analysis. To clarify this point we added, “The LOXA Work Plan (Brandt et al. 2004) states that, “data will be reviewed… and the list of parameters reduced as appropriate” (p. 12) with the objective to optimize the marsh monitoring network”. Second bullet: “… It is important to undertake an initial data analysis to assess whether some sites can be eliminated for optimizing the expanded network .” Status: Final Date: December 15, 2005 Report No. LOXA05-004 Response – As stated in the cover memo when the draft report was circulated for review and comment, we are currently undertaking a site optimization exercise. Responses to “more detailed technical comments” from SFWMD. Page 3, Line 2 (missing text and need for editing?) Response – The revised text now reads: “Another characterization approach was the spatial analysis approach which was performed from canal to the interior of the marsh on several parameters.” Alkalinity: “Calcium, alkalinity and hardness correlated well because two of the parameters are reported in units as CaCO3. It seems that the correlation analysis of this parameter was unnecessary. The reason to keep ALKA is not because of its correlation with calcium or hardness. It was kept because it had a numeric criterion pursuant to Section 62-302.530 F.A.C. and is important to the basic ecology of a softwater marsh.” Response – The revised text now reads: “Alkalinity correlated well with four parameters – calcium (r = 0.98), hardness (r = 0.98), specific conductance (r = 0.96), and total dissolved solids (r = 0.95). The high correlation between ALKA, calcium, and hardness is most likely an artifact of their dependence on the concentration of CaCO3. Although each of these parameters share this dependency, they are still determined through unique analytical procedures and explain different conditions of the analyzed water parcels and as such yield different values and patterns through time, which is why they are not 100% correlated. Further, these correlations allow the observer to quickly qualify data points as outliers (prior to a rigorous quantitative assessment) with respect to specific conductance, which is exceptionally important for tracking water movement from the canal to the interior marsh. Alkalinity has a Class III criterion of = 20 mg L-1. In October and November 2004, five sampled sit es had values below the criterion. This criterion is inappropriate for rainwater- dominated wetlands where alkalinity is naturally below 20 mg L-1 (Weaver, 2005). A water quality goal of the Refuge is to maintain this low alkalinity condition. Because ALKA provides the ability to quickly determine specific conductance outliers; because there is an established numeric criterion pursuant to Section 62-302.530 F.A.C.; and because ALKA is important to the ecology of softwater marshes, ALKA will not be eliminated from the analytical suite. Alkalinity remains a monitored parameter.” Chloride: “Chloride is strongly correlated with sodium and specific conductance (and strongly correlated with TDS). If it has a strong correlation with specific conductivity and cost needs to be reduced, specific conductivity might be an acceptable surrogate and chloride could be reduced. If justification is needed, why isn’t mass balance modeling mentioned?” Response – We decided to keep chloride as it provides a check for specific conductance. We have found a number of occasions when specific conductance has been entered incorrectly in DBHYDRO. Further, these two parameters are the only parameters that act as true biologically Status: Final Date: December 15, 2005 Report No. LOXA05-004 and chemically conservative parameters out of the monitored suite of parameters and as such we are maintaining both of them to ensure we will be able to track canal water penetration in the future. Dissolved Organic Carbon (DOC): “DOC is very important to the biogeochemical cycling in the marsh. It may not be appropriate to eliminate DOC as a monitoring parameter.” Response – We have revisited DOC and agree with the conclusion to continue monitoring DOC. The revised text now reads: “Dissolved organic carbon has been linked to microbial respiration and fulvic acid regulated mercury methylation in the southern Everglades (Reddy and Aiken, 2001). One of the experts in preliminary lab experiments revealed microbial respiration to be strongly carbon limited instead of phosphorus or nitrogen limited as for most system. Additional influxes of labile carbon can affect marsh soil processes. Reddy and Aiken (2001) demonstrate strong correlations between Hg bioaccumulation and DOC concentrations. Correlation analysis for DOC did not reveal significant