4
Ecological Assessments

In the Phase I study, six St. Johns River Water management District (SJRWMD) workgroups used hydrodynamic and hydrologic modeling data, existing monitoring data, and literature reviews to provide preliminary assessments of potential biogeochemical and ecological impacts in the St. Johns River from withdrawing 262 million gallons per day (MGD) of surface water. These efforts will be expanded in Phase II to include additional modeling and data collection to delineate environmental effects boundaries and assess the potential for crossing response thresholds at various levels of continuous withdrawal. This chapter provides assessments of the Phase I report chapters on biogeochemistry; plankton, nutrients, and total maximum daily loads (TMDLs); the littoral zone; benthos; fish; and wetlands and wetland-dependent species. For each workgroup, the chapter includes recommendations to improve the Phase II efforts.

BIOGEOCHEMISTRY

The biogeochemistry workgroup of the Water Supply Impact Study (WSIS) identified seven potential effects of additional water withdrawals on biogeochemical processes and conditions in the St. Johns River drainage basin. All seven effects are related to the possibility that soil accretion will be reduced and/or oxidation of organic soils (histosols) will be enhanced in riparian wetlands associated with the extensive floodplains of the St. Johns River as a consequence of changes in river stage induced by additional water withdrawals. The floodplains have swampy herbaceous wetlands with deep organic soils, and withdrawals could increase the number of days that organic wetland soils are exposed to air. Exposure promotes oxidation and diagenesis of organic matter and release of various substances that can be exported to the river when the soils are inundated again.

During Phase I, the workgroup concluded that the potential effects of changes in pH and increased releases of organic inhibitors, labile dissolved organic carbon, and metals from oxidizing organic soils had unknown significance. However, three other potential effects were considered to have potentially high significance: (1) reduced nutrient sequestration, (2) increased release of colored dissolved organic matter (CDOM), and (3) increased production and reduced sequestration of greenhouse gases (carbon dioxide, methane, nitrous oxide) produced within inundated organic soils. The workgroup decided to focus on nutrient and CDOM release because they were thought to have the greatest potential for effects on the river.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 43
4 Ecological Assessments In the Phase I study, six St. Johns River Water management District (SJRWMD) workgroups used hydrodynamic and hydrologic modeling data, existing monitoring data, and literature reviews to provide preliminary assessments of potential biogeochemical and ecological impacts in the St. Johns River from withdrawing 262 million gallons per day (MGD) of surface water. These efforts will be expanded in Phase II to include additional modeling and data collection to delineate environmental effects boundaries and assess the potential for crossing response thresholds at various levels of continuous withdrawal. This chapter provides assessments of the Phase I report chapters on biogeochemistry; plankton, nutrients, and total maximum daily loads (TMDLs); the littoral zone; benthos; fish; and wetlands and wetland- dependent species. For each workgroup, the chapter includes recommendations to improve the Phase II efforts. BIOGEOCHEMISTRY The biogeochemistry workgroup of the Water Supply Impact Study (WSIS) identified seven potential effects of additional water withdrawals on biogeochemical processes and conditions in the St. Johns River drainage basin. All seven effects are related to the possibility that soil accretion will be reduced and/or oxidation of organic soils (histosols) will be enhanced in riparian wetlands associated with the extensive floodplains of the St. Johns River as a consequence of changes in river stage induced by additional water withdrawals. The floodplains have swampy herbaceous wetlands with deep organic soils, and withdrawals could increase the number of days that organic wetland soils are exposed to air. Exposure promotes oxidation and diagenesis of organic matter and release of various substances that can be exported to the river when the soils are inundated again. During Phase I, the workgroup concluded that the potential effects of changes in pH and increased releases of organic inhibitors, labile dissolved organic carbon, and metals from oxidizing organic soils had unknown significance. However, three other potential effects were considered to have potentially high significance: (1) reduced nutrient sequestration, (2) increased release of colored dissolved organic matter (CDOM), and (3) increased production and reduced sequestration of greenhouse gases (carbon dioxide, methane, nitrous oxide) produced within inundated organic soils. The workgroup decided to focus on nutrient and CDOM release because they were thought to have the greatest potential for effects on the river. 43

OCR for page 43
44 Review of the St. Johns River Water Supply Impact Study: Report 1 Assessment of Phase I The Phase I study involved calculations based on literature values for release rates of various constituents from flooded and exposed (i.e., dewatered) organic soils. No new field- based measurements or studies were performed. The additional duration of soil exposure was estimated for nine segments of the middle and upper St. Johns River basins based on hydrologic stage-frequency curves, and the areas of organic soils that would be affected were estimated from geographic information systems (GIS) data. Riparian wetlands in the lower St. Johns River were not included in these calculations because stage is controlled in that region by tide rather than river flow. The incremental loading to the river was based on a mass calculation of the change in release rates of substances between the inundated and drained states: ∆L = (FD – FI) × A × T where ∆L is the change in load to the river (g yr-1), FD is the areal release rate when drained (g m-2 d-1), FI is the areal release rate when inundated (g m-2 d-1), A is the affected area of organic soils (m2), and T equals the additional days of exposure (d yr-1). A positive change in loading indicates increased release of nutrients or organic matter to the river (or atmosphere), and a negative change indicates increased retention. Implicit in the above equation is the assumption that net primary production in the plant community associated with affected organic soils does not change between the inundated and drained states. The approach thus assumes that net change in flux is a reflection solely of changes in oxidation rates between inundated and drained states. The committee is concerned that no supporting evidence is provided for this assumption. Preliminary results were presented in the Phase I report for inorganic phosphorus release only. The authors suggested that increased releases caused by additional exposure (dewatering) of organic soils could be significant for Lake Winder (in the upper St. Johns River) relative to a current TMDL that requires a substantial decrease in phosphorus loading from the lake. No results were presented for the other major nutrient (nitrogen), or for CDOM, labile and inhibitory dissolved organic carbon (DOC), and metals because the workgroup concluded that insufficient information was available in the literature on release rates of these substances from exposed histosols. However, the committee notes that Table 1.1 of the Phase I report appears to include as much information on nitrogen release rates from flooded and drained peat lands as it does for phosphorus. The committee concludes that the focus on phosphorus in the Phase I studies was as much a result of the short amount of time available to the workgroup as a reflection on the adequacy of the release rate constants for various constituents. The work plan for the Phase II study indicates that these other substances will be addressed by undertaking appropriate data collection, and the committee concurs with this decision. The committee discussed at considerable length the issue of potential additional CDOM loadings as a result of increased water withdrawals, and it concluded that the Phase I report did not provide sufficient evidence that this would occur. The committee accepts the possibility that under some circumstances increased water withdrawals could lead to substantial increases in drying and oxidation of wetland soils and this could lead to losses in organic soils and increases in release of CDOM. Nonetheless, a convincing case was not made that this would happen in the St. Johns River as a result of the proposed additional water withdrawals. Oxidation of air- exposed organic soils degrades high molecular weight (insoluble) humic material into smaller, soluble humic and fulvic acid molecules that are exported to streams and rivers during soil reflooding, and chemical transformation of CDOM continues to occur under the influence of ultraviolet light and microbial processes in aquatic environments. Such processes already occur