correlation with other parameters. Because neither Hg nor microbial respiration are monitored in the Refuge, because of the potential of DOC concentration changes to cause Hg bioaccumulation rates to change, and because no other parameters can serve as surrogates of DOC, continued monitoring DOC is recommended. DOC remains a monitored parameter.” Dissolved Oxygen (DO): “Since DO follows a diel cycle influenced by light and temperature, the usefulness of DO grab samples is highly debatable. As described in the Refuge’s draft paper current legal mandates were adjusted by a Site Specific Alternative Criteria (SSAC). The value of grab sample monitoring of DO at an expanded list of stations is questionable. A more practical approach would be deployments at selected stations for several days.” Response – Along the central transect in the marsh we do monitor DO for several consecutive days during the wet season. However, this does not replace the entire monitoring network. Grab sample monitoring for DO will continue. Hardness: “Hardness is calculated from calcium and magnesium: Hardness=2.497 (Ca, mg/L) + 4.118 (Mg, mg/L). If magnesium is eliminated from the parameter list, hardness can not be calculated. The argument presented for keeping hardness stems from the f act that it is used to calculate criteria for a variety of trace metals. However, since trace metals are not monitored in the LOXA program, is there another reason for keeping this parameter? From the viewpoint of tracking the softwater aspects of the marsh, alkalinity should provide all necessary information. A reconsideration of the value of hardness appears worthwhile.” Response – We have revisited hardness based on the above suggestion and concur with the recommendation to remove hardness from the monitored parameters. The revised text for hardness is: “HARD: Hardness is measured as an equivalent concentration of calcium carbonate (CaCO3) and was expected to correlate well with alkalinity (which is also measured as an equivalent concentration of CaCO3). Hardness correlated well with ALKA (r = 0.98), Ca (r = 0.99), magnesium (r = 0.97), specific conductance (r = 0.97), and TDS (r = 0.97). Although these Status: Final Date: December 15, 2005 Report No. LOXA05-004 strong correlations are observed, the Florida Class III water quality criteria for various metals (Cd, Cr, Cu, Pb, Ni, Zn) require hardness in order to calculate the metal criterion (Bechtel, 2000). These metals have never been measured as a part of the LOXA project and were discontinued from the EVPA project in 2000. From the perspective of tracking softwater movement interior of the canals, ALKA and specific conductance appear to provide the necessary information, thus the elimination of the hardness appears to be warranted. Hardness is on the list of parameters to eliminate.” NH4 (ammonium): “The EPA in the context of south Florida refers to the Everglades Protection Area not the Environmental Protection Agency. It is worthwhile to note also that NH4 is rapidly recycled and instantaneous concentrations are not usually helpful in understanding N availability in a functioning marsh with relatively high levels of nitrogen. However, NH4 is the inorganic portion of the TKN value and it provides (through subtraction) an estimate of the organic nitrogen content. By dropping this parameter, it will not be known if that make-up shifts in the future.” Response – We corrected the editorial mistake. NH4 is low in the Refuge marsh and TKN is comprised mostly of organic material. As such we do not find it necessary to continue the monitoring of this parameter. Nitrate, nitrite, and NOx: “The only reason that should be given for keeping NOx is that it is necessary to calculate total nitrogen. The rationales for eliminating NO3 and NO2 do not flow logically and some of the facts presented could argue to keeping them. It could be noted in these sections that NO2 and NO3 are not particularly useful as individual parameter and their value is maintained by measuring NOx.” Response – The revised text now reads: “There were no strong correlations associated with nitrate. Nitrate is generally a significant portion of nitrate-nitrite (NOX) measurement and in combination with the other components of total nitrogen provides information about nutrient lability. Because NO2 is generally only a small fraction of the NOX, it was expected that NOX values can be used to estimate the NO3 concentration. This rationale holds true for station LZ40 in Lake Okeechobee such that NO2 was ~11% and NO3 was ~90% of the NOX concentration. However, this pattern was not observed in the Refuge marsh. Observation of the data showed NO2 at ~43% and NO3 at ~58% of the NOX concentrations. Thirty-one percent of the NO3 values for the Refuge were below detection limits, which may have complicated this assessment and potentially confound any trend analysis attempted for the Refuge. Independently, NO3 is not particularly useful and the value of this nitrogen fraction is maintained in measurements of NOX. Nitrate is on the list of parameters to be eliminated.” OPO4 (Orthophosphate) and TDP: “OPO4 is the bio-available fraction of P. It is the P fraction that can be compared to Si or inorganic N to look at nutrient availability and potential limitation. Since it may have a strong correlation to TDP and cost less to analyze, there is no compelling reason to eliminate it and keep TDP. TDP could be eliminated because it costs more and it does not provide a quick estimate of the biologically available P. Since OPO4 is extremely strongly correlated to TDP (r=0.99), then particulate P could be estimated by subtracting OPO4 from TP. Reconsideration of these parameters would seem appropriate.” Status: Final Date: December 15, 2005 Report No. LOXA05-004 Response – The argument presented here is very strong, but the only issue is the number of times OPO4 is below the detection limit. TDP and OPO4 are already not analyzed when the water depth is below 20 cm. But even when OPO4 is below detection limit, TDP values generally are measurable. If we eliminate TDP to keep OPO4, we will have even larger data gaps. As such, we decided to estimate OPO4 from TDP and eliminate OPO4. pH: “Point measurements ‘for’ pH can hint at interesting processes but cannot resolve the actual issue. This parameter is better suited for diel deployments combined with research to understand the processes that drive pH below 6.” Response – Presently, we monitor pH for several consecutive days during the wet season along the central transect through marsh. We do not have a research portion of this program dedicated to understanding what drives pH below 6. Our water is poorly buffered and even monthly snap shots help us understand the marsh and when conditions may be changing, particularly if pH deviates from the normal pH values we have observed. Silica (SiO2): “The reference below is incorrect as written: … There is no mention of hard water altering ecosystem structure, only biotic communities. Has the author noticed the changes in Si:P ratios in the Refuge or are these changes associated with diatoms anticipated in the near future? The argument is not easily followed as written and not very compelling.” Response – The citation has been referenced appropriately in the revised text. “There were no strong correlatio ns associated with silica. Silica shows a seasonal pattern for the interior sites of the Refuge. Silica is higher in the wet-warm season (late April – early November) and lower during the dry-cool season (November to early April). Independent research has demonstrated that silica concentrations decrease during the cooler season when algae populations (dominated by diatoms) are dying off and increase during the warmer season when these populations grow in again. Changes in these patterns can serve as a good indication of eutrophication and even biota shifts (Biggs, 1990). These assertions are generally for flowing waters (i.e., rivers) and historically appear to be of less importance for the Refuge wetland ecosystem which has a low diatom abundance associated with the periphyton population dominating as the primary producer of the Refuge (Kadlec and Knight, 1996). However, with STA-1E becoming operational in September 2005, there will be larger volumes of higher mineral content water discharge into the L-40 Canal. Canal water penetration into the marsh has been documented (Harwell et al., 2005). The introduction of higher mineral content water into a soft-water ecosystem is expected to impact the biotic community. The combination of temperature, nitrogen to phosphorus, and silica to phosphorus ratios has been shown to relate to algae species distributions (Adamus et al., 2001). This suite of indicators can theoretically be applied as an indicator of changes in periphyton community dynamics as the high mineral content water impact the Refuge. Presently, WCA-2 has a higher diatom abundance associated with the periphyton communities (relative to the Refuge) and the Refuge perimeter canals are major sources of water for WCA-2. Because of the potential of silica concentration change to cause biotic shifts, and because the canals of the Refuge provide water to other areas of the Everglades, silica is not on the elimination list. Silica remains a monitored parameter.” Status: Final