OCR for page 43
Ecological Assessments 45 to a substantial extent in the St. Johns River basin. The river and its tributaries and associated lakes are highly colored with humic substances (CDOM) under existing conditions. The Phase I report did not make a persuasive case that changes in CDOM concentrations or loadings under conditions of additional water withdrawals would have significant ecological or water quality impacts. If water flow rates decrease and CDOM loadings remain constant (or increase), the concentration of CDOM would increase, but the Phase I report did not provide evidence that increased water withdrawals would enhance generation of CDOM in the riparian wetlands or that flows during reflooding events would be lower than under present conditions. Nonetheless, the committee recognizes that increased CDOM concentrations could affect light penetration (even though it already is low in this highly colored river) at least for short periods of time. Because primary production in the river may be light-limited (this is especially likely for submersed aquatic macrophytes), this potential effect deserves further attention. None of the literature values of substance release rates from drained and inundated soils used in the Phase I study (Table 1.1 of the biogeochemistry chapter) appears to be from soils in the St. Johns River basin. Many of the rate constants are from studies in South Florida, specifically the Everglades Agricultural Area (EAA), an area heavily affected by agricultural activities for many decades. Nutrient release rates from EAA soils are not likely to be applicable to riparian organic soils in the St. Johns River basin, which have not been heavily impacted by agriculture and other human activities. There also are many questions regarding the environmental conditions under which release rates reported in the literature were measured and whether those conditions are representative of conditions that would occur if St. Johns River basin organic soils were subjected to additional drainage. Recommendations for Phase II The workgroup intends to address the shortcomings described above regarding substance release rates from drying soils in the Phase II studies. To the extent that experimental studies are undertaken to obtain nutrient and CDOM release rates from drained soils, the committee encourages the workgroup to use procedures that will yield data reflective of environmental conditions. The committee understands that the Phase II biogeochemistry studies will address the question of enhanced nitrogen release from drained organic soils, and it agrees that they should, in addition to enhanced phosphorus release. It is essential that these release rates be placed in the proper context that can come only from having knowledge about the total nitrogen and phosphorus loadings to the river—specifically to the major lakes along the river channel. This information also needs to be obtained and analyzed at a sufficiently fine temporal scale to be relevant in assessing impacts on algal blooms; in a highly dynamic system like the St. Johns River, algal blooms are not controlled by total annual nitrogen and phosphorus loading, but by loadings at critical times of year, particularly those during and preceding low-flow conditions during warm weather periods. Furthermore, the ratio of concentrations of these two limiting nutrients is an important determinant of phytoplankton community composition (Lowery, 1998; Conley, 2000). The committee appreciates the difficulties and complexities involved in quantifying the additional loading of nutrients, metals, and natural organic substances to the St. Johns River that could result from the sequence:

OCR for page 43
46 Review of the St. Johns River Water Supply Impact Study: Report 1 Additional Changes Increased exposure Increased oxidation Increased transport water → in water → of soil organic matter → and release → to the river withdrawals stage to air of substances upon reflooding. In addition to large uncertainties in translating predicted changes in stage-duration to area- duration changes in exposure of soil organic matter to air, there are important uncertainties in the extent to which substances produced during air exposure will be transported to the river. In addition, there are substantial issues with regard to predicting the actual extent of substance oxidation and release during exposure to air. The District plans to conduct laboratory experiments during Phase II to determine substance release rates from drained soils (SJRWMD, 2009c, pp. 7-9). However, the committee is concerned that small-scale (laboratory) experiments will not be able to simulate field conditions adequately. For example, organic soils in field conditions probably receive moisture from upward capillary flow on a continuing basis even when there is no standing water above the soil surface. Mimicking this phenomenon in the laboratory could be difficult. Temperature, duration of drained conditions, and the degree of drying are likely to vary temporally under field conditions. These variations will affect oxidation rates in ways that would be difficult to duplicate in the laboratory. In this context, in situ mesocosm studies indicated in the work plan (p. 8) take on special importance. The work plan does not provide information on the size of the mesocosms, but Figure 2 (p. 9), which is a sketch of the proposed mesocosms, suggests they will be quite small. Given the well-known heterogeneity of soils, it is probable that many samples will need to be tested. The Phase II work plan provides various numbers in this regard: 150 general sites will be sampled; 36 soil cores will be collected from 12 sites; in situ studies will be conducted at an unspecified subset of the 12 sites. Unfortunately, no rationale is presented in the work plan for these numbers. Finally, there are likely to be important “scale factors” complicating the interpretation of results from laboratory microcosm experiments. For example, the presence of vegetation and extensive root systems in field soils may change soil moisture conditions, as well as the extent of oxic conditions within surficial soil layers. The committee encourages the District to carefully consider these issues in designing experiments so that the results can be of more than academic interest and in fact can be useful in predicting changes at the spatial and temporal scales of interest in the impact assessment. The above concerns lead to the following recommendations regarding Phase II biogeochemical studies. First, it is critically important that sufficient samples be analyzed to address the well-known large heterogeneity found in soils. If the proposed sampling design was not based on the known or expected diversity of soil types and within-plot heterogeneity of the study area, District staff should revisit their sampling plans. Second, experimental studies should be done at as large a spatial scale as possible to avoid artifacts caused by trying to extrapolate results from small sample sizes and small containers to the ambient environment. Specifically, samples incubated to assess effects of drying on rates of soil oxidation and nutrient release must include roots and vegetation and not just peat soil. In this regard, the committee is not convinced that samples of the size implied in the work plan for Phase II are adequate to give realistic and representative results for soil oxidation and nutrient release from riparian wetland soils in the St. Johns River under the conditions likely to be engendered by enhanced consumptive-use water withdrawals. Finally, the committee is not convinced that the Phase I studies have provided a sufficient predicate for the laboratory and in situ studies proposed for Phase II. In particular, it is

OCR for page 43
Ecological Assessments 47 premature to conduct extensive laboratory and field experiments to evaluate rates of nutrient and CDOM release from drying soils of riparian wetlands in the St. Johns River Basin because the Phase I studies were not adequate to demonstrate that drying and oxidation will occur on a spatial scale to cause significant impacts to the river ecosystem with the proposed additional water withdrawals. The committee is not concluding that such impacts will not occur, nor is it implying that additional water withdrawals will not result in drying and oxidation of riparian soils. Rather, the committee concludes that the Phase I studies were inadequate to demonstrate that this represents a likely impact of sufficient magnitude to warrant further study and analysis. Consequently, the committee concludes that a sequential approach would be more effective and efficient for the Phase II studies. First, additional analyses and experiments need to be undertaken to establish the areal extent of wetland soils that would be dried to a sufficient extent and for a sufficient duration to enhance oxidation of soil organic matter and subsequent release of nutrients and CDOM. The duration of drying needed to enhance oxidation likely will vary with season, and so timing of drying events needs to be considered. Second, if these studies indicate that further studies are warranted on transformation rates of soil organic matter during drying, then it would be appropriate to undertake experimental studies at mesocosm rather than microcosm scales to measure rates of nutrient and CDOM release under environmental conditions relevant to the proposed water withdrawal scenarios. Studies at this larger (mesocosm) scale are recommended to minimize experimental artifacts and enhance the ability to extrapolate results to the actual St. Johns River system. Summary Although the biogeochemistry workgroup identified and ranked several potential effects of additional water withdrawals on biogeochemical processes, their analysis did not use local release rates of constituents from flooded and exposed organic soils. Results were only presented for inorganic phosphorus release, although the work plan for Phase II indicates that nitrogen will be addressed. If laboratory experiments are undertaken to obtain local nutrient and CDOM release rates from drained soils, the workgroup should use sufficient sample numbers to address the well-known large heterogeneity found in soils. Experimental studies should be done at as large a spatial scale as possible to avoid artifacts caused by trying to extrapolate results from small sample sizes and small containers to the ambient environment. Finally, the District should consider the worth of collecting such data before there is better information on the areal extent of wetland soils that would be substantially dried due to water withdrawal. PLANKTON, NUTRIENTS, AND TMDLs The plankton and TMDLs workgroup was tasked with identifying and quantifying possible environmental impacts of water withdrawals on plankton communities and existing TMDLs in the lower and middle St. Johns River. In the Phase I report (SJRWMD, 2008), the group attempted to address four principal questions related to plankton and TMDLs using existing data: 1. What are the potential impacts and which impacts are likely significant? 2. What mechanisms or empirical relationships connect direct hydrological consequences of

OCR for page 43
48 Review of the St. Johns River Water Supply Impact Study: Report 1 water withdrawal to direct or indirect environmental impacts? 3. What environmental and hydrologic criteria are appropriate to measure impacts? 4. What environmental and hydrological boundaries define significant adverse impacts? In Phase II, the workgroup will complete the evaluation of potential effects on plankton and related conditions in the river and evaluate specific water withdrawal scenarios. According to SJRWMD (2009c), three main tasks will be to: (1) characterize relationships between hydrologic variables and direct and indirect impacts on plankton, (2) establish environmental boundaries for levels of adverse effects, and (3) identify hydrologic regimes compatible with environmental boundaries. The primary objective is to determine acceptable limits of change (environmental boundaries) in plankton and nutrient-related conditions that are affected by water withdrawals. Assessment of Phase I The strategy for addressing the potential impacts to plankton and TMDLs from reduced discharge is logical, sound, and clearly presented, and Table 4 in SJRWMD (2008) provides a comprehensive overview of the progress made in Phase I of the WSIS. Overall, the workgroup did a commendable job summarizing and interpreting archival data from a variety of studies conducted in the middle and lower reaches of the St. Johns River over the past 25 years. However, during Phase I the District was unable to satisfactorily answer the four main questions listed above because several critical issues were not considered. The Phase II studies need to be designed to address the following critical issues. First, additional water withdrawals may increase the likelihood, duration, and areal extent of water column density stratification and induce bottom water hypoxia in the lower St. Johns River under low flow conditions, which was not discussed in the Phase I report. Such conditions, even if they occur for only short periods of time, could have major effects on water quality in the river, in addition to affecting benthic organisms, nekton, and nutrient cycling processes. Second, the Phase I report does not adequately address the type or frequency of additional water quality and biological monitoring data needed to adequately assess the impacts of water withdrawals on TMDLs and plankton. This suggests the need for a comprehensive monitoring strategy and protocol to assess the potential impacts of water withdrawal to support the Phase II efforts. Such a strategy would include a detailed list of critical variables to be measured, as well as the frequency and locations of measurements. According to SJRWMD (2009c), Phase II plans neglect large segments of the river that could be affected by water withdrawals by including monitoring only in Lake George. More comprehensive monitoring during Phase II would include key freshwater and estuarine regions and encompass benthic meroplankton (mentioned later in conjunction with the benthos workgroup as an important group of taxa to investigate). Third, based on presentations by District staff at the January 2009 meeting and on various statements in the Phase I report, it appears that the District has assumed that phosphorus is the limiting nutrient for algal growth in the freshwater portions of the St. Johns River. That assumption may not be valid. The occurrence of substantial blooms of nitrogen-fixing cyanobacteria in Lake George, for example, is strong evidence that nitrogen limitation occurs at least at some times of year and in some locations within the river.

OCR for page 43
Ecological Assessments 49 Recommendations for Phase II One important issue of concern that cuts across the nutrient and plankton subject areas is whether additional water withdrawals would increase the likelihood of water column density stratification and bottom water hypoxia in the lower St. Johns River under low flow conditions. Analysis of this issue in Phase II will require close collaboration with, and careful consideration of the findings from, several other workgroups including the hydrodynamic and hydrologic and biogeochemistry workgroups. Because of the short timeframe of the Phase I study and the fact that the groups were working in parallel to produce their Phase I documents, integration of findings across the workgroups may have been fragmentary and incomplete. Strong interactions between the workgroups must be established in Phase II so that the findings related to the effects of water withdrawals on the physical and chemical driving forces can be thoroughly integrated into the analysis of plankton dynamics. As discussed in greater detail below, more attention should be given to the role of CDOM, heterotrophic bacterioplankton, and microbial loop processes in the analysis of plankton dynamics in the Phase II study. The effects of high concentrations of CDOM on phytoplankton ecophysiology were not explored during Phase I. CDOM alters the light environment and reduces the amount of useful light available for photosynthesis. Although Secchi depth was included in the Phase I analysis, it is at best a crude indicator of light attenuation (Kirk, 1994). Under conditions of increased water withdrawals, CDOM concentrations could increase if export of CDOM from riparian wetlands remains constant (or possibly increases) and river flow is decreased. This could lower light levels, possibly limiting phytoplankton and benthic macrophyte production. For low-light acclimated phototrophs, small alterations in ambient irradiance can result in exponential changes in rates of photosynthesis. The combination of change in both light quality and quantity due to changes in CDOM concentrations may have a variety of effects on plankton and associated conditions in the river and its lakes. The potential effects of water withdrawal on heterotrophic bacterioplankton and microbial loop processes also were not addressed in the Phase I report. Given the high CDOM and nutrient concentrations in this system, bacterioplankton likely play an important role in oxygen dynamics, water quality, and biogeochemical cycling in the river (Tranvik, 1990; Schultz, 2000; Joint et al., 2002; Alonso-Saez et al., 2008). Although data on this community in the St. Johns River may not be available, the potential implications of water withdrawal should be explored using data from similar river systems. Nutrients Historical data for this system clearly demonstrate that the ratios and concentrations of nitrogen and phosphorus are important determinants of phytoplankton biomass and community composition. System biological responses to water withdrawal will depend directly on the concentrations and ratios of these primary limiting nutrients. In Phase II, there needs to be an explicit differentiation between discharge, nutrient concentration, and nutrient loading. For example, as flow decreases, nutrient concentrations should increase due to lower dilution rates. Will nutrient loadings remain constant while nutrient concentrations increase? As mentioned in the section on biogeochemistry, this information needs to be obtained and analyzed at a sufficiently fine temporal scale to be relevant in assessing impacts on algal blooms, which are