Date: December 15, 2005 Report No. LOXA05-004 Total Dissolve Solids (TDS): “With the ability to measure specific conductivity, this parameter is unnecessary. Chloride (rather than TDS) could be used to verify that specific conductivity is measured correctly…” Response – As we have historically used TDS to identify outliers and erroneous data, we have decided to keep TDS. Temperature: “The reasoning for keeping TEMP is intriguing. All multi-parameter sonde units automatically correct specific conductance whether temperature is recorded or not. Just because it is used to adjust conductance is not a compelling reason to keep temperature. There are costs associated with data entry, data review, equipment calibration, etc., even if there are not analytical costs for temperature. If data are entered into the database, it should have value; point measurements for temperature have limited value.” Response – There were no strong correlations associated with temperature. Temperature is an integral parameter that changes seasonally. It is used to adjust conductance, dissolved oxygen measurements, determine DO saturation, and the DO-SSAC. As noted above, cost savings was not the primary factor in this exercise, and the minor cost savings alone does not provide enough justification to eliminate this parameter. Temperature remains a monitored parameter. Corps Comments 1. Most of the parameters that were eliminated, the logic is compelling. A few of them you may wish to reconsider. These are NH4 and DOC. You may wish to include these for the following reasons: 2. Relative to NH3- NH4, : certainly You can identify TN with TKN and NOx but you may not fully understand any future transformations or changes to the Refuge due to ECP modifications and CERP. Consider the following: a. NH3- NH4: Ammonium is a product of soil organic matter and NO-3 consumption per the simplified formula : (CH2O) 106 (NH3) 16 (H3PO4) + 84.8NO3- ?106 CO2 + 16 NH3 + H3PO4+148.4 H2O b. Increases in this may be increases in these patterns. (Davidson2000) c. This may contribute to substantial TN losses for poorly buffered waters in high photosynthetic areas (due to associated changes in pH),(DeBusk ). d. This is of particular interest to the down stream of the Refuge and will contribute to understanding the TN changes taking place in the refuge. 3. Relative to DOC: a. Hg interaction with DOC may be important in regulating ecosystem mercury methylation patterns (Reddy and Aiken 2001). b. Certainly, sulfate serves as the electron donor for bacteriamediated methylation of mercury c. but fulvic acid competes with the inorganic sulfide ion for mercury binding. d. Fulvate is a substantial fraction of the DOC. e. In my estimation, you really need both parameters to effectively track this. f. I am attaching some the pictorial plates from REMAP as a file. Status: Final Date: December 15, 2005 Report No. LOXA05-004 g. Note TOC (I couldn't find anything for DOC) corresponds to some of the methylation patterns. Response to the NH4 comment– Though we recognize the potential value of the information lost when we eliminate NH4, NH4 is relatively low in the Refuge compared to most marsh systems. NH4 is also a very small component of TKN, which is mostly organic and is a parameter we continue to monitor. The rationale for eliminating NH4 is that total nitrogen encompasses the NH4 fraction and because nitrogen is not considered a limiting nutrient in the Refuge. The small NH4 fraction of total nitrogen provides limited ecologically meaningful information. As a result, NH4 is being eliminated from this monitoring network. Response to the DOC comment – Please see comments for DOC above. ENP Comments “Parameter reduction can also be achieved through application of "Principle Component or Factor Analysis" which are more robust statistical tools for such applications. Other rationale still can be applied to bring in or throw out parameters after that. Some links below can be referred to check whether these procedures would apply or not for your report.” Response – We agree that PCA techniques can be valuable for general data exploration and analysis. For the purposes of our exercise, we chose to stick with the presented degree of analysis which allowed us to look at variables from both individua l and multivariate approaches. In our more robust analysis of the data sets (independent of the presented parameter reduction report) we do employ PCA and other similar techniques to include canonical correspondence analysis and multidimensional scaling approaches to assessing the datasets. Status: Final Date: December 15, 2005 Report No. LOXA05-004 APPENDIX D Comments from reviewers