OCR for page 43
50 Review of the St. Johns River Water Supply Impact Study: Report 1 controlled by nitrogen and phosphorus loading at certain times such as low-flow conditions during warm weather periods. Thus, the Phase II efforts should emphasize the importance of both nitrogen and phosphorus and the effects of water withdrawals in the context of a dual nutrient management strategy (Paerl, 2009). The Phase II work plans do include both nutrients, but not a dual management strategy. Existing data suggest that phytoplankton biomass and community composition in Lake George are regulated by the relative concentrations of nitrogen and phosphorus. Plots of nitrogen:phosphorus ratios as a function of potential controlling variables such as discharge rates, residence time, and water age could be useful for assessing potential impacts, especially as related to nitrogen-fixing cyanobacterial blooms. Regressions of chlorophyll levels versus nutrient ratios may provide useful empirical functions for assessing responses to changes in loading and/or concentrations. However, in many cases significant correlations between nutrient concentrations and phytoplankton biomass can be difficult to obtain because the ambient dissolved nutrient concentrations in the water column reflect only the residual nutrients—the nutrients not already assimilated by phytoplankton. In this sense, inorganic nutrient concentrations indicate what is available for future growth and may not be related to phytoplankton standing stock at the time of measurement. For this reason, nutrient loading estimates, in addition to measurements of concentrations, are essential for developing realistic budgets and predictions of system responses. Water Age The “water age” approach described in the Phase I report may be useful in dealing with nutrient-biomass problems if it can include quantitative characterization of phytoplankton growth responses to ambient nutrient concentrations as the water becomes older. Water age, however, needs to be defined more clearly in the Phase II effort, and a clearer explanation should be provided of how it helps to address the project objectives. Can confidence intervals be calculated for water age to provide some measure of the variability in this parameter? Will water age be location-specific? A graph of water age versus discharge for the main reaches of the river may provide insights into how plankton will respond to water withdrawal scenarios and reduced river discharge. Proposed Phase II Studies A clear description was not provided in the Phase II work plan concerning how two proposed activities are related to the overall project goals. First, the study of the effects of DOC on the inhibition of microbial activities (Appendix C) seems experimental and peripheral to assessing the potential impacts of water withdrawal. Changes in the loadings and composition of DOC should be determined prior to experimental assessments of its potential effects on the activity of microbial communities, in order to provide guidance and a clear justification for the proposed studies. The dinoflagellate toxin studies (both the cyst and animal surveys) are exploratory and do not offer a mechanistic (either theoretical or empirical) justification that can be linked to water withdrawal scenarios. Second, mesocosm studies to assess the effects of cyanobacteria on zooplankton grazing are also proposed in Phase II. These studies could provide

OCR for page 43
Ecological Assessments 51 useful insights into the effects of cyanobacterial blooms on trophodynamics in affected waters. However, the mesocosm seem too small and the incubation period too long. Artifacts increase with the duration of such experiments because of algal growth on the walls of the tank, nutrient depletion, mixing conditions, and other factors (Carpenter, 1996; Chen et al., 2000; Porter et al., 2004). Scaling up results from the mesocosms to the river system will be complicated by these potential artifacts. All of these factors must be carefully considered before proceeding with mesocosm studies. The workgroup is encouraged to continue the development and validation of three- dimensional water quality simulation models (e.g., CE-QUAL-ICM) for the major reaches and lakes of the St. Johns River. These models will be invaluable for scenario simulations. Summary The key issues for the plankton, nutrients, and TMDLs workgroup to consider in Phase II involve (1) determination of nutrient and CDOM loading estimates for segments of the middle and lower St. Johns River under the various water withdrawal scenarios, (2) tighter integration with the hydrodynamics and biogeochemistry groups, (3) consideration of potential impacts on bacterioplankton, and (4) implementation of dynamic simulation modeling (CE-QUAL-ICM). Furthermore, the proposed mesocosm experiments could provide useful results, but potential problems with artifacts and scaling need to be considered. BENTHOS Benthic macroinvertebrates live in or on the bottom substrate of aquatic environments and have been used for decades to understand the ecological dynamics of streams, wetlands, and coastal marine environments as well as to study the effects of human activity on aquatic ecosystems. Knowledge of these facts led the District to define a benthic workgroup as a crucial program within the WSIS. Both freshwater and brackish waters were to be included in their analyses. The first phase of the effort emphasized three major activities: 1. Explore the state of knowledge on the ecology of benthic environments, including both the organisms themselves and aspects of physical environments upon which these species are dependent. Contracts with several scientific leaders in the field facilitated the efforts of this workgroup. 2. Develop a logical approach to the study of benthic macroinvertebrates as a way to understand the biological consequences of changes in flow or related factors caused by water withdrawal, and 3. Examine available data for insight about the effects of hydrologic alterations from water withdrawals on benthic macroinvertebrates. The Phase I report reviewed past work in the watershed, began analysis of an existing data base, and briefly described several conceptual models to guide thinking about how to study and understand the effects of water withdrawal on the trophic organization of benthic invertebrates. The conceptual models also described how changes in water level might influence trophic levels above and below the dominantly plant-feeding benthic invertebrates. In addition to a focus on trophic organization, the Phase I report also targets a few invertebrate taxa (e.g., crayfish, apple

OCR for page 43
52 Review of the St. Johns River Water Supply Impact Study: Report 1 snail, blue crab, and penaeid shrimp) considered to be of special interest to the WSIS. Assessment of Phase I Benthic macroinvertebrates are used as indicators of habitat condition, including the effects of hydrologic alteration on habitat, and more generally to assess the biological integrity of aquatic ecosystems. Biological integrity, an explicitly framed objective of water resource management under the Clean Water Act (Section 101(a)), refers to the condition or character of living systems in the relative absence of modern human activity. The authors of the Phase I report noted that, at least in general terms, much is known about the effects of water withdrawal on freshwater macroinvertebrates in the St. Johns River. However, little detail was provided on the lessons of the studies cited in Appendix 1 of the Phase I report to the St. Johns River situation. For example, how are invertebrate assemblages changed (e.g., shifts in taxa richness, composition, density, and trophic organization) as a result of water withdrawal? What papers document and demonstrate those or other patterns? Are the situations (benthic macroinvertebrate species composition; waterbody type and size; environment type, biogeographic context) for those papers similar to those in the St. John River? Without these details, it is not possible for the committee to evaluate the relevance of the assertion that much is known about the effects of water withdrawal on freshwater macroinvertebrates to the situation in the St. Johns River. Those kinds of insights and interpretation need to be clear in the body of the report, rather than buried in an extensive, unsynthesized appendix tabulation. One subject not touched in the Phase I report is the effect of water withdrawal on macroinvertebrates with meroplanktonic larval stages, a group of taxa with high relevance in a comprehensive ecological assessment. The potential impacts of water withdrawals for these early life history stages were not addressed adequately in the Phase I report, a subject that might be addressed jointly with the plankton workgroup during Phase II. Questions that could be address include: what proportion of the species in the St. Johns system have meroplanktonic larvae in freshwater or estuarine regions? How much emphasis needs to be placed on study of meroplanktonic larvae in those two environment types? An important insight acknowledged in the Phase I report is the need to explore existing data. One such data set was explored but the results were not very robust and, in the committee’s view, cannot be generalized without analysis of more detailed data. The need to collect more data is recognized in the Phase I report and a first level plan is fleshed out in the Phase II work plan. A sampling design for the upper reaches of the river is described briefly on the Phase II work plan (SJRWMD, 2009c, pages 54-55) as a first effort to move forward in this important area. In contrast to freshwater areas, the Phase II work plan concludes that no new data will be collected in the estuarine segments of the watershed. Instead, data from existing studies will be further analyzed in 2009/2010 with the goal of evaluating the effect of salinity and other water quality variables on invertebrate community structure. Without more information on the context of those historical data collections (e.g., timing, spatial distribution, duration, kinds of data, data collection protocols), it is impossible to judge the merit of that decision. Rather, it would be better to defer any decision about additional data collection from the estuary until a careful evaluation of the utility of existing data is completed. Other factors—such as new dredging by

OCR for page 43
Ecological Assessments 53 the port authority or the Navy—that might affect salinity, sedimentation, or other conditions in the lower river may also require reevaluation of estuarine monitoring decisions. Recommendations for Phase II Several of the WSIS workgroups described in general terms the need to identify indicators that can provide reliable and easily interpreted signals about the condition of sites influenced by water withdrawals. Often, those descriptions focused on individual species as indicators. The selection of species for special study seems logical. A second approach for monitoring benthic invertebrates is the study of trophic groups. This promising approach is warranted, and the outside experts selected by the district are the leaders in this field. The committee strongly supports the development of indicators beyond individual species in the WSIS, and commends the benthos team’s commitment to use of more integrative measures of biological condition to judge the effects of water withdrawals. In addition to the study of functional or trophic groups, the Phase II work plan suggests the exploration of a broader array of measures to characterize biological condition (SJRWMD, 2009c, page 58). Examples of such measures include taxa richness or relative abundance of tolerant and intolerant taxa, taxa richness of selected taxonomic groups such as mayflies or caddisflies, taxa richness of ecological groups such as clingers, and dominance by a few taxa. Although the committee endorses a Phase II effort to use such broad measures, it notes that sufficient detail was not provided on how relevant measures will be defined and used. It is not clear how this activity will be guided by the large body of work in this area developed over the past three decades (see Table 4-1 for selected examples relevant to this project from the literature). There is precedent for the use of multiple biological metrics as indicators to understand the impact of water withdrawals. For example, a recent study (Freeman and Marcinek, 2006) evaluated fishes in 28 streams used for municipal water supply in the Piedmont region of Georgia. Increasing the withdrawal rate increased the odds that a site’s Index of Biotic Integrity score would fall below a regulatory threshold indicating biological impairment. Estimates of reservoir and withdrawal effects on stream biota revealed by such broad measures can be used in predictive landscape models to support adaptive water supply planning intended to meet societal needs while conserving biological resources. Summary The committee supports the District’s commitment to study of the effects of water withdrawal on benthic invertebrates. Both the Phase I report and the Phase II work plan demonstrate an evolving approach to the study of the effects of water withdrawal on invertebrates. The literature and approaches used by the workgroup should be extended from selected individual species to integrative views of invertebrate assemblages. To date, the available data are insufficient to precisely define what the effects of water withdrawals on the benthos of the St. Johns River will be. However, the District is commended for sketching the rudiments of a sampling and data analysis program as part of the Phase II work

OCR for page 43
Ecological Assessments 61 variability in the future, when the effects of water withdrawals may be coupled with additional river dredging by the port authority and the Navy to allow ship access further upstream. Managing for any range of variability is not the same as managing for the range within which these species have evolved. If the range of variability changes, it may influence future year-class strength of some species and assemblage structure in these same geographic areas along the transition zones from saltwater to freshwater. Shifting the saltwater–freshwater transition zone to different areas may also influence normal migratory routes and timing for some taxa (fish and other nekton) with negative effects on survival during various life-cycle stages. Finally, the fish data analyses need to be integrated with other components of the regional biota (e.g., benthos, decapods, SAV)—factors that also may be influenced by salinity shifts. The cumulative interaction of the assemblage will be critical to understanding fish responses to salinity (Peterson, 2003) and thus to water withdrawals. Indeed, a program could be developed that integrates the FFWCC fish and decapod data and the benthic macroinvertebrate and aquatic vegetation components using an approach similar to that of Peebles and colleagues from the University of South Florida. Peebles et al. (2007) developed an approach for evaluating the effects of water withdrawals on the egg, larvae, juvenile, and adult stages of the bay anchovy, Anchoa mitchilli, in a number of Florida estuaries. This approach could be modified to incorporate taxa (beyond fish), using data from the WSIS and FFWCC projects. Such an approach would allow a more realistic examination of how much the nursery areas of the Lower St. Johns River might expand, degrade, or be lost due to saltwater intrusion (e.g., Kimmerer et al. 2009). Water Level Changes in Transition Zones and Floodplains A major concern with the Phase I report is the lack of consideration of water level decreases (projected to be as much as 4 cm in the upper St. Johns River basin) and the influence that those water level changes will have on fish population dynamics and distribution. Many important factors for fish survival from spawning habitat to foraging opportunities and the availability of refuges may be influenced by water level declines in complicated ways. Indeed, much is known about shallow vegetated areas in lake and stream systems that are important fish habitats and that drive life history parameters such as reproduction, growth, foraging, and distribution of fish biomass by habitat type (Keast et al., 1978; Keast and Eadie, 1984; Keast, 1985). Many but not all of these life history parameters are strongly linked to emergent, floating, and submersed aquatic vegetation. Centrarchid fishes (sunfishes and black bass) provide excellent opportunities to understand the dynamics of these relationships and dependencies. This diverse family in the St. Johns River system includes many valuable sport fish throughout their range. Aquatic vegetation mediates numerous ecological processes in aquatic habitats, specifically reproduction, foraging and predator–prey interactions (Theel et al., 2007; Kovelenko et al., 2009). Aquatic vegetation could be negatively affected by water level reductions in the shallow and relatively flat floodplains of the middle and upper St. Johns River. These studies also noted that many of the impacts to trophic pathways of sunfishes, beyond direct changes in interstitial space in aquatic vegetation, are indirect due to changes in macroinvertebrate assemblages and substrate heterogeneity (Kovelenko et al., 2009).

OCR for page 43
62 Review of the St. Johns River Water Supply Impact Study: Report 1 Given the spatial variation in geomorphology and microtopography of the floodplains along the St. Johns River, important habitat characteristics are likely to be modified by a 4-cm water level drop spread across a large area. Therefore, the District should carefully study potential impacts to all fishes in the middle and upper St. Johns River, but particularly those that require shallow areas for spawning and foraging, including but not limited to the centrarchids (Breder and Rosen, 1966). The District should also develop a comprehensive study of spawning habitat in the upper St. Johns River relative to the projected 4-cm water level drop and use GIS to evaluate how much area will actually be lost in terms of drying out or converting to upland habitat. In the Phase II work plan the District has made great progress explaining its approach to examining how water level changes might impact floodplain habitat for fishes. It will develop a detailed literature review on fish use of floodplains in Florida and then couple those literature patterns with selected minimum flow and level (MFL) profiles in the St. Johns River for which it has calculated the loss of floodplain habitat based on predictions of water level change. The District suggests in its Phase II work plan that it will conduct these habitat availability studies in areas being considered for consumptive use withdrawals during months when reproduction of a number of fish species is documented to occur (although details of this “documentation” are not given). This is reasonable, but the committee questions how the District will go from change in available floodplain area for a given water level change to the fish production metrics that are vital to sustainability. Some fish species use floodplains to actually spawn, while the adults of other species use it to forage prior to spawning—a very big difference that may not be captured by the proposed analysis. Impacts to Anadromous fish In the Phase II work plan, the District added an ongoing cooperative (University of Florida, FFWCC, and the District) component to their collection activities addressing tracking juvenile and adult herring, as the populations of three herring species are depressed. The goal of this activity is to determine migratory and spawning habitat of herrings, and then to compare the needed habitat characteristics with what is actually available in the St. Johns River before and after potential water withdrawals. This approach appears reasonable, but few details are provided to allow for a critical evaluation; if possible, it would be appropriate to link these data with the larval herring data sets for a more complete picture. Other Issues In the Phase I report, the District categorized fishes as commercial, recreational, or forage fishes. Although some fish species are commercial or recreational (or both), this somewhat arbitrary classification narrowly views the remaining species in the St. Johns River as important only as food for the commercial and recreational species. The District should take a broader view in Phase II. Dozens of other fish species are permanent residents or seasonal visitors to the St. Johns River. An understanding of their ecologies and how their distributions and abundances will be affected by proposed water withdrawals is central to the evaluation of natural resource effects that comprises the bulk of the WSIS. For example, depending on exactly where the water

OCR for page 43
Ecological Assessments 63 intake structures will be located and how many structures will be built, the narrowly framed forage category may be the most affected group. A number of other fish species listed in Appendix A of the Phase I Report, such as two sturgeon species (shortnose and Atlantic), rare and habitat-specific sunfishes (3 Enneacanthus spp.), rare pygmy sunfish (Everglades and Okefenokee), and the River goby (Awaous banana) could be sentinels to any withdrawal project as they may be some of the first species to be affected. In the Phase I report, the District developed a graded impact score approach to judging the “likelihood of impact” (e.g., low, moderate, and high) from water withdrawal. The report, however, does not provide any information about the origin and foundation of those scores or the nature of the background criteria used to judge these impacts, other than to say that the “likelihood of impact” is based upon the District’s analysis of the risk compared to the preliminary hydrologic modeling results. Are these scores based on some quantitative or semi- quantitative analysis or are they a reflection of expert opinion? A description of the process underlying these “likelihood of impact” scores is needed to understand the predicted outcomes. Of the 15 potential impacts related to all categories outlined in Phase I (freshwater influences, water quality, river-floodplain interactions, and estuarine influences), only one “likelihood of impact” is classified as moderate-to-high (current reversals) and one as moderate (effects on winter spawning migrations of the American shad); the projected likelihood of the remaining factors are low-moderate or low. Can these scores be justified given the lack of solid data (including spatial and temporal patterns) on fishes of the St. Johns River system? For example, a considerable number of references are based on studies of clupeids in reservoirs and lakes, for which the characteristics may vastly differ from those of riverine fish. Of the 22 species listed in Table 1 (pages 11-12 of the fish section in the Phase 1 report), the “likelihood of impact” for 13 of the species is categorized as “low,” “negligible” for four, “low-moderate” for one, and “moderate-to-high” for one. Overall, the Phase I report concludes that almost none of the identified fish will be significantly affected either directly via entrainment and impingement or indirectly through habitat loss or degradation and salinity intrusion. The committee is not convinced that the information available justifies this interpretation, and it looks forward to the more complete risk analysis that is slated to be part of Phase II. Summary A number of decisions need to be made by the District that are crucial to the development and clarity of a final fish program and study design. Based on the Phase I report, the District should evaluate the merit of at least three additional program activities before final sampling protocols are developed: (1) analyze the larval fish data collected to date to determine if the sampling protocols are adequate to address the intended goals, (2) integrate existing fish data sets with benthic macroinvertebrate and SAV data to address impacts to fish assemblages, and (3) comprehensively study the impact of a 4-cm water level drop to the spawning and feeding habitat in the middle and upper St. Johns River. In the Phase II work plan, the District did not address adequately the first item mentioned above, but it did enhance and clarify the second and third items. The second item will now be addressed, in terms of fishes and selected decapods, by the FFWC FIM program contract now in place. It should be kept in mind that while the FIM program is a great enhancement to the

OCR for page 43
64 Review of the St. Johns River Water Supply Impact Study: Report 1 overall WSIS, it does not collect samples in small creeks in the lower St. Johns River and thus will miss important nursery areas. In terms of the third item, the District needs to elaborate on how the loss of available floodplain area will translate into loss of fish production, which is vital to a more complete understanding of the effects of water level changes. WETLANDS AND WETLAND-DEPENDENT SPECIES Changes in hydrology can alter the structure and function of wetlands. Therefore, the wetlands workgroup is examining the potential impacts of the proposed surface water withdrawals to wetland vegetation and a few species of wetland-dependent fauna. The workgroup also plans to address indirect effects resulting from changes in water quality following withdrawal, including the value of wetlands as habitat for wetland-dependent species. The District predicts that impacts could range from changes in vegetation community type or structure (including species composition), altered productivity, and shifts in the position of boundaries between communities. Although hydrologic alterations are expected to impact a broad array of wetland functions, only the habitat function was addressed in this section. In general, the Phase I wetlands work is descriptive and conceptual in nature. It is planned that the conceptual model developed in Phase I will be developed into a fully operational GIS-based model in Phase II of the study. Assessment of Phase I Report This section includes an assessment of the Phase I wetlands work along with some related recommendations for improvements in the proposed Phase II. Issues discussed include the methods used to identify floodplain wetlands, the GIS-based determination of sensitive wetland areas, the use of wetland-sensitive fauna as indicators of hydrologic change, and maximizing the use of available data. Identification of Floodplain Wetlands The WSIS is limited to wetlands located on the floodplain of the St. Johns River (i.e., wetlands directly influenced by the river’s hydrology). A GIS analysis was conducted in Phase I to define the floodplain and identify wetlands that lie within it. With a few exceptions,1 the floodplain was delineated using the 5-foot contour line (the line representing points of equal elevation at 5-ft above sea level) on 7.5-minute USGS quadrangle maps, with the assumption that the 5-ft elevation captures much of the 50-year floodplain. The District considered this 1 At the river mouth, where the 5-ft contour line was not available, a digital elevation model was used to estimate the 5-ft contour, which was then checked for accuracy using aerial photography and wetland maps. At the southern end of the basin, first the 10-ft and then, at higher elevations, the 15-ft contours were used. Where these contours did not coincide with the upland edge of the floodplain, the report states that the floodplain boundaries were drawn in manually.

OCR for page 43
Ecological Assessments 65 elevation to overestimate the portion of the floodplain likely to be affected by the river, although no data were provided to support this conclusion. Given the current limited availability of GIS data for wetland mapping, the District had little choice but to use the relatively coarse topographic data. However, this approach lacks the resolution necessary to predict the ecological effects of hydrologic change. For instance, the floodplain, particularly in the lower segments, is relatively flat with microtopographical features that create and maintain diversity as hydrological conditions (e.g., water levels, frequency and duration of inundation) vary. Fine-scale elevation data in the form of a digital elevation model (DEM) are needed to produce accurate maps in flat, wetland-rich areas, particularly in efforts to characterize hydrology and map wetland diversity (Maxa and Bolstad, 2009). The District has recognized the need for a DEM in order to adequately evaluate wetland response to the proposed withdrawals, and it plans during Phase II to produce DEMs for the portions of the watershed where appropriate data are available. Recent advances in wetland mapping are available to update maps and improve their accuracy and are recommended for use in this study. One of the best options is the use of LIDAR (Light Detection and Ranging) imagery, which, if used, could greatly increase the sensitivity of models to assess the effects of hydrologic change. LIDAR provides elevation data with high resolution (15 cm to 1 m elevation) and accuracy and can improve efforts to map vegetation structure (Hyde et al., 2006). This also will help characterize changes due to altered hydroperiod, in which small changes can lead to large alterations in wetland function and extent. Thus, the use of LIDAR imagery would allow the systematic monitoring and assessment of hydrologic changes (Lang and McCarty, 2008). According to the Phase II work plans, the District is planning to test the use of LIDAR imagery using data obtained from several counties (see Figure 4-1). Much of the lower and middle St. Johns River floodplain either has LIDAR data available or it is pending. This should provide insight into the usefulness of LIDAR in the District’s GIS analysis and DEM development. The District is encouraged to acquire LIDAR imagery for the entire study area as soon as possible, understanding the constraints imposed by limited resources. Use of previously collected data on plant communities and elevation available from transects sampled as part of the MFL study (as described in Neubauer et al., 2008)—but not included in the Phase I report—also may provide ground truth information for the existing wetland maps and a means to determine their degree of accuracy. Where transect data are available, they can help characterize the extent of hydrologic connectedness (river to floodplain) and identify the locations where impacts from withdrawals are expected to occur. These data can serve as a baseline against which future conditions are measured. The Phase II work plan indicates that the transect data will be used to accomplish this through appropriate data analysis. Identifying Sensitive Wetland Areas To cope with the distribution of wetland types found in the watershed, the wetlands workgroup divided the river into nine segments (see Figure 3-3) deemed relatively homogeneous in terms of soils, vegetation, hydrology, water quality, and fauna. For each segment, information was provided on channel length, hydrologic features, salinity, predominant wetland plant species, shoreline ratio, soils, wetland-dependent fauna, and the relative likelihood of impacts

OCR for page 43
66 Review of the St. Johns River Water Supply Impact Study: Report 1 FIGURE 4-1: St. Johns River Basin LIDAR Data (June 2009). SOURCE: SJRWMD. from water withdrawals to all wetlands in the segment. This approach is analogous to the River Continuum Concept used to model the relatively predictable changes assumed to occur in river ecosystems from source to mouth (Vannote et al., 1980), and it is an excellent start to address the expected physical and biological diversity that exists within the basin. However, this conceptual, qualitative approach made no effort to discriminate the varying response of different wetland types within the river segments. Wetlands vary widely in terms of hydrology, landscape position, and the functions they provide. Adopting a classification scheme, such as the hydrogeomorphic (HGM) approach (Brinson 1993), which the District now plans to

OCR for page 43
Ecological Assessments 67 employ as part of the Phase II work, has been shown effective in stratifying the diversity in species composition and ecosystem function. In conjunction with a more detailed and accurate DEM, this might help the District better cope with the landscape-scale diversity of wetland types and their hydrologic drivers, and it could replace the notion of the “landscape hydrological milieu” used in Phase I. The HGM approach is well established in wetlands research and monitoring programs and has been shown to be effective in reducing wetland variability in order to evaluate the relative response of wetland classes to hydrologic change (Rheinhardt et al., 1997; Whigham, 1999; Wardrop et al., 2007). The District used GIS data layers along with best professional judgment to evaluate potential wetland response to hydrologic alterations and to categorize the nine river segments according to the “likelihood of hydrologic impacts” (i.e., changing stage, salinity). The “likelihood of impacts” was classified qualitatively (i.e., high, moderate, low, none) as follows: • Weights were assigned to the various GIS data layers (e.g., wetlands, soils, topography, hydrology) to indicate their relative importance to wetland response. • A stressor layer was created to represent water level lowering or other environmental changes, including changes in salinity. • Where sensitive areas overlapped with areas impacted by stressors, some likelihood of effects was assigned. The committee is concerned that the criteria used to determine the likelihood of impacts are not provided in the report. Specifically, the weights assigned to the GIS data layers were not presented in the Phase I report, nor was there an explanation of how the weights were developed. No information was presented about how the stressor layer was created, nor was a comprehensive list provided of the stressors considered. The committee appreciates the difficulty of this type of modeling effort, but without a description of the weights assigned to the different types of data, details on the stressors included in the GIS database, and the criteria used to estimate the likelihood of impacts, it is impossible to judge the results. Results of the Phase I analysis are summarized in Box 4-1. This approach was developed and used as a coarse screen to identify river segments that warrant the most attention, which is logical. In large part it is meant to identify wetlands on highly permeable, highly organic soils that are hydrologically linked to the river and dominated by vegetation that is likely to be sensitive to drawdown. However, it is difficult to assess the reliability of this analysis without a detailed description of the rationale, data, and procedures. Note that many of the conclusions listed in Box 4-1 go beyond impacts to wetlands, but rather pertain to the other workgroups. Wetland-Dependent Fauna as Indicators The wetlands workgroup presented a literature review of four bird species (wading birds and raptors) known for their sensitivity to hydrologic change that may be useful as indicators of wetland impacts: the wood stork (Mycteria americana), white ibis (Eudocimus albus), limpkin (Aramus guarauna), and snail kite (Rostrhamus sociabilis). The wood stork and the snail kite are federally endangered species, and their water-level requirements for successful foraging and nesting are well known. The white ibis and limpkin are listed as Florida species of special concern. The report lists specific hydrologic ranges required for foraging and nesting of these

OCR for page 43
68 Review of the St. Johns River Water Supply Impact Study: Report 1 Box 4-1 Preliminary Assessment of Impacts along Nine Segments of the St. Johns River The Phase I report includes a broad assessment of potential impacts in nine river-wetland segments (see Figure 3-3): 1) Mayport to Fuller Warren Bridge (River Kilometer [RK] 0 – 39.6): this reach is dominated by high- energy tidal exchange. It is unlikely that there will be impacts due to water withdrawals. 2) Fuller Warren Bridge to Flemming Island (RK 39.6 - 65): This area could suffer moderately from salinity, although stage effects are likely to be low. Altered salinity may impact SAV in this area. 3) Flemming Island to Little Lake George (RK 65 - 163.1): Few impacts due to water level changes are anticipated, but there may be impacts due to altered salinity. The likelihood of effects due to altered salinity is judged to be moderate. 4) Little Lake George to Astor (RK 163.1 - 204.3): This area is the focus of the plankton workgroup because of the increased probability of algal blooms. The District reports that wetland effects are likely to be minimal, in part because the area already has high salt inputs from springs. 5) Astor to Wekiva River (RK 204.3 - 253.7): Wetlands in this area are dominated by hardwood swamps. This is the first segment above tidal influence, so it is predicted that salinity will not be affected (unless this reach becomes tidally influenced). Stage changes with water withdrawal could be an issue. This segment concerns the biogeochemistry group because of possible changes in the duration of inundation for wetland soils. Small changes in stage may lead to the dewatering of large floodplain areas, and the likelihood of effects due to altered stage is judged to be moderate. 6) Wekiva River to St. Road 46 (RK253.7 - 310): This segment is part of the middle river and includes Lakes Monroe, Jessup and Harney. Stage and salinity impacts are likely to be low to low-moderate. The area is important for shad spawning and any potential for impacts to fish communities are of concern. 7) St Road 46 to St Road 520 (RK 310 – 378): Segment 7 is predominantly wet prairie and includes Puzzle Lake. There are no tidal effects in this segment, so the potential for stage effects are relatively high (the likelihood of effects due to altered stage is judged to be moderate-high). Some plant communities here may become stressed. 8) St Road 520 to Three Forks Marsh (RK 378 - 442.8): This area is dominated by shallow marsh and includes Lake Poinsett. This is considered the uppermost region that could be affected by withdrawals; the likelihood of effects due to altered stage is judged to be moderate-high. 9) Three Forks Marsh to Blue Cypress Lake (RK 442.8+): This segment will not be subject to water withdrawals. four species, but it does not address the details of how these species might be used as indicators (i.e., what aspects of the populations will be measured, how will data be collected, what metrics might be developed?), or whether a field-based monitoring program is even planned. In addition, links between these populations and their predicted response to hydrologic change are not clearly presented. For example, a recent study by Darby et al. (2008) concluded that Florida apple snail (Pomacea paludosa) survival rates can be quite high in drought conditions, contrary to earlier reports of their low tolerance to drawdown. This calls to question the assumption that the snail kite can serve as an indicator of lowered water levels. Another concern is that many wetlands in the floodplain are not used as habitat by these avian species; therefore, these species may not provide much relevant information on wetland impacts. This is compounded if the bird populations occur in low densities, making it difficult to detect patterns in their population numbers or habitat use at the local scale. The District reports high faunal diversity within the floodplain, including many species of fish, amphibians, reptiles, birds, mammals, and invertebrates. Many of these taxonomic groups are valuable indicators of hydrologic impacts to wetlands in monitoring programs (Micacchion,

OCR for page 43
Ecological Assessments 69 2002) and might be considered as indicators in Phase II. For instance, reptiles and amphibians are valuable as monitoring targets because many exploit transition areas (e.g., wetland–upland) where the effects of dewatering may be seen. Amphibians are closely tied to wetland habitat. They depend on wetland pools for recruitment and serve as an important trophic link, preying upon insects and acting as a food source for predacious insects, larvae, fish, birds, and snakes. Amphibian-based indicators are effective assessment tools because of their sensitivity to even minor anthropogenic disturbance (Welsh and Ollivier, 1998; Adamus et al., 2001; Sparling et al., 2001). Depending on its future use of indicators, the range of taxonomic groups used by the District in monitoring wetland impacts could be broadened. Ideas can be drawn from the considerable literature on the use of both plant and animal communities to assess wetland ecological condition (Sparling et al., 2001, Fennessy et al., 2002, Lopez and Fennessy, 2002). The district proposed investigating additional species in the Phase II work plans, but no details or methods were provided on how the different assemblages (e.g., reptiles, amphibians, etc.) might be used to indicate impacts. Rather than spending time on a literature review as proposed in Phase II, the District might consider field trials of existing biological indexes to determine their usefulness (for example, see Sparling et al., 2001). Many such indexes use multispecies assemblages as sources of interpretable signal that may be more robust than focusing on rare or other special interest species. Data Availability and Analysis Beyond the GIS data layers, very few data were presented in the wetland section of the Phase I report. As a result, it is difficult to understand how many of the conclusions of the wetlands workgroup were reached because of the many unstated assumptions embedded in the report. The report lacks details, partly because the Phase I hydrological modeling necessary for this analysis was still underway when the wetlands report was written. There are plans to address this in the Phase II study, although once again, few details are provided in the Phase II work plans on the methods that the district will use to accomplish this. One concern is that the District appears to lack data against which future conditions can be compared. A strategic, field-based monitoring program to track key elements of the system that are expected to respond to hydrologic alterations could address this issue. Aside from the biological measures mentioned above, several chemical-physical measures could provide useful information. For example, to address concerns about the oxidation of organic soils due to dewatering, one approach is to monitor wetland subsidence or accretion using sediment elevation tables (Day et al., 1999, Cahoon et al., 2002). These measurements are both easy to make and, once equipment is installed, are not expensive. An integrative method such as sediment elevation tables may provide information that is better able to reveal the impacts of hydrologic change than, for example, the soil analysis planned for Phase II. This proposed analysis involves the collection of soil samples at 150 locations throughout the basin to characterize the dominant soils in each river segment, particularly the organic soils vulnerable to drawdown. The approach is to perform a series of field and lab analyses on each sample to determine physical and chemical characteristics such as pH, conductivity, organic carbon and nutrient content, and concentrations of several metals. While these data may provide interesting insight into soil chemistry and will augment the work

OCR for page 43
70 Review of the St. Johns River Water Supply Impact Study: Report 1 of the biogeochemistry workgroup, the committee has questions about (1) how these data will be incorporated into the GIS model, and (2) in conjunction with the biogeochemistry workgroup, how 150 soil cores, collected over the length of the St. Johns River, will be scaled up to provide information on the response of the system overall. Additional Recommendations for Phase II In addition to the recommendations provided above, several other issues need to be addressed in Phase II. One fundamental need is a complete methods section describing all assumptions, data collection techniques, analytical approaches, and metadata for the GIS layers. Hydrologic modeling will be completed in Phase II, and the District plans to select a set of specific hydrological metrics believed to be most important for predicting vegetation community response. Indeed, the selection of these metrics should be one area where early integration of ideas and insights from multiple disciplines (in this case the hydrology and wetlands workgroups) is essential, as discussed in Chapter 2. Flooding depth, duration, and frequency, and water source, as well as water chemistry, are some key variables to consider. Additional analyses are needed to test the assumption that the effects of drawdown in the river propagate only a limited distance upslope in the wetlands and to identify how far upslope effects are felt (Doherty et al., 2000). This has important implications for predicting impacts to wetland functions such as providing spawning and nursery grounds for fish, amphibians and other species, and the maintaining of organic soils. It also can affect survival of vegetation in largely terrestrial areas near the edge of the floodplain. The committee is concerned that the landward edge of the floodplain may experience more dewatering than a 4-cm drop in river water level implies. How a drop in water level at Lake Jessup translates into the dewatering of adjacent floodplain and areas upstream should be addressed during Phase II. The proposal to collect 150 additional soil samples should also be clarified to reflect that the biogeochemistry workgroup will be using these same soil cores in their analysis. This also speaks to the general need for better integration between the workgroups. For example, in the biogeochemistry sections of the Phase I report, soils are referred to as “floodplain” soils while the wetlands workgroup refers to “wetland” soils. The lack of consistent terminology has ramifications for the integration of the study as a whole and provides evidence that the groups are not communicating as much as they should. The District has collected data on vegetation communities and elevation at a substantial number of floodplain transects throughout the basin. These data could be used as ground truth information to evaluate the GIS model output and verify the planned HGM classification of wetlands on the floodplain. Consideration could be also given to the use of these data as part of the Floristic Quality Assessment Index (FQAI), an index developed to assess the impact of human disturbance on vegetation communities. This index, akin to a plant-based Index of Biotic Integrity, assigns a repeatable and quantitative value to vegetation community composition and could be used to detect shifts in vegetation community composition that might occur with water withdrawals. Use of the index requires that the local flora be identified (including invasive species) and that the coefficients needed for each species be available; fortunately, these coefficients are available for Florida (Cohen et al., 2004). The FQAI has been shown to be an effective plant-based bioassessment tool in many regions (Lopez and Fennessy, 2002) including

OCR for page 43
Ecological Assessments 71 Florida (Cohen et al., 2004), and it could prove useful if the District intends to monitor wetlands to track wetland response to management activities. Summary In order to move beyond the descriptive Phase I work, the wetlands workgroup will need to enhance the methods used to identify and delineate wetlands that are influenced by the river’s hydrology. Fine-scale elevation data in the form of a LiDAR-based DEM are needed to produce accurate maps in flat, wetland-rich areas, particularly in efforts to characterize hydrology and map wetland diversity. The District plans during Phase II to produce DEMs for the portions of the watershed where data are available. It will also be important for the workgroup to better define the criteria used to determine the likelihood of impacts of water withdrawals to each wetland segment. Finally, the District should consider broadening the range of taxonomic groups used in monitoring wetland impacts. Amphibians, reptiles, invertebrates, and plants have proven to be valuable indicators of hydrologic impacts to wetlands in monitoring programs and might be considered as indicators in Phase